Scientists have sequenced the oldest DNA yet, smashing through a symbolic barrier in the study of ancient genomes and opening an unprecedented window into the evolution of North America’s extinct Ice Age giants—the Columbian and woolly mammoths.
The feat is unlikely to spark a mammalian Jurassic-Park style recreation; the study isn’t the first to sequence a mammoth’s genome, nor does it bring humankind any closer to resurrecting a mammoth. Instead, the study of DNA more than a million years old, published in Nature on Wednesday, sets a milestone for the rapidly growing study of ancient DNA, nearly doubling the record for the oldest genome ever sequenced.
The DNA comes from three mammoth molars found in Siberia in the early 1970s by Russian paleontologist Andrei Sher, a legend in the field for his mammoth research. Researchers estimate that the youngest of the three teeth is about 500,000 to 800,000 years old, while the older two are between one million and 1.2 million years old. The next-oldest DNA ever sequenced came from a nearly 700,000-year-old horse fossil found in Canada’s Yukon Territory.
“Breaking this somewhat magical barrier of more than one-million-years-old opens a new time window, so to say, and evolutionary perspective,” says lead study author Tom van der Valk, an Uppsala University bioinformatician who worked on the study while at the Centre for Palaeogenetics in Stockholm, Sweden.
The findings add surprising detail to scientists’ picture of how North America’s mammoths evolved. For one, the teeth’s ancient DNA strongly suggests that North America’s Columbian mammoth, one of the major North American mammoth species, is a hybrid that arose 400,000 to 500,000 years ago—a fact only revealed because the study’s older DNA dramatically precedes this interbreeding. “If we look at higher-order organisms like vertebrates, I can’t think of a single example where people have sampled before the origin of a species,” says study coauthor Love Dalén, a geneticist at the Centre for Palaeogenetics.
The further back in time DNA records can go, the more scientists can learn about how evolution works. The study’s success also implies that in the perfect conditions, even deeper glimpses into the evolutionary past may be possible, potentially back a few million years, its authors say. (Any older than that, and the DNA would be broken into pieces too small to reassemble.)
Work on the teeth began in 2017, when the Centre for Palaeogenetics received samples of the teeth from the Russian Academy of Sciences. Clad in protective suits now grimly familiar in the age of COVID-19, a team led by geneticist Patrícia Pečnerová, a postdoctoral researcher now at the University of Copenhagen in Denmark, ground 50 milligrams of bony powder off of each sample. Pečnerová then carefully extracted small amounts of DNA from each pinch of powder with a series of chemical baths, which concentrated the DNA in small drops of fluid no bigger than peppercorns.
“Basically I’m like in a cocoon—with a face mask and face shield—really trying to minimize contamination,” Pečnerová says. “One single [human] cell could fall down into the tube” and ruin the sample.
Sequencing this DNA was just the first step. Next, van der Valk and his colleagues had to ensure that they focused on only the DNA snippets that were authentically old, and authentically mammoth in origin. After all, the teeth had been buried for upwards of a million years in permafrost teeming with microbes, and they had been dug up and handled by countless scientists over nearly five decades. Despite best efforts to prevent contamination, the researchers had to contend with whatever extra DNA the teeth had picked up in their travels.
After weeks of computationally crunching through the sequenced DNA, the team could accurately identify snippets of mammoth DNA as short as 35 base pairs long and map them onto a genome that, in life, was more than three billion base pairs long.
A sequence surprise
Already, the new study is shedding light on how North America’s mammoths evolved. To the researchers’ shock, the new study’s DNA sequences are so old that they predate the origins of the Columbian mammoth, one of two major mammoth species that once roamed North America—giving scientists fresh insight into how mammoths evolved.
By 1.5 million years ago, relatives of Europe and Asia’s steppe mammoth had arrived in North America from Siberia, crossing a land bridge now covered by the Bering Strait. These fresh arrivals later gave rise to the Columbian mammoth. By about 100,000 to 200,000 years ago, North America was home to at least two main types of mammoths: woolly mammoths in the north, and Columbian mammoths as far south as Mexico. Researchers also knew from past genetic studies that Columbian mammoths and woolly mammoths interbred.
