Recently we’ve discussed the flurry of attention being paid to resveratrol, a natural product derived from grape skins. Several current papers have demonstrated that the compound extends lifespan, increases exercise tolerance, and decreases some types of inflammation — at least, when administered in very high doses to rodents.

Most of the excitement has focused on resveratrol’s role as an activator of the sirtuins, proteins that have been implicated in lifespan regulation across evolution, in organisms ranging from yeast to worms to mammals. Perhaps as a consequence, one of the molecule’s other properties has gotten short shrift: thanks to the central olefin bond joining the two hydroxyphenyl rings, resveratrol is a potent antioxidant:


The ability to efficiently scavenge reactive oxygen species such as peroxide makes resveratrol protective against oxidative stress, specifically in the vascular endothelia. From Ungvari et al.:

Epidemiological studies suggest that Mediterranean diets rich in resveratrol are associated with reduced risk of coronary artery disease. However, the mechanisms by which resveratrol exerts its vasculoprotective effects are not completely understood. Because oxidative stress and endothelial cell injury play a critical role in vascular aging and atherogenesis, we evaluated whether resveratrol inhibits oxidative stress-induced endothelial apoptosis. We found that oxidized LDL (ox-LDL) and TNF{alpha} elicited significant increases in caspase 3/7 activity in cultured endothelial cells, which were prevented by resveratrol pre-treatment (10-6 to 10-4 mol/). The protective effect of resveratrol was partially reversed by glutathione peroxidase inhibitor mercaptosuccinate suggesting a role for antioxidant systems in the anti-apoptotic action of resveratrol. Indeed, resveratrol treatment protected cultured aortic segments and/or endothelial cells against increases in intracellular H2O2 levels and H2O2-mediated apoptotic cell death induced by oxidative stressors (exogenous H2O2, paraquat, ultraviolet light). Resveratrol treatment up-regulated the expression of glutathione peroxidase and catalase in cultured arteries, whereas it had no significant effect on expression of SOD isoforms. Resveratrol also effectively scavenged H2O2 in vitro. Thus, resveratrol seems to increase vascular oxidative stress resistance by scavenging H2O2 and preventing oxidative stress-induced endothelial cell death. We propose that the anti-oxidant and anti-apoptotic effects of resveratrol, together with its previously described anti-inflammatory actions, are responsible, at least in part, for its cardioprotective effects.

While the authors didn’t investigate the role of sirtuins in this protective effect, so it’s possible that activation of sirtuins result in upregulation of cellular antioxidant defenses — but that adds an extra step to the story, one that the in vitro radical scavenging experiment demonstrates isn’t essential to explain the phenomenon. (Likewise, the anti-apoptotic activity discussed early in the abstract could be the result of sirtuin activation, but that is a less parsimonious explanation than that oxidized proteins trigger caspase activity and resveratrol decreases protein oxidation, thereby down-modulating apoptosis).

Thus the cardioprotective effects of resveratrol (or, if you prefer, the wine in a Mediterranean diet) could be mediated not by the regulatory-biological output of a ligand-receptor interaction but rather by an unassisted chemical reaction between small molecules.

Why the fuss?

Because (as I’ve argued before with respect to the catechins in green tea) in order for scientists to make the best choices about how to spend our time, it’s important to know how a beneficial molecule is exerting its effects. If it’s the receptor-mediated outcome that is most desirable, then we should be working on better ligands: molecules that bind more tightly to the proteins that actually generate the relevant output. On the other hand, if the growing family of salubrious natural products turn out to be valuable primarily for their antioxidant activity, then we should be focusing on generating highly effecting antioxidant compounds that can target every tissue and subcellular structure in the body.

Of course, the answer could be “both,” in which case either path could bear fruit, but before a huge amount of effort is invested in drug development, it would be nice to know for sure.