Human-chimp Gene Study Upsets Long-held View

But new research at the University of Michigan challenges that human-centered view. "We often think that we're unique and superior to other species, so there must be a lot of Darwinian selection behind our origin," said Jianzhi (George) Zhang, associate professor of ecology and evolutionary biology. "However, we found that more genes have undergone positive selection in chimpanzee evolution than in human evolution."

The U-M work is not the first attempt to examine positive selection (natural selection that promotes the fixation of advantageous mutations) in humans and chimps. However, earlier efforts focused on identifying specific genes under positive selection, not on comparing their numbers, Zhang said. In the current analysis of nearly 14,000 genes, Zhang's group not only looked at numbers of genes, but they also made three key improvements over previous approaches.

The first improvement allowed the U-M researchers to more accurately determine which human-chimp differences were due to genetic changes in the human lineage and which were due to changes in the chimp lineage. This was possible because they compared both groups to the recently-sequenced macaque monkey genome, whereas other researchers had to rely on the mouse genome for comparison.

"If we only compare human and chimp, we can see differences, but we can't tell whether a particular difference is due to a change in the human or a change in the chimp," Zhang said. "But if we compare both to another species—say the monkey—and if chimp and monkey are identical for a particular trait such as brain size, but human is different, then we can infer that

something must have changed during human evolution." Using a more closely related comparison species—monkey instead of mouse—makes the inference more reliable.

The second improvement was using a different statistical method than previous analyses employed, one that is less likely to indicate that positive selection has occurred when it has not.

Third and most important, the U-M researchers took steps to assure the "quality" of the gene sequences they analyzed.

"Sequence quality is an indication of how close the obtained sequence is to the true sequence," Zhang said. The human genome published in 2004 was a high-quality, "finished" version, whereas the chimp genome published in 2005 was a lower-quality "draft" version. However, each position in the chimp sequence was assigned a quality score, and the Zhang team used only high-quality portions of the genome sequence in their analysis.

The finding that chimps had substantially more positively selected genes than did humans surprised Zhang, but he has a possible explanation. There is evidence that over the past one to two million years, human populations have gone extinct in certain areas, only to be replaced through recolonization. Such a pattern makes for a smaller "effective population size," a term that refers to the number of individuals contributing to the next generation. According to population genetics theory, selection is more effective in large populations than in small populations, so the lower number of positively-selected genes in humans may be a reflection of humans' smaller long-term effective population size, Zhang believes.

In addition to tallying positively selected genes, Zhang's group also looked at which genes in humans and chimps were under positive selection. Again, the results were a surprise. It's been suggested that genes expressed in the brain underwent rapid evolution by positive selection in humans. "But we didn't see that," Zhang said. In fact, the researchers found no discernable trends in where within the body positively selected genes were expressed.

That finding doesn't negate the role of positive selection in human brain development, Zhang noted. "I believe that human brain evolution is due to changes in a small number of genes, not large numbers, and that is why we do not see a genome-wide signal."

One long-held tenet did stand up to scrutiny. The "thrifty gene hypothesis," proposed by U-M human genetics pioneer James Neel in 1962, postulates that certain genes in humans were advantageous in the evolutionary past when food was scarce, but have become disadvantageous in times of plenty, predisposing their carriers to diseases such as diabetes and obesity. Zhang's team tested this hypothesis by checking to see whether genes under positive selection during human evolution are more likely to be associated with disease. "We did find some evidence of that," he said.

Zhang's coworkers on the project were doctoral student Margaret Bakewell and postdoctoral fellow Peng Shi. The researchers received funding from the University of Michigan and the National Institutes of Health.

The researchers' results were published in the Proceedings of the National Academy of Sciences the week of April 16-20.