In 1975, Mary-Claire King and the late Allan Wilson, both then at the University of California, Berkeley, showed that the genetic distance between humans and chimpanzees is simply too small to account for the dramatic anatomical and behavioral differences between the two species.1 No matter what method scientists used to measure genetic distance--protein electrophoresis, DNA hybridization, immunology, or amino acid sequencing--the result was always the same: Humans and chimpanzees are 98.7% genetically similar.
"The molecular similarity between chimpanzees and humans is extraordinary because they differ far more than sibling species in anatomy and way of life," King and Wilson argued. "Is it possible, therefore, that species diversity results from molecular changes other than sequence differences in proteins?" In particular, they suggested that differences between humans and chimps are perhaps based, not on dissimilar gene sets, but rather on gene expression divergencies. The thought was, maybe the degree to which a gene is turned on or off is just as important, if not more so, than the gene's composition. Even a very small difference in the level or timing of gene expression could influence the phenotype in major ways.
Finally, 27 years later, a multinational team of scientists has provided evidence that supports King and Wilson's so-called regulatory hypothesis.2 Led by Svante Pääbo at the Max Planck Institute for Evolutionary Anthropology, Leipzig, researchers used DNA microarray technology to show pronounced differences in gene expression levels between humans and chimpanzees. "People have been thinking about this for some time," says molecular evolutionist Caro-Beth Stewart, State University of New York, Albany. "But this is the first time anyone has done anything but hand-waving. It really is a landmark piece of work."
LOTS OF FIRSTS Why was the wait so long? As molecular anthropologist Maryellen Ruvolo, Harvard University, explains, "As technology changes, we can ask different questions." In the 1960s, scientists were not even sequencing genes yet; they were just looking at proteins. They have come a long way since then. Researchers have already sequenced the entire human genome, and now they are talking about a comparable primate genome project. Moreover, investigators have taken genomics one step further by using DNA microarrays to measure the expression levels (that is, mRNA levels) of thousands of genes simultaneously, which is exactly what Pääbo and his collaborators did.
The researchers compared mRNA levels in the brains and livers of humans and chimpanzees, and used the data to construct distance trees representing the overall expression differences. "This is the first use of these DNA microarrays to look at human-chimp differences," says Ruvolo, "in particular, gene expression patterns. So it's the first time we're seeing gene expression differences." Pääbo and colleagues also assessed mRNA levels in the brains and livers of comparably related murine species to see whether the differences between humans and chimpanzees typified those between other mammals. With one notable exception, they did.
Image: Courtesy of Ajit Varki
Although gene expression levels in both the brain and liver were different between human and chimps, this difference was much more pronounced in the brain. This shows that the human brain has been evolving at a much faster rate than the liver. In contrast, the evolutionary rates of gene expression levels in murine brains and livers are the same. Pääbo's coauthor, Ajit Varki, professor of medicine and director of the Glycobiology Research and Training Center at the University of California, San Diego, says that he chose to include the liver in part because it is a large, easy-to-access organ, and that structurally and functionally speaking, it is generally conserved among mammals.
Varki says it is gratifying to have evidence that gene expression changes are especially pronounced in the brain, but it is not surprising. "It is well known that the human brain has shown one of the highest rates of evolutionary change of all organs," says human geneticist Luca Cavalli-Sforza, Stanford University. "It has increased about four times in the last five million years, with respect to what it probably was at the beginning, and what it is still in other primates."
NEW DATA, NEW QUESTIONS The real eye-opener in this study, according to Ruvolo, is the interindividual differences that Pääbo and his team reported. As the authors claim, "The variation in gene expression between individuals within the species is substantial, relative to the differences between humans and chimpanzee." The distance between one of the human samples and the others is greater than the overall distance between humans and chimpanzees, Ruvolo explains. "That's a very intriguing difference. It cries out that we need to do more experiments so that we can understand these differences among individuals. And it 'ain't' just in humans. You also see it in mice." Edwin McConkey, University of Colorado, agrees: "We really need a lot more information on the effect of extrinsic factors on brain gene expression in individuals."
