By Rick Weiss
Washington Post Staff Writer
Friday, April 25, 2008
Protein retrieved from a 68 millon-year-old Tyrannosaurus rex bone closely resembles the main protein in chicken and ostrich bones and is only distantly related to lizards', strengthening the popular idea that birds, and not reptiles, are the closest living relatives of dinosaurs.
The new work builds on a 2007 analysis showing remarkably close similarities between T. rex collagen and collagen from modern-day chickens, but that work did not include comparisons to other living species. Collagen is the primary protein in bones.
"We had made a very loose connection at first," said John M. Asara of Harvard Medical School and Beth Israel Deaconess Medical Center, who led both studies. "Now we're able to make out robust evolutionary relationships and, with very high confidence, basically group the T. rex dinosaur with birds."
More is at stake than T. rex's prehistoric pedigree. Asara and his colleagues say their novel approach has the potential to redraw the evolutionary tree based on molecular data instead of the traditional comparisons of skeletal remains. Bones can be deceiving, because unrelated animals can have similar structures.
Key to the new finding -- and the cause of some controversy -- is the Asara team's assertion that it retrieved a smidgen of intact collagen from the fossilized thigh bone of a T. rex. Biological materials generally degrade in the environment, and scientists who work with ancient DNA feel lucky when they find a sample that is 100,000 years old. Yet the T. rex protein specimen is more than 100 times older than that, leading some scientists to wonder whether it might be a more recent contaminant.
Mary H. Schweitzer, a molecular paleontologist at North Carolina State University in Raleigh who oversaw the protein's extraction, said she was confident that what they were dealing with was dinosaur collagen, preserved because of favorable conditions in the Montana soil where the bone was buried.
"There is no evidence of contamination," Schweitzer said, emphasizing the painstaking method developed by the team, which captures less than a millionth of a gram of protein for every five grams of bone.
But not everyone agrees.
Pavel Pevzner, director of the Center for Algorithmic and Systems Biology at the University of California at San Diego, said his own research, soon to be published, refutes Asara's work. He said he cannot describe details until they are published, but he was blunt in his response to the new study, which appears in today's issue of the journal Science.
The findings are "a joke," Pevzner wrote in an e-mail. "Serious evolutionary biologists will laugh reading this piece."
It was unclear whether Pevzner disputes the link between birds and dinosaurs or simply distrusts the methods the team used.
Proteins are strands of amino acids, and the order of those various amino acids determines a protein's identity and function. In the new analysis, the team compared the order of 89 amino acids from the T. rex sample to the equivalent collagen sequence from a chicken, an ostrich, an alligator and a green anole lizard, a reptile commonly used in laboratory research.
The results indicate that T. rex, chickens and ostriches are evolutionary siblings, all descended from a single unidentified predecessor. Alligator collagen is more distantly related, and lizard collagen is more distantly related still.
"Birds are a type of dinosaur the way Saint Bernards are a type of dog," Schweitzer said, "or like a poodle is a type of wolf."
A second analysis by the team backed up scientists' long-standing suspicions that extinct mastodons and modern-day elephants are close cousins.
Eddy Rubin, a geneticist and expert in ancient DNA who heads the Department of Energy's Joint Genome Institute in Walnut Creek, Calif., said he is "pleased to see" that the Harvard group was doing the hard work of duplicating and expanding upon its initial results, which were unusual not only because of the starting material but also because of the team's reliance on a novel method for analyzing tiny samples.
"Extraordinary results demand extraordinary evidence," Rubin said, conceding that in the past, at least, he "had some concerns" about the data.
"It's very exciting if they really are getting proteinaceous material from these terrifically ancient samples, because in some ways these sequences are time machines," Rubin said. "We can't go back in time, but these samples tell us things about these creatures that we can't tell from their bones and other features."
If further tests confirm a capacity to extract and accurately analyze smidgens of protein from fossilized remains, he and others said, it would be a boon to scientists who are trying to work out the branching pattern of the evolutionary tree of life.
Schweitzer said she welcomed challenges to their results. "If they can show our data are wrong, that's great," she said. "That is the way science progresses."
For his part, Asara said he is so sure of the results he can almost taste them.
"Based on this data," he said, "you can be very confident that T. rex would taste more like chicken than it did last year."