Palaeoshellomics - proteins from old shells

An Italian team are taking modern science and applying it to the past, both to understand biology better and to solve an important archaeological riddle. From the University of Turin, speaking with Chris Smith, Beatrice Demarchi...

Beatrice - The first question we had was we wanted to know if we could apply new molecular tools to invertebrate calcified tissues such as shells. We really know very little about the biochemical evolution of these organisms. But at the same time we also had a very specific archaeological question from some colleagues in Denmark: they asked me whether I could identify the material, the shell that had been used, to make some special ornaments called double-buttons. Everyone thought that the shell that they used was an oyster, because at that time in Europe - and we're talking about 6,000 years ago - there were different cultures and different ways of making ornaments, and understanding which materials they were using would give us an understanding of the view that they had of the world.

Chris - And you couldn't get that information from anything else? The context and so on wouldn't tell you that?

Beatrice - Actually interestingly the context led us astray if you will, because this was a coastal site, a shell midden of oyster shells. And so we all thought that this was a sensible material to use for these ornaments, and of course that turned out not to be the case. And morphology couldn't help us anymore because these ornaments - they're really really tiny, they've been worked really heavily, in this case they were worked to resemble pearls. And so all the morphological features, they're just gone.

Chris - How did you resolve that then? You've got something where there's not enough morphology to tell you what sort of shell it's come from, you just know it's come from some kind of mollusc or something. So how did you take it forward?

Beatrice - That's right. We only knew that it was shell, and only after we really looked at it with a microscope. We used a series of different techniques that could tell us broadly whether these shells were marine or freshwater, and that was important. The same technique also told us that these shells were likely to have been collected locally. Then we looked at the microstructure using electron microscopy. And then finally we had to develop a whole new strategy based on molecular analysis: we looked at proteins. Proteins in shells are very special; they can survive for thousands or even millions of years, but we don't know much about the sort of sequences that the shells have.

Chris - Now when you say the sequences, do you mean as in what the proteins are actually made of, the building blocks, the amino acids that are in there, and therefore what the gene sequences that the organism would have used to assemble them in the first place? Because obviously if we know what the protein's made of we have some idea as to what the organism was, because we can identify almost genetically.

Beatrice - That's right, yes. By looking at these sequences we hope to get some through genetic information. The problem is that we know very little about the DNA sequences even of mollusc shells. We only have about 15, 20 genomes of mollusc shells; a phylum that has something like 100,000 species.

Chris - I was going to ask you that! Because yes you can get the protein out, yes you can potentially - assuming it's not too degraded  - work out what the sequence of the amino acids is in that; but without a database, a reference database of all of these proteins in all the different molluscs and shellfish and shelled creatures, you're not really much further forward are you. So how how did you get that database?

Beatrice - That's right. We were actually stuck with it for a few years. And then we were very lucky because a transcriptome of a freshwater mollusc was released and suddenly our sequences made sense. After a couple of years of just having sequences without any reference we could map them on some sort of scaffold. And it was very clear that our main proteins were freshwater-mussel-type-of shell proteins, and so we could take the work forward. But the other thing that we did do was to build our own reference dataset. So we looked at modern and archaeological shells from the same sites and we had to develop some fancy bioinformatics approaches for that.

Chris - Serendipity is a wonderful thing in science, isn't it, when it happens like that! What does that therefore tell you about the archaeological question? Because once you managed to get this needle in a haystack and you knew, right, this is a kind of freshwater mussel, does that move you forward?

Beatrice - Well we couldn't really pinpoint them geographically, but the big surprise was that these freshwater molluscs, they were found lonely in this shell midden made of millions of marine shells. So we were on a winner in that sense, that suddenly we had a whole new perspective on our archaeological question. Because in general our expectation is that to make prestigious objects you need prestigious materials, and in Central European sites where these double-buttons are quite common, people used... well we thought that people would have wanted to use exotic shells, so marine shells for example from from the Mediterranean, but our coastal Danish site was the exact opposite, we had freshwater shell probably coming locally. And so this was telling us that wherever we were in Europe, people had a very precise idea on how to make these double-buttons and they didn't really care about whether the material was exotic or not in that sense. They cared about the actual properties of this material. Mother-of-pearl is a beautiful material, it's really hard, it's really resistant, but it's also really easy to work. And so this is giving us a whole new insight really on how people were thinking about materials around their environment.