From the opening sentence of the Federal Circuit's opinion, it is clear that the Court believed that Bayer had claimed more broadly than it was entitled to and claimed species it had not described, thus failing to satisfy the written description requirement. Regents of Univ. of Cal. v. Eli Lilly & Co., 119 F.3d 1559, 1566, 1569 (Fed. Cir. 1997). The Court's language hies back to the Eli Lilly decision from 15 years go, where the Court speaks about limitations on "functional" claiming rather than structure-based claiming as emphasized by the Federal Circuit after the Eli Lilly decision. "Bayer got ahead of the science" according to the Court, and moreover, the functional terminology used in the claims was inaccurate and Bayer knew it. Indeed, "[w]hen [Bayer] ultimately sued Dow AgroSciences for infringement, Bayer recognized that the term used, in its established scientific meaning, did not cover the accused product (itself different from the particular enzyme whose gene Bayer's inventors had sequenced), so it argued for a broad functional claim construction."
The issue here comes down to claim construction: the only way Bayer could prevail was to obtain a claim construction broader than its specification supports, which the District Court refused to do and granted summary judgment of non-infringement to defentant. The Federal Circuit was not about to upset that judgment on appeal.
The patent at issue, U.S. Patent No. 6,153,401, relates to "genetically modifying plants in order to confer resistance to a commonly used herbicide called 2,4-dichlorophenoxyacetic acid ("2,4-D")." The technology involves producing a transgenic plant whose genome encodes an exogenous 2,4-D gene wherein the plant expresses the protein product. The genes encompassed by the claims arises in bacteria that can convert 2,4-D into 2,4-dichlorophenol (2,4-DCP), which is a non-toxic carbon source for bacteria expressing 2,4-D. Bayer scientists cloned the 2,4-D gene from Alcaligenes eutrophus JMP134 and provided the nucleotide sequence of the gene. (This was the only gene disclosed in the patent.) Genes having activity against 2,4-D were then isolated by functional cloning, wherein bacteria unable to grow on 2,4-D were transfected with DNA from a library made with DNA from a bacteria that could grow on 2,4-D, and bacteria comprising exogenous DNA that had transferred the growth phenotype was selected.
The '401 patent granted in 2000 but the Court notes that Bayer did not change its nomenclature in the face of this scientific evidence. Claim 1 of he '401 patent reads as follows:
A recombinant gene, comprising
a DNA sequence encoding a polypeptide having the biological activity of 2,4-D monooxygenase which is capable of being expressed in a plant, operably linked to
a heterologous promoter capable of promoting the expression in a plant of a structural gene operably linked thereto.
The Court focuses on the lack of understanding of the molecular mechanism of the catalyzed reaction, because those mechanistic considerations informed the language the patentee used to describe the generic class of proteins (and the genes that encode them) in their claims: monooxygenases. However, as it turns out -- and the evidence showed that Bayer became aware of the true nature of the enzymes at issue well before the '401 patent granted -- the enzyme having the desired activity is actually a dioxygenase; the distinction is that a monooxygenase results in one molecule of oxygen being incorporated into the product and another molecule being incorporated in water, while dioxygenase catalyzes incorporation of both atoms of molecular oxygen being incorporated into products (and not water), citing Fukumori & Hausinger, Alcaligenes eutrophus JMP134 "2,4- Dichlorophenoxyacetate Monooxygenase" Is an a- Ketoglutarate-Dependent Dioxygenase, 175 J. Bacteriology 2083 (1993).
Dow's accused infringing articles are genetically modified seeds resistant to 2,4-D. Dow's technology depends on two genes, termed aad-1 and aad-12 that encode aryloxyalkanoate dioxygenase enzymes. It was undisputed that these are different genes than the genes disclosed in Bayer's patent. Bayer's argument at trial was that these genes "fall within the broader claim 1, which encompasses enzymes that 'hav[e] the biological activity of 2,4-D monooxygenase.'" Bayer defined this claim term functionally, taking the position that "the quoted phrase covers any enzyme that triggers cleaving of the side chain of 2,4-D to produce 2,4-DCP, even if it is a dioxygenase and even if it does not share other biological activities of the particular enzyme whose gene Bayer sequenced."
At the District Court, the Markman hearing resulted in judgment in Dow's favor, based on evidence that the Court's Markman decision is consistent with (or at least more consistent with) how the skilled worker would interpret the claim. The Court's basis was "plain and ordinary meaning" if the term is interpreted consistent with "the established scientific meaning of the term," as opposed to the claim construction the District Court adopted, which is that a monooxygenase means "the enzymatic activity of an enzyme, in a biological system, that causes a reaction with 2,4-D, and two molecules of oxygen, where one molecule of oxygen is added to 2,4-D and the other ultimately forms water."
