Gathering Scientific Evidence that Traditional Medicines Can Work

Illustration for article titled Gathering Scientific Evidence that Traditional Medicines Can Work

Traditional medicine doesn't just fill up the health food aisle at the supermarket — it could help make everybody healthier. But how can we figure out which ancient herbal remedies actually work, and which ones are just hype? An estimated 10,000 to 53,000 plant species were traditionally used as medicines, and only some of those could have bioactive molecules with actual molecules. That's a lot of plants to sort through.


But now, a new study in the Proceedings of the National Academy of Sciences suggests that we can find out — by comparing the plants that multiple different cultures adopted as remedies.

Top image: Smoobs/Flickr.

For example, say a plant often used to cure headaches in an ancient Nepalese culture was closely related to a plant used for the same purpose in South Africa. The communication-crippling geographical distance between these two cultures indicates that they discovered their headache cures independently. So they probably began using these plants because the flora had real pain-killing abilities. This plant family may produce valuable bioactive compounds.

But studies attempting to analyze plants in this way face a tricky obstacle: that large geographical distance between the cultures also means that they don't share many plant species. As the scientists write in their paper, "The disparate regions that have experienced limited cultural contact are floristically disparate too, so different cultures will not be exposed to the same species, genera, or even families."

To get around this problem, the latest study ignored the system of taxonomy and instead directly compared plants' DNA. Researchers analyzed gene sequences from 20,000 plant species native to Nepal, New Zealand, and South Africa. Genetic differences between species helped the scientists reconstruct the plants' evolutionary relationships and place each species in its proper place on a type of family tree.

1,500 of the species studied were also used in traditional medicine. And the researchers found that these plants tended to cluster around the same sections of the tree, areas dubbed "hot nodes." Hot nodes also contained a significant number of known bioactive plants, which have already become the basis for modern medical drugs. When the researchers divided the traditionally used species into 13 categories, based on the type of ailments the remedies were supposed to treat, they found that plants in each category also clustered together.

The authors conclude that the plants near hot nodes on the cross-cultural plant family tree are more likely to have biological effects on the human body, and should become targets for drug development. "More than 80% of plant species have not been investigated for bioactivity and methods to distinguish those plants most likely to be bioactive when selecting species for further testing are needed," they write. "The finding that medicinal plant use shows strong phylogenetic clustering indicates targeting close relatives of plants with known bioactivity or phylogenetic medicinal hotspots identified as hot nodes is a good strategy for focused screening."



Cursing at the Astronette

Can't tell from the abstract, but would be good to know if they identified well established plant-derived therapeutics - aspirin, taxanes, digitalis, etc - from the analysis. Perhaps further comparative genomic analyses will enable identification of therapeutically active molecular species, by identifying common gene networks.