In a venture thought to lie outside the reach of science, HMS researchers have captured and sequenced tiny pieces of collagen protein from a 68-million-year-old Tyrannosaurus rex. The protein fragments—seven in all—appear to most closely match amino acid sequences found in collagen of the present day chicken, lending support to a recent and still controversial proposal that birds and dinosaurs are evolutionarily related. The findings appear in the April 13 Science.
“Most people believe that birds evolved from dinosaurs, but that’s all based on the architecture of the bones,” said John Asara, director of the mass spectrometry (MS) core facility at Beth Israel Deaconess Medical Center and an HMS instructor in pathology, who sequenced the protein fragments over the course of a year and a half using highly sensitive mass spec methods. “This gives you a chance to say, ‘Wait, they really are related because their sequences are related.’ We didn’t get enough sequences to definitively say that, but what sequences we got support that idea.”
In another study appearing in the same issue of Science, Mary Schweitzer, of North Carolina State University, and colleagues found that extracts of T. rex bone reacted with antibodies to chicken collagen, further suggesting the presence of birdlike protein in the dinosaur bones.
The mere existence of such exceedingly ancient protein defies a longstanding assumption. When an animal dies, protein immediately begins to degrade and, in the case of fossils, is slowly replaced by mineral. This substitution process was thought to be complete by one million years.
“For centuries it was believed that the process of fossilization destroyed any original material. Consequently, no one looked carefully at really old bones,” said Schweitzer, who is also at the North Carolina Museum of Natural Sciences. She is a co-author on the Asara study.
That may change, said Lewis Cantley, HMS professor of systems biology and professor of medicine at BID, who also participated in the study. “Basically, this is the breakthrough that says it’s possible to get sequences beyond one million years,” he said. “At 68 million years, it’s still possible to get sequences.” In addition to the seven dinosaur sequences, Asara and his colleagues isolated and sequenced more than 70 protein fragments from a 160,000- to 600,000-year-old mastodon, providing further evidence of the staying power of ancient protein.
“I think what this says,” Cantley continued, “is that when people make new discoveries now, if they want to get maximum information out, they have to immediately handle material in a way that first of all will avoid contamination and, second, ensure that whatever is there gets well preserved because it can be interrogated.”
The scraps of dinosaur protein were wrested from a fossil femur discovered by John Horner, of the Museum of the Rockies, and colleagues in 2003 in Hell Creek Formation, a barren fossil-rich stretch of land that spans several states, including Wyoming and Montana. Schweitzer and colleagues reported in 2005 that they had found evidence of soft tissue inside the fossilized femur, a discovery widely covered in the news. After seeing one such story in the New York Times, Cantley contacted Asara, who, in 2002, had sequenced collagen fragments from 100,000- to 300,000-year-old mammoth bone samples sent by Schweitzer’s team. More recently, Asara has been working to develop mass spectrometry techniques capable of sequencing minute amounts of protein from human tumors.
“I realized when I read the New York Times and saw Mary Schweitzer’s story of having uncovered this dinosaur that this is exactly the sort of thing that would appeal to John and hence ripped him off an e-mail from my BlackBerry,” Cantley said. “John, of course, took the bait.” Schweitzer readily agreed to provide T. rex bone samples.
“She knew from our last collaboration that I was not going to stop until I found something,” Asara said.
At the outset, he faced two challenges. The first was to gather enough protein to sequence. The bone extract sent by Schweitzer arrived in the form of a gritty brown powder that had to be rid of contaminants. Using techniques and tricks perfected while working on the mammoth sample, Asara purified the protein, identified as collagen, and with the enzyme trypsin, broke it down into fragments 10 to 20 amino acids long. These peptides were passed over a liquid chromatography (LC) column, where they were separated from one another and then sprayed at extremely low, nanoliter flow rates, for optimal sensitivity, into a mass spectrometer.
Typically, a mass spectrometer measures the mass, specifically the mass-to-charge ratio, of peptides as they come off the LC column. To maximize his yield, Asara used an ion trap mass spectrometer, which captures and holds peptides through time. The collected peptides were measured for mass and, in a second step, isolated and fragmented to reveal their amino acid sequence. Using this tandem (MS/MS) procedure, Asara netted seven separate strings of amino acid.
What’s It All About?
He now faced his second challenge, namely, to interpret the amino acid sequences. Normally, when a sequence comes out of a mass spectrometer, it is compared to a database of existing amino acid sequences. Collagen is a highly conserved protein, so it was highly likely that some of the dinosaur peptide sequences would match those of an existing species. Of the seven T. rex peptides, five were for a particular class of collagen protein, collagen alpha 1. The majority of these were found to be identical matches to amino acid sequences found in chicken collagen alpha 1, while others matched newt and frog.
For extinct species, the real goal is to find sequences unique to that organism. Asara generated a set of theoretical collagen protein sequences representing the kinds that might have been present around the time of T. rex. None of his dinosaur peptides matched the theoretical set, which is not surprising. “If you’re only finding seven sequences in T. rex, you’re not going to find novel ones, which we didn’t,” said Asara. He also tested mastodon bone, sent by Schweitzer’s group, against a database of existing amino acid sequences and against a set of mastodon theoretical sequences. He identified a total of 78 peptides, including four unique sequences.
“As we get better at doing the extractions, and the sensitivity of the instruments and techniques improve, I anticipate that we’ll be able to get from comparably aged species much more extensive sequences and to get novel sequences unique to that species, which will give us ideas about the relationship between species,” said Cantley.
Still, it may be the rare fossil that is as pristinely preserved by the environment as the T. rex and mastodon specimens analyzed in the current study. “Nature has to give you the opportunity to do this first,” Asara said.