The field of protein microarrays is moving — slowly at least. Despite complaints from some experts that the same protein chip companies have been giving the same old talks at recent conferences, a number of protein array products have actually made it to market over the last year, or are close to getting there. At least one company, Protometrix, is bucking the trend in the industry toward capture arrays and is set to begin manufacturing a yeast protein chip later this year.
Research reports maintain their bullish outlook for the sector: A report from Select Biosciences that came out last week predicts that the market for protein biochips will grow to more than $400 million in 2007, from $100 million last year.
A lot of companies — among them Zyomyx, Molecular Staging, BD Biosciences Clontech, Schleicher & Schuell, and Beckman Coulter — have developed antibody or capture arrays, most of them for profiling cytokines, which have wide applications for a number of diseases. “Those are the most useful circulating or secreted targets,” said Brian Haab, a scientific investigator at the Van Andel Institute in Michigan. Also, antibodies for profiling cytokines are readily available. However, “once you start working on intracellular signaling, new applications have to become very specialized, and then your market becomes much smaller,” Haab said. In addition, “it takes a long time to validate and develop these sandwich assays.” According to Haab, PerkinElmer is also working on a cytokine array, however the company would not confirm this.
At the recent Pittcon conference in Orlando, Fla., a Zyomyx representative told ProteoMonitor that his company is working on a mouse cytokine chip, expected to come out this summer, as well as a human serum protein chip that would include antibodies to chemokines and growth factors, targeted for the end of the year. In addition, Zyomyx is planning a signal transduction chip to profile components of different signaling pathways.
In the long term, “the real bottleneck is content,” said Thomas Joos, head of biochemistry at the Natural and Medical Sciences Institute in T bingen, Germany. Creating antibodies for capture arrays is mostly a question of cost, he said, estimating that it costs about $5 million to generate 1,000 monoclonal antibodies.
But content is not a problem for all protein chip companies — not all of them are working on capture arrays. Last week, ProteoMonitor caught up with Protometrix of Branford, Conn., a company that was founded almost two years ago to commercialize protein microarray technology it exclusively licenses from Yale University, where it was developed in the laboratory of Michael Snyder (see ProteoMonitor 1-28-02).
According to CEO Hollis Kleinert, Protometrix has just completed quality controls on its first product, a yeast proteome array carrying more than 4,000 different yeast proteins, after scaling up the technology for manufacturing them. At the moment, the company is in discussions with potential worldwide distributors and is hoping to launch the chip later this year.
Although it does not carry human proteins, the yeast proteome chip could have applications in drug discovery, said Protometrix vice president for R&D Paul Predki. About one third of the roughly 6,000 yeast proteins have counterparts in humans, he said, suggesting that their functions — at a cellular level — are conserved. For example, company researchers have taken a yeast homolog of a human protein of unknown function that is involved in an undisclosed human disease and have mapped its interactions with proteins on the chip. The results allowed them to place the protein into a functional cellular pathway, which they are currently validating in a human system.
They also profiled the interaction of small molecules with the chip proteins, a possible way to map targets — intended or unwanted ones — a drug is likely to interact with in the cell.
Finally, the company has used the arrays to profile the specificity of antibodies, and obtained some surprising results that could not have been predicted from the protein sequences alone, Predki said. This application could prove useful for selecting the most suitable antibodies, both for diagnostic tests and for large-scale antibody arrays, he added.
However, not every human protein has a yeast homolog. Also, posttranslational modifications might play a crucial role in protein-protein interactions, and often differ between yeast and human proteins, so “there is a necessity to follow up what you discover in yeast with human systems,” said Predki. Nevertheless, he and his colleagues showed that certain human proteins and their yeast equivalents interacted with a similar set of proteins on the yeast chip.
Protometrix is also working to expand its range of products over the next few years to include not only human protein chips but also arrays with proteins from “animal models that are used for drug discovery and development,” said Kleinert. Rather than the entire human proteome, these arrays will contain certain protein classes, or proteins expressed in a specific tissue.
Besides selling protein arrays, the company plans to offer array-based research services, and to use the chips for its own in-house research in order to create intellectual property.
Since last year, the company has nearly doubled the number of its employees, to 24, and has moved from Guilford to Branford into 14,000 square feet of space.
Protometrix received $8 million in funding from venture capital firms Orbimed and Collinson, Howe, and Lennox in 2001 and another $850,000 from Connecticut Innovations, a state agency for development, at the end of last year. According to Kleinert, the company is hoping to close a series B round by year’s end.
— JK