Paleontologists have long used mammoths’ distinctive upper molars to help tease apart different species. Based on fossil mammoth teeth, paleontologists traditionally had surmised that the mammoths present in North America after about 1.5 million years ago were Columbian mammoths. But whereas the fossil tooth record shows continuity, the genetic record in the new DNA study reveals change.
Two of the new study’s mammoth genomes fall into the lineage that later gave rise to woolly mammoths. But DNA from the oldest of the three teeth, nicknamed Krestovka by the scientists after the river near which it was found, seems to fall into a previously unknown genetic lineage, one that about 1.5 million years ago split from the lineage containing the other two teeth.
When van der Valk’s team compared the mysterious mammoth genome to previously sequenced Columbian mammoth DNA, the researchers came to a startling conclusion: The Columbian mammoth is a hybrid that arose 400,000 to 500,000 years ago, after interbreeding between the Krestovka mammoths and Siberian woolly mammoths occurred somewhere in Siberia, North America, or Beringia, the land bridge that once connected the two.
After a second interbreeding event that took place in North America roughly 200,000 years ago, the Columbian mammoth gained another 11 to 13 percent of its genome from woolly mammoths. By the time the Columbian mammoth went extinct some 12,000 years ago, about three-fifths of its genome traced back to the woolly mammoth, while the other two-fifths traced back to the enigmatic Krestovka mammoth, which is known only from the DNA contained within a single tooth.
The study also shows how well—and how early—mammoths adapted to the cold. Past ancient DNA studies had delved into the genetic details of how the woolly mammoth thrived in low temperatures. But many of the gene variants behind the woolly mammoth’s ability to endure the cold first appeared in much earlier mammoths. The new study finds that upwards of 85 percent of these woolly variants were already in Siberia’s steppe mammoths, woolly mammoths’ ancestral cousins, more than a million years ago.
By that million-year mark, mammoths were already living at high latitudes, based on fossil evidence, so it’s unsurprising that these frosty titans adapted to weather the cold. However, the study provides a unique glimpse into the pace of this winterizing process. Mammoths seem to have evolved these cold-adapted gene variants at a more or less steady pace, not in bursts.
Details in the DNA
Paleontologists say the revelation that Columbian mammoths were hybrids will further stoke an ongoing reevaluation of the North American mammoth fossil record.
Recent research comparing fossil mammoth teeth against genetic family trees has found that—far from being dead ringers for different mammoth species—the shape and form of teeth overlapped considerably from region to region across North America. The new study accentuates this point: There’s no huge change in North America’s fossil mammoth teeth before and after 500,000 years ago, even though the genetic changes that yielded the Columbian mammoth were immense.
“Without the genetics, we’re usually looking at morphology, or changes in shape, and without those changes in shape, we can’t document changes in species,” says Lindsey Yann, the paleontologist at Waco Mammoth National Monument in Texas. “When you add in that genetic component, we’re able to actually separate things, and we have the data to show that.”
For study coauthor Adrian Lister, a paleontologist at the Natural History Museum, London, and one of the world’s foremost mammoth experts, the study also highlights a lingering tension: how to define North American mammoth teeth in cases where DNA is still missing. If, genetically speaking, Columbian mammoths don’t show up until 400,000 to 500,000 years ago, how should paleontologists define older mammoth teeth that otherwise look identical? So far, nobody has published any DNA from North American mammoth teeth more than half a million years old.
To fill in more of the puzzle, Dalén says, he and his colleagues want to try applying their record-breaking skills on North American mammoth teeth. Already, the team has identified a 500,000-year-old mammoth tooth from Canada, as well as a 200,000-year-old tooth that probably belonged to a woolly mammoth, as possible candidates for future sequencing.
And now that scientists have broken through the million-year barrier, it’s just a matter of time before even older DNA reveals its secrets. “That’s the million-dollar question,” Dalén says. “We’ve seen the data we have, and I think it would be relatively easy to go beyond two million, if we just had a good specimen.”