Natural History, Dec 2002 - January 2003
In the past decade, as molecular biologists have learned to read DNA sequences rapidly, the chimpanzee has clearly emerged as humanity's closest living relative. Our DNA is astonishingly similar. You can see for yourself by visiting the "Silver Project" Web site of Japan's National Institute of Genetics, which is home to a growing database of chimpanzee DNA. With a couple of clicks you can compare the sequence of DNA nucleotides for a particular chimpanzee gene-molecular fragments whose identity is given by one of the four "letters" in the DNA "alphabet"-with the sequence in the corresponding human gene.

Say, for instance, you decide to look at a gene known as CHRM2, which codes for a kind of receptor on nerve cells. Stretched out alongside the human sequence is the sequence of nucleotides for the same gene in the chimpanzee. If you read the letters in the sequence-A, C, G, and T, shuffled and repeated line after line-you'll find that the two sequences are identical for hundreds of nucleotides at a stretch. Only here and there are they punctuated by a rare difference. Such differences would have evolved sometime after the chimpanzee and human branches split from their common ancestor, and one lineage or the other (or perhaps, even more rarely, both) picked up a mutation.

Molecular biologists still don't know exactly how many such differences exist between the human and chimpanzee genomes. In spite of the hoopla in June 2000 over the "essentially complete" sequencing of the human genome, enough mopping up still remained that the sequence will be truly complete only as of this coming April. As for the chimpanzee genome, only isolated fragments have been determined-though it is a top priority for the next full sequencing of an organism's genome. So far, the biggest published surveys extrapolate from less than 1 percent of the two species' DNA. Nevertheless, most of the surveys have converged on the conclusion that the two genomes are about 98.7 percent identical.

What is one to make of this? If humans and chimpanzees are really so similar genetically, why have they turned out to be so different "in the wild"? Is genetic similarity a mismeasure of the species? Is there something else, biologically, that accounts for our obvious differences? Or are chimpanzees and human beings genuinely the close cousins that the genetic counts suggest, making the differences we notice at the level of the organism little more than skin deep? The answers to these questions are still up for grabs, but however they come out, they promise to echo far beyond the halls of genetics and on to the centuries-old debate about what is unique and what is not about Homo sapiens.

Chimpanzees have been disconcerting Western scientists since at least the 1690s, when the English anatomist Edward Tyson became the first to dissect one. He noted that the chimp and the human brain bore a "surprising" resemblance, writing that "one would be apt to think, that since there is so great a disparity between the Soul of a Man, and a Brute, the Organ likewise in which 'tis placed should be very different too."

In the nineteenth century, the English anatomist Richard Owen begged to differ. The contrast between the brain of the chimpanzee and that of a person, he maintained, was actually quite sharp; no human could ever be thought of as merely a modified chimpanzee. Owen was a fierce opponent of Darwin's evolutionary ideas, as well as of those espoused by the French naturalist Jean Baptiste Lamarck, one of Darwin's intellectual predecessors. He looked intensively for something unique in the human brain that would set us off from other primates. The only thing he found was a small fold in the back of the brain, which he could not identify in any ape. But that was enough for Owen-he made it the seat of human reason and, on that sole basis, assigned humanity to a class of its own, which he dubbed Archencephala, or "ruling brain."

When Darwin heard about Owen's claims, he wondered, in the shorthand of a notebook, "what a Chimpanzee wd say to this." But Darwin left it to the biologist Thomas Henry Huxley, Owen's nemesis and a tireless defender of Darwinism, to publicly demolish the idea of Archencephala. (Not one to mince words, Huxley privately declared that Owen had built his classification like "a Corinthian portico in cow dung.") Huxley maintained that Owen had blatantly overlooked the presence of Owen's own "fold of humanity" in the chimpanzee brain. In fact, Huxley argued, a human differs much less from an ape, such as a chimpanzee or gorilla, than an ape does from a baboon.

Huxley had essentially one source of information on the evolution of people and apes: their living anatomy. Neanderthal fossils first attracted scientific attention in the 1860s, but it wasn't until after Huxley's death in 1895 that a steady stream of hominid fossils emerged. In contrast, today's paleoanthropologists count some twenty precursor species that were early relatives or direct ancestors of modern people. In fact, as recently as July 2001, investigators in Chad began to find fossil fragments of the oldest hominid precursor yet known, the six- to seven-million-year-old Sahelanthropus tchadensis. Many of these species display a mix of both humanlike and chimplike features, blurring the distinction between the two.