The District Court found non-infringement when it refused to construe the claim as Bayer advocated. The Court also noted that, had it adopted Bayer's construction, the claims of the '401 patent would have been invalid for failing to satisfy the written description requirement of 35 U.S.C. § 112.
The Federal Circuit affirmed, in an opinion by Judge Taranto joined by Judges Prost and Bryson. The panel characterizes the dispositive aspect of the issue before it as claim construction, which is reviewed de novo. The Court further said that the question before it "turns on whether Bayer's proposed construction of the term 'the biological activity of 2,4-D monooxygenase' in the only independent claim (claim 1) is correct." Bayer's position (its only position according to the opinion) was that the District Court erred in not construing the term to mean "bringing about the cleavage of the side chain of 2,4-D," period. The panel agreed with the District Court that this reading is incorrect according to "familiar aspects of textual analysis," i.e., the rubrics of claim construction arising from Markman v Westview and Cybor v FAS, and reaffirmed in Phillips v. AWH Corp., 415 F.3d 1303 (Fed. Cir. 2005) (that claims should be construed according to the intrinsic evidence that is the plain meaning of the claim terms, the specification and the prosecution file history). The opinion also notes that to interpret the claims as Bayer advocated "would read independent claim 1 so broadly as to raise serious doubts about validity." Trapped between this patent law Scylla and Charybdis, the Court came to the conclusion that Bayer could not prevail.
The Court characterizes Bayer's argument regarding its preferred claim construction as comprising two steps, both of which the Court finds erroneous. First, according to the Court, Bayer would have the Court ignore the ordinary and plain meaning of the term "monooxygenase" in the art, as indicating that the enzyme splits molecular oxygen into its constituent oxygen atoms and then incorporates one of these atoms into an organic reaction product and the other into water. The second step in Bayer's construction is to broaden the scope of the claim by arguing that the Court should define the term "biological activity" regarding the claimed nucleic acid as encoding any enzyme capable of cleaving the 2,4-D molecule. Both these steps "encounter serious textual difficulties," according to the opinion.
The first and most basic difficulty the Court finds with Bayer's construction is that this meaning of the disputed phrase "fight a facially straightforward textual analysis" insofar as there is no disagreement that the term "monooxygenase" is understood in the art to encompass enzymes whose enzymatic activity results in one atom of molecular oxygen being incorporated into water. Taking the phrase as a whole, "biological activity of 2,4-D monooxygenase" is most faithfully construed as encompassing only those enzymes that cleave 2,4-D in the presence of molecular oxygen wherein one atom of said oxygen molecules is incorporated into the cleavage product and the other into water. In this construction, "the full phrase works as an integrated unit in a way that fits its structure and the ordinary meaning of its words."
On the contrary, Bayer's construction requires that the phrase "2,4-D monooxygenase" be "stripped" of the accepted meaning of the term "monooxygenase." If this is what Bayer intended to do, Bayer had ample opportunity to do so, inter alia, in its specification or prosecution history, because part of the right of a patentee to be its own lexicographer includes the right to use a term contrary to its accepted meaning. (There is nothing sinister here; for example, the term "mouse" had a much different meaning before graphical user interfaces requires a device for manipulating the cursor.) There is a need for clarity, however, when a patentee does this according to the opinion; that need encompasses at least public notice and the requirement that a claim not be indefinite. But the Court recognizes that any such intentional change in how the term "monooxygenase" is interpreted never arose; rather, "Bayer chose the language based on an unverified belief that it accurately described its enzyme, [and] learned that the belief was false while its application was pending." Moreover, Bayer "had seven years before its patent issued to alter the language, but never did," a consideration clearly implicating the public notice function, or its breach, by Bayer.
Turning to the intrinsic evidence, the Court finds no clear evidence that Bayer intended any variation in the conventional meaning of the term "monooxygenase"; at best, the Court says, the evidence is "inconsistent," and "[m]uch of it actually reinforces the straightforward descriptive meaning of the claim terms [adopted by the District Court]." In addition to failing to find clear evidence in the specification as filed of the meaning for the term now advocated by Bayer, the court notes that the claims as filed ("a 2,4-D monooxygenase gene ... encoding a polypeptide having the biological activity of such a monooxygenase") "reinforce rather than undermies the ordinary descriptive meaning" of the term.
