Name: Richard Laursen
Position: Professor Emeritus of Bioorganic and Protein Chemistry, Boston University
Background: PhD, University of Illinois at Urbana-Champaign; Fellow, American Association for the Advancement of Science; Honorary Professor, Hunan Normal University.
For more than a decade, Richard Laursen has used LC-MS for textile analysis, developing methods to identify dyes used in historic textiles based on their distinctive sugar patterns.
Last year, Laursen, who retired in 2009, attended the 20th anniversary celebration for the China National Silk Museum where he presented a talk on his textile research and helped arrange a collaboration between his former postdoc, Liang Songping, now a professor of biology at Hunan Normal University, and CNSM researchers to use mass spec-based proteomics to identify different species of silkworm in ancient silk samples.
ProteoMonitor spoke recently with Laursen about the collaboration and more broadly about his use of LC-MS for textile analysis.
Below is an edited version of the interview.
How did you come to arrange the collaboration between Hunan Normal and the CNSM using proteomics to study ancient silk samples?
I met [Zhao Feng,] the director of the [CNSM,] about a year ago, and then last summer one of his people worked in my lab all summer so I sort of got to know what was going on there, and I learned that one of the things they were interested in was trying to figure out what kind of silk had been used [in various archaeological samples].
Silk was developed in China some 5,000 years ago at least, and there is one silkworm that is used now primarily in China... but in ancient times there may have been other ones. What they would like to do is try to figure out by analyzing the silk if other [types of] moths or silkworms have been used. And they have a collection of quite a few types of silk – different ones from India, from different parts of Southeast Asia, reference materials, basically. The idea was to try – if they got an ancient sample of silk – to figure out if it was made by this same conventional silkworm.
And the way they started doing it was by amino acid analysis, which is an old technique I used to use 40 years ago, and I know it doesn't give you much information. But now with the development of proteomics, with very small amounts of material you can figure out what a protein is, get the protein sequence information. And so I thought this would be a perfect way to do the silk thing, because most species [of silkworm] have different amino acid sequences [to their silk fibroin]. If you could chop up the proteins and get a few sequences using mass spectrometry, if you had some reference compounds you could figure out what silk was used or if it was a different [silkworm].
My background is in protein chemistry; for 35 years that's what I did. So I know about proteins and the people at the silk museum really don't, and then I have a friend in China who does proteomics and has a lab there with a lot of nice equipment. I was sort of the middleman to get this protein chemist talking to the silk people.
The protein chemist is Liang Songping at Hunan Normal?
Yeah. He worked in my lab around 20 years ago, so we visit each other fairly often. He is doing proteomics, among other things, on spider toxins. He has the equipment – he has a lot of really nice stuff. I'm actually kind of envious of what he's got. So he's done this sort of thing on other proteins.
Are there proteomic reference databases the researchers can compare these silk samples to?
I think the whole genome has been sequenced [for two species of silk], so the Bombyx mori [silkworm], which is the one that everybody uses, is known, and then there is at least one other species. There may not be more than those two, though. It's a lot of work to do a genome sequence, so unless there's a good reason to do it people won't.
The other thing about this project is you can't just take the protein and chop it up. You can do that with some proteins, but [with] silk fibroin, probably 95 percent of it is a repetitive sequence, so it might be the same for most species. Also, you can't just chop it up with enzymes the way you can most proteins. But there are bits of it that I think you can get to that are non-repetitive. So I guess my contribution is kind of knowing what to go after. Because I think that if you know what to go after using the right techniques you can chop off pieces of this protein and get a small piece and then subject that to the proteomics.
So you would have to do something other than a typical trypsin digestion?
I'm not sure trypsin would be the best thing, but there are some other enzymes that might work that are pretty widely used. Silk fibroin consists of a heavy chain and a light chain, which are connected by a disulfide bond, and I think you could easily separate those two chains and then work with the light chain, which I think would have more information.
Because the light chain is more variable across different species?
Yeah, it's more variable. It's not repetitive at all. Of course, we don't know whether every variety of silk has a light chain, but I bet they do.
How difficult would it be to detect those differences in the light chain via mass spec?
Well, there is a lot of silk, so there's a lot of material you can get, even with an archaeological sample. Suppose you have a couple micrograms; for people who do this work, that's a lot of protein. You consider the whole textile is made out of silk, so one little fiber like that is not very much. I think you could get plenty of material to do this and then if you chop off the light chain – the light chain is only 260 amino acid residues or something, it's very small – you chop that up, and I think it's quite doable. My guess is if everything is set up and going you could probably do it in a couple of days. But no one has done the controls and things like that.
So the real challenge is putting together the reference samples to compare the ancient samples to?
Well, what you would have to do is get three or four reference samples of silk, which is easy to do, because those exist already. Then you have to figure out the best conditions for separating out the heavy and light chains. And when I talked to the people last year they tried some things, but they aren't protein chemists, so I think you have to work out the right conditions. It might take you a little time to work those out, but once you had that then I think you could probably analyze a whole bunch of things very quickly.
What specific questions do the researchers hope to answer with this project?
Well, they get fragments of silk textiles from all over the place, and I think one question is – especially in the old days, thousands of years ago – did they always use the same species of silkworm or were there others? Nobody knows. And I think you could take these samples and it may turn out they are all the same, but maybe not.
So this is an open question in the field?
I don't think anybody has asked the question. Or, you can ask the question, but if you don't know how to go about solving it then you can't do much.
How did you move into this area of research?
In 1999 my research group was getting smaller, and I decided I wanted to do something different, and I had always had an interest in art and so I did a sabbatical at the Metropolitan Museum of Art in New York. Then I came back and a new graduate student came to talk to me, and I told her I wasn't going to take any more graduate students, but it turned out she was interested in art, too, and so we sort of cooked up a project, something we could do that combined chemistry and art. That's when we got into using LC-MS for dye analysis. No one had used LC-MS [for that before.] And it kind of blossomed. We did a lot of stuff using that technique and discovering things about dyes that nobody knew about before.
What sorts of things did LC-MS help you discover about dyes?
Well, one thing is, these are natural dyes, not synthetic dyes, and most of them come from plants. And during the biosynthesis of these colored compounds there usually are present glycosites and sugars attached to them, but the standard method people had used in the past for analyzing dyes was to treat it with strong acid, which took off all the sugars, and so you lost the information that you would get if you analyzed the whole molecule with the sugar attached. To analyze the sugars you can't use the old methods, and this is where the mass spectrometry came in. Using mass spectrometry makes a big difference in what you can learn about these mixtures, so we introduced a way of extracting the dyes and maintaining the sugars and also analyzing them using mass spectrometry.
So one of my goals is to create sort of a fingerprint library. We've collected plant materials from all over the place. We have several hundred of them now. And we can get profiles, fingerprints of all these dyes, so if you have something from a textile you can figure out where it came from.