Dinosaur proteins from T. rex and hadrosaurs

The controversy over the finding of Tyrannosaurus rex proteins in a fossilized bone
continues, with a long article in Wired magazine this week exploring the issue. The Wired reporter, Evan Ratliff, takes a refreshingly skeptical view through much of the article. It's a good read, and I recommend it.

But first, an update on the science: Mary Schweitzer, John Asara and colleagues published a new report in Science on May 1 describing their analysis of an 80-million-year-old fossil from the dinosaur Brachylophosaurus canadensis, a hadrosaur that is 12 million years older than T. rex. Once again, they found fragments from collagen, the protein that is a major component of bone, and as with their original T. rex study, they claim that these represent original dinosaur proteins. And as before, they found that the dinosaur proteins most closely resemble modern birds – in particular, ostrich.

The new study spends a significant amount of effort addressing the question of contamination. This question has been raised in at least two ways. Tom Kaye and colleagues reported in PLoS ONE that the soft tissue found by Schweitzer could be explained as a bacterial biofilm, rather than as preserved dinosaur tissue. I still like the biofilm explanation, and I don’t see how the new study refutes it. The study contains many microscope photos showing how the fine structure of the fossilized bone resembles modern ostrich bone, but preserved physical structure is not in question. The real question is, did dinosaur proteins survive for 80 million years in these bones?

The second contamination question emerges from this: Marty McIntosh discovered – after Asara released his mass spec data to the public – that the original T. rex sample contained hemoglobin. This finding has been discussed in the mass spec community, and privately among a group of scientists (including me) following this story, but it has not been published (as far as I know) until the Wired article.

Hemoglobin! Now, from what I know, we don’t expect to find much hemoglobin in bone tissue, and we sure don’t expect to find it in 68-million-year-old fossils. Could this be another big discovery? Or could it, perhaps, be that the sample is contaminated with modern proteins, perhaps from ostrich?

Upon further inquiry, a number of scientists learned that John Asara’s lab had indeed used its mass spec equipment to analyze samples of ostrich bone. Asara reports (in the Wired article) that the ostrich sample had been analyzed long before either of the dinosaurs:
“Asara conducted his ostrich and T. rex experiments a year and a half apart, separated by roughly 1,500 mass spectrometry runs. According to Asara, none of those spectra, nor samples of the soil surrounding the fossils, nor his daily control runs—in which he sequences known solutions to check for contaminants—turned up any ostrich hemoglobin.”
The new Science paper on the hadrosaur proteins doesn’t mention the (unpublished) hemoglobin fragments in the T. rex data, so this important question was not addressed. Well, if Asara’s lab handled ostrich material, then I remain skeptical of their assurances that ostrich couldn’t have possibly contaminated the original T. rex samples. And Marty McIntosh explained to the Wired reporter that “a chemical modification on the hemoglobin makes it more likely to be contamination.” McIntosh submitted a paper on his results to Science, but they rejected it – perhaps because it called into a question a finding that Science’s editors have obviously endorsed. That’s too bad.

If the T. rex sample was contaminated, that throws all the results into question. It also throws into question the new hadrosaur results – if contamination was a problem before, it might be still. The best way to resolve this is for an independent group to take the original fossils and repeat the study. I know that there are groups trying to do exactly this, but the fossils are controlled by paleontologist Jack Horner, who thus far has refused to share them with anyone but Schweitzer.

Extraordinary claims require extraordinary evidence, as any good skeptic knows. The finding of intact collagen protein in dinosaur fossils is certainly an extraordinary claim, and the finding of hemoglobin is even more stunning. Alas, the alternative explanations (statistical artifacts - see my earlier blogs and Pavel Pevzner's article in Science - and contamination by bacteria and/or ostrich tissue) are much more likely. I'd love for these results to be true - intact dinosaur proteins! Think of all the evolutionary comparisons we could do if we can reconstruct the past 80 million years directly from the molecular record! But I'm afraid I remain skeptical.


  1. What about the haemaglobin issues that the "scientists" do not discuss (perhaps only in private) in their papers. Clearly, this must be discussed in their "papers" as it destroys their credibility. these papers really should be withdrawn, or "asterisked"--that is yet another example of academic gone awry--publish or perish for one's self grandisement.

