‘Dragon Man' DNA revelation puts a face to a mysterious group of ancient humans

CNN

19-06-2025

Sign up for CNN's Wonder Theory science newsletter. Explore the universe with news on fascinating discoveries, scientific advancements and more.
CNN —
An enigmatic skull recovered from the bottom of a well in northeastern China in 2018 sparked intrigue when it didn't match any previously known species of prehistoric human. Now, scientists say they have found evidence of where the fossil fits, and it could be a key piece in another cryptic evolutionary puzzle.
After several failed attempts, the researchers managed to extract genetic material from the fossilized cranium — nicknamed Dragon Man — linking it to an enigmatic group of early humans known as Denisovans. A dozen or so Denisovan fossilized bone fragments had previously been found and identified using ancient DNA. But the specimens' small size offered little idea of what this shadowy population of ancient hominins looked like, and the group has never been assigned an official scientific name.
Scientists typically consider skulls, with telltale bumps and ridges, the best type of fossilized remains to understand the form or appearance of an extinct hominin species. The new findings, if confirmed, could effectively put a face to the Denisovan name.
'I really feel that we have cleared up some of the mystery surrounding this population,' said Qiaomei Fu, a professor at the Institute of Paleontology and Paleoanthropology, part of the Chinese Academy of Sciences in Beijing, and lead author of the new research. 'After 15 years, we know the first Denisovan skull.'
Denisovans were first discovered in 2010 by a team that included Fu — who was then a young researcher at the Max Planck Institute for Evolutionary Anthropology in Leipzig, Germany — from ancient DNA contained in a pinkie fossil found in Denisova Cave in the Altai Mountains of Russia. Additional remains unearthed in the cave, from which the group gets its name, and elsewhere in Asia continue to add to the still-incomplete picture.
The new research, described in two scientific papers published Wednesday, is 'definitely going to be among, if not the, biggest paleoanthropology papers of the year,' and will spur debate in the field 'for quite some time,' said Ryan McRae, a paleoanthropologist at the Smithsonian National Museum of Natural History in Washington, DC, who was not involved in the studies.
The findings could help fill in gaps about a time when Homo sapiens weren't the only humans roaming the planet — and teach scientists more about modern humans. Our species once coexisted for tens of thousands of years and interbred with both Denisovans and Neanderthals before the two went extinct. Most humans today carry a genetic legacy of those ancient encounters. Neanderthal fossils have been the subject of study for than a century, but scant details are known about our mysterious Denisovan cousins, and a skull fossil can reveal a great deal.
Long in the tooth
A laborer in the city of Harbin in northeastern China discovered the Dragon Man skull in 1933. The man, who was constructing a bridge over the Songhua River when that part of the country was under Japanese occupation, took home the specimen and stored it at the bottom of a well for safekeeping.
The man never retrieved his treasure, and the cranium, with one tooth still attached in the upper jaw, remained unknown to science for decades until his relatives learned about it before his death. His family donated the fossil to Hebei GEO University, and researchers first described it in a set of studies published in 2021 that found the skull to be at least 146,000 years old.
The researchers argued that the fossil merited a new species name given the unique nature of the skull, naming it Homo longi — which is derived from Heilongjiang, or Black Dragon River, the province where the cranium was found. Some experts at the time hypothesized that the skull might be Denisovan, while others have lumped the cranium in with a cache of difficult-to-classify fossils found in China, resulting in intense debate and making molecular data from the fossil particularly valuable.
Given the skull's age and backstory, Fu said she knew it would be challenging to extract ancient DNA from the fossil to better understand where it fit in the human family tree. 'There are only bones from 4 sites over 100,000 (years old) in the world that have ancient DNA,' she noted via email.
Fu and her colleagues attempted to retrieve ancient DNA from six samples taken from Dragon Man's surviving tooth and the cranium's petrous bone, a dense piece at the base of the skull that's often a rich source of DNA in fossils, without success.
