Transversely cut piece of 195-million-year-old Lufengosaurus rib

Dark, iron-rich hematite particles may have preserved protein fragments in this 195-million-year-old dinosaur rib, according to one of two independent studies of dinosaur proteins.

Robert Reisz

Scientists retrieve 80-million-year-old dinosaur protein in ‘milestone’ paper

It’s not quite Jurassic Park: No one has revived long-extinct dinosaurs. But two new studies suggest that it is possible to isolate protein fragments from dino­saurs much further back in time than ever thought possible. One study, led by Mary Schweitzer, a paleontologist from North Carolina State University in Raleigh who has chased dinosaur proteins for de­cades, confirms her highly controversial claim to have recovered 80-million-year-old dinosaur collagen. The other paper suggests that protein may even have sur­vived in a 195-million-year-old dino fossil.

The Schweitzer paper is a “milestone,” says ancient protein expert Enrico Cap­pellini of the University of Copenhagen’s Natural History Museum of Denmark, who was skeptical of some of Schweitzer’s ear­lier work. “I’m fully convinced beyond a reasonable doubt the evidence is authen­tic.” He calls the second study “a long shot that is suggestive.” But together, Cappellini and others argue, the papers have the po­tential to transform dinosaur paleontology into a molecular science, much as analyz­ing ancient DNA has revolutionized the study of human evolution.

Back in 2007 and 2009, Schweitzer re­ported in Science that she and her colleagues had isolated intact protein fragments from 65-million- and 80-million-year-old dino­saur fossils. But the claims were met with howls of skepticism from biochemists and paleontologists who saw no way that fragile organic mol­ecules could survive for tens of millions of years, and wondered whether her samples were contaminated with modern proteins.

Then last year Cappellini and Matthew Col­lins, a paleoproteomics expert at the Uni­versity of York in the United Kingdom, and colleagues managed to identify protein fragments from 3.8-million-year-old ostrich egg shells, a claim that most of their colleagues found convincing. Now, the case for dramatically older proteins seems to be firming up, too. Last week in the Jour­nal of Proteome Research, Schweitzer, her postdoct Elena Schroeter, and colleagues report that they did a complete makeover of their 2009 experiment to rule out any possible contamination. They took new samples from the same 80-million-year-old fossil, of a duck-billed dinosaur called Brachylophosaurus canadensis. They reworked procedures for extracting would-be proteins from the bone, identi­fied protein fragments with a more sensi­tive mass spectrometer, and compared the recovered protein sequences to those from many more living animals. Schroeter even went so far as to break down the mass spectrometer piece by piece, soak the whole thing in methanol to remove any possible contaminants, and reassemble the machine. “About the only thing that is the same [as the 2009 experiments] is the di­nosaur,” Schweitzer says.

In their 2009 paper Schweitzer’s team had identified three fragments of a protein called collagen 1 from their fossil. Collagen is the main protein in connective tissue and is abundant in bone. Each fragment contained about 15 amino acids strung to­gether, which the mass spectrometer was able to identify. In their current study, Schweitzer’s team identified eight protein fragments, two of which matched those identified originally. “If [both sets] are from contamination, that’s almost impos­sible,” Schweitzer says.

The three protein fragments originally recovered most closely resembled the col­lagen found in living alligators and other reptiles. But the new data show that B. canadensis collagen was a better match to that of birds. That’s just what paleontologists, who consider birds to be descendants of extinct dinosaurs, would predict.

Just how those collagen sequences sur­vived tens of millions of years is not clear. Schweitzer suggests that as red blood cells decay after an animal dies, iron liberated from their hemoglobin may react with nearby pro­teins, linking them together. This crosslink­ing, she says, causes proteins to precipitate out of solution, drying them out in a way that helps preserve them. That’s possible, Collins says. But he doesn’t think the pro­cess could arrest protein degradation for tens of millions of years, so he, for one, re­mains skeptical of Schweitzer’s claim. “Pro­teins decay in an orderly fashion. We can slow it down, but not by a lot,” Collins says.

<p>Researchers claim this scanning electron microscope image captures collagen fibrous material from an 80-million-year-old dinosaur fossil.</p>

Researchers claim this scanning electron microscope image captures collagen fibrous material from an 80-million-year-old dinosaur fossil.

Mary Schweitzer North Carolina State University, and ICAL facility, Montana State University

The second paper, published this week in Nature Communications, goes back even fur­ther in time but offers weaker evidence, Cap­pellini says. In this work, researchers led by paleontologist Robert Reisz at the University of Toronto in Canada reported finding what they believe is collagen in a 195-million-year-old fossil rib from a large plant-eating dino­saur called Lufengosaurus that lived in what is now southwestern China. Reisz says his team’s methods, called Raman spectroscopy and synchrotron radiation Fourier trans­form infrared microspectroscopy (SR-FTIR), can probe the chemical makeup of a sample without the need to purify it first, which low­ers the risk of contamination. The rib, he and his colleagues report, absorbed infrared light in wavelengths that match those of collagen from mod­ern animals.

Schweitzer and Cappellini caution that while SR-FTIR is good at spotting the so-called amide chemical bonds that link suc­cessive amino acids in proteins, it can’t pin down exactly which protein is present, or the protein's sequence. Thus it isn't useful for evolutionary studies. This method also can’t rule out that the amide bonds are in other compounds, such as the epoxy used to assemble microscope slides. “Synchrotron data is very powerful, but it’s limited,” Schweitzer says. “I would like to have seen confirmatory evidence,” such as exposing the fossilized material to an antibody that binds solely to collagen to see whether it targeted the fossilized material. Reisz agrees “that certainly would be the next step.” But he’ll have to team up with other specialists to carry that out.

Still, his work, too, suggests that colla­gen fragments can survive for astonishing periods of time. Meanwhile, Schweitzer’s team is going beyond collagen. In a 2015 paper in Analytical Chemistry, her group reported isolating fragments of eight other proteins from fossils of dinosaurs and extinct birds, including hemoglobin in blood, the cytoskeletal protein actin, and histones that help pack­age DNA. Comparing those se­quences from many different species could reveal evolution’s handiwork over geological time, much as studies of an­cient DNA do today.

Her team won’t be the only ones exploring these methods. Now that it seems that very an­cient protein fragments can, in fact, be isolated and examined, it’s a safe bet that many new collaborations will soon take shape to pin down the evolu­tionary relationships among different dinosaurs, as well as among ancient mammals and other extinct creatures. Says Schweitzer: “The door is now open.”