Resveratrol History: Hellebore, Knotweed and the French Paradox

Resveratrol was first isolated from white hellebore root in 1939, according to a 2019 review. This page follows the record through knotweed, the 1992 French paradox paper, early yeast and mouse studies and human trials, and marks where the sources stop.

The history of resveratrol starts with a 1939 isolation from white hellebore root and runs through wine, mouse studies and trials that have not yet settled whether it helps people. This page follows that path through the sources we could verify in PubMed, and it marks where those sources stop. For the compound's reference profile, see the [resveratrol ingredient page](/ingredients/resveratrol). ## Two plants in the early record According to PubMed, a 2019 review by John Pezzuto reports that Takaoka first isolated resveratrol in 1939 from the root of white hellebore ([DOI](https://doi.org/10.4062/biomolther.2018.176)). Japanese knotweed (*Polygonum cuspidatum*), called Huzhang in Chinese, has long been recognized in traditional Chinese herbal medicine. A 2019 genome paper says its roots are the main medicinal tissue and links an expanded family of stilbene synthase genes to resveratrol biosynthesis ([DOI](https://doi.org/10.3389/fpls.2019.01274)). ## The French paradox and the wine question The 1992 *Lancet* paper by Renaud and de Lorgeril describes low coronary heart disease mortality in France despite high saturated fat intake, and it partly credits high wine consumption. Its abstract does not name resveratrol ([DOI](https://doi.org/10.1016/0140-6736%2892%2991277-f)). A 2019 review says interest in resveratrol largely stemmed from its presence in grapes and wine ([DOI](https://doi.org/10.4062/biomolther.2018.176)). A 1997 *Science* paper reported cancer chemopreventive activity for the compound, and the same review says sales of grape products containing it rose after that report ([DOI](https://doi.org/10.1126/science.275.5297.218)). We did not find a PubMed source that ties the French paradox headline itself to resveratrol. ## What the early yeast and mouse studies tested A 2003 *Nature* paper reported that, in yeast, resveratrol stimulated Sir2 and extended lifespan by 70%. The paper described the compound as a polyphenol found in red wine ([DOI](https://doi.org/10.1038/nature01960)). A 2006 *Nature* paper by Baur and colleagues gave it to middle-aged mice on a high-calorie diet and reported significantly better survival ([DOI](https://doi.org/10.1038/nature05354)). Later work narrowed the picture. A 2008 *Cell Metabolism* study found that mice on a standard diet did not live longer when treated from 12 months of age ([DOI](https://doi.org/10.1016/j.cmet.2008.06.011)), and the 2019 review describes the earlier SIRT1 activation results as an assay artifact. For a separate look at NMN and NR, see the [visual research article](/research/nmn-nr-synergistic-stack-visual). ## What human trials show, and how doses compare The 2019 review counted around 130 human clinical trials with varying results, and said expert opinion did not yet support its use for treatment or prevention ([DOI](https://doi.org/10.4062/biomolther.2018.176)). In a 30-day double-blind crossover trial, 11 men took 150 mg of resveratrol a day or placebo, and the authors said the metabolic changes mimicked calorie restriction ([DOI](https://doi.org/10.1016/j.cmet.2011.10.002)). The label of [NMN + NR with Resveratrol and TMG](https://mtangelvitamins.com/products/nmn-nr-cellular-wellness-supplement?utm_source=learn_hub&utm_medium=content&utm_campaign=blog_post&utm_content=history-of-resveratrol-from-hellebore-to-french-paradox) lists 250 mg of resveratrol per two-capsule serving, and its directions call for two capsules a day. That is more than the trial's 150 mg, but the trial tested resveratrol alone, so the comparison says nothing about the rest of the capsule. For the article on that combination, see [NMN, NR, TMG, resveratrol and zinc](/research/nmn-nr-tmg-resveratrol-zinc-synergistic-stack). ## Bottom line The history of resveratrol is a chain of links that each need a check: a 1939 isolation, a plant that makes the compound, a wine association that drew attention, and mouse and cell results that later sources questioned. The human evidence is small and mixed. When a label lists a dose, compare it with the trial it cites, and check whether that trial tested the compound alone. ## Frequently asked questions ### Which plant was resveratrol first isolated from? According to PubMed, a 2019 review