Genomic study of Africa's hunter-gatherers elucidates human variation and ancient interbreeding

That history of inheritance is written in the DNA of modern Africans, but it takes some investigative work to interpret. In a report to be featured on the cover of the Aug. 3 issue of the journal Cell, University of Pennsylvania geneticists and their colleagues analyze the fully sequenced genomes of 15 Africans belonging to three different hunter-gatherer groups and decipher some of what these genetic codes have to say about human diversity and evolution.

The study, led by Sarah Tishkoff, a Penn Integrates Knowledge Professor with appointments in the School of Arts and Sciences' biology department and the Perelman School of Medicine's genetics department, tells several stories.

It identifies several million previously unknown genetic mutations in humans. It finds evidence that the direct ancestors of modern humans may have interbred with members of an unknown ancestral group of hominins. It suggests that different groups evolved distinctly in order to reap nutrition from local foods and defend against infectious disease. And it identifies new candidate genes that likely play a major role in making Pygmies short in stature.

"Our analysis sheds light on human evolution, because the individuals we sampled are descended from groups that may have been ancestral to all other modern humans," Tishkoff said. "A message we're seeing is that even though all the individuals we sampled are hunter-gatherers, natural selection has acted differently in these different groups."

Joining Tishkoff in the work from Penn was first author Joseph Lachance as well as Clara Elbers, Bart Ferwerda and Timothy Rebbeck. Their collaborators include Benjamin Vernot, Wenqing Fu and Joshua Akey of the University of Washington; Alain Froment of France's Musée de L'Homme; Jean-Marie Bodo of Cameroon's Ministère de la Recherche Scientifique et de l'Innovation; Godfrey Lema and Thomas B. Nyambo of Tanzania's Muhimbili University College of Health Sciences; and Kun Zhang of the University of California at San Diego.

The researchers sequenced the genomes of five men from each of three hunter-gatherer groups: the Hadza and the Sandawe of Tanzania and the Western Pygmies of Cameroon. The three differ greatly from one another in appearance, in language, in the environments they occupy and in cultural practices, though the Hadza and the Sandawe live just 200 kilometers apart.

"We purposefully picked three of the most diverse hunter-gatherer groups," Tishkoff said, "because they have not been very well represented in other genome sequencing projects, which tend to focus on majority populations in Africa. This is a unique and important dataset."

The researchers used a method that involves sequencing each strand of DNA more than 60 times on average. This redundancy makes the sequencing highly accurate, giving the geneticists confidence that any mutations they identify are real and not errors.

Scanning these sequences, the researchers found 13.4 million genetic variants, or locations in the genome where a single nucleotide differed from other human sequences; at the time they were discovered, 5 million were new to science.

"It was awe-inspiring," Lachance, a postdoctoral researcher, said, "to find millions of new variants that we never knew existed in our species. It's humbling but invigorating to think about how to make sense of all this diversity."

Only about 72,000 of these variants were in regions of the DNA that code for genes. The rest were in non-coding regions, which may influence how and whether genes are expressed.

"Our study underscores the importance of noncoding regions of the genome, particularly for regulating gene expression," Tishkoff said. "That has important implications for anyone doing biomedical research because, if they're only looking at the coding regions, they're missing information that may be critically important for normal human variation as well as disease susceptibility."

The researchers also used the genomes to investigate ancient human interbreeding.

Using a statistical method, the team detected partial sequences in all three populations that appear to have derived from a hominin different from Homo sapiens. Much as recent studies have found evidence that modern humans interbred with Neanderthals, these new findings suggest that the ancestors of modern humans in Africa mated with individuals from another hominin lineage. This archaic lineage appears to have diverged from the modern human lineage several hundred thousand years ago, around the same time that Neanderthals diverged from Homo sapiens.

"Fossils degrade fast in Africa so we don't have a reference genome for this ancestral lineage," Akey said, "but one of the things we're thinking is it could have been a sibling species to Neanderthals."

Evidence of interbreeding with an archaic lineage, known as introgression, was found in all three groups tested.

"Given that introgression is present in these very diverse groups," Vernot said, "I think we can now say that this seems to be a pretty universal aspect of human history."

In an additional analysis, the researchers looked for major differences among the genomes of the three groups, signs that the populations evolved differently to adapt to their specific environments. Among the variations that stood out were genes related to immune-system function, smell and taste. These findings suggest that adaptations to local diseases and local foods were important in how the groups evolved.

Finally, building on previous research from Tishkoff and colleagues that sought an explanation for why Pygmies are all short -- men standing only 4'11" on average -- the researchers looked for variants that were common in the Pygmies of Cameroon but rare or absent in the other groups. They found several genes located near these mutations that act on the pituitary gland, a "master regulator" responsible for metabolism, growth, sexual development and immunity.

"We would have never found these strong candidate genes for short stature," Lachance said, "if we hadn't looked at multiple genomic sequences from these isolated populations. It hints at the power of what you can do when you sample multiple genomes and compare them at a population scale."

Tishkoff's group plans to sequence the genomes from more Africans -- a task that new technology has made faster and cheaper than ever before -- to increase their sample size. With more genomic data, scientists will better their understanding of how evolution acted on humans throughout the last several hundred thousand years and even how certain mutations may predispose certain people to disease.

"Our study emphasizes the critically important role of next-generation genome sequencing for elucidating the genetic basis of both normal variable traits in humans as well as identifying the genetic basis of human disease risk," Tishkoff said.