Research Article

Dinosaurian Soft Tissues Interpreted as Bacterial Biofilms

  • Thomas G. Kaye mail,

    Affiliation: Department of Paleontology, Burke Museum of Natural History, Seattle, Washington, United States of America

  • Gary Gaugler,

    Affiliation: Microtechnics Inc., Granite Bay, California, United States of America

  • Zbigniew Sawlowicz

    Affiliation: Department of Geology, Jagiellonian University, Krakow, Poland

  • Published: July 30, 2008
  • DOI: 10.1371/journal.pone.0002808
  • Published in PLOS ONE

Reader Comments (4)

Post a new comment on this article

Amino acids support Schweitzer et al.

Posted by Broadlands on 08 Aug 2008 at 11:20 GMT

As I recall, in addition to their other evidence, Schweitzer et al. were able to detect the presence of both hydroxyproline and hydroxylysine in their T rex preparations. These two amino acids are molecular "fingerprints" for collagens and are not found in prokaryotes making them difficult to explain from a bacterial source.

RE: Amino acids support Schweitzer et al.

Paul_Orwin replied to Broadlands on 13 Aug 2008 at 03:07 GMT

I think that you (and Schweitzer et al) should be cautious about "prokaryotes don't have this" arguments - we don't know nearly enough about environmental prokaryote biochemistry to be sure of that. An article in Mol. Micro showed that Bacillus anthracis spores contain a collagen like protein as their major component (BclA), although they did not address your specific point about hydroxyproline and hydroxylysine.

RE: RE: Amino acids support Schweitzer et al.

Broadlands replied to Paul_Orwin on 13 Aug 2008 at 16:07 GMT

We know quite a bit more than you might think. With one exception that I know of (a protozoan foraminiferan) neither of these amino acids has been found in the Procaryota. And, as regards the collagen-like materials in bacteria, see, e.g.,

"...bacterial CSMs form collagen-like trimers, even though these organisms cannot synthesize hydroxyproline, a critical residue for the stability of the collagen triple helix."

Any bacterial origin for the material found preserved in the dinosaur bones must be able to explain the presence of these two amino acids.

RE: RE: RE: Amino acids support Schweitzer et al.

Paul_Orwin replied to Broadlands on 14 Aug 2008 at 22:16 GMT

I am not an expert in this topic by any stretch! However, I saw the article you pointed out, but I'd also point out that genome sequences of Bacillus strains and other environmental bugs have shown potential homologs of P4H.
Also, protozoa are not procaryotes! But more to the point, the soil contains eukaryotes as well as prokaryotes, so the hydroxylated residues might be from them rather than the proks.
Finally, I think (and this is an outsider's decidedly non-expert opinion for this type of research) that categorical statements like "hydroxyproline isn't found in bacteria" are a bad idea, ESPECIALLY when it comes to as diverse a phylogenetic group as the eubacteria. If it is biologically useful to hydroxylate proline and lysine, then it is reasonable to suspect that bacteria do so - it doesn't mean that they must have done so in this instance, but I don't think it is safe to rule it out based on the small (relative to the size of the microbial biosphere) level of biochemical data that is present. With all due respect, we may collectively know more than I do (I certainly hope so!), but what we know is a drop in a massive, massive bucket!

RE: RE: RE: RE: Amino acids support Schweitzer et al.

Paul_Orwin replied to Paul_Orwin on 14 Aug 2008 at 22:17 GMT

Just as an added follow on, I did a quick microbial genomes search based on the HIF-prolyl-hydroxylase, and found several putative homologs (not very strong identity, but enough for a machine annotation) in Bacillus strains and Ralstonia strains - If it were my field of inquiry, I might invest some time and effort in acquiring these strains and seeing if these enzymes can perform this function in vitro.

Amino acids still support Schweitzer et al.

Broadlands replied to Paul_Orwin on 16 Aug 2008 at 16:49 GMT

Orwin is quite correct that making all-encompassing categoric statements is a bad idea, but so far neither peptide-bound hypro nor hylys has yet been found in a bacterium. This doesn't mean that they might not exist somewhere in some bacterium. However, it seems very unlikely, even with a massive bacterial "bucket". Why? Neither amino acid is in the genetic code. Both are post-translational requiring mixed-function enzymes that have four cofactors: molecular oxygen, ferrous iron, ascorbic acid (vitamin C) and alpha-ketoglutarate. Remarkably, this complicated pathway is the same for both the collagens (in all animals from sponges to vertebrates) and the extensins (hypro in higher plants). Such an identical pathway in animals and higher plants implies a common evolutionary origin and the fossil record of multicellular life shows that it is one that took billions of years to evolve. In collagens, both amino acids are necessary for functional integrity and for strengthening by cross-linking (biologically useful). If Orwin, Kaye et al. (or anyone) can point to any bacteria where even ONE of these two amino acids has been found it would be a startling find, one much more important than showing that there may be bacterial biofilms is a dinosaur bone. In the meantime, although they are very labile, they might be (surprisingly) expected in exceptionally well-preserved collagens. But BOTH are not to be expected in either a higher plant residue nor a bacterial biofilm. To point out that science hasn't studied every bacterium on the planet as regards these amino acids and then extrapolate this to imply that Kaye et al. may actually have stumbled on some that have this ability in a dinosaur bone is really grasping at straws. The simple statement that these amino acids support Schweitzer's conclusion remains. Their presence in the T. rex preparations makes them difficult to explain from a bacterial source, regardless of whether there are biofilms present as well.

RE: Amino acids still support Schweitzer et al.

