Ancient DNA reveals secrets of human history

For a field that relies on fossils that have lain undisturbed for tens of thousands of years, ancient human genomics is moving at breakneck speed. Barely a year after the publication of the genomes of Neanderthals and of an extinct human population from Siberia, scientists are racing to apply the work to answer questions about human evolution and history that would have been unfathomable just a few years ago.

The past months have seen a swathe of discoveries, from details about when Neanderthals and humans interbred, to the important disease-fighting genes that humans now have as a result of those trysts.

Neanderthals were large-bodied hunter-gatherers, named after the German valley where their bones were first discovered, who roamed Europe and parts of Asia from 400,000 years ago until about 30,000 years ago. The Neanderthal genome — shepherded by Svante Pääbo at the Max Planck Institute for Evolutionary Anthropology in Leipzig, Germany — indicates that their evolutionary story began to split from the lineage of modern humans less than half a million years ago, when their common ancestor lived in Africa. In December last year, Pääbo's team released the genetic blueprint of another population of ancient humans — unlike ourselves or the Neanderthals — that was based on DNA recovered from a 30,000–50,000-year-old finger bone found in a cave in Denisova in southern Siberia². Palaeoanthropologists call these groups archaic humans, distinguishing them from modern Homo sapiens, which emerged in Africa only around 200,000 years ago.

Pääbo is amazed at how quickly the Neanderthal genome has been mined. At a genomics meeting last year, for example, Cory McLean, a graduate student at Stanford University in California, was scheduled to talk immediately after Pääbo presented the Neanderthal genome. Inspired, McLean had trawled through the just-released genome in the days before his talk. He discovered that Neanderthals, like humans, lacked a stretch of DNA that orchestrates the growth of spines on the penises of other primates, and promptly presented the find just after Pääbo presented his.

Since then, scientists have fleshed out the details of one of the biggest surprises from the Neanderthal genome: humans living outside Africa owe up to 4% of their DNA to Neanderthals. One explanation might be that humans migrating out of Africa mated with Neanderthals, probably resident in the Middle East, before their offspring fanned out across Europe and Asia.

By comparing individual DNA letters in multiple modern human genomes with those in the Neanderthal genome, the date of that interbreeding has now been pinned down to 65,000–90,000 years ago. Montgomery Slatkin and Anna-Sapfo Malaspinas, theoretical geneticists from the University of California, Berkeley, presented the finding at the Society for Molecular Biology and Evolution meeting in Kyoto, Japan, held on 26–30 July.

Slatkin says that their result agrees with another study presented at the meeting that came from the group of David Reich, a geneticist at Harvard Medical School in Boston, Massachusetts, who was involved in sequencing both the Neanderthal and Denisova genomes. The dates also mesh with archaeological finds bookending early human migrations out of Africa to between about 50,000 and 100,000 years ago. Reich's team is now developing tools to find signs of more recent interbreeding that might have occurred after humans arrived in Asia and Europe.

The denizens of Denisova also bred with contemporary humans, according to Pääbo and Reich's analysis². But the only traces of their DNA to be found in modern humans were in residents of Melanesia, thousands of miles away from Denisova, suggesting that the Denisovans had once lived across Asia. In 2008, Pääbo's team set up a lab in Beijing to screen fossils that might contain Denisovan DNA, in the hope of learning more about them and their interactions with modern humans. Currently, the bone that yielded the Denisovan genome, and a single molar from the same cave, are their only known fossil remains, but other archaic human fossils from Asia could bear traces of this group.

Even before the Neanderthal genome made its debut in May 2010, scientists had argued that humans may have acquired not just DNA from archaic humans, but useful traits too. Human gene variants linked to brain development and speech were proposed as candidates, only to be scotched after closer inspection of the Neanderthal genome. However, a study presented at a Royal Society symposium in London in June suggests that humans owe important disease-fighting genes to Neanderthals and Denisovans. Interbreeding endowed humans with a 'hybrid vigour' that helped them colonize the world, said Peter Parham, an immunogeneticist at Stanford University School of Medicine, California, at the symposium.

Parham's team compared a group of diverse immune genes — the human leukocyte antigen (HLA) genes — in Neanderthals, Denisovans and human groups from around the world. In several cases, Neanderthals and Denisovans carried versions of HLA genes that are abundant in modern humans in parts of Europe and Asia, but less common in Africans. Varying degrees of interbreeding could explain the mismatch, Parham says. He estimates that Europeans owe 50% of variants of one class of HLA gene to interbreeding, Asians 70–80%, and Papua New Guineans up to 95%.