  2. Dinosaur Study Backs Controversial Find
    By Robert F. Service
    ScienceNOW Daily News
    31 July 2009

    When scientists reported 2 years ago that they had discovered intact protein fragments from a 68-million-year-old Tyrannosaurus rex, the skeptics pounced. They argued that one of the main lines of evidence, signatures of the protein fragments taken by mass spectrometry, was flawed. But now a reanalysis of that mass-spec data from an independent group of researchers backs up the original claim that dinosaur proteins have indeed survived the assault of time.
    In 2005, a team led by Mary Schweitzer of North Carolina State University in Raleigh reported in Science that it had discovered an unusual T. rex fossil, in which some of the soft tissues, including blood vessels and other fibrous tissue, seemed to have been preserved. Two years later, Schweitzer teamed with mass-spec expert John Asara of Harvard Medical School in Boston and colleagues to report that mass-spec studies identified seven peptide fragments that appeared to come from dinosaur collagen and that those sequences were closely related to analogous sequences from the chicken and other modern birds, as would be expected given the many lines of evidence that birds evolved from dinosaurs. But skeptics argued that the mass-spec signals barely hovered above the data's background noise. And Schweitzer and Asara, they argued, couldn't rule out that the signals were caused by contaminants.

    The controversy has continued in letters and follow-up papers. It also prompted Asara to release his complete mass-spec data set to other experts to allow them to judge for themselves. So researchers from the Palo Alto Research Center in California and the University of California, Davis, decided to do just that. They reanalyzed Asara's mass-spec data using a different set of bioinformatics tools and statistical tests.

    In a paper published this month in the Journal of Proteome Research, the California researchers report that three of the peptides are a strong match to what looks to be an ancient form of collagen, whereas others matched with less statistical significance. "In summary, we find nothing obviously wrong with the T. rex mass spectra: the identified peptides seem consistent with a sample containing old, quite possibly very ancient, bird-like bone, contaminated with only fairly explicable proteins," they conclude.

    "That's pretty good news," says Asara. He doubts that the new result will put the entire controversy to rest, because soft tissue from dinosaur is such an extraordinary find. However, he argues, "I think it puts the mass-spectrometry interpretation to rest." The case for T. rex proteins is also helped by the fact that it is no longer a unique discovery. In May, Schweitzer, Asara, and colleagues reported a similar result from an 80-million-year-old dinosaur and raised further hopes that scientists may soon have a window into the molecular makeup of dinosaurs.

    ( skip to comments for this article )
    Next Article


    The editors suggest the following Related Resources on Science sites:
    In ScienceNOW

  3. Yes, I saw that study. It does not address, in any way, the possibility of contamination. If a tiny fragment of collagen from modern ostrich (or a fragment of hemoglobin!) was in the sample, then of course you'll still find collagen when you re-analyze the data. What is needed is for an independent group to replicate the result - to extract material from a dinosaur fossil, under very careful conditions, and to then find collagen. This hasn't happened.

  4. ? I don't see how the contamination issue isn't address by their subsequent Hadrosaur research showing a closer resemblence between the Hadrosaur and the T.Rex at the protein level than with modern animals. The Wired article you cite even mentions this, with Dr. McIntosh calling it "a nice bit of work." How would they find similar results using two different samples processed in two different ways, the second being more rigorous, using sterlized instruments, if the first was solely the result of contamination? They may not be "an independent group" but it looks to me like they've already done exactly what you say "hasn't happened". Are you suggesting that the Hadrosaur was contaminated also?

  5. Yes, I'm saying exactly that. The mass spec work for both samples was done at the same facility (John Asara's lab in Boston). The claim that the instruments were "sterilized" doesn't convince me - if ostrich material is in the lab, it can get on the samples.
    Furthermore, I've learned more about Mary Schweitzer's field methods, and it's absurd to claim that the samples were sterile in the field. So the collagen could also be fragments from other modern birds. All sorts of modern organic matter can leach through the soil and into fossilized bone over the course of millions of years. Schweitzer's claims are laughable.
    And by the way, Marty McIntosh isn't convinced either, despite the brief quote from him in Wired.

  6. Interesting, thanks for the response. I'm no expert (I have degrees in Physics, not biochemistry, etc.); are there two explanations to explain Schweitzer's results then? 1) The "tissues" observed are actually bacterial biofilms and 2) the proteins found are accounted for by contamination? I have to admit the contamination theory seems a little tough to me since, assuming it was there during the T.Rex testing, they would have been pretty careful before testing the Hadrosaur, given the heavy criticism of the T.Rex results (and knowing the scrutiny they would get)? Is it that hard to remove contaminates? Is Asara doomed to have ostrich chunks in his MS from here on out? I'm kidding, but I do think I tend to agree with you that they need a lot more than they have to convince the world that dino protein lasted for 10's of millions of years. Can bacteria in the bone account for the proteins observed or is contamination just that hard to deal with?


Markup Key:
- <b>bold</b> = bold
- <i>italic</i> = italic
- <a href="http://www.fieldofscience.com/">FoS</a> = FoS

Note: Only a member of this blog may post a comment.