Assistant professor, Department of Biochemistry & Biophysics
University of North Carolina at Chapel Hill
At A Glance
Name: Christoph Borchers
Position: Assistant professor, Department of Biochemistry & Biophysics, University of North Carolina at Chapel Hill, since 2001. Faculty Director of UNC Michael Hooker Proteomics Core Facility.
Background: Staff scientist, research fellow, visiting fellow and guest researcher, University of North Carolina Chapel Hill, 1998-2001.
Visiting fellow, laboratory of structural biology, National Institute of Environmental Health Sciences, 1997.
PhD in analytical chemistry/biochemistry, University of Konstanz, Germany, 1996.
Christoph Borchers is scheduled to give a talk at next week's Peptalk conference in San Diego, on a peptide chip-based technology for the absolute quantitation of cancer-related proteins involved in the signaling pathway of breast cancer. ProteoMonitor spoke with Borchers to find out more about his background and work.
What is your research background, and how did you get into proteomics?
In 1991, I joined a group at the University of Konstanz in Germany. They had mass spectrometers, and they were using this to study proteins and peptides. We called it biological mass spectrometry.
When I came into this field, I was using mass spectrometry to study the tertiary structure of proteins. Then, during my PhD time, I used mass spectrometry to identify a drug binding site in a protein. This was a collaboration between the university and a pharmaceutical company called Altana.
I used a combination of photo-affinity labeling and mass spectrometry, and I figured out that the binding site. That was pretty much the end of my dissertation.
I then came to the United States and did a postdoc at the National Institute of Health Sciences. After that, I had several offers in Europe, and they also offered me a staff scientist position at the NIHS. One of my responsibilities at the time was to build up a proteomics facility. That was in 1999.
Then in 2000, my chairman called me up from the University of North Carolina. He asked me if I wanted to come in and give a talk and said they were looking for people. It worked out — he offered me a job, and one of my responsibilities was also to start up a proteomics facility, really from scratch: there was nothing there.
We did it, and we did a good job, and since the middle of this year, Duke University actually joined us. So now we have a joint UNC-Duke facility. Duke actually closed their facility, and now Duke investigators have joined us, so we have joined forces and combined strength.
Also, I'm an assistant professor here in biochemistry. I have several interests. One of the fields I'm interested in is quantitative proteomics and, in particular, clinical proteomic diagnostics.
We developed a peptide chip technology that allows us to absolutely quantify proteins. Except we are actually not using proteins, we are using peptides. This peptide chip technology is a combination of immunoaffinity and mass spectrometry. What we do is we use standard immobilization techniques to immobilize antibodies on beads. The antibodies are not against intact proteins — they are against specific tryptic peptides. Then when we want to quantify the amount of peptide in the sample, we digest the entire proteome, incubate that with the immobilized antibody, and pull it out.
Once we have the epitope peptide — if we add a synthesized peptide that is isotopically labeled, then we have a standard. This is the technology. Then we add the standard. We know exactly the amount of the standard, so based on the ratio of the intensity of the two ions of the doublets, we can quantify this. The quantification is at least linear over two orders of magnitude.
We're using MALDI-MS because it's fast, and with MALDI TOF/TOF, we can sequence the peptide that we pull out. So this gives us absolute specificity. These are the main advantages of the technology: it's fast, it's pretty inexpensive, and you have great specificity, and you can use it for absolute quantitation.
And also it's not restricted to any protein class because there are always peptides, even in membrane proteins, that have great solubility and which have great sensitivity in the MALDI mass spec.
Why were you interested in absolute quantitation, as opposed to relative quantitation?
Because absolute quantification really allows you to compare numerous samples. It's very tough to get the right control. How do you define the state of good health, for a 'healthy' control? So in this case, what we can do is just get a number out — we can say, 'so many femtomoles or micrograms of this protein are in there.' Then we can compare the concentration to other samples.
It's like when you do a blood test. You get a concentration out. You're not comparing something. This is pretty much the same principle as what we do.
What kinds of diseases are you investigating using this technology?
I have a small grant using this technology in the biodefense field. We are developing right now a diagnostic tool to detect Francisella tularensis, which is one of the Category A pathogens. A is the most dangerous type of pathogen.
We can absolutely quantify this bacteria. We can tell [researchers], 'OK, in your sample you have this many bacteria.' We are also very specific, because we have a peptide from one of the proteins in this bacteria, which is absolutely unique. So we can use a specific antibody against this peptide, and add this standard, and we can tell how many bacteria are in there, in direct, quantitative numbers.
What we also want to do next is to use this technology for quantifying the expression and the modification levels of proteins which are involved in the signaling pathway of breast cancer development. What we're doing is we're making antibodies against all the proteins which are involved in the AKT pathway, which is a common signaling pathway that is involved in breast cancer development.
And not only that — we're also making antibodies against all the phosphorylation sites that are involved. This is something you can not do with RNA. You can not quantify and detect post-translational modifications.
Sometimes the protein expression level will be the same, but kinases start phosphorylating these proteins like crazy, and the pathway is then activated, but the expression levels of the protein are pretty much the same.
So we wanted to use this technology to quantify the changes in the modifications of the protein.
Ideally, what we want on the chip is [antibodies] that cover all the AKT pathways. This also can be used for drug development.
We also have, together with a company here — Biomachines — a small business innovation research grant funded by the National Cancer Institute's Early Detection Research Network program. With this company, we're developing an automated robotic system that spots these beads on the MALDI target.
Again, we're using peptides because the detection sensitivity of peptides compared to proteins in a mass spectrometer is much better. Also, we can much better sequence them.
What are you concentrating on for the future?
We have right now an FTICR mass spec, which I was awarded this year by the NIH. These are the mass spectrometers that have the highest accuracy and highest resolution. We purchased this machine from Bruker, which has so far manufactured only four machines of this kind in the world. So the sticker price of this machine is over $2 million. There is one of [these machines] in Japan, one in Germany, and two in the United States: One is on the West Coast, and UNC has the second one.
We're using this high-end instrumentation for all kind of proteomics — bottom up and top down — and as well for metabolomics. We have shown that you can detect hundreds of metabolites without the necessity of an LC system in front of it.
The resolution of this machine is phenomenal — it has a resolution bigger than a million, compared to a Q-TOF, which has a resolution of about 20,000. What we think is we want to use this for high-throughput metabolomics. If we don't need LC, we can reduce the analysis time dramatically. For example, our collaborator will take a batch of 25 samples, and it will take them two full days to analyze. With this machine, it's done in less than an hour.
We also use the machine to analyze blood samples, but blood is too complicated, so you need to have LC/LC beforehand. The high resolution does allow us to deal better with the complexity.
For our proof-of-principle experiment using this machine, we analyzed blood samples from cystic fibrosis patients before and after therapeutic treatment. What we want to do is find an early marker for infection, using proteomics. Ideally, there should be a test like the one for diabetes. In diabetes, you have a small needle that takes blood, and then the test checks out the person's blood sugar, and if it's too high or too low, they get something.
We want to have the same idea with CF — if an infection is coming, you give the patient antibiotics. Because you can not have the patient on antibiotics all the time, or they will develop resistance.
We just got recently got a grant to develop this from the Cystic Fibrosis Foundation. I proposed using the FTICR to analyze this, because our preliminary results have shown that we can detect using LC/LC beforehand over 10,000 peptides. We have labeled every peptide before and after infection, and about 20 percent of these are differentially expressed by greater than a factor of 10.