Myriad Proteomics of Salt Lake City and its new CEO Thomas Ingolia have jumped on the drug discovery bandwagon — and are now riding it full speed. Armed with an array of new and established technologies, most notably a target pull-down approach known as chemiproteomics, Myriad has abandoned its quest to map the entire human proteome and now hopes to carve out a niche in the increasingly crowded field of collaborative target validation and discovery.
“What we’re looking at now is whether our partners’ needs could be met best by adding extra value to [our database] by getting a target identified,” said Ingolia, who joined Myriad last month.
The company, which began as a joint venture among Myriad Genetics, Hitachi, and Oracle in April 2001, originally set out to create a complete database of protein-protein interactions using its yeast two-hybrid and mass spectrometry-based protein complex analysis technology (see PM 3-18-02). It hoped to sell access to the database on a subscription basis.
Two years later, it is clear that this approach doesn’t sell. “If you have all the information, that’s great, but you still need to find compounds that you can work on to develop the drug,” explained Peter Friedli, president of the Swiss public investment company New Venturetec, one of Myriad Proteomics’ investors.
To do this, the company is now pursuing a dizzying variety of target discovery and validation methods, ranging from the use of protein interaction interference agents such as antibodies, RNAi and gene deletion to confirm protein interactions found through the yeast or mass spec methods already in place, to focused yeast two-hybrid assays aimed at finding clinically significant binding partners for beta-catenin, a signal transduction protein implicated in prostate cancer.
Myriad Proteomics’ crown jewel of the moment, however, is clearly chemiproteomics. The method, already part of rival Cellzome’s platform since last year (see PM 3-24-03) and recently added to MDS Proteomics’ platform as well, uses tethered small molecules to capture protein aggregates. A protein source such as cell culture extract is fed through the pull-down assay and a mass spec then identifies all of the proteins in the resulting bound complexes. Both novel small molecules and known drugs can be used as the “bait.” Thus new targets can often be found from known leads. “If you know a phenotype associated with the administration of an agent, but you don’t know what the target is, it’s very attractive then to use the chemiproteomic approach to pull down the protein,” Ingolia said.
Although competitors Cellzome, of Heidelberg, Germany, and MDS Proteomics, of Toronto, also use chemiproteomics, Ingolia insists that Myriad will stand out from the crowd. “The breadth of the approach we are using gives us an advantage to partners,” he said.
At the recent BIO 2003 conference in Washington, DC, Ingolia says he discussed possible collaborations not only with large pharma companies such as Glaxo, Merck, and Aventis, who he says are interested in “backing up genomics information with protein interaction information,” but also smaller companies who could provide Myriad Proteomics with protein interference technology. Ingolia cited Dyax and Cambridge Antibody Technology as potential partners.
Myriad Proteomics has already begun collaborations with two partners to look at specific disease pathways, although no leads have yet been found. In work under a nearly $500,000 research contract for The Hereditary Disease Foundation, the company is screening libraries for agents that would block protein interactions that cause neurodegenerative diseases such as Huntington’s and Alzheimer’s. The company is also working with Howard Hughes Medical Institute investigator Stanley Fields, the inventor of yeast two-hybrid technology, to identify protein interactions formed due to infection by the malaria parasite Plasmodium falciparum. As for its work on the role of beta-catenin, for which Ingolia says Myriad has already identified “dozens” of proteins involved in binding, there are not yet any takers.
For now, Ingolia insists, no takers is still no problem. He emphasized that Myriad Proteomics still has “well over two years of cash at our current burn,” citing that $30-40 million remains from the initial 2001 investment of $70-80 million, with a burn rate of $1 million per month. Ingolia expects more revenue to come in within six to 12 months, and hopes to have financially beneficial partnerships ready in the next 18 months. He confirmed that he is pursuing undisclosed academic grants as well as Defense Advanced Research Projects Agency and Advanced Technology Program grants from the US government.
Still, becoming profitable will not be easy, says James Featherstone, head of pharmaceuticals and biotechnology consulting at Wood Mackensie in London. “I speak to a number [of companies] every week who are starting out with a fee-for-service model and whose wish it is to use that as a platform to moving into drug discovery,” Featherstone told ProteoMonitor in an e-mail. “It is early days to say if any of the proteomics companies will make it as drug discovery houses.”
— KAM