Clocking the Human Exodus Out of Africa
Like bloodhounds on the fading scent of an escaped convict, researchers have tried for decades to trace the ancient footsteps of the first modern humans who left Africa. Even though this exodus was one of the most important events in human evolution, scientists have been unable to pinpoint when and where it began. Now, using ancient DNA for the first time from ancient Europeans such as Ötzi, the famous Iceman, and from earlier fossils, a team of evolutionary geneticists has dated the start of this legendary journey to less than 95,000 years ago and, possibly, as recently as 62,000 years ago.
The findings fit with the evidence from fossils and stone tools, but they contradict a spate of recent genetic studies. Those studies determined the mutation rate for the entire genome from living humans by counting the number of new mutations that arise in the nuclear DNA of a newborn baby compared with its parents. The number of mutations per generation—the mutation rate—can be used to fine-tune a molecular clock that has long been used to date key events in human evolution. This supposedly better molecular clock has pushed back several important dates, such as when the ancestors of humans and chimps split and the exodus of modern humans out of Africa. Indeed, one team revised the date of the migration out of Africa from less than 80,000 years ago to at least 90,000 to 130,000 years ago.
Evolutionary geneticist Johannes Krause of the University of Tübingen in Germany, however, wasn't so sure that a mutation rate calibrated for living humans could be applied so far back in time. He and his colleagues decided to test the idea by sequencing the DNA from the maternally inherited mitochondria (mtDNA), or powerhouses of the cell, from fossils of modern humans who lived in the past 40,000 years and whose age was reliably known from calibrated radiocarbon dating methods. If the age of the fossil was 40,000 years, for example, it would be missing 40,000 years of evolution that took place in the lineage of a living person—and, therefore, missing mutations that would have arisen during the time since the fossil human died.
The team analyzed 10 well-dated fossils, including a medieval man who lived in France 700 years ago; the 4550-year-old Iceman; two 14,000-year-old skeletons from the tombs of Oberkassel in Germany; three related, modern humans from 31,000 years ago in Dolni Vestonice in the Czech Republic; and an early modern human from 40,000 years ago in Tianyuan, China. When the researchers applied this ancient DNA-derived mutation rate to the out-of-Africa migration, they got a range of dates from 62,000 to 95,000 years ago for the start of the migration, which is almost half the age of the migration out of Africa that was calculated using the "de novo" mutation rate, the group reports online today in Current Biology. "The nice thing about this is it was similar to the archaeological evidence," Krause says.
The team's method for checking the mutation rate is clever, says geneticist Aylwyn Scally of the Wellcome Trust Sanger Institute in Hinxton, U.K., co-author of one of the studies that calculated the slower mutation rate in living humans. "It's excellent that they have been able to get a better baseline for calibrating the mtDNA mutation rate by looking at ancient DNA."
However, Scally notes, mtDNA is a single genetic lineage, which is not typical of the genome, partly because the mutation rate of mtDNA could be higher because it has a higher proportion of genes under selection than the entire nuclear genome. Krause and one of his collaborators, paleogeneticist Svante Pääbo of the Max Planck Institute for Evolutionary Anthropology in Leipzig, Germany, agree that future work will be needed to resolve the differences in mutation rates in the mtDNA and nuclear genomes. "It is possible that there are things we do not understand about mitochondrial inheritance and mutation patterns," Pääbo says.
Or the problem may be with undercounting nuclear mutations in living humans. When it comes to precisely counting about 50 new mutations out of 3.2 billion bases in a newborn's genome, current sequencing methods are at the limit of their ability to filter out true mutations from mistaken ones and may be discounting a statistically significant number of actual mutations, Krause says. "The way forward is to truly master accurate sequencing of nuclear genomes," Pääbo says.
And that matters, Krause says, because a sense of timing is critical in human evolution. Knowing when modern humans spread out of Africa and into Europe and Asia, for example, allowed Krause and his collaborators to show that the same modern humans were in Europe before and after the glaciers covered that continent—and had the ability to adapt to changing climates. They found that modern humans before and after the last major ice age in Europe share the same mtDNA lineage, making them direct descendants of the same linage. "Out of Africa is one of the major events within human evolution," Krause says. "We need to know when it happened."