Huxley also had no way to compare the biochemistry, much less the genes, of people and chimpanzees. The first glimmerings of our biochemical bond with chimpanzees came at the dawn of the twentieth century. Biochemists figured out how to make antibodies tailored to recognize proteins that occur in human blood. They then tested the antibodies on proteins from other animals. It turned out that the antibodies for human proteins formed bonds with chimpanzee and gorilla proteins far more readily than with the proteins of other animals. That suggested that chimpanzee and gorilla proteins had a shape similar to that of human proteins. In the 1960s Morris Goodman, a geneticist at Wayne State University in Detroit, and his coworkers learned how to compare the building blocks of the proteins in humans and apes. They found that many proteins in people and chimpanzees appeared to be nearly identical.

By the 1980s scientists were able to switch their gaze from proteins to the genes that encode them. (Each gene serves as a template for making what can become many copies of a single kind of protein.) At first they could only make simple comparisons between chimpanzee and human DNA. They unzipped the twin strands of a human DNA fragment, then zipped each one back up with a strand of DNA from another animal. Workers measured the strength of the bonds by measuring the amount of heat needed to make the hybrid DNA fall apart. (The weakest bonds began to let go at about 60 degrees Celsius; the strongest ones held together at 90 degrees.) They reasoned that the more closely related the two species, the more stable the bonds between two strands of hybrid DNA. The experiments supported the idea that apes and people have highly similar DNA. And that general conclusion has been borne out by the revolution in DNA sequencing of the past decade.

But just how similar is "highly similar"? This past October Roy J. Britten, a geneticist at the California Institute of Technology, published a study asserting that estimates in the range of 98.7 percent are too high. Too much attention, he argues, has focused on mutations that change a single nucleotide in the genome. DNA also mutates when an entire stretch of the molecule gets copied and inserted somewhere else in the genome, or is simply deleted altogether. When Britten searched for these additional kinds of mutations, he concluded that the human and chimpanzee genomes are 95 percent identical, nearly quadrupling the previously estimated difference of 1.3 percent.

Britten has discovered a real shortcoming in previous estimates, but other biologists may still be reluctant to start adopting his figure. Most of the insertions and deletions Britten studied occur in long stretches of so-called junk DNA, which includes no functioning genes. The parts of the human genome that actually carry codes for our body's proteins are much more similar to the genes of chimpanzees. Britten's work also shows how tricky it is to describe our similarity to chimpanzees with a single number. Suppose a stretch of our DNA 6,000 base pairs long disappeared a million years ago. Britten would count that as 6,000 separate changes, yet other geneticists would count it as a single evolutionary event.

Even if ten years from now biologists can state the exact percentage of DNA we humans share with chimpanzees, the number by itself won't really mean very much. Our DNA most resembles that of chimpanzees, but we share a lot of DNA with other animals as well. Geneticists recently surveyed a chromosome in mice, looking for genes related to the ones in the human genome. Of the 731 genes they checked in the mice, all but fourteen had a human match. Yet no one would-or should-conclude on that basis that we have a primordial taste for cheese.

Putting a number on our genetic similarity isn't pointless, though. It helps biologists in their quest to pinpoint the genes, and the biochemistry those genes control, that make us truly human. The quest is a hard one because most of the mutations our ancestors acquired probably had no effect on human evolution at all. Most of the harmful mutations were weeded out over time, and many of the rest had no effect one way or another. Some of these harmless mutations may have spread thanks only to chance-a process known as neutral evolution (see my column "Tuning In," September 2001).

The only changes that really mattered were the ones that altered the proteins that were themselves responsible for changing the reproductive success of the people who inherited them. These mutations spread thanks to natural selection. But distinguishing neutral from selected changes has proved a tricky statistical challenge. On the basis of early results Derek E. Wildman, a biologist also working at Wayne State University, and his colleagues project that between 2,800 and 4,000 of our genes underwent substantial change after the human branch of the evolutionary tree split off from that of the chimpanzee. A lot of those altered genes could well turn out to specialize in regulating other genes. Regulatory genes code for proteins that help switch on other genes or shut them off, thereby promoting or inhibiting the production of the proteins those other genes are responsible for. Hence a single regulatory gene can "leverage" its effects, and if a regulatory gene evolves into a different form, it could alter not only whether but also when and where various proteins are transcribed. Small genetic alterations can create an avalanche of changes in an animal's anatomy and way of life.

Svante Pbo of the Max-Planck-Institute for Evolutionary Anthropology in Leipzig has uncovered the first solid evidence that our regulatory genes set us apart from chimpanzees. Pbo and his coworkers surveyed the genes that are active in human and chimpanzee cells. They found that the cells in a human liver use many of the same genes that are active in the cells of chimp livers. But in the neurons of the brain, very different sets of genes become active in the two species.