The team also tried to retrieve genetic material from the skull's dental calculus — the gunk left on teeth that can over time form a hard layer and preserve DNA from the mouth. From this process, the researchers managed to recover mitochondrial DNA, which is less detailed than nuclear DNA but revealed a link between the sample and the known Denisovan genome, according to one new paper published in the journal Cell.
'Mitochondrial DNA is only a small portion of the total genome but can tell us a lot. The limitations lie in its relatively small size compared to nuclear DNA and in the fact that it is only inherited from the matrilineal side, not both biological parents,' McRae said.
'Therefore, without nuclear DNA a case could be made that this individual is a hybrid with a Denisovan mother, but I think that scenario is rather less likely than this fossil belonging to a full Denisovan,' he added.
Mounting evidence
The team additionally recovered protein fragments from the petrous bone samples, the analysis of which also suggested the Dragon Man skull belonged to a Denisovan population, according to a separate paper published Wednesday in the journal Science.
Together, 'these papers increase the impact of establishing the Harbin cranium as a Denisovan,' Fu said.
The molecular data provided by the two papers is potentially very important, said anthropologist Chris Stringer, research leader in human origins at London's Natural History Museum.
'I have been collaborating with Chinese scientists on new morphological analyses of human fossils, including Harbin,' he said. 'Combined with our studies, this work makes it increasingly likely that Harbin is the most complete fossil of a Denisovan found so far.'
However, Xijun Ni, a professor at the Institute of Vertebrate Paleontology and Paleoanthropology in Beijing who, along with Stringer, worked on the initial Dragon Man research but not the latest studies, said that he is cautious about the outcome of the two papers because some of the DNA extraction methods used were 'experimental.' Ni also said he finds it strange that DNA was obtained from surface dental calculus but not inside the tooth and petrous bone, given that the calculus appeared to be more exposed to potential contamination.
Nonetheless, he added that he thinks it is likely the skull and other fossils identified as Denisovan are from the same human species.
The goal in using a new extraction approach was to recover as much genetic material as possible, Fu explained, adding that the dense crystalline structure of dental calculus may help prevent the host DNA from being lost.
The protein signatures Fu and her team recovered indicated 'a Denisovan attribution, with other attributions very unlikely,' said Frido Welker, an associate professor of biomolecular paleoanthropology at the University of Copenhagen's Globe Institute in Denmark. Welker has recovered Denisovan proteins from other candidate fossils but was not involved in this research.
'With the Harbin cranium now linked to Denisovans based on molecular evidence, a larger portion of the hominin fossil record can be compared reliably to a known Denisovan specimen based on morphology,' he said.
A name and a face for Denisovans
With the Dragon Man skull now linked to Denisovans based on molecular evidence, it will be easier for paleoanthropologists to classify other potential Denisovan remains from China and elsewhere. McRae, Ni and Stringer all said they thought it was likely that Homo longi would become the official species name for Denisovans, although other names have been proposed.
'Renaming the entire suite of Denisovan evidence as Homo longi is a bit of a step, but one that has good standing since the scientific name Homo longi was technically the first to be, now, tied to Denisovan fossils,' McRae said. However, he added that he doubts the informal name of Denisovan is going anywhere anytime soon, suggesting it might become shorthand for the species, as Neanderthal is to Homo neanderthalensis.
The findings also make it possible to say a little more about what Denisovans might have looked like, assuming the Dragon Man skull belonged to a typical individual. According to McRae, the ancient human would have had very strong brow ridges, brains 'on par in size to Neanderthals and modern humans' but larger teeth than both cousins. Overall, Denisovans would have had a blocky and robust-looking appearance.
'As with the famous image of a Neanderthal dressed in modern attire, they would most likely still be recognizable as 'human,'' McRae said.
'They are still our more mysterious cousin, just slightly less so than before,' he added. 'There is still a lot of work to be done to figure out exactly who the Denisovans were and how they are related to us and other hominins.'