Domestication: The birth of rice

From a wild Asian grass to a refined crop that is the staple diet of half the world's population, the domestication of Oryza sativa spans centuries, but the grain's ancestry is hotly contested.

Asian civilization was built on rice — on Oryza sativa, to be exact. The crop, which today is the primary food source for half of the world's population, transformed nomadic hunter-gatherers into stay-at-home farmers, spawned the first urban centres and built empires and dynasties. “Probably more so than any crop, it drove societies and economies to become densely populated, potentially more urbanized, and it also transformed landscapes,” says Dorian Fuller, an archaeobotanist at University College London.

Chinese farmers planting rice — humans domesticated grass many centuries ago but archaeologists are still investigating the crop's origin. Credit: Mary Evans Picture Library/Grosvenor Prints

Despite — or possibly because of — rice's primacy, the history of the grain remains controversial, with little agreement on where, when and how many times humans tamed O. sativa in Asia to create the world's most important crop. (The only other domesticated rice species, Oryza glaberrima, has its roots in Africa. See 'The second story'.) “Almost every part of Asia had been pinpointed as the area where rice originated,” says Michael Purugganan, an evolutionary geneticist at New York University who studies rice domestication. Unravelling the history of rice in Asia would illuminate a turning point in human civilization and give scientists fresh insight that could help improve the crop for the future. Thanks to advances in genetics and to new archaeological finds, that history is becoming clearer — and it is a lot more complicated and convoluted than anyone thought.

Competing claims

Oryza rufipogon, the Asian wild grass that is most closely related to O. sativa, is a sinewy, weedy plant. Its red grains are edible, but some modern rice growers consider it a pest. As humans started intentionally planting rice around 8,000 to 9,000 years ago, they sought out plants with the most desirable traits. Over time, the cultivated grass became stouter and straighter, producing heftier grains in greater quantities and that clung to the plant instead of tumbling to the soil — or 'shattering' — to facilitate spread. Claiming the title of the birthplace of rice would be a matter of great pride for a nation. This, along with the wide geographical range of O. rufipogon, has led to competing claims for domestication — including the Ganges valley in northern India, several locations in China, southeast Asia, the southern slope of the Himalayas and numerous other places. “Because rice is embedded within cultural identities within different nations in Asia, every-body wants to have had rice first,” says Fuller.

For the past few decades, much of the focus on rice domestication has centred on China and India. In the 1960s and 1970s an influential Taiwanese agriculture scientist, the late Te-Tzu Chang, collected and cultivated wild rice from across Asia and proposed a band of domestication that stretches from northern India to southern China. As Fuller explains, this work encouraged Indian archaeologists to continue maintaining that rice came out of India, and the Chinese to claim that it came out of South China.

Perhaps both camps are correct. Han dynasty records dating back more than 2,000 years distinguish between two varieties of rice, Keng and Hsien, now known as japonica (short-grained) and indica (long-grained) respectively. Research¹ comparing dozens of wild and domestic strains has suggested that japonica and indica are more closely related to distinct wild varieties than they are to each other, pointing to two separate domestications: japonica in China and indica in India.

Indica and japonica clearly come from different genetic stock, says Tao Sang, a plant geneticist at the Institute of Botany of the Chinese Academy of Sciences in Beijing. But a closer look at those genes shows some surprising — and telling — overlap. In 2006, Sang's team identified a variation in a gene in domesticated rice known as sh4 that prevents rice grains from shattering². Japonica and indica share identical non-shatter mutations in sh4, suggesting that the two varieties had a shared history. Later studies suggested that the mutation arose in an ancestor of japonica rice first then found its way to indica. Since then, other teams have identified domestication genes that follow the same pattern: they appeared first in japonica before showing up in indica. One example is a mutation in a gene called Rc that lightened the seed coating, or pericarp³.

These discoveries sent rice scientists scrambling to explain how indica and japonica could be genetically distinct across most of their genomes yet share mutations responsible for the key traits that made rice easier to cultivate. Although the specifics are still hotly contested, most researchers have concluded that human-directed hybridization played an important part in the very early history of domesticated rice.

One leading theory comes from a 2011 study⁴ by Puruggunan and his colleagues that modelled the relationship of dozens of wild and domestic rice varieties. The study suggested that the domestication of japonica, most probably in China, was the watershed event in the history of Asian rice. From there it seems that japonica spread to India, where farmers intentionally hybridized it with their local rice to produce indica. The exact nature of that local rice is still a mystery. It could have been an early domesticated variety, or it could have been wild.