"It does mean that some of us owe part of our immune-system function to Neanderthals," says Pääbo. However, John Hawks, a biological anthropologist at the University of Wisconsin-Madison, notes that many HLA genes pre-date humans' split from Neanderthals and Denisovans, and that the differences may have arisen by chance as the groups evolved.

Hawks, too, has been digging into the archaic genomes, and his team has already discovered that Neanderthals and Denisovans lack certain forms of genes that may help modern humans to fend off epidemic diseases, such as measles. This is hardly surprising: the low population density of hunter-gatherers meant that epidemics were unlikely, so they probably would not have benefited from these immune genes.

But Hawks's team is now using the find to test whether the defensive genes are linked to autoimmune diseases. In September, Hawks and his colleague Aaron Sams are scheduled to present data at a meeting of the European Society for the Study of Human Evolution in Leipzig, Germany, showing that the Denisovans lacked nearly all of the gene variants linked to coeliac disease, a gut autoimmune disorder present in modern humans. Hawks suspects that the variants may actually be in the same genes that are linked to epidemic resistance — if they are, further study could reveal how recently such autoimmune diseases arose in humans.

Unlike most scientists mining the ancient genomes, Hawks has reported some of his more prosaic findings — Denisovans didn't have red hair, for example — on his blog.  "These genomes are publicly available. There's nothing stopping high-school students from doing this, and the kind of stuff that I'm putting out on my blog is the stuff that a smart high-school student could do." More significant (and closely guarded) insights will come from developing new methods for analysing ancient genomes to test hypotheses about evolution, he says.

Pääbo, Reich and the other scientists involved in sequencing the ancient genomes are eager to see others run with their data, but caution that they need to be aware of the limitations. "They're really terrible-quality genomes", chock-full of gaps and errors and sections in which short stretches of DNA sequence have been put in the wrong place, says Reich. "There are a lot of traps in using these data, and if people are not careful they'll find all sorts of interesting things that are wrong." Pääbo's team is working on improving the quality of the sequences and including data from more Neanderthals and — he hopes — Denisovans.

Pääbo says that he and his team regularly receive e-mails from scientists asking them questions about using the ancient genomes, which they have attempted to make as user-friendly as possible. But if the first year of ancient human genomics is any indication, these requests will multiply as scientists find new applications for the genomes. "Maybe we should write a little booklet called archaic genomics for dummies," Pääbo says.

Australopithecus Sediba could be direct ancestor of Homo

Last year Lee Berger from the University of the Witwatersrand and his team discovered the skeletal remains of two specimens they determined to be a new species of human called Australopithecus sediba. The skeletons had characteristics of previous species of Australopithecus, but also of Homo, leading the researchers to believe they may have found an evolutionary connection between the two. This became a very controversial idea, with many believing there was no connection to Homo and that what they had discovered was really an ancestor of later Homo species.

At the annual meeting of the Paleoanthropology Society on April 12 and again on April 16 at the annual meeting of the American Association of Physical Anthropologists, Berger and his team presented new findings on their most recent bone analysis.

Kristian J. Carlson discussed the size and shape of A. sediba's brain, showing that by synchrotron scanning of the interior brain case, they were able to determine the estimated capacity to be around 420 cubic centimeters. This led to a very small brain size and is the reason researchers first determined these new skeletal findings to be in the Australopithecus genus. However, they also discovered that the frontal lobe of this small brain contained organization more similar to that of humans, showing that contrary to what was previously thought, organization and brain size with human characteristics may not have been a simultaneous change.

The pelvis of the A. sediba is what researchers believe show the strongest link toward the beginning of an evolutionary change to the Homo. Researchers have always linked the larger brain size of the Homo to the evolutionary change in the pelvic structure between the two. However, even with the small brain size and cranial structure of A. sediba, the pelvic structure has changed from previous Australopithecus to much closer to that of Homo.

Researchers consider ancestry of recent fossil finds

Someday a future intelligent organism could sweep away a million years of dust and find the bones of a Homo sapiens and wonder what he was.

Further research would show Homo sapiens walked upright, lived in communities and buried their dead. But this future intelligent organism might be faced with an old puzzle--determining where Homo sapiens came from.

"If their cognitive world induced them to ask the same sort of questions, the problems we pointed up would still be there," said Bernard Wood, professor of Human Origins and of Human Evolutionary Anatomy at George Washington University in Washington DC.

He was speaking of a recent paper published in the journal Nature in which he and physical anthropologist Terry Harrison of New York University argued it's not so easy to determine whether relatively new fossil finds are early members of the human evolutionary family or prehistoric apes.