Paul_Orwin replied to Broadlands on 17 Aug 2008 at 16:44 GMT

As I noted above, homologs to one of the main enzymes have been found in common soil bacteria - I don't expect all that much research effort has been made to LOOK FOR these post-translational mods in environmental bacteria, the field has simply not been looking in this direction. The complex pathway you note does not require anything that a common microbe wouldn't be able to access/make. The evolutionary argument falls apart when we unpack your statements about the conservation to note that microbial evolution has been going on for ~3 by, while this pathway is conserved over groups that stretch back ~0.6 by (I might be a little off with that number, don't quite remember).
I think finally that we are avoiding something I mentioned earlier, that admittedly wasn't part of the PlosONE paper, which is that there are certainly eukaryotes (protozoa similar to the one you mentioned in the first post on this) present in any soil biofilm.
In the end, without more information - I just don't see this argument holding up - especially with what you just said about them being extremely labile - why then don't we expect there to be a different source?
Finally, there might be contamination of the tissue that would introduce collagen from the handlers - I don't want to get into that, because it isn't fair of me to assume the experimenter's error in this way, and I'd much prefer to assume they did everything right!
So, I guess in the final analysis the question is whether you think it is more likely that an unknown chemical process maintained the integrity of these highly labile compounds, or an unknown biological process allowed the accumulation of them - I vote for the latter :)

Unknown bacterial process?

Broadlands replied to Paul_Orwin on 18 Aug 2008 at 11:18 GMT

Fair enough. However, if an unknown biological process among bacteria is the answer it would be an amazing coincidence, a finding hitherto completely unknown in paleobiological work, and one (two, actually) that showed up at just the right time. If you really believe this I urge you and the others to follow that up ASAP. As I indicated earlier, if hypro and hylys are truly bacterial in origin it would be an historic scientific finding...much more significant than dinosaur speculations. Contamination is, of course, the usual "out" when things don't seem to fit. Perchlorate on Mars? For me, the very low amounts Schweitzer found (picograms?) would tend to support their explanation because an accidental contaminant might be expected to show up in larger amounts. This has happened in the past in paleobiological work.

You wrote: "The evolutionary argument falls apart when we unpack your statements about the conservation to note that microbial evolution has been going on for ~3 by, while this pathway is conserved over groups that stretch back ~0.6 by..."

This is beside the point for the current discussion, but I think you may have missed my point. Because bacterial evolution had, indeed, been going on for over 3 billion years with no significant evolutionary advancement in size and structural complexity it required the evolution (choanoflagellates? some fungal protist?) of new amino acids (these two) not among the 20 in the genetic code to develop structural proteins for multicellular life to make an appearance. In combination with an increase in atmospheric oxygen, this led to the evolution of collagen which was the key innovation that led to the explosion of multicellular life ~0.6 byr ago. After 3-plus billion years of trying, no combination of the 20 AAs in the code (oxygen or not) was able to come up with a structural moiety able to do the job. This is one reason why it is most unlikely that there exists any bacterium that has even one of these AAs now.
If you have internet access, go here: SCIENCE 30 May 2003, vol. 300, no. 5624, pp. 1370 - 1371. "Evolution of Protein Amino Acids".
Another reference:

You also write: "...why then don't we expect there to be a different source?"

I DID expect a "different source", and contamination comes quickly to mind (and may even be the right answer in the long run). I've explained why I think contamination is probably not the answer. On the other hand, high-sensitivity amino acid analyses on beautifully preserved collagens in 450 Myr-old graptolite periderm (Nature (1972) vol 237, p. 443-445) showed that there were little or no amino acids preserved in spite of the stunning physical preservation seen in TEM preparations. The same holds true for organic matter trapped within calcified Paleozoic trilobite cuticles. The preservation of proteins is a tricky process and there don't seem to be many iron-clad rules. Obviously, keeping water out to protect the peptide bonds is essential.

In the present instance there seem to be three choices: (1) the material is real dinosaur stuff (2) it is a contamination, most likely of metazoan origin, (3) it is a completely unknown bacterial material. Hopefully, further research will sort it all out, but for the moment, #3 remains a real long shot.

RE: Unknown bacterial process?

WvanSchaik replied to Broadlands on 22 Aug 2008 at 07:43 GMT

First let me make it absolutely clear that I am no expert in paleontology or MS. Still I found this paper very interesting because I have a truly hard time believing the original claims by Schweitzer et al. As a microbiologist I cannot really believe that macromolecules from an animal source can be preserved for such an incredibly long period of time. It seems much more likely that these proteins should have been degraded long ago by bacterial or uni-cellular eukaryotic scavengers.
With this in mind, I find Paul Orwin's comments right on target. Bacterial biofilms are often communities where both prokaryotic and eukaryotic organisms are living together (often not exactly in close harmony, but that's a different story: see for example http://www.ncbi.nlm.nih.g...). I would say the presence of eukaryotes in a biofilm may be a perfectly acceptable explanation of finding hypro and hylys in the samples.
Like Orwin I would also not be very surprised to find hypro or hylys in some eubacteria. However one could argue that this is besides the point, because there are much more plausible explanations for the presence of hypro and hylys in a biofilm.

Both AAs in a biofilm?

Broadlands replied to WvanSchaik on 22 Aug 2008 at 18:26 GMT

As I said earlier, I had difficulty accepting these two amino acids in fossil materials that old. But if metazoans in biofilms is "a perfectly acceptable explanation"(remember BOTH of these amino acids were found), it is unique, not having been reported in other paleontological materials. Hypro OR hylys, one or the other in some bacterium? Maybe, but, again, BOTH? Also unique, never been seen before in a bacterium. Not impossible, but most unlikely. It is probable that Schweitzer et al. will prepare a response to this paper. I'm willing to wait to see if the data are reproducible and that contamination can be ruled out. Amino acid racemization studies might help.