At A Glance
Name: Peter Juhasz
Age: 48
Position: Director of Proteomics Research Center at Applied Biosystems in Framingham, Mass., since 2000
Background: Senior scientist at PerSpective Biosystems from 1994 to 1997, when company was acquired by Applied Biosystems. Has remained at Applied Biosystems until now.
Research scientist and postdoc at Massachusetts Institute of Technology, 1990-1994.
Postdoc University of Alberta, Edmonton, 1989
PhD and postdoc at Central Research Institute for Physics, Budapest, Hungary, completed 1988.
You started out with a degree in physical chemistry from Hungary. How did you get into proteomics?
It was in 1995 when there was this conference, an annual genomic sequencing conference, and at that conference I saw a couple of presentations from Steven Fey’s group from Denmark looking at the effects of gene knock-outs on 2D gel protein maps. That was the time when the technology captured my imagination. We were working at PerSeptive Biosystems developing mass spectrometry and the core technology for identifying those proteins from the gels was mass spectrometry. I just realized that this is a great application and the company should focus on these developments.
What is the relationship between PerSeptive Biosystems and Applied Biosystems?
PerSeptive Biosystems was acquired in 1997 by Applied Biosystems
How did you get involved in setting up the Proteomics Research Center?
So at that point [in 1997], the major revenue for PerSeptive Biosystems was development and sale of mass spectrometers. The entire concept of developing mass spectrometers by focusing on applications - that was a philosophy that was championed by Steve Martin. Actually he was my boss at that time, the director of R&D at PerSeptive. We realized that the proteomics as an application will be a major driver of the mass spectrometry business.
At that point, what were you working on at Applied Biosystems?
Basically, we worked on doing instrument development with an application focus. We were trying to practice by working on DNA applications such as genotyping applications. At that point, using mass spectrometry for DNA sequencing was seriously considered as an attractive method. The applications in question for mass spectrometry were mainly DNA applications for genotyping and DNA sequencing. Then after 1995, it clearly started to shift towards protein applications and proteomics.
How were you involved in developing these new applications?
We developed a new technology at PerSeptive called delayed extraction that made PerSeptive Biosystems the market leader in the MALDI-TOF. Delayed extraction is a particular way the electronics operates — it just dramatically improves the performance of the instrument. And the combination of those two — those really had a huge impact on the mass spectrometry of DNA. But what you realized in the middle 90’s was that clearly proteomics was the emerging application, not DNA analysis.
What would you say are some of the most popular tools at Applied Biosystems?
Here’s a little historical perspective — at that time [in the mid-90’s], the MALDI-TOF was the technology of choice for analyzing gel spots from 2-D gels, where the mass accuracy and the resolution of these instruments was just perfect for the applications. You needed very high mass accuracy, and the MALDI-TOF could deliver that. Those were the most important early developments, adapting this so-called peptide mass fingerprinting method to the new delayed extraction technology. So since then, I would say that the most important innovations at Applied Biosystems are the commercialization of various protein quantification technologies. First ICAT, which is a cystein-specific reagent. It facilitates a binary analysis of different samples, like healthy versus diseased or any biological system in two different stages. And more recently this year, the chemistry called iTRAQ — that is a general purpose, an amino-reactive chemistry that labels every peptide on the N-terminus and the lysines. This reagent allows for comparing four different samples in a single experiment. It’s a four-plex.
Are people using the iTRAQ now more than the ICAT?
It is rather a complementary technique. So whereas ICAT targets only a specific class of peptides, and it can reduce complexity of the sample, enhancing the dynamic range of the analysis, iTRAQ doesn’t reduce complexity, but increases the sequence coverage of proteins and improves the statistics of quantification and allows for comparing four different samples in a single experiment. Since it is a generic reagent, it allows the quantification of not only proteins, but also the levels of modifications on proteins, for example stoichiochemistry of phosphorylated peptides. In general it’s accurate to say that the applications in proteomics by Applied Biosystems are mainly focused on protein quantification. This is brand new. The iTRAQ was commercialized last spring. ICAT has been around for over two years.
In terms of quantification, are you looking to develop something even more advanced?
The core requirement of biomarker validation and discovery is to analyze as many samples parallel as possible. We are pushing for higher multiplexing as long as we can.
Aside from reagent techniques, what other products are most popular?
The major source of revenue is actually instruments. So we have more recently two major instrumental developments — the hybrid mass spectrometer called the Q TRAP, which is a hybrid, quadrupole linear ion trap mass analyzer. In addition to being a very sensitive MS/MS technology, it allows for all the advanced scanning functions that people were familiar with on the triple quadrupole instruments. For instance, if you’re looking at a very complex sample of phosphorylated peptides, there is a scanning function of this instrument that can look at every peptide that has a phosphate group on it. It has a scanning function called the precursor ion scanning that can detect a phosphate group. And then the instrument can look at which peptide that phosphate came from and can sequence that peptide. So you can look at specific classes of peptides, for example phosphopeptides and you can detect them from very complex systems where otherwise you can’t find these peptides because in just straight MS-mode, when you just measure the masses of the components in the mixture - the signals from these peptides are greatly suppressed. That instrument was introduced at year and a half ago in the summer of 2003 and it has been very popular.
What about other instruments?
The other instrument that was more recently introduced is the 4700 instrument, which is a MALDI-TOF/TOF instrument. So it’s an MS/MS instrument. You may remember that MALDI-TOF was the instrument of choice for analyzing 2D gel spots. Now in addition to measuring the masses very accurately, this instrument can also sequence those peptides in a very efficient way. The combination of the MALDI MS/MS with HPLC is gaining popularity. It gives a good flexibility to analyze peptide separations deposited on MALDI targets. In total at Applied Biosystems, there are a total of five MS products, including those that are sold mainly for drug metabolite studies.
Who would you say are your biggest competitors?
In proteomics/mass spectrometry, it’s Thermo Finnegan — they have been very successful with their latest product line LTQ-FT — the combined ion-trap Fourier transform instruments, and the other company, there is Waters micromass. They have some really innovative chromatography products lately introduced. And on the MALDI area, it’s Brucker. I would like to make the point that where we put our money, where we think that we’ll be successful, is with the multiplex protein quantification strategies. That technology will help us to sell instruments. This technology is not working on Finnegan instruments. We think that the quantification sets us apart.
Is Applied Biosystems also working on software and informatics?
Yes, Applied Biosystems is marketing Celera Discovery Systems. That’s an informatics repository of information accumulated by Celera Genomics on their human, mouse and rat genome sequencing efforts.
What do you see in the future as the area of development that the Proteomics Center will concentrate mostly on?
The most important area is in robust data validation so that really the quality of protein IDs and the protein quantitation numbers are as robust as it can be. Protein identification as it is practiced today is a very high error-rate process, and it requires real mass spectrometry and data processing expertise to find your way in the jungle of the data that comes out of a proteomics experiment. I think it is most important that biologists who are not expert in mass spectrometry or data processing can have robust data processing systems that always gives an error-free answer, like an automated way of giving qualitative and quantitative results.