The panel is similarly unpersuaded by Bayer's proposed construction of the term "the biological activity of." Ignoring the structure of the phrase (which the Court says provides the context wherein the biological activity of a monooxygenase should be that activity that makes the enzyme a "monooxygenase" (emphasis in opinion)), Bayer attempts to expand the definition of the term to encompass any enzymatic activity that "bring[s] about the cleavage of the side chain of 2,4-D." The Court rejects this construction, relying on the but two instances where the specification recites a '"biological activity." In one instance, upon which Bayer relies, the Court finds support for Bayer's construction to be ambiguous at best. And the other instance "actually works against Bayer's argument," wherein the Court interprets the disclosure in the specification (coding for a protein which has the biological activity of the protein encoded by tfda, e.g., its 2,4-D-monooxygenase activity") to mean that there is (or can be) more than one activity and that the activity encompassed within the scope of the claim is its monoxygenase activity. Concluding this portion of the opinion, the Court says:
In short, as the district court explained, the claim language has a strong accepted scientific meaning. Bayer's alternative construction strips the monooxygenase half of the claim phrase of its accepted descriptive meaning and then asserts a specification "definition" of the biological-activity half. We do not find enough in the specification or prosecution history to justify those steps.
But this isn't Bayer's only problem, according to the opinion. Even if the Court were to adopt Bayer's construction it would not prevail, says the Court, because the construction Bayer advocates lacks adequate written description support in the specification. Typically, the Court notes, patentees advocate a narrow claim construction in order to minimize the risk of invalidation at trial. Here, Bayer argues for a broad interpretation as it must in order for Bayer to prevail against a defendant making, using, selling or offering to sell, or importing a species that is not encompassed within the scope of the claim if construed according to the plain meaning of the terms in the claim. However, this interpretation is inconsistent with the Court's precedent regarding what must be disclosed to support a broad genus claim limited, as in Bayer's proposed construction, only by the function of the claimed molecule or accused infringing item. The opinion recites the past 15 years of the Court's own jurisprudence, including Ariad Pharms., Inc. v. Eli Lilly & Co., 598 F.3d 1336, 1351 (Fed. Cir. 2010); Carnegie Mellon Univ. v. Hoffman-La Roche Inc., 541 F.3d 1115, 1121, 1124 (Fed. Cir. 2008); Univ. of Rochester v. G.D. Searle & Co., Inc., 358 F.3d 916, 925 (Fed. Cir. 2004); Enzo Biochem, Inc. v. Gen- Probe Inc., 323 F.3d 956, 964 (Fed. Cir. 2002) and Regents of Univ. of Cal. v. Eli Lilly & Co., 119 F.3d 1559, 1566, 1569 (Fed. Cir. 1997). In all these cases, the Court determined that for at least some claims there was insufficient support for the claims in the specification.
And part of that conclusion stems from the Court's understanding that adopting Bayer's construction "would leave the '401 patent far from providing even an indirect structural identification of all that would be within the claim's scope." The functional language encompasses all enzymes capable of cleaving 2,4-D, but the specification discloses only one species capable of producing an enzyme having plaintiff's claim biological activity. This would defeat the public policy-based patent quid pro quo, because it would provide claims far in excess of the scope of what the patentee had invented. This would necessarily include genes from all species, not just the bacterial species set forth in the specification. However, the Court noted that although "Bayer has sought to mitigate this concern by expressly arguing that any genes not derived from soil bacteria would fall outside of the claimed genus[,] Bayer did not present a claim construction based on that trimming effort to the district court."
Finally, the Court did not deign to perform its own independent claim construction hearing. The opinion notes that Dow proposed two alternative constructions but that the Court did not rule on them; however, the panel did find that construction of the claim term "monooxygenase" as it is commonly accepted in the art is "the most natural reading" of the term. The difficulty with this approach is that it "reads out" of the claim the expressly disclosed species (because there is no dispute that the disclosed enzyme is a dioxygenase). The Court notes that such a decision would be "disfavored" but not prohibited.
This case illustrates the continuing tension between claim construction and sufficiency of disclosure that has arisen in claim drafting. There is not an easy answer or cookie-cutter approach to producing a specification and claims that satisfy Section 112 while at the same time can be construed in such a way that the claims broadly recite the subject matter at issue. This is a balance that is continually shifting, as claims having proper scope are the subject of efforts to design around the protections offered by a patent. In this case the Federal Circuit breaks no new ground but in rendering its decision provides a reminder for courts and litigants alike by affirming the District Court's proper refusal to broaden the scope of the '401 patent claims outside the four corners of its written description.
Bayer Cropscience AG v. Dow Agrosciences LLC (Fed. Cir. 2013)
Panel: Circuit Judges Prost, Bryson, and Taranto
Opinion by Circuit Judge Taranto