The patent at issue related to genetically modifying plants to confer resistance to the herbicide “2, 4-D” by inserting a gene that is capable of turning the herbicide into “2, 4-DCP.”  During development, the scientists characterizing the process believed the enzyme to be a “monooxygenase,” which uses one oxygen atom in the process of breaking down the herbicide, and incorporates the other oxygen atom into a water molecule. As such, the initial patent application was drafted to include the term “monooxygenase” in certain claims. During the pendency of the patent application, however, the Bayer scientists discovered that the enzyme responsible for the herbicide resistance was not a monooxygenase, but a dioxygenase. The patent claim language was not changed, and the patents were issued with the term “monooxygenase.”

The accused product contained DNA encoding for a dioxygenase enzyme. As a result, Bayer pursued a broad claim construction of “biological activity of 2, 4-D monooxygenase,” which would have encompassed any enzyme that cleaved the side-chain of 2, 4-D regardless of whether it was a monooxygenase or dioxygenase. The district court rejected this claim construction and granted summary judgment of noninfringement.

The Federal Circuit affirmed the district court’s ruling. The court held that Bayer’s proposed construction ignored the strong accepted scientific meaning of the phrase “2, 4-D monooxygenase” and nothing in the intrinsic record supported such a broad construction.

The court also observed that Bayer’s proposed construction raised serious concerns about the validity of the asserted claims under 35 U.S.C. § 112(a). Specifically, the court observed that Bayer’s proposed construction would broadly cover a class of enzymes while the written description identified only one gene sequence and the enzyme it encodes. This significant invalidity concern was viewed by the court as substantiating its rejection of Bayer’s proposed construction.

A copy of the case can be found here.