AT A GLANCE
Name: Raj B. Parekh
Position: Chief Scientific Officer, Oxford GlycoSciences
Prior Experience: Founding Research Director of OGS, a
spin-off of the Oxford University Department of Biochemistry
How did OGS first become involved in proteomics?
Originally the interest was in the area of glycobiology, which is still a core area of expertise at OGS. The interest was in doing two things: finding new technologies to analyze proteins and protein carbohydrates, and understanding how glycoproteins change in human disease. So it was both a technology focus, and a molecular medicine focus. Out of this came a series of techniques that allowed us to make the study of protein carbohydrates quite systematic, in a way that they hadn’t been done before, and it was on that basis that we actually founded OGS. So OGS was founded initially as a platform technology company, devoted to producing instrumentation that would allow relatively straightforward analysis of protein glycosylation.
However, the business didn’t grow; it didn’t seem to be quite as large as we had anticipated early on because the early ‘90s were really focused on DNA and gene sequencing. A lot of the capital equipment budgets were consumed [by that]. In 1995 we decided to expand the technology platform beyond glycobiology into what is now proteomics. Looking around we felt the most productive way of building a meaningful proteomics platform was to improve the 2D gel technology and integrate it clearly and coherently with mass spectrometry. That was the first platform that we built, and it took us about three years to do it. It was on that basis that we built a revenue business, through partnerships with Pfizer, Incyte, and others, and listed the company on the London Stock Exchange.
In 2000, we decided that we were ready to go beyond the initial focus of the company — which was platform technologies and a service business — into using the same technologies to build a protein patent estate that would serve as the basis for a therapeutic pipeline, and also to leverage the glycobiology expertise into clinical candidates. That’s the third phase in our evolution.
How did you build expertise in 2D gels?
There were two elements to it. The first was that in our business model, as a carbohydrate instrumentation and reagent and services business, we had had to build quite a lot of infrastructure and expertise in engineering, biochemistry, chemical engineering, and in automating, improving, and turning [biochemical techniques] into processes. So we had a lot of the expertise in-house. Secondly, we decided to acquire the 2D gel business from Millipore so as to get up the learning curve most quickly. Acquiring the business from Millipore helped us understand what the merits, and indeed what the deficiencies of 2D gel technologies were, and we had an in-house R&D team that was capable of addressing those difficulties. So one of the first things we did was come up with a process that allowed us to covalently back the gels onto glass, so that instead of manipulating the gels, which is a troublesome business, we manipulate plates of glass. That makes the processing, barcoding, archivability, throughput, and reproducibility all much easier. The team also decided that silver staining and Coomassie — the conventional ways of detecting proteins — were inadequate, so we designed, devised, and implemented a fluorescence-based technology. And in this way we were just able to work our way up the process, starting with [detection methods for] 2D gels, image generation and analysis, automation for recovering proteins, proteolysis, and interfacing that to mass spectrometry. It was about a four year period. It was a fairly lengthy, time-consuming, and expensive campaign, but the kind of thing that small companies can do in a really focused way.
Were there others at the time following the same route?
A couple of years after we started, proteomics started to become all the rage. There’s now a lot of companies. Just as with genomics, there [are] a small number of early pioneers, and then there are a large number of second- and third-generation companies. Most technologies don’t allow there to be too many companies. I think we’re seeing that now; there are a lot of companies that are just surviving, [and there is] a lot of consolidation taking place. As with genomics, I think we’ll end up with a few dominant companies who will fill the spaces that need to be filled. One is clearly provision of instrumentation and materials — I think companies like ABI, Bruker, and Amersham are establishing a position there. There’ll be other companies that are contract service providers, and I think the most valuable group of companies will be those that have truly been able to use proteomics and genomics to build therapeutic product pipelines. Because there’s nothing, nothing like a new therapeutic entity, or a clinical development pipeline, to build shareholder value.
How wide a lead do you think you have over other companies?
We’re on phase two and three of our evolution as a company, when I think most people are on phase one. So if I think of building the technology platform and validating it through corporate partnerships — that’s phase one. If we then think of phase two as [transforming] that technology platform into an IP position, and if we think of phase three as turning the IP position into clincial candidates — a drug pipeline — then we’re focused solely now on phase three. Most of the other proteomics companies are still at phase one. If you look at GeneProt, they’re still setting up their factory. If you look at [Large Scale Biology] and MDS Proteomics, that’s where they’re at. So I think we just had a first mover advantage, and a temporal advantage that’s allowed us to move on down the line. There’s no doubt that these other companies — first class companies — [will] be able to move this way too, but I think just the fact that we started so much earlier has allowed us to mature ahead of them.
Will we see companies licensing your 2D gel technology, as GeneProt did recently?
There are relatively few companies that have the kind of technology patent position that we do. Therefore to the extent that companies want to get into contract proteomics, or into selling materials for 2D gel-based proteomics, they’ll run into that patent. We don’t see enforcing that patent as the main value driver at OGS. It’s just a sort of fallout of having been pioneers in this field. Having said that, patents are important and we will defend it. But the key value for us now is in proving that proteomics and glycobiology — the two enabling technolgies that we have — that those two technologies together can really allow us to build a powerful pipeline of new medicine.
How long do you have to wait before your proteomics platform is responsible for new medicines?
There are compounds going into clinical trials over the next 12 months that will have come out of our proteomics platform, and I think that will be something that’s been done much more quickly than any of the genomics companies managed to do. Drug discovery is all about decision making: which target to take, which of the targets are involved in a pathological process, and how you want to screen them. There’s no perfect information that allows those decisions to be made as a mathematical science. It’s really the quality of the information that one has and the breadth of the information at each stage of the decision-making process, and we feel that a drug discovery platform, based on proteomics, but including all these additional technologies and processes, has a fairly high content of valuable information through which one can make the kinds of decisions one needs to.