It wasn’t easy to make a meal of teosinte, a grass that was the ancient precursor to maize. Each cob was shorter than your little finger and harbored only about 12 kernels encased in rock-hard sheaths. But in a dramatic example of the power of domestication, beginning some 9000 years ago people in Mexico and the U.S. Southwest transformed teosinte into the many-kerneled maize that today feeds hundreds of millions around the world.
Researchers had already identified a handful of genes involved in this transformation. Now, studies of ancient DNA by two independent research groups show what was happening to the plant’s genes mid-domestication, about 5000 years ago. The snapshot reveals exactly how the genetics changed over time as generations of people selected plants with their preferred traits.
“These results sharpen the focus of what we know at this early period,” says Michael Blake, an anthropologist at the University of British Columbia in Vancouver, Canada, who was not involved in the work. “They have implications for understanding later developments in maize domestication and help us to see what people were selecting for at the time.”
The first glimpses of maize domestication came in the 1960s, when esteemed U.S. archaeologist Richard MacNeish excavated at caves in Mexico’s Tehuacán Valley, a center of early Mesoamerican agriculture. In the dry, dark environment there, he found tiny, well-preserved maize cobs dated to roughly 5300 years ago and harboring only 50 kernels each, compared with the 1000 on modern cobs.
Nearly 60 years later, after the advent of modern sequencing tools, geneticist Jean Philippe Vielle-Calzada at the Center for Research and Advanced Studies of the National Laboratory of Genomics for Biodiversity in Irapuato, Mexico, and his colleagues wanted to find out which genes the ancient domesticators had unwittingly been selecting. But he worried that MacNeish’s specimens, now in museums, might have been damaged by handling or improper storage. So he and his team decided to go back to the caves in Tehuacán Valley. Macneish had died, but one of his former students, Angel Garcia Cook, served as guide. “He had all the maps, he knew where to dig,” Vielle-Calzada says. “He went back with us at 73 years old. When he went the first time he was 21.”
The team discovered several new specimens, dated to about 5000 years ago, from San Marcos cave. They applied shotgun sequencing to three cobs, extracting DNA and breaking it up into short fragments for sequencing. Computer software then reassembled these DNA snippets, eventually reconstructing more than 35% of the ancient maize genome.
Vielle-Calzada’s team identified eight genes influencing key traits, as they wrote in the Proceedings of the National Academy of Sciences this week. The cobs carried the modern variants of tb1, which simplified the plant’s branching for easier harvest, and bt2, which helped boost the starch content and sweetness of the kernels. But the cobs had the teosinte variant of tga1, which encloses the kernels in those hard sheaths—a sign that domestication was only partial.
Meanwhile, archaeologist Nathan Wales of the University of Copenhagen and his colleagues discovered MacNeish’s original samples, stored for 60 years in a museum in Andover, Massachusetts. He and his colleagues shotgun sequenced the genome of a 5300-year-old cob called Tehuacan162.
Wales’s team was able to sequence 21% of this cob’s genome. Their results confirmed and complemented those of Vielle-Calzada’s team. The museum cob also had modern variants of td1 and bt2, as reported in Current Biology last week. But Tehuacan162 also had a more modern variant of the gene tga1, which partly released kernels from their rigid shells, making them easier to eat. Wales’s team also found a teosinte gene not seen by the Mexican team: zagl1, which makes kernels fall from the cob very easily. That’s useful for wild plants spreading their seeds, but frustrating for humans trying to harvest them. These differences may reflect the fact that Tehuacan162 came from a different population of maize, and show that domestication was still in progress in the valley, the researchers say.
Vielle-Calzada was shocked at the teams’ similar results. “I’m really amazed to see how convergent the results are,” he says. “This is unusual in paleogenomics where it’s difficult to get good data from old DNA. This is encouraging.”
Robert Hard, an archaeologist at the University of Texas in San Antonio, agrees: “It’s remarkable how these studies support each other,” he says. That’s a good sign that future sequencing can fill in more details, he says. “It’s really important that we recognize the significance of transformations in maize,” Blake says, adding that knowledge of how certain traits helped maize adapt to drought and disease in the past could help save it from disasters in the future.
With reporting by Lizzie Wade.