Using a combination of techniques from immunoblots to mass spectrometry, a group of researchers led by North Carolina State University's Mary Schweitzer confirmed that proteins from Cretaceous-period dinosaurs could be preserved and sequenced. In 2007, Schweitzer and her colleagues had reported on peptides recovered from Tyrannosaurus rex remains, though those findings were controversial, as proteins are generally thought to degrade quickly.
For the current project, researchers studied a femur from Brachylophosaurus canadensis, a hadrosaur that lived 80 million years ago, to see if it too contained protein remnants. "We basically went looking for a dinosaur preserved deeply in sandstone," Schweitzer says. "If we want the best preservation possible, it's going to come from dinosaurs that have been deeply buried very rapidly after death in sandstone."
Due to the controversy surrounding their earlier T. rex study, particularly regarding contaminants, Schweitzer and her colleagues were extra careful in the collection and analysis of B. candensis. Instead of excavating the skeleton completely in the field, the researchers took it back to the lab surrounded in six inches' worth of sandstone. "If you think about it, this dinosaur has been sitting in equilibrium with its environment for 80 million years. Then we come in and we dig it up and sweat on it and eat lunch over it and drink beer on it and degradation picks up where it stopped," says Schweitzer. "We wanted to keep it as stable as we could."
When the sample arrived in the lab, the team set to demineralizing it, looking at it under scanning and transmission electron microscopes. They tested to see if hadrosaur tissues and tissue extracts bound to chicken and ostrich collagen antibodies — and they did. Then they analyzed the remains with mass spec, still trying to limit the effects of contamination. "While [one colleague] produced mass spec sequences from extracts of the bone, we also did some in situ mass spectrometry and were able to identify amino acid residues that are specific for collagen and localize those to the tissues," she says.
From those amino acid residues, Schweitzer and her colleagues determined that the collagen sequence from the hadrosaur samples fell in the dinosaur-bird clade. From their phylogenetic tree, they predict that B. candensis is more closely related to birds than to alligators and that T. rex is more closely related to birds than to B. candensis.