Whether the creation of indica from japonica counts as a separate domestication event “is somewhat a semantic issue”, says Jeffrey Ross-Ibarra, an evolutionary geneticist at the University of California, Davis. The take-home message remains: the earliest fields of japonica — again, most probably in China — provided the key materials for the rice revolution in Asia.

Origin stories

Exactly where in China japonica was domesticated is another lively debate. The lower stretches of the Yangtze River valley in eastern China had long laid claim to early domestication, based on the discovery of roughly 8,000-year-old grains in archaeological sites there. And so a team led by Bin Han, a geneticist at the Shanghai Institute for Biological Sciences of the Chinese Academy of Sciences, raised eyebrows in 2012 when they proposed the Pearl River valley in deep southern China as the birthplace. Han's team had compared the genomes of more than 1,000 domesticated strains of rice and another 446 wild varieties from across Asia. They found that both indica and japonica were most closely related to wild varieties growing in the Pearl River valley. There is no archaeological evidence supporting early rice cultivation in the region, which is one reason why the claim remains controversial. But Han says that this could be because of the comparative poverty of its ancient inhabitants, who did not keep large stores of rice, unlike the richer Yangtze valley dwellers.

A pot from the Han dynasty, about 162 BC, for steamed food. Credit: The Museum of East Asian Art/HIP/TopFoto

Archaeological work remains focused on the Yangtze. Since 2004, Fuller has excavated at a site in the lower Yangtze River valley that has been touted as an early centre of rice domestication. “I had always been taught, and all the textbooks will tell you, that there's early rice domestication here in 5000 BC,” he says. But the history, he says, turns out to be a bit more nuanced. “My impression of visiting this site is that it's full of storage pits that are full of acorns. Nobody mentions the acorns.”

In Fuller's view, the early inhabitants of this and other Yangtze valley sites were hunter-gatherers who dabbled in rice cultivation but mostly ate acorns, water chestnuts and fish. In a 2009 paper⁵ that analysed thousands of microscopic rice remains from the site, Fuller's team documented a slight increase in the presence of domestic-looking rice, beginning around 5000 BC. But these rice grains kept changing shape — a sign of a crop not yet fully domesticated. Not until the shape stabilized thousands of years later did acorns disappear from Chinese archaeological sites. Rice domestication, as Fuller sees it, was a slow, haphazard enterprise.

In the hope of untangling rice domestication further, researchers are turning to the same ancient-DNA technologies that have revolutionized the understanding of human evolution. If they can recover DNA from ancient rice remains, researchers may be able to determine when and where the key domestication genes emerged. Most rice from archaeological sites is charred, and past efforts to glean DNA from grains have failed. But modern sequencing machines that can decode very short strands of DNA — likely to be the only ones left in rice that is thousands of years old — mean it may be possible to get DNA and even whole genomes from ancient rice. “The techniques are here, and the revolution is going to happen,” says Fuller, who has begun a project with a team at the University of Warwick, UK, to sequence ancient rice DNA from China, India and Thailand.

Humans continue to shape the rice genome, but some researchers worry that intensive breeding for high yield and pest resistance has narrowed the genetic diversity of cultivated rice, leaving crops more susceptible to disease and less adaptable to the effects of climate change. Sang and other scientists studying rice domestication argue that contemporary rice breeders should take advantage of the handiwork of the humans who transformed wild grasses into the world's most successful crop. Ancient rice varieties no longer under widespread cultivation could provide attributes to help high-yielding strains to feed a world of nine billion people, Sang says. “There's great potential in those ancient cultivars.”

Box 1: The second story: The roots of African rice

A few thousand years after Asian farmers fully domesticated rice, ancient African farmers created a species of their own. African rice (Oryza glaberrima) generally does not produce as many grains as Asian rice, and although the crop never became a dominant staple, it was historically an important addition to some local diets (see page S64). In July, a team co-lead by Rod Wing, a geneticist at the University of Arizona in Tucson, reported⁶ the genome sequence of African rice, along with an analysis of 20 domestic strains and 94 varieties of Oryza barthii, its closest wild relative. Their analysis suggested that the African rice was domesticated just once, probably near the Niger River in West Africa. What is more, they discovered that African rice has mutations near the same genes that control shattering in Asian rice, as well as other domestication traits. Although humans in Africa and Asia domesticated rice independently and thousands of years apart, they altered the crop's genome in similar ways, Wing says.