At A Glance:
Name: Edward Dratz
Age: 63
Position: Professor of Chemistry and Biochemistry, Montana State University
Founder of Montana State University proteomics center
Prior Experience: Professor of Chemistry and Biochemistry, University of California, Santa Cruz, 1969-1986
Postdoc at MIT in Molecular Biology, 1967-1969
Postdoc at University of California, Berkeley, in Biophysical Chemistry, 1966-67
PhD in Chemistry, UC Berkeley, 1966. Worked with Nobel Prize winner Melvin Calvin.
BS in Chemistry, Carleton College, 1961
How did you get into the field of proteomics?
I had been working on membrane signal receptor proteins, particularly rhodopsin, for about 35 years. Working on signaling proteins, we found that they function by coupling to other proteins and that we can block these protein-to-protein interactions with peptides. We had to do peptide mass spec to characterize the synthetic peptides we used because there are an awful lot of ways that a peptide can go wrong when you try to make one. Then we started getting into heterologous expression of proteins in one cell type versus another. We needed to use mass spec to characterize these proteins.
So as our rhodopsin system [was] starting to get a little mature, I decided to step back and start looking at signal transduction in a more global way. Since I’m pretty familiar with all the mass spec tools, I thought I was in a pretty good position to get a little more general and global than I had been doing all this time. It seems like kind of the next level.
And so now you’re looking to establish a proteomics center?
Yes. Now we have approximately 10 proteomics projects on the campus that have either started or are very soon to start. We are starting to pick up some momentum and we’re needing a support structure. We just got a big boost from an NSF major research instrumentation grant and [a] private foundation grant, and so we’re about to acquire a lot more equipment supporting proteomics and more staff — altogether we’ll get about $750,000. Like all state universities these days, the university doesn’t have a lot of resources, but they are doing what they can. The campus has a lot of active biological research, so it’s expanding in genomics and bioinfor-matics — all the good things that are taking hold these days and helping to drive an explosion in information in biology.
What projects will the center work on?
One of the big strengths on the campus here is in developmental neurobiology. We also have a thermal biology institute on the campus, which studies hyperthermal organisms, largely from Yellowstone Park, which is close to us. And we have a biofilm engineering center on the campus. Most bacteria in the environment don’t live as individual bacteria. They live as films, collections of bacteria. Some of the films are important medically because they infect implants and form things like tooth plaque. These communities of bacteria are much more resilient than individual bacteria. There are several groups trying to figure out the mechanisms by which the bacteria communicate with each other and how these interactions could be disrupted — and also the mechanisms by which they become more resistant to antibiotics.
My own personal big interest is working on developing methods to revitalize a classic 2D gel methodo-logy. One basic approach to proteomics now is to leave the proteins intact and try to separate all the proteins [from each other], and the major technique for that is 2D gels. But the other big thrust is to digest the proteins into peptides and try to separate all the peptides and identify them, and this second method has, I think, got too many limitations. What we ultimately want to do is understand the posttranslational modifications of the proteins. So we’d like to focus in on a relatively few number of proteins and try to get as close to 100 percent coverage of those proteins as we can. I think that it’s important to keep the proteins whole until very late in the game, so you can, first of all, get different forms of the protein. A typical bacterial protein can have two or three posttranslational modifications; a eukaryotic protein might have four, five or 10 or 12. We’d like to get these different forms. Everyone who knows the field knows the limitations on 2D gels, and so we’re working on methods to overcome those limitations.
Can you describe the methods that you are developing?
Our big thrust is developing some new detection dyes that are much more sensitive [than current dyes] and have better properties for 2D gels. The idea is to use many different colored dyes that are very similar chemically, and then try to get lots of information off [of] single gels. We’d like to look at proteins from cells grown under four to five different biological variables all on the same gel and stained different colors. The goal is to get a lot more information out of one experiment and eventually also be able to get specific stains for different kinds of posttranslational modifications that are sensitive enough to be useful. Using the dyes, we’re so far looking at protein sensitivity increases of several hundred fold, and that’s an important number because right now, classical 2D gel methods really only see about half the proteins that are there and the lower-level ones being missed are often the most interesting ones. So by bumping up the sensitivity several hundred fold, we’re starting to get at those other proteins. We think with this technique we’ll be able to get down to detecting proteins that have just a few copies per cell. Also, most people these days are still running proteins from different conditions on different gels. With this, gel-gel reproducibility is a problem. This method overcomes that.
Basically what we then want to do is to take these gels, identify interesting proteins, and then efficiently recover those proteins from the gel. The other goal is to then get them very efficiently into the mass spectrometer. The mass spectrometer has got plenty of sensitivity to see these low levels of protein, but the recovery between the gel and the mass spectrometer hasn’t been as good as it could be. So ultimately, one would use the 2D gel to see which proteins are interesting, and then you isolate those proteins from the gel and digest them. You then recover the peptides, and study those with LC tandem MS. So, out of the 10,000 proteins that might be in your sample, you focus them down to the few or few hundred that look like they might be the most interesting. And then ultimately you’d like to get protein-protein complexes out of these proteins, so we’re also working with various ways to more efficiently get antibodies against proteins. We’re not developing new technology there, we’re just using random antibody libraries to rapidly get antibodies against relatively small amounts of protein.
What do you hope to do with the new technology?
Ultimately, my personal interest is more to focus on the biochemistry of health and nutrition rather than so much the biochemistry of disease. Probably only 10 to 20 percent of human health problems have genetic origins, and the rest have to do with lifestyle and nutrition and things of that nature, so that’s what I’m interested in — what can we say about how to understand each individual’s nutritional needs and how to optimize them.
Our initial experiments revolve around the brain. Omega-3 and omega-6 fatty acids are two families of fatty acids. Our diets are flooded with too much omega-6 and we don’t get enough omega-3. Omega-3 comes from green plant chloroplasts. So fish have a lot of omega-3 fatty acids because they are at the top of the food chain that starts with algae. Grass-fed animals — cows, chickens, etc. — that eat a healthy grass diet have high levels of omega-3 fatty acid and low levels of omega-6. But seed oils like corn and soy and safflower and sunflower are almost entirely omega-6. So when animals are taken from grass-fed range conditions and into feed lots then their fatty acids change and become corn-like instead of grass-like. And of course we use a lot of refined oils in food. So basically our brains desperately need omega-3 fatty acids, and there have been studies of rats, monkeys, even some in humans that show that learning and memory is improved by higher levels of omega-3 fatty acids. What’s behind this is that the membranes in the synapses in your brain are extremely rich in omega-3 fatty acids. So if you don’t have enough of them, then that impairs the performance of those synapses. We’re collaborating with a group at the NIH that has been raising rats on high and low levels of omega-3, and they found learning and memory changes in these animals. They are supplying brains and other tissues to us, and we’re looking at the proteins that are being affected by these fatty acids. Omega-3 fatty acids appear to actually control protein expression. We’re trying to sort this all out and figure out what the mechanisms really are, and what the optimum levels are. If we make progress on the fatty acid front, we’ll branch out to other nutrients and try to build a body of knowledge to understand better how to optimize health. That’s a big goal.
Aside from improving 2D gels, do you see any other big technological challenges in the field?
They’re everywhere you turn. One of the big problems is characterization of posttranslational modifications. There’s a tremendous amount of work going on in this area, but it’s still pretty dicey. Also, there’s a need to get high affinity antibodies in a more high-throughput manner for protein-protein complexes. Probably the biggest void is bioinformatics. We want to take the information we’re finding and kind of hang it on a framework to better understand how it all works. So I think that the biggest deficiencies right now are software and models to understand the metabolic and genetic regulation itself.