Out of Africa

Evy Potochny
January 01, 2001

Modern humans are thought to have originated in Africa. From there bands of hominids migrated first to the Middle East, then throughout Europe and into Asia.

But exactly who moved away? A single population of already-evolved Homo sapiens? Or did several groups of more primitive humans migrate separately, then evolve independently into the modern variety?

sunset in Kenya

Evolutionary geneticists struggle with this question, scrutinizing DNA samples from around the world for tell-tale variations. Until recently, they have relied heavily on mitochondrial DNA (mtDNA). Now, new studies using nuclear DNA are changing the debate.

Mitochondrial DNA is found outside the cell nucleus in the organelles that produce a cell's energy. MtDNA is useful to geneticists, explains Sarah Tishkoff, because it is plentiful (hundreds of copies of the mitochondrial genome exist in each cell), it does not recombine (portions of the mother's DNA are not exchanged with the father's), and it mutates quickly (so there is a lot of genetic variation to compare).

"But mtDNA only tells us half the story," adds Tishkoff, who did postdoctoral research in genetics at Penn State. Only the mother passes on mtDNA to her progeny; the father's contribution is lost.

The amount of genetic material in the nucleus is immense compared to what is in the mitochondria: some 80,000 genes versus only a few. And each gene can exist in several versions, or alleles. That is, there can be subtle changes in the sequence of A,C,T, and G, the four bases that make up DNA, without changing the gene's function. For instance, a two-base sequence like TG might be repeated five times in a row (TGTGTGTGTG)—or six times, or four—without affecting the gene's function. These "short tandem repeats" tend to mutate a lot. But that's good: mutations are useful for comparing populations over time. TG repeated five times would be considered one allele, while TG repeated six times would be another allele. Tishkoff also looks at alleles caused by less frequent types of mutations—alterations by insertion or deletion of a DNA section several hundred bases long.

For one study, Tishkoff selected three human genes: CD4, which produces a cell-surface protein that enables HIV to enter and infect certain immune cells; DM, which causes myotonic dystrophy, a neuromuscular disease; and PLAT, short for tissue plasminogen activator locus, a gene involved in tissue remodeling and destruction.

Tishkoff compared these genes in DNA samples donated by collaborators from 45 different populations worldwide, including Europe, the Pacific islands, Africa, Asia, and the Middle East— making hers one of the largest data sets on human nuclear variation.

She found that while non-African populations were relatively similar genetically, the variations among African populations differed widely. In the CD4 gene, for instance, Tishkoff found only three major variants in populations outside of Africa. Among African samples, it was common to have 24 variations within a single population.

This lack of genetic diversity in non-Africans suggests that they are more closely related than the African populations, and that their differences evolved over a much shorter period of time. "The only variants that made it out of Africa," says Tishkoff, "have both a characteristic deletion and a repeat of six on the chromosome with the CD4 gene."

By calculating how much time it would have taken for these and the other mutations to accumulate, Tishkoff estimates the migration out of Africa occurred approximately 130,000 years ago, rather than over 300,000 years ago, as was previously thought. "In the non-African populations," she explains, "there's only been enough time for a few shuffled sets of genes to arise."

Taken together, Tishkoff's results provide strong new evidence that modern humans descended fairly recently from a single ancestral population, one that was already fully modern when it left its African home.

Sarah Tishkoff, Ph.D., completed her post-doctoral research fellowship in genetics and is currently an assistant professor of biology at the University of Maryland. Her adviser was Andrew Clark, Ph.D., professor of biology, the Eberly College of Science, 208 Mueller Bldg., University Park, PA 16802; 814-863-3891; c92@psu.edu. Kenneth Kidd of Yale University and Trefor Jenkins of the University of the Witwatersand, South Africa, collaborated on this study, which was funded by the National Science Foundation and a Burroughs-Wellcome Fund Career Award.

Last Updated January 01, 2001