About TCAW - Subscription Info
September 2001
Vol. 10, No. 09,
pp 19–20.
Instruments & Applications
Proteins in the Pot, Nine Eons Old

opening artImmunological analysis of ancient ceramics could fill gaps in prehistory.

After a stressful day behind the bench, some of us might struggle to recall what we had for breakfast that morning, but what about trying to unearth what our ancient ancestors ate for their prehistoric meals?

The analysis of fatty compounds (lipids) absorbed in ancient ceramics from pots and storage vessels have, in the recent past, helped researchers reveal the uses to which archaeological (more recent) and prehistoric pottery may have been put—distinguishing between vessels that carried water and those used for storing food or cooking.

But the analysis of lipids will only take us so far in identifying which vessels were originally food-bearing. They do not provide much of an indicator as to what particular foodstuffs were held in the vessel. Proteins, on the other hand, being far more complex, can leave a telltale fingerprint that narrows down the search for the missing contents quite rapidly.

Indeed, according to Peggy Ostrom of Michigan State University (East Lansing), proteins act as an “enormous reservoir of information.” For instance, they contain fundamental genetic information, just as DNA does, thus providing the key to rebuilding phylogenetic relationships between ancient creatures. Whether this information might be applied to human bones remains to be seen.

The advantage of protein analysis over DNA analysis of archaeological samples is that there is usually more protein in a sample and it is generally in better condition than DNA. (See also, Amino Acid Racemization” by Ed Brignole and Julie McDowell in the February 2001 TCAW.)

But, the challenge is how to extract and identify the exceptionally small amounts of protein and determine what their presence means.

Milk Does a Relic Good
Oliver Craig and Matthew Collins of the University of Newcastle-upon-Tyne (Newcastle, England) have developed an assay similar to an enzyme-linked immunosorbent assay (ELISA) to detect protein remnants from archaeological pottery specimens and paleontological specimens. In part, their results can tell us about the kinds of foods people were eating during the Iron Age, but they can also answer important questions about whether or not those ancestors farmed cattle for milk or beef, which has implications for understanding prehistoric societies. Dairy farming, as any farmer will tell you, is a high-input, high-output, high-risk operation. According to Craig and Collins, if our ancestors were capable of raising and using cattle, then they must also have sustained a sophisticated economy. The only trouble is that the archaeological record does not provide many clues that would allow researchers to distinguish between dairy or beef farming. The jawbone of a steer, after all, is nothing more than the jaw bone of a steer—and they all eat grass.

One crucial clue could lie in the presence of bovine milk proteins in prehistoric storage vessels. Such residues have not, until recently, been detected, but Craig, Collins, and colleagues have used an immunological detection method to unlock the protein secrets from ceramic samples.

FIGURE 1: Grind and Bind.
FIGURE 1: Grind and Bind.
The DACIA system has two stages: digestion (left) where samples are degraded in concentrated hydrofluoric acid (HF) to release the antigen (Ag), and immunoassay, where the antigen is bound by a monoclonal antibody (mAb), which is itself bound by a biotinylated secondary antibody (2° Ab). The biotin moiety binds streptavidin conjugated to an enzyme that converts a colorless substrate into a colored product.
(Adapted from Craig, O. E.; Collins, M. J. J. Immunol. Meth. 2000, 236, 89–97.)
The team points out that their digestion-and-capture immunoassay (DACIA) overcomes the problem of first extracting the protein by simply dissolving the whole sample ceramic in 4 M hydrofluoric acid and detecting the liberated proteins using an immunoassay (Figure 1).

“The method works by simply removing the mineral phase, thus liberating bound organic molecules, which are immediately captured by the special coating on the sample tube,” explains Collins.

The team studied fragments, or sherds, from nine ceramic cooking vessels, which have been dated to the middle of the first millennium BC. They were unearthed from an early Iron Age house in the Outer Hebrides, Scotland.

To find out whether the vessels had ever contained milk, the team raised a monoclonal antibody to bovine -casein, which is found only in cow’s milk. “The captured organics [proteins] are assayed using a normal ELISA technique,” adds Collins. “The method has lots of potential applications, especially as it can be modified to detect a wide range of molecules by altering the antibody.” The team also carried out several reference tests in order to preempt the criticism that would result from a lack of negative controls. Seven of the nine sherds tested positive for casein and, says the team, revealed amounts comparable with sherds soaked with milk and buried by the researchers. No casein was found in the surrounding sediment and this, coupled with a large number of calf bones at the site, suggests that young animals were culled to help maintain fodder for milk-producing cows in this harsh landscape.