That the gene expression data reveal so much variation within a single species is another example of the microarray's extraordinary potential to detect individual differences, Ruvolo says. But other scientists extend varying cautionary notes. Microarrays have already proven their power in individualized medicine, but Cavalli-Sforza forewarns, "It's a little premature to say how useful it can be in understanding human evolution." McConkey agrees that the technology will have a major eventual impact on evolutionary studies, but opines, "Interpretation of gene expression patterns, whether obtained via microarrays or otherwise, has to be correlated with knowledge about the genes themselves." Plainly speaking, researchers still need sequence data to make sense of all this.
It is helpful, says Varki, to "think of the genome as a recipe book and gene expression as the meal." The meal does not provide much detailed information about the ingredients, but with the recipe in hand, a researcher is in a more powerful position to decipher the underlying genetic complexities that explain how and why humankind is so different from its closest relative, the chimpanzee.
Microarrays have another capability, says Ruvolo: They could be used to look at the timing disparities of gene expression between chimps and humans. Even tiny timing differences in gene expression early in development could result in major phenotypic impacts further down the line. "This story is purely looking at adults--middle-aged humans and chimps," says Varki, in reference to the group's paper. "Obviously, there could be a lot of dramatic differences developmentally, postnatally. But getting the samples is very, very difficult." For several practical and ethical reasons, primate tissue samples, especially fetal tissue samples, are not easy to come by. Other scientists, like Kevin White, Yale University, are avoiding this problem by studying species that are much easier to work with. White uses microarrays to study gene expression timing in Drosophila.
A PLETHORA OF DIFFERENCES Varki, who finds it "bemusing" that the study received so much attention, says that gene expression differences are just one of many reported in the last decade. "The phenotypic differences between humans and chimpanzees are likely due to a combination of various genetic factors," he says. Others agree. For example, chimpanzees probably do not have as many active transposable elements as humans do, according to SUNY's Stewart. Moreover, Varki explains, the microarray approach complements the classical, candidate gene method, but it is not necessarily better; they are simply two different ways of looking for dissimilarities. "It's an interesting new kind of data and probably very rich in some ways," says population geneticist David Reich, of the Whitehead Institute for Biomedical Research, Cambridge, Mass. With microarrays, an investigator can pick out patterns and overall differences in gene expression and then try to discern the biological and phenotypic significances. In contrast, with the candidate gene approach, researchers pick out phenotypic differences and then try to discover the underlying genetic basis.
Others agree that gene expression data, as interesting as they are, beg the question: Which of these differences are functionally important in terms of making people human? Says Ruvolo, "That will be a whole different set of experiments. But how are we going to study that? You can't make knockout primates."
WHAT MATTERS The next step using the microarray approach is sorting through this long list of differently expressed genes and determining which ones are significant. A difference in gene expression does not necessarily imply dissimilarity in protein expression, Varki explains. Likewise, commonalities in gene expression do not necessarily preclude differences in protein expression. Varki studies one of the few known biochemical differences between humans and chimps; N-Glycolylneuraminic acid (Neu5Gc), a sialic acid, is active in chimpanzees but completely dormant in humans.3 Yet, its mRNA is still expressed in people. So, using microarray analysis, no differences in Neu5Gc expression would be seen between humans and chimps. But, in fact, some very important structural and implied functional differences exist. Sialic acids are the terminal molecules of the glycan (sugar) chains that coat the surfaces of all mammalian cells.
Perhaps Varki was bemused by the attention the team's work received, because the study, he says, could have been denigrated. "It would be very easy to criticize this study because it is the first reconnaissance mission just to see what's going on," says Varki. "'You could have done it differently,' or 'You could have had more samples.' You could go on and on with that sort of criticism." But he and his coauthors have been relatively spared, he says, probably because most people know how much hard work was involved. Maybe in 50 years people will look back and find fault in it, he says. "But right now, it is the best that can be done."
"The bottom line," says McConkey, is that Pääbo and colleagues "have offered a surprising set of data whose significance cannot yet be interpreted. But their results will surely stimulate more research on the same topic, and that is good."
Leslie Pray (firstname.lastname@example.org) is a freelance writer in Leverett, Mass.
1. M.C. King, A.C. Wilson, "Evolution at two levels in humans and chimpanzees," Science, 188:107-16, 1975.
2. W. Enard et al., "Intra- and interspecific variation in primate gene expression patterns," Science, 296:340-3, April 2002.
3. A. Varki, "Loss of N-glycolylneuraminic acid in humans: mechanisms, consequences, and implications for hominid evolution," Yearbook of Physical Anthropology, 44:54-69, 2001.