Japanese polymer giant Mitsubishi Rayon is working to weave its plastic fibers into the microarray market. The company is planning to launch its first product, a fiber-based custom DNA array for gene expression analysis, in Japan by the end of the year, and overseas in 2003.
Like Motorola and Corning, which made brief advances into the array arena but both backed out, Mitsubishi's core business lies outside of biology - in chemicals and plastics as well as various kinds of fibers. And just as Motorola said almost two years ago that it was looking to build its biochips business as a "third leg" of the company, Mitsubishi is also considering life sciences business, including microarrays, as one possibility for a "third pillar" of its business strategy - according to Hirofumi Uno, general manager of Mitsubishi Rayon America.
But two factors set the company apart from the other large corporate entrants into microarrays, potentially giving it a leg up in its endeavor: For once, unlike the main microarray players, it is based in Japan, giving it a possible home-field advantage in that market. Although the presence of Affymetrix is felt strongly even across the Pacific, a household name will likely carry a lot of weight. "Mitsubishi is a very famous company in Japan," said Uno. Secondly, the company's technology to produce its arrays appears to be easy and cheap, making it attractive to high-throughput users if the quality of the chips lives up to that of its competitors. Mitsubishi's arrays are constructed in an entirely different way from traditional printed or in situ synthesized chips: hollow plastic fibers, which the company also uses to make filtration membranes, for example for water purification, are filled with oligonucleotides or cDNAs, then 100-10,000 of these fibers are bundled up and fixed with a resin. After that the bundle is sliced crosswise. The result is a large number of uniform arrays with up to 10,000 different probes.
The idea for these "sushi-style" chips was originally conceived by Genox Research, according to Uno, but was independently developed by Mitsubishi, which has meanwhile applied for several patents. The company began working on the technology about five years ago, in part at its Chemicals Development Laboratories in Yokohama, where a production facility capable of cranking out up to 100,000 chips per year was installed last fall. Mitsubishi has also been presenting the technology to potential users in an "Open Laboratory" since the spring of last year. Uno said that several companies are currently evaluating the chips, which Mitsubishi also showcased at last month's Bio Expo meeting in Tokyo.
Researchers at the NIH and at Large Scale Biology Corporation have each taken similar approaches to make protein chips but are currently far from commercialization (see BioArray News 6/14/02).
For Mitsubishi, the main remaining obstacle to get to the market is its lack of expertise in biology. Uno said the company is currently looking for a partner in the life sciences to help with such tasks as designing and manufacturing oligonucleotides and dealing with bioinformatics issues, and has been talking to several players so far. It also hopes to find a consultant who could help broker such a partnership deal. On the instrument side, although Mitsubishi has its own CCD-based scanners, it might consider a partnership as well, according to Uno. After launching custom-made gene expression arrays for research purposes in Japan this year, the company is planning to expand its market range to the US and Europe in 2003. Depending on the needs customers express, ready-made chips may follow, Uno said, and Mitsubishi is already considering genotyping arrays as well as diagnostic DNA arrays and protein chips.