Neanderthal had flings with relatives, other human ancestors

• DNA analysis shows three kinds of humans had relations leading to mixed heritage
• It was not uncommon for Neanderthals to mate with relatives as close as half-siblings
• Study paints complicated picture of connections in the human family tree

Maybe those long-ago couples exchanged glances across a campfire. Maybe they had brief encounters, or long, tranquil cohabitations, or stormy on-again, off-again relationships.

However it happened, a new DNA analysis shows that the modern and archaic humans living in the Late Pleistocene did more than flirt. The analysis reveals that at least three kinds of humans – Neanderthals, the enigmatic Denisovans and our own species – had amorous trysts leading to children of highly mixed evolutionary heritage. The research also hints that a mysterious fourth kind of human was playing the prehistoric field and suggests that some Neanderthals mated with relatives as close as half-siblings.

"What we were struck by most was just how complicated it is and how much interbreeding there was among all these human relatives," says study co-author Montgomery Slatkin, a population geneticist at the University of California, Berkeley. "Every new fossil … shows more evidence of more kinds of interbreeding."

The study paints a picture of "complex migrations … and interrelationships amongst these archaic people, much more so than we ever imagined," says human geneticist Sarah Tishkoff of the University of Pennsylvania, who was not involved with the new research. "There may be more out there that we're not even aware of."

This steamy story begins with a toe bone discovered in 2010 in Denisova Cave in the Altai Mountains of Siberia. The scientists extracted DNA from the bone, which is at least 50,000 years old, and found it belonged to an adult female Neanderthal. Her short, stocky species lived side-by-side with our own Homo sapiens species for millennia before disappearing some 30,000 years ago.

The Neanderthal woman's mother and father probably knew each other well — very, very well. The exact relationship is unclear, but according to the DNA, the woman's parents could've been grandparent and grandchild, uncle and niece or even half-siblings. The woman's DNA also testifies to past couplings between relatives, showing the woman's parentage was not an isolated incident, the team reports in this week's Nature.

Such interaction is taboo in many societies today, but Neanderthals would've had a hard time finding a date. It's thought the Neanderthal population was extremely small to begin with, and climate fluctuations probably isolated bands of Neanderthals from each other, paleoanthropologist Katerina Harvati of the University of Tuebingen in Germany, who was not involved with the study, says via e-mail.

Under such conditions and "in the absence of other candidates, mates may be selected from close kin," Harvati says. The small population may also explain another finding of the DNA analysis: Neanderthals show some of the lowest genetic diversity of any life form studied. That's presumably because there weren't many of them, leading to significant inbreeding.

But a few Neanderthals did look further afield for amore. The study confirms that our own species, Homo sapiens, occasionally had assignations with both Neanderthals and Denisovans, another form of early human known from only a few tiny bits of skeleton in Denisova Cave. The new results show that Denisovans and Neanderthals interbred, as well.

Most striking of all, the analysis found that Denisovans mated every so often with an archaic human that had arisen a million or more years earlier. The most likely candidate, the researchers say, is Homo erectus, an early human that began to spread out of Africa roughly 2 million years ago – many hundreds of thousands of years before Denisovans began to emerge as a separate branch of the human family tree.

The identification of Homo erectus as the Denisovans' mystery partners "makes sense," Harvati says, but the team behind the new findings also concedes there are other possible explanations for the DNA patterns they saw, rather than interbreeding between Denisovans and an ancient human. So far Tishkoff is not totally convinced.

Mixing between Denisovans and a more ancient species "is just one of many, many possible scenarios that could account for the present data," she says. "We need to get a lot more data."

The authors of the new study say they're reasonably confident their results are correct. Besides, with clear evidence for so much mixing between species, "it doesn't seem so strange to me that there would be one more mixing event," says study co-author Svante Pääbo, a paleogeneticist at the Max Planck Institute for Evolutionary Anthropology in Germany. "I have a suspicion that almost always, when groups meet, these things do happen."