KIRKLAND, Wash.--In a study published in the February 4 edition of Science, researchers from Rosetta Inpharmatics, Queen’s University, and Mount Sinai Hospital described the use of Rosetta’s FlexJet DNA Microarray technology and Resolver Expression Data Analysis System for genome-wide transcript profiling to track signal transduction pathways in yeast.
Rosetta officials called the project, which was carried out over a two-year period by a team of a dozen scientists, a scientific breakthrough and a "major advance in the field of genomics." The new application for microarray technology, they said, has the potential to significantly expedite drug discovery by providing new information on gene and protein function.
Stephen Friend, president and chief scientific officer of the two-year old company that is known for building DNA microarrays by synthesizing oligos using inkjet printers, remarked, "Until now, it has been assumed that to study all proteins in the cell simultaneously one has to use proteomic approaches such as mass spectrometry. We’ve shown that much can be learned about protein activities
simply by looking at the expression patterns of all genes in the organism."
The company contended that while current methods view signaling pathways as a group of independent, non-cohesive snapshots, its approach "provides the full-featured film of gene expression cascades."
Friend explained, "If you take individual pictures so you look at cells in one state and then another state, all you can do is look at them separately. If you end up taking many snapshots, we’ve learned that you can actually connect those pictures together and know what’s happening within the cell."
Mary Drummond, spokeswoman for Rosetta, which employs more than 100 people, noted that the company not only sells its tools for use by life science industry customers, but performs research in-house for pharmaceutical partners. "Resolver is probably the best tool out there for this analysis and this validates that it is the best," Drummond remarked. She added, "From an array perspective, what makes Rosetta’s technology unique is not just the speed and flexibility with which we can make arrays and decide what we want to put on them, but how we decide what we want to put on them."
Friend, a coauthor of the study, spoke with BioInform on the day it was published in Science.
BioInform: Please explain the significance of the study you have published.
Friend: Until now, arrays have been good at following levels of genes. The reason this paper is in Science is that it is one of the first times when those arrays have been used to follow protein function. This is important because if you are a company interested in finding drugs, you don’t really care what’s going on at a transcript level. This paper was an example by which you could use transcript arrays to do something that people had not expected you could do: to monitor how proteins talk to each other, the signals they send back and forth to each other. This, of course, does not change the transcript of the single gene, but the entire sensors that are in the cell act differently enough that we found that you could go through very classic kinase signaling pathways and monitor what you’d actually done to those pathways.
BioInform: You’ve discovered a new application for a product you’re already marketing?
Friend: Exactly, and it’s one that shows why you would use Resolver, and this is why you would use arrays. It allows us to go out and look for partners who clearly would want to be able to engage this technology and showcases one of the strengths of using Resolver.
BioInform: What does this mean for Rosetta as a company trying to market this technology?
Friend: It means three things for us. It shows the power of using sensitive arrays. It allowed us to showcase how, when you have accurate information from good arrays, you can learn things about the cell you wouldn’t know.
Second, it highlighted using entire patterns, or what we call a matrix approach, to the information, that I think three or five years from now will be the common way arrays are used. That is, not just to look at them as parallel indicators of single genes, but to use the network of signals that you get back as a matrix to give you a signature of what’s happening.
Thirdly, it showed pharmaceutical companies that these approaches that use the Resolver and use the Array are ones that could help them solve the problems that they’re really stuck with. That is, for particular pathways, let’s say the obesity pathway, it showed a way to ask what might be a good target to choose in order to make a drug.
BioInform: Are there other companies with microarray technologies applying them to this type of analysis?
Friend: We don’t know of anyone who has applied their technology to follow the circuitry of signaling pathway like this. That’s where the excitement is.
BioInform: Why was yeast used for this study?
Friend: We did it in yeast because there we could build hundreds, thousands of profiles of different proteins in order to verify that the signatures we were getting were correct. The strategy that was used is one that applies to rodent cells, human cells. It’s the same method, but the particular cell type that we used in this was yeast.
BioInform: Will this method also be applicable for agricultural research?
Friend: Absolutely. The work is directly applicable in terms of using this type of yeast. But you know that the agricultural and agrichemical companies are quite aware of how to use model organisms. So they see this as a method by which to do discovery for products that go beyond what pharmaceutical companies would think about.
BioInform: Now that you’ve proven this method, would it take as long for a customer to repeat the experiment?
Friend: No. Three quarters of that work was learning how to do it, one quarter doing it. It would be something that would take a limited number of people three to six months to do now. It’s not an overnight experiment, but it’s not a shoot at the moon either.
People had not expected transcript arrays to help you follow protein function. This is a big wakeup call to everyone. It’s like someone finding out that the enzymes that you used to duplicate DNA can now be used to do PCR; you always knew you could amplify, but the concept of PCR was really a major one. It opens up an entire field of discovery. No one has ever before said, I’ll use arrays to monitor protein functions. We just said, Yes you can, and here’s how.
Putting arrays together with analyzing many conditions of a cell is important. A single array will not give a clue as to what’s going on in a cell. But to do an experiment where you use a series of arrays and look at a cell under different conditions adds a lot of value as you build up a library of patterns. That library allows you to identify function.
BioInform: Will your research be of benefit to other makers of microarray technology--your competitors?
Friend: We’re hoping that people buy lots and lots of arrays. One of our major focuses is on how you interpret the data. So [the research] highlights our tools for how you interpret the data.
The whole game is going to be a bioinformation game here soon, where the array prices are going to go significantly down and what’s really going to be valuable is not what you learn on an array but what you do with the data. This actually accelerates that, so it’s a nice business opportunity.