Pbo concluded that the livers of chimpanzees and people still work much the same way they did seven million years ago. espite the similarity of the human and chimp genomes, however, evolution has created a big difference in the way genes work in our brains. That fits well with what we can see with our own eyes: in many ways people look a lot like other apes, but they have large, intricately wired brains that enable them to do things no other primate can do. Pbo's group is now trying to figure out exactly how these genes function in the brain.

In spite of Wildman's estimates for thousands of altered human genes, geneticists have identified only a few of them. They barely understand what those few genes do, or how they do it differently from related genes in chimpanzees. One of these genes (called CMAH) codes for an enzyme that helps make a sugar that coats the surfaces of cells. More precisely, this sugar, known as Neu5Gc, coats cells in chimpanzees and other mammals studied so far, but not in humans. In us the gene that makes the enzyme is useless. At some point in hominid history a long stretch of junk DNA was substituted for part of the gene, rendering its code nonsensical. The useless version now occurs in every human being.

Ajit Varki, a biologist at the University of California, San Diego, who discovered this inactivated gene, teamed up with Pbo and several other investigators to figure out exactly how long ago the gene shut down. They were able to extract a sugar from the fossils of Neanderthals that is closely related to Neu5Gc, but is not Neu5Gc itself, suggesting that Neanderthals also lacked a functioning CMAH gene. If that's true, the gene most likely shut down before our two lineages parted. Many leading experts now think the last common ancestor of Neanderthals and modern people lived at least 500,000-and perhaps as much as 800,000-years ago [see "Requiem for a Heavyweight," by Juan Luis Arsuaga, page 42]. If that's true, the gene must have shut down before then.

Varki, Pbo, and their colleagues found two other ways to estimate the gene's shutdown date. Once the stretch of junk DNA precluded the action of the CMAH gene, the junk DNA picked up a few mutations in various groups of people. A lot of evidence suggests that such mutations pile up at a steady, clocklike rate. On that basis, the investigators estimated that the gene shut down about two million years ago.

The second estimate came from a comparison of the junk DNA in the human CMAH gene with related stretches of junk DNA present elsewhere in the genomes of chimpanzees and other apes. The differences in the stretches of junk DNA also suggested that the human gene became inactive two million years ago.

That date turns out to be fraught with significance in hominid history. Until then, hominids weren't all that different from chimpanzees. They were bipedal and lived in more open habitats than chimpanzees, but their brains were still small and they showed no ability to make sophisticated tools. Then, around two million years ago their brains began to expand, a process that continued, in fits and starts, until about 100,000 years ago. And when hominid brains began to expand, the first stone axes also began to become part of the archaeological record.

Varki and his colleagues suspect that losing Neu5Gc sugar may have had something to do with that change. In chimpanzees and other mammals, Neu5gc can be found on the surfaces of cells throughout the animals' bodies. But in the brains of the same animals, other genes slow down the production of Neu5Gc, so little is found on neurons. The biologists speculate that such inhibition evolved because Neu5Gc has some harmful, though unidentified side effect on brain cells. So our ancestors may have been lucky to lose the functioning of the CMAH gene two million years ago. The loss may have entailed some kind of disadvantage, but it enabled the evolution of the brain to proceed in unique ways.

A second gene that seems to have taken on a unique form in humans is a gene that may have been crucial for the evolution of language. Linguists think the advent of language depended on new genes for forming and controlling the vocal tract and for the abstract thought needed to string words together meaningfully. In 2001 a team of geneticists from the University of Oxford discovered that mutating a single nucleotide of a gene called FOXP2 robs people of fine motor control of the mouth, and of their ability to understand some aspects of grammar.

The geneticists teamed up with Pbo's group to study the evolution of the gene. Many mammals, they noted, have related versions of FOXP2, though no one knows what it does for them. But the investigators found that after the human lineage split from that of the chimpanzee, FOXP2 underwent rapid evolution by natural selection. In fact, minor variations in the present-day forms of the gene point to its rapid evolution and subsequent spread throughout our species less than 200,000 years ago-just around the time when modern people emerged.

Human evolutionary genetics is an infant science. FOXP2 and CMAH are only two out of thousands of genes that likely set human beings apart from chimpanzees. And biochemists have yet to understand how the two genes operate in either species-much less the workings of any other genes. Even with the growing sophistication of genomic technology, biologists are still, fundamentally, as confused about the differences between chimpanzees and people as Edward Tyson was 300 years ago. But the fact that science can begin to address such questions at the level of brain chemistry is cause for amazement; it gives us reason to hope that the confusion won't last forever.

Carl Zimmer's latest book, Evolution: The Triumph of an Idea, is available in paperback from HarperCollins.