Neandertals 'hardly differed at all' from modern humans

Although we are both hominids, the fossil record told us long ago that we differ physically from Neandertals, in various ways. But at the level of genes and the proteins that they encode, new research published online May 6 in the journal Science reveals that we differ hardly at all. It also indicates that we both -- Neandertals and modern humans -- differ from the chimps in virtually identical ways.

"The astonishing implication of the work we've just published," says Prof. Gregory Hannon, Ph.D., of Cold Spring Harbor Laboratory (CSHL), "is that we are incredibly similar to Neandertals at the level of the proteome, which is the full set of proteins that our genes encode."

Collaboration with a paleogenetics pioneer

Hannon, who is also an Investigator of the Howard Hughes Medical Institute and is well known for his work on small RNAs and RNA interference, was invited this past year to help examine Neandertal DNA by Dr. Svante Pääbo, a pioneer in paleogenetics, a field that employs genome science to study early humans and other Paleolithic-era creatures. In a separate paper, Pääbo's team today publishes in the same issue of Science the first complete genome sequence for Neandertal, an achievement that builds on work he has led since 2006 at the Max Planck Institute for Evolutionary Genomics in Leipzig.

"Dr. Pääbo's publication of the full Neandertal genome is a watershed event, a major historical achievement," Hannon says. "The work we conducted in collaboration with his team is only a small part of the larger effort, but it helps us put the Neandertal genome into better perspective, relative to the modern human genome and those of our nearest common ancestor among the apes, the chimpanzees, from whom we diverged about 6.5 million years ago."

The CSHL team contributed a technology developed by Hannon, postdoc Emily Hodges, Ph.D., and others at CSHL in 2007. "We call it "Array capture re-sequencing," says Hannon, "and it enables us to extract from genomes important information, on a very selective basis, rapidly, very accurately, and at low cost. We always anticipated that it might help in the analysis of evolutionary relationships, so when Svante offered us the opportunity to apply it to a Neandertal sample, we were very excited and grateful for the opportunity."

The technique enabled Hannon's CSHL team, working with Pääbo's team in Leipzig, to greatly amplify intact bits of DNA from a Neandertal sample that was 99.8% contaminated -- mainly by bacterial DNA-- and regarded by Pääbo as not likely to yield useful data. The sample studied was considerably more impure than that used as the basis for Pääbo's full Neandertal genome sequence. "Our technology is particularly useful in enabling us to work with the most contaminated samples," says Hannon. "We identify and then greatly amplify just those portions of the target DNA called exons. Exons are stretches of DNA that encode proteins. They comprise only a small fraction of the total genome of modern humans, about 1%."

The Neandertal genome, like that of modern humans, contains about 3 billion base-pairs of nucleotides -- often referred to metaphorically as "letters" in the genome's "book of life." The Hannon-Pääbo collaboration focused on obtaining the most accurate possible sequence of only 14,000 protein-coding segments within the full genome. "These," Hannon explains, "are exons that give rise to the 14,000 proteins that we know are different in modern humans and chimpanzees." The question was what those 14,000 proteins would look like in our Neandertal relatives.

"The overwhelming majority of chimp proteins -- about 75% -- are different from ours in at least one amino-acid 'letter," according to Hannon. These amino-acid changes are in most instances slight, but the resulting functional differences -- the way they affect what proteins do in cells -- can be great, and presumably help to explain many of our differences from chimpanzees.

Eighty-eight amino-acid differences -- and what they might signify

Hannon's team applied its focused sequencing method on those areas in the Neandertal sample obtained from Dr. Pääbo, and, after several rounds of refinement, they arrived at the number 88: they found only 88 changes in Neandertal protein sequences compared with the modern human. Hannon calls this number "astonishing."

At an early stage of the study, the team identified many more protein differences -- about 1000 -- between modern man and the specific Neandertal individual sampled, a male who died about 49,000 years ago in a cave called El Sidrón, in Spain. But that initial figure was based on comparing the Neandertal sequence to that of the modern human reference genome. When the teams incorporated into their calculations variations in the modern human code that they catalogued in 50 individuals from a range of modern ethnic groups, the number of human-Neandertal protein differences dropped from over 1000 to only 88.

Although Hannon says it will be important to study the functional role of the 88 proteins, he expects that many may prove "neutral," functionally. These would be changes in the genetic code that do not issue in any difference in the function of the associated proteins. If even more human genome samples -- say, from 500 contemporary individuals rather than 50 -- were included in the comparison, the number of differences might drop again, Hannon believes. And if additional Neandertal samples were factored into the comparison, he says, "it's possible that the number of differences could approach zero."

In short, Hannon says, "the news, so far, is not about how we differ from Neandertals, but how we are so nearly identical, in terms of proteins." In addition to following up on the functional associations of the 88 proteins identified in the current study, Hannon says new research is likely to address other portions of the genome -- particularly those segments responsible for regulating what genes do. In effect, the search for what distinguishes us from our nearest hominid ancestors will shift to from differences in sequence to differences in function.