They write, "All the organisms alive today are the terminal twigs of the crown of the tree of life." Though not mentioning it directly, they hint that one day even Homo sapiens--anatomically modern humans--might disappear thus leaving some future intelligent organism to determine Homo sapiens place in the tree of life. But not only that, such disappearance might leave these future organisms with the same species classification problems that exist today.

The paper, "The Evolutionary Context of the First Hominins" is a thought experiment. Its authors think through the various consequences of placing recently discovered fossils Orrorin tugenensis, Sahelanthropus tchadensis and Ardipithecus ramidus so firmly on the human branch of the evolutionary tree.

The fossils may not be related to Homo sapiens at all, despite their discoverers' claims. Instead, they may be extinct distant cousins, with a relationship to modern humans much like that of living chimpanzees, gorillas or orangutans, say Wood and Harrison.

"Our null hypothesis is that the more morphology that is shared, the closer the relationship between the organisms," said Wood explaining the default position generally accepted by scientists when classifying species.

Evolutionary biologists and physical anthropologists traditionally use a species' physical characteristics--or morphology--to determine which groups of organisms are most closely related to one another. But, in some cases, common physical traits may not indicate a common evolutionary history.

The central theme here is something called homoplasy, a character shared by a set of species but not present in a common ancestor. A good example is the evolution of the eye which originated independently in many different species.

Sometimes these homoplasies and/or common morphologies may mean common ancestry and sometimes they may not.

Wood cautions that if morphology or homoplasy can be shared for reasons other than a recent common ancestor, such as common environmental pressures, then anthropologists are stumped.

"Supposing two distantly-related fruit-eating apes were living at the same time and facing the consequences of the same climate change," Wood explained. "Then suppose this climate change resulted in the forest's fruiting trees being replaced by open woodland and increased grassland, forcing the distantly-related apes to adapt to eating tough tubers instead of soft fruit.

"They both would undergo an increase in the size of their chewing teeth and an increase in the size of their mandibles, but they would not have inherited this morphology from a shared recent ancestor."

Instead they would have gotten it from a common environmental event and convergent evolution, developing the same or similar biological traits in unrelated or distantly related lineages.

In the case of Ardipithecus and other recently discovered fossils, traits shared in common with Homo sapiens such as bipedal locomotion or small canine teeth may indicate a close relationship with anatomically modern humans or, as Wood and Harrison suggest, may be the result of convergent evolution.

"Homoplasy makes it unwise to rely on just a few morphological similarities to declare fossils are early hominins," Wood contends.

Wood and Harrison underscore their concern with two cautionary tales of errant early human ancestor classification.

In an example from more than 30 years ago, scientists classified 12-million-year-old Asian fossils named Ramapithecus punjabicus as a likely human ancestor. Now, after further research, Ramapithecus is thought to belong to one or more species of extinct primates, and perhaps most accurately regarded as a distant cousin of modern orangutans.

In another example, a species called Oreopithecus bambolii, represented by a cache of 7 to 8 million-year-old fossils found at sites in Tuscany and Sardinia, Italy and first described nearly 140 years ago, was thought to be a potential early human ancestor because of skeletal similarities with modern humans. Research suggested Oreopithecus had many anatomical similarities with the group of organisms that includes great apes and modern humans. It also had characteristics generally considered to be uniquely associated with bipedal behavior.

But only after additional discoveries by a Swiss paleontologist in the late 1950s, researchers began to rethink Oreopithecus' initial classification. Scientists still debate its precise position on the evolutionary tree and whether it is more accurately regarded as a descendant of a European ape or as an African anthropoid.

"We emphasize that we are not claiming that the presence of homoplasy in and around the hominin clad ... doom(s) all efforts to recover evolutionary relationships to failure," the researchers write. They take the same stance regarding other methodological and analytical limitations that arise when investigating evolutionary relatedness.

They stress that their paper doesn't claim Orrorin, Sahelanthropus and Ardipithecus are not early human ancestors. But they encourage paleoanthropologists to acknowledge the potential shortcomings of their data when it comes to generating hypotheses about relationships.

"We urge researchers, teachers and students to consider the published phylogenetic interpretations of these taxa as among a number of possible interpretations of the evidence," they write.

For now, they hope paleoanthropologists will heed the warning and leave an accurate record that perhaps some future intelligent organism will be able to decipher.

The National Science Foundation funds Wood's and Harrison's research through its Division of Behavioral and Cognitive Sciences.

Provided by the National Science Foundation