A previously unknown human ancestor may have coexisted with Neanderthals and early modern humans, German researchers report online in Nature today (March 24).

Denisova cave from the outside
Image: Bence Viola

For the first time, the scientists identified the novel hominin using mitochondrial gene sequencing of bone fragments, not fossils. The genomic analysis also revealed a hitherto-unknown migration from Africa to Eurasia just under one million years ago. "It's such a surprise," said Terence Brown, a biomolecular archaeologist at the University of Manchester, who was not involved in the study. "You start to think, how complete is our knowledge of ancient human ancestors?"

"While most of us are convinced that other humans existed, the idea that we might be able to identify them is rather a different thing," said Guido Barbujani, a population geneticist at the University of Ferrara, Italy, also not a co-author. "In this case, the...

Svante Pääbo of the Max Planck Institute for Evolutionary Anthropology in Leipzig, Germany, and his colleagues sequenced mitochondrial DNA they extracted from a fragment of bone found in Denisova Cave in the mountainous steppe of Siberia. Based on its size, the fragment appears to be the pinky finger of a small child, and carbon dating of other artifacts found at the same archaeological level date it at around 40,000 years old. 

Researchers use mtDNA to date ancient samples because it is much more abundant than nuclear DNA -- a cell has just two copies of the nuclear genome but thousands of copies of the mitochondrial genome. Still, extracting mtDNA pure enough for sequencing from samples that old, and making sure the samples aren't contaminated by the researchers themselves, is no easy task. 

Last year, Pääbo's lab published a method for extracting and enriching ancient mtDNA, as well as the complete mtDNA sequences of five Neanderthals. They applied the same technique to the present sample. Mitochondrial DNA acts like a molecular clock; by measuring accumulated mutations, researchers can assign approximate ages to sequences. 

The researchers compared the new mtDNA to the sequence in 54 humans from the present time; a sequence from an early modern human sample, about 30,000 years old, found in Russia; six compete Neanderthal mtDNA sequences; a bonobo mtDNA sequence; and a chimpanzee mtDNA sequence. Their analysis showed that the new fragment's sequence differed significantly from modern human sequences, differing by almost twice as much as the Neanderthal mtDNA differed from humans'. The results suggest the hominin diverged from a common ancestor well before Neanderthals and modern humans did -- about one million years ago. 

"At the moment with existing data there seems no doubt that this is a new human form," said Barbujani.  

Homo erectus, the first hominin to move from Africa to Eurasia, did so 1.9 million years ago. After that, two known migrations occurred: Neanderthals are thought to have left Africa between 300,000 and 500,000 years ago, while early modern humans -- immediate ancestors of Homo sapiens -- took a later wave, migrating some 50,000 to 60,000 years ago. However, because the lineage of the hominin whose finger was found in the cave split off before Neanderthals and early modern humans arrived, the researchers speculate that that this new human form must have moved to Eurasia after H erectus but before Neanderthals -- probably a couple of hundred thousand years after the divergence. 

"It was amazing to clearly see something else coming out of Africa between these two points," said Pääbo in a press briefing yesterday. In hindsight, though, he said, this should perhaps not be surprising. "There might have been more or less continuous gene flow" out of Africa, he suggests, rather than distinct migration. "The picture might be a much more complex one" than the data had so far suggested. 

"I think the really interesting bit is, what the hell is it?" said David Lambert, an evolutionary biologist at the Griffith University in Brisbane, Australia. Lambert added that he is not yet convinced the sample is an entirely new hominin. Pääbo noted in the briefing that it's impossible to say whether the new hominin is a new species. "There's no metric where you can say, this percentage of divergence is a new species," he said. 

The researchers are now trying to sequence as much nuclear DNA as they can recover from the bone fragment, and expect to have results in a few months, said Pääbo. That analysis would help clarify the relationship between the Denisova hominin, Neanderthals, and early modern humans. "This is high tech stuff," said Brown, who wrote an accompanying commentary on the study in Nature, adding that just four or five labs around the world had the expertise to perform such feats of sequencing on ancient DNA. "The ability to discover these things using DNA really opens things up."

Interested in reading more?

Become a Member of

Receive full access to more than 35 years of archives, as well as TS Digest, digital editions of The Scientist, feature stories, and much more!