After mulling over microarrays in sunny San Diego two weeks ago, scientists had a chance to continue discussing the “pitfalls, prospects, and promises” of protein and peptide arrays in frosty Munich last week at Cambridge Healthtech Institute’s “Proteins to Profits” meeting.
Judging by the plethora of competing technologies presented, it is too early to say which company — if any — will become the Affymetrix of protein arrays and realize the profits. In contrast to the DNA array field, there are many different assays, ranging from miniaturized sandwich immunoassays to antibody specificity tests on denatured proteins or assays for interactions between native proteins and ligands.
“There is profit from the DNA arrays and interest in the protein arrays,” said Jörg Hoheisel of the German Cancer Research Center (DKFZ) in Heidelberg. “But in terms of excitement, it is the time of protein arrays.”
The excitement comes partly from the fact that a number of technologies in the nascent field have moved from mere concepts to real applications, according to several participants. But a number of significant challenges remain. A common complaint, both in the talks and in panel discussions, was the lack of good capture reagents — antibodies and other binding molecules. Expressing and purifying proteins — especially membrane proteins, the main class of drug targets — in a functional form with correct post-translational modifications is the other big challenge. Following is a summary of some notable talks.
High-Density Protein Chip?
Ian Humphery-Smith, of the University of Utrecht and Glaucus Proteomics in the Netherlands, gave an update on the “human proteome look-alike” chip, which should contain at least 100,000 proteins in a few years, and a human antibody chip, which he said could take up to seven more years to develop. So far, he said he had arrayed about 1,000 different recombinant proteins on a chip, hoping to get this number up fivefold soon and to test it in drug screening. Using a parallel immunization strategy for making monoclonal human antibodies, he predicted he will be able to keep 100,000 hybridomas in cell culture by this fall.
Meanwhile, Dolores Cahill from the Max Planck Institute of Molecular Genetics in Berlin reported that she has already made a chip containing over 10,000 different but denatured human proteins expressed in bacteria. In a proof-of-principle study she detected auto-antibodies from sera of patients with autoimmune disease binding to the arrays. Another application of the arrays could be the characterization of antibodies, Cahill said. Protagen, a Dortmund, Germany-based company Cahill co-founded, is currently looking for a partner to make the protein arrays commercially available.
Mike Taussig, head of the Technology Research Group at the Babraham Institute in Cambridge, UK, offered a cell-free alternative to the traditional method for making proteins. He talked about protein in situ arrays (PISA), which are based on cell-free protein expression that does not even require any DNA cloning. Cambridge, UK-based Discerna, a company founded by Babraham Bioscience Technologies and KS Biomedix Holdings last June, is currently trying to commercialize PISA, which uses a microtiter plate format at the moment, and combine it with ribosome display, a protein expression and screening technology, for library screens.
Another talk introduced an unusual technique for transferring proteins to arrays: electrospray deposition (ESD). Victor Morozo, of the Russian Academy of Sciences near Moscow described this method, wherein a protein solution is dispersed into small droplets which dry in the air; the protein then passes through a dielectric mask onto the array substrate. Two articles to be published this year in Analytical Chemistry and in the Journal of Biochemical and Biophysical Methods describe first applications of these arrays in immunoassays and in screens for the binding of small radio-labeled metabolites.
In still another twist on protein preparation methods, Cambridge, UK-based Sense Proteomic’s co-founder Jonathan Blackburn presented the company’s expression cloning approach, called COVET, which adds a tag to the protein and allows for rapid and parallel cloning. The company launched its first two chips this year — a P53 array and a potassium channel array containing beta-subunits of Kv channels.
Addressing the challenges of making antibody arrays, Hoheisel pointed out some problems he has encountered while creating cancer-specific arrays. For example, he said he has had difficulties in creating highly specific antibodies, spotting them onto arrays, and keeping them functional. So far, he said, he has arrayed 90 antibodies to cancer-associated proteins that allow him to capture protein complexes. Hoheisel is also developing arrays of peptides synthesized in parallel and mentioned converting a light-induced oligo synthesis system marketed by Mannheim-based febit into a peptide-synthesizer.
Array of Early Applications
Markus Templin of the Natural and Medical Sciences Institute in T bingen, Germany, discussed a number of protein array technologies, some of which he and his colleagues are co-developing with other companies. Like Cahill, he presented data applying chips for the diagnosis of autoimmune diseases. The institute is also collaborating with Swiss Novartis spinoff Zeptosens in developing protein arrays that employ Zeptosens’ planar waveguide technology for analysis of protein binding.
Several protein array companies said they had chosen cytokines, which can serve as markers for a variety of diseases, as a training ground to test their miniaturized ELISAs. PerkinElmer Life Sciences chose cytokines from patient samples to test its Packard HydroGel coated slides, which have been on the market since last fall, and found them to be comparable to ELISAs, according to company representative Robert Cavallo. Zyomyx, of Hayward, Calif., which is hoping to come out with its first antibody chip early next year, profiled a panel of cytokines from human blood as well.
Meanwhile, Michael Egholm of New Haven, Conn.-based Molecular Staging presented evidence showing that its rolling circle amplification technology increases the sensitivity of a cytokine assay. The company has defined sets of cytokines for cerebral palsy and autism and will publish some of the results of its research soon, said Egholm. Both disorders are difficult to diagnose early using current approaches, and Egholm told the audience he has a personal interest in seeing this change: his 3-year-old son suffers from cerebral palsy.