reports that Takaoka first isolated resveratrol in 1939 from the root of white hellebore ([DOI](https://doi.org/10.4062/biomolther.2018.176)). The abstract names the source only as "O. Loes," so the plant comes from the review's full text. We did not find the 1939 original in PubMed, so this answer rests on one secondary review, and the page treats it that way. ### How did Japanese knotweed enter the story? We did not find a source that explains how knotweed entered the resveratrol story. The sources do support a plant with a long record in traditional Chinese herbal medicine. A 2019 genome paper reports that *Polygonum cuspidatum*, known as Huzhang in Chinese, is recognized there, with the roots as its main medicinal tissue ([DOI](https://doi.org/10.3389/fpls.2019.01274)). The same paper links an expanded family of stilbene synthase genes to resveratrol biosynthesis in the plant. ### Why did the French paradox put resveratrol in the headlines? The sources we checked do not show that it did. The 1992 *Lancet* abstract describes low coronary heart disease mortality in France despite high saturated fat intake, partly credits wine, and does not name resveratrol ([DOI](https://doi.org/10.1016/0140-6736%2892%2991277-f)). The documented links are different. A 2019 review says interest in resveratrol stemmed from grapes and wine, and that grape product sales rose after a 1997 *Science* report ([DOI](https://doi.org/10.4062/biomolther.2018.176)). ### Why do wine and grape skins come up in the story? The sources name grapes and wine, not grape skins. A 2019 review says resveratrol occurs naturally in grapes and grape products, mainly wine, and that a liter of wine contains only low milligrams of it ([DOI](https://doi.org/10.4062/biomolther.2018.176)). The 2003 yeast paper described it as a polyphenol found in red wine ([DOI](https://doi.org/10.1038/nature01960)). We did not find a PubMed source that ties the compound to grape skins specifically, so this page does not make that claim. ### What did the early yeast and mouse studies test? The 2003 *Nature* study tested resveratrol in yeast and reported a 70% lifespan extension linked to stimulation of the Sir2 enzyme ([DOI](https://doi.org/10.1038/nature01960)). The 2006 study gave it to middle-aged mice on a high-calorie diet and reported significantly better survival ([DOI](https://doi.org/10.1038/nature05354)). A 2008 study found that mice on a standard diet did not live longer when treated from 12 months of age ([DOI](https://doi.org/10.1016/j.cmet.2008.06.011)). None of these studies measured human outcomes, so they describe the animal record only. ### How do supplement doses compare with the research doses? The 2011 human trial used 150 mg a day for 30 days in a small crossover study ([DOI](https://doi.org/10.1016/j.cmet.2011.10.002)). The 2019 review says nearly all supplements on the open market provide high milligram to gram levels, and that most animal and human studies use concentrations far above what diet could supply ([DOI](https://doi.org/10.4062/biomolther.2018.176)). Dose alone does not settle the question. A 2026 dose-response meta-analysis found no significant link between daily dose and the inflammatory outcomes it measured ([DOI](https://doi.org/10.1007/s10787-026-02424-x)). ### What do the human trials show so far? The evidence is small and mixed. The 2019 review counted around 130 human trials with varying results, and said expert opinion did not yet support using resveratrol for treatment or prevention ([DOI](https://doi.org/10.4062/biomolther.2018.176)). A 2026 systematic review and dose-response meta-analysis of 25 randomized trials in people with metabolic disorders found a statistically significant but modest drop in C-reactive protein ([DOI](https://doi.org/10.1007/s10787-026-02424-x)). Effects on the other inflammatory and oxidative stress markers were null or inconclusive, and the authors say C-reactive protein's status as a surrogate marker limits how far the finding can be read. ## Next reads - [Resveratrol ingredient page](/ingredients/resveratrol): the compound's reference profile. - [History of creatine, from meat extract to sports shelf](/blog/history/history-of-creatine-from-meat-extract-to-sports-shelf): another origins story in this series, starting from a food source. ### Related reading - [the piperine origins story](/blog/history/history-of-black-pepper-from-spice-trade-to-piperine) - [the pomegranate origins story](/blog/history/history-of-pomegranate-from-ancient-orchards-to-urolithin-a)