Whether you think it’s the devil’s lettuce, nature’s medicine or a conduit to the divine, cannabis and humans have a relationship stretching back thousands of years that has now spawned hundreds of varieties. But millennia of cultivation, breeding and the plant’s relatively recent status as a cultural taboo have obscured where exactly cannabis went from being a wild weed to being picked up by humans and put on a path toward becoming the multi-billion-dollar crop it is today.
Now, after sequencing and analyzing more than 100 whole Cannabis sativa genomes, researchers say that the wild plants that gave rise to today’s psychoactive strains as well as the fibrous varieties used in textiles likely came from East Asia—or present-day China. The analysis also suggests that cannabis took its first steps toward domestication somewhere in China about 12,000 years ago.
The findings, published today in the journal Science Advances, could spark the development of new varieties using the Chinese cannabis strains shown to be more genetically similar to the crop’s wild progenitors. This could provide plant breeders with a new genetic toolset to boost cannabis’ disease resistance, production of certain compounds or growing efficiency.
Prior to this paper, the predominant notion was that Central Asia was where cannabis was first domesticated.
“This was mostly based on the fact that in Central Asia, it’s very easy to find lots of feral plants growing on their own, even along roadways,” says Luca Fumagalli, an evolutionary biologist at the University of Lausanne in Switzerland and co-author of the study. “Observational and historical data have suggested this might have been the original domestication site for cannabis.”
These roadside plants, sometimes referred to as “ditch weed” in the United States, may look wild but they’re more accurately classified as feral. Despite propagating without direct human assistance, DNA analysis shows these upstarts are descended from escaped domesticated plants. Because the species is wind pollinated, these escapees can readily mix with any other nearby cannabis plants. In some locations this would have set up a scenario in which domestic escapees likely swapped genes with their undomesticated ancestors, potentially diluting or even, in a slightly Oedipal turn, eliminating truly wild cannabis. This, on top of continuous artificial selection and intentional hybridization by farmers and breeders as well as the plant’s still-checkered legal status, which hamstrung research for decades, has made the tale of cannabis’ origins a tangled one.
To figure out where and when cannabis was domesticated, researchers spent around four years assembling what they say is a comprehensive group of 110 different cannabis varieties from around the world. The international collaboration gathered seeds, leaves and other shreds of plant material in the field in countries including Switzerland, China, India, Pakistan and Peru, as well as from botanical collections and commercial sources.
The researchers used all this greenery to sequence 82 brand new genomes which they then combined with 28 genomes that were already publicly available.
“They harvested 12 million SNPs [single nucleotide polymorphisms] for analysis,” says John McPartland, a botanist and physician at the University of Vermont who was not involved in the paper. A SNP is a single location within a genome that differs compared to a reference genome, and SNPs can be sources of variation within a species. Getting 12 million SNPs for cannabis is “astounding,” says McPartland, “previous studies were in the thousands.”
Armed with this trove of genetic information, the team analyzed the genomic data to figure out the evolutionary relationships among these cannabis plants from around the world. The analysis revealed that a group of feral plants hailing from East Asia are more closely related to cannabis’ wild ancestors than any of the varieties grown today for fiber or medicine and recreation.
“If you wanted to make a world map of where all the genetic pieces of cannabis come from, this paper shows us where all the weird and wild stuff is,” says Jonathan Page, a plant biologist at the University of British Columbia who was not involved in the work.
The researchers also found that this so-called basal lineage of cannabis split off from the more heavily domesticated varieties grown today about 12,000 years ago, and that the psychoactive and fibrous strains of cannabis didn’t diverge from one another until about 4,000 years ago. This places the initial stages of cannabis’ domestication in a time and place that is already known as a hotbed of agricultural innovations—where modern crops including rice, broomcorn and foxtail millet, soybean, foxnut, apricot and peach were launched.
For the researchers behind the paper, the results came as a surprise. “We thought we would find two main lineages, one with plants for fiber use and then plants developed for cannabinoid production,” says Fumagalli. “We didn’t expect to find this third independent and basal lineage among the samples from East Asia.”
The two main lineages Fumagalli mentions are commonly referred to as hemp, which is grown for its fibrous stems, and marijuana, which has mostly been bred to produce psychoactive compounds called cannabinoids that include THC and CBD. This sometimes-fuzzy linguistic distinction has become an official legal designation in places like the European Union and the U.S., where marijuana is still illegal at the federal level. Any plant with more than 0.3 percent THC content per dry weight is officially considered by the authorities to be the drug marijuana, while plants below this threshold skate through the dragnet as hemp.
Though the third lineage uncovered by the study is more closely related to cannabis’ wild ancestors than any of the drug or fiber varieties grown today, these basal cannabis varieties all appear to be feral rather than truly wild—meaning humans had some hand in guiding the plants’ evolution. This, along with the sheer number of genomes they sequenced, led Fumagalli and his co-authors to the somber conclusion that the pure wild progenitors of cannabis are probably extinct.
But other researchers wonder if this research might lead to individuals and companies looking for wild cannabis in East Asia anyway. “I wonder if there are plant explorer types who will go hunting for a Valley of Shangri-La for wild cannabis,” ponders Page. “East Asia is a big place, it may exist.”
Unless a cannabis Shangri-La materializes, the study’s most tangible contribution is likely to be the 82 newly sequenced genomes it makes publicly available.
“These additional genomic data are a phenomenal resource that adds a huge amount to our existing knowledge,” says Nolan Kane, a plant geneticist at the University of Colorado who wasn’t involved in the paper. “There really hadn't been much in the way of publicly available sequences from many of the countries they sampled—I’ll certainly be downloading their data and reanalyzing it.”
But despite the seemingly comprehensive nature of the sampling effort behind this paper Kane says it leaves geographical gaps that could yield new wrinkles in cannabis’ domestication story. For instance, the study contained no samples from Afghanistan, which is renowned for its profusion of cannabis, or Russia, which encompasses a vast swath of territory, much of which has little history of cultivating the crop and thus has potential as a bastion for wild cannabis.
Page also notes that the current study is based only on living samples, and that drawing on dried plant materials preserved in herbarium collections could provide a source of old and rare varieties. “There’s a whole other dimension to be explored there, but the onus is on us as a research community to extend the work,” says Page.
“This paper fills in a big knowledge gap on a plant that was sort of ignored by research because of legal restrictions, which really speaks to the ascendancy of cannabis science,” he says, “it shows that serious research on cannabis isn’t off limits anymore.”