The Proteins of Prehistory

The hit movie Jurassic Park sparked broad interest in paleontology by raising the tantalizing possibility of bringing dinosaurs back from extinction. In the film, this was accomplished by extracting dinosaur DNA from mosquitoes preserved in ancient amber. That, however, remains a critical plot element and nothing more; the general scientific consensus is that DNA has a half-life of hundreds to thousands of years at most.

Proteins, on the other hand, have much longer half-lives – hundreds of thousands to over a million years under optimal storage conditions. Plus, we have long known that amino acids can be extracted from fossilized hard tissues. This knowledge, and a desire to look deeper, gave rise to paleoproteomics, an interdisciplinary field that examines ancient proteins to study the molecular-level adaptations of species – and evolution itself.

Crucial evolutionary changes emerged during the age of the dinosaurs: gigantism, endothermy, and the development of feathers, to name just a few. Due to protein degradation, however, conventional wisdom asserts that paleoproteomic investigations carry a very low probability of success. But what if the proteins did not degrade? Dinosaur fossils are intriguing reservoirs of ancient protein because the fossilization process can impede degradation and loss. In fact, reports detail the discovery of proteins such as collagen (the most abundant protein in vertebrates and fundamental to animal evolution) in dinosaur fossils using MS – but, because of the potential for exogenous protein contamination, these reports have been met with skepticism.

To have detected ancient dinosaur protein is indeed an extraordinary claim – one that must be supported by extraordinary evidence. But we feel that the criticism overlooks one important point: that evidence of a protein’s existence does not require the detection of a full protein sequence. Distinctive post-translational modifications (such as extensive proline hydroxylation in collagen) can support its identification in fossils. Blanks and independent methods of analysis can lower the probability of false positives or misinterpreted results due to contamination. Here, we suggest additional criteria to bolster positive reports of protein detection from dinosaur fossils.

Because the genomes of dinosaur species are not known, every putative protein identified from fossils is a unique chemical entity. But even ancient proteins should have molecular relationships to those in modern animals – particularly birds and reptiles – which makes sequence homology with other species essential to these studies. We also advocate the introduction of fresh, immobilized enzyme microreactors to digest extracted protein in the microreactor environment and maximize digestion efficiency; using multiple enzyme microreactors will enhance sequence coverage for such identifications.

What’s more, with time, chiral amino acids in proteins will racemize. Although insufficient protein recovery from dinosaur fossil specimens is an issue, we believe that (where sufficient sample exists) measuring the D:L ratios in acid hydrolysates is critical. Low D:L ratios suggest that a protein is not ancient and is likely the result of exogenous contamination.

The greatest discoveries in science are achieved by those willing to challenge conventional paradigms. In our view, there is sufficient, carefully collected evidence from fossils to suggest that protein molecules from dinosaurs are detectable – and just waiting to be discovered!