The researchers believe that dairy farming during the Iron Age was indeed possible and suggest that the early inhabitants of the harsh, marginal lands of the Scottish Atlantic coast had a surprisingly well-developed economy. The fact that these protein residues are some 2500 years old also highlights the power of the technique for archaeological study, and the team thinks its high-resolving power will allow researchers to determine how other archaeological ceramics were used.

Make No Bones
Ostrom also points out that the structural characterization of proteins from archaeological finds, particularly bones and ancient fossils, is difficult, because the samples retrieved are typically quite small. It is hard to purify the protein that is present, and classical Edman degradation is usually almost impossible.

She and her team have adapted a microbore reversed-phase HPLC method to help them purify small (picomolar) quantities of the bone protein osteocalcin (OC), the presence of which was suggested by gel electrophoresis and radioimmunoassay studies of bone samples—ranging in age from modern to 450,000 years old. OC is often very well preserved in ancient samples because of its strong affinity for the hydroxyapatite mineral phase of bones and teeth. It is important to note that OC is found only in mineralized vertebrate tissues; therefore, contaminators such as microbes, fungi, and plants can be ruled out. The team applied matrix-assisted laser desorption ionization mass spectrometry (MALDI-MS) to the purified extracts from modern cow bones and the fossilized bones of 800- and 10,000-year-old bison. Several rather ancient (up to 53,000 years old) bison bones were also studied.

This is the first time MALDI-MS with peptide mass mapping (PMM) and post-source decay (PSD) analysis have been used to sequence OC from ancient samples. According to the Ostrom team, MALDI-MS with PMM or PSD analysis provides protein sequence information, which means that delineating taxonomic relationships between organisms is possible. Ostrom’s work confirms that OC is indeed preserved both in old samples and in a variety of fossils. The team also suggests that their technique should be equally applicable to other proteins from preserved hard tissues.

Boning Up on Dinosaurs
Web Resources
The Ancient Biomolecules Group at Newcastle maintains a Web site at http://nrg.ncl.ac.uk/research/resareas/ancient-biomols.html

A poster regarding the evidence for prehistoric dairy farming can be found at http://nrg.ncl.ac.uk/news/poster.pdf

Peggy and Nathanial Ostrom maintain a Web site at http://www.glg.msu.edu/people/ostrom/ostrom.htm

Ostrom’s evidence of the preservation of OC stretches the conventional wisdom about this protein back as far as 50,000 years, but Collins and his colleagues believe that success is also possible with bones going much further back than that. He and his Newcastle team, working with biochemists Cees Vermeer of Maastricht University (Netherlands) and Peter Westbroek of the State University of Leiden (Netherlands), demonstrated their immunoassay techniques with OC extracted from dinosaur bones.

This earlier work was contested, however, because there seemed to be a paradox in that the OC in archaeological bone specimens is not always as well preserved as that found in fossils. To resolve this problem, the team heated bone samples to simulate deterioration over geological timescales and found that OC was more resistant to deterioration than had been expected. Moreover, kinetic analysis further suggests that survival of proteins in dinosaur bones cannot be ruled out. “The key seems to be that association of OC with mineralized bone provides a unique preservation environment for the protein,” explains Collins. “When examined carefully, we found that unlike dinosaur bone, most archaeological samples have been attacked by microorganisms, causing rapid alteration of the mineral.”

The gold standard for identifying and exploiting ancient proteins is sequencing. Ultimately, the timely development of Ostrom’s MALDI-MS technique ought to provide researchers with the evidence they need to back up their claims about what prehistoric people ate for breakfast and what types of dinosaurs roamed the earth.

Further Reading

  • Collins, M. J.; Gernaey, A. M.; Nielsen-Marsh, C. M.; Vermeer, C.; Westbroek, P. Geology 2000, 28, 1139–1142.
  • Craig, O.; Mulville, J.; Pearson, M. P.; Sokol, R.; Gelsthorpe, K.; Stacey, R.; Collins, M. Nature 2000, 408, 312.
  • Craig, O. E.; Collins, M. J. J. Immunol. Meth. 2000, 236, 89–97.
  • Muyzer, G.; Sandberg, P.; Knapen, M. H. J.; Vermeer, C.; Collins, M.; Westbrock, P. Geology 1992, 20, 871–874.
  • Ostrom, P. H.; Schall, M.; Gandhi, H.; Shen, T-L.; Hauschka, P. V.; Strahler, J. R.; Gage, D. A. Geochim. Cosmochim. Acta 2000, 64, 1043–1050.


David Bradley is a freelance science writer based in Cambridge, U.K. Send your comments or questions regarding this article to tcaw@acs.org or the Editorial Office 1155 16th St N.W., Washington, DC 20036.

Return to Top || Table of Contents


 CASChemPortChemCenterPubs Page