Novozymes A/S et al. v. DuPont Nutrition Biosciences APS et al., No. 2012-1433 (Fed. Cir., July, 2013)

The present case concerns improvements in alpha-amylase enzymes, which catalyze the breakdown polysaccharides to simple sugars, and are widely used in in detergent formulations, sugar refining, and ethanol production, among other uses. Alpha-amylases derived from Bacillus bacteria are especially attractive for industrial use because of their exceptional activity.

Industrial uses of alpha-amylases often involve high temperature and/or high acidity, and both conditions can degrade the performance of the enzymes. In an effort to produce a more robust commercial product, Novozymes embarked on a program to identify particular amino-acid residue sites within alpha-amylase enzymes and particular mutations at those sites that would increase the thermal or acid stability of the enzyme. One arm of the research involved rational protein design, in which knowledge about the three-dimensional shape of the protein, combined with the known properties of the different natural amino acids, could in theory be used to predict how particular amino acid mutations at given positions might affect a protein property of interest, such as temperature or acid stability. A second arm of the research used random mutagenesis of a alpha-amylase parent enzyme and screening the resultant variants to identify enzymes with improved properties. Using these two approaches, Novozymes ultimately identified 33 residue positions (out of a total of about 500 positions) as promising targets for mutagenesis, 17 of them predicted by the rational design approach and 16 by random mutagenesis.

Based on these studies, Novozyme filed a patent application in November, 2000, covering Bacillus alpha-amylase variants with enhanced stability. The application disclosed seven different Bacillus parent species and 33 separate amino acid positions as targets for mutations, and also taught that one or more of the 33 sites could be altered by amino-acid deletions, additions or substitutions. Included in the application were examples with data confirming the enhanced stability of selected variants identified through random mutagenesis. There was no empirical data for mutations at any of the 17 positions predicted by rational protein design, and in fact, later experiments showed that some of these 17 residue positions did not yield any thermostable variants, and even for many of the others that did, only a minority of amino acid substitutions were effective in improving measurable enzyme properties. For example, a serine-to-tryptophan substitution at position 239 (designated S239W) was specifically disclosed in the application, but this substitution proved ineffective in later testing.

Beginning in 2006, DuPont began a program of its own to develop alpha-amylase enzymes with enhanced temperature stability. Using a selected Bacillus (BSG) alpha-amylase enzyme, DuPont generated about 1,500 alpha-amylase variants with amino acid substitutions at 150 different residue positions and screened these for thermostability. The best performer identified among these had a serine to glutamine substitution at position 239 (S239Q). In November, 2008, DuPont filed an application on the S239Q variant of BSG alpha-amylase, and the application issued as US Patent No. 7,541,026 in June, 2009.

The DuPont patent prompted Novozymes to file, in December, 2009, a continuation of its 2000 application, seeking claims drawn specifically to an alpha-amylase variant having (i) at least 90% sequence identity to BSG alpha-amylase, (ii) an amino acid substitution at position 239, and (iii), increased thermostability relative to the native enzyme. A patent issued from this application on May 11, 2010.

The same day that Novozymes received its patent, it sued DuPont in the Western District of Wisconsin, alleging infringement of several of the patent claims. The case went to trial before a jury, which found willful infringement by DuPont, and that DuPont had failed to establish by clear and convincing evidence, that the Novozymes patent was invalid for failure to meet either the written description of enablement requirements. DuPont filed JMOL motions on various issues, and the district court granted the DuPont’s motion that the claims in Novozymes’ patent were invalid under §112 for lack of written description in the original 2000 application.

In finding lack of written description, the district court reasoned that the Novozymes specification covered "a potentially enormous number of alpha-amylase variants, encompassing possible combinations among the seven disclosed parent enzymes, the thirty-three disclosed positions for mutation, the numerous different mutations possible at each position, and the various possible combinations of individual mutations." Although the original application expressly named the limitations recited in the claims, there was no disclosure that would "inform the reader which member of that [claimed] group was the right one."

On appeal to the Federal Circuit, Novozymes argued that that the original application discloses each of the three limitations of the asserted claims—(i) a parent BSG alpha-amylase, (ii) a substitution at the S239 position, and (iii) increased thermostability--and that a person of ordinary skill in the art would therefore have understood the original application as clearly describing the claimed invention.

DuPont defended the district court’s judgment, arguing that "the written description requirement precludes premature claims to a research plan and requires the disclosure of an actual invention." The Novozymes application, DuPont argued, discloses no more than a theory or a laundry list of potential solutions, and in fact, fails to disclose a single alpha-amylase variant substituted at position 239 that actually exhibits increased temperature stability, noting that the only substitution at this position disclosed in the Novozyme specification (S239W) does not yield increased thermostability.

In affirming the district court ruling, the Federal Circuit was guided particularly by In re Ruschig, 379 F.2d. 990 (CCPA 1967); Boston Scientific Corp. v. Johnson and Johnson, 647 F.3d 1353 (Fed. Cir. 2011); Purdue Pharma L.P. v. Faulding Inc., 230 F.3d 1320 (Fed. Cir. 2000); and University of Rochester v. G.D. Searle & Co., 358 F.3d 916 (Fed Cir 2004). In Ruschig, the CCPA affirmed a POBA decision that a claim to a specific drug molecule, added after filing, lacked sufficient description because the specification "failed to provide sufficient ‘blaze marks’ to guide a reader through a forest of disclosed possibilities toward the claimed compound, which resided among the myriad others that also could have been made."

In Boston Scientific, the patents in suit covered a drug-eluting stent coated with either rapamycin or a macrocycline triene analog of rapamycin. In finding that the patent specification failed to demonstrate that the inventors were in possession of the claimed invention, the court noted that the specification "made passing reference to the term ‘macrocyclic triene’ but failed to describe or identify any member of the claimed sub-genus of macrocyclic triene analog encompassed by the claim.

The disputed claims in Purdue Pharma recited an extended release drug formulation requiring a certain ratio between the drug’s maximum blood concentration and its concentration 24 hours after administration. The Federal Circuit upheld the district court’s conclusion that one of ordinary skill in the art would not been directed to the claimed ratio, because neither the data nor any anything else in the disclosure provided guidance to the claimed ratio.

The University of Rochester case represents one of the more striking failures to provide written description for a claimed invention. Here the Federal Circuit affirmed a summary judgment of invalidity of a claimed treatment method that required administering a drug having a certain target-specific activity, where the specification failed to disclose any suitable drugs for use in practicing the method, and none were known in the art at the time of filing.

Applying this guidance to the case at hand, the court noted that, although the three individual limitations in the patent claims are expressly stated in the Novozymes specification, there is no disclosure of any variant that actually satisfies all the limitations in the claim. First, the specification focuses on BLA alpha-amylase variants, rather than the claims BSG variants, and the two alpha-amylase share only 64% sequence identity, far less than the 90% required by the claim. Additionally, position 239 is disclosed as one of 33 possible "target" mutation sites, and the only specific substitution disclosed for this site—S239W-- was subsequently shown not to confer any increased thermostability. In short, noted the court, "one searches the [Novozymes] application in vain for the disclosure of even a single species that falls within the claims or for any ‘blaze marks’ that would lead an ordinary investigator toward the species among a slew of competing possibilities."

In his dissent, Chief Judge Rader placed heavy emphasis on the jury’s determination that DuPont had failed to meet its burden of invalidity with clear and convincing evidence. “The jury received expert testimony, heard from skilled protein engineers, reviewed visual aids and publication experts, and examined the patent document as guided by those skilled in the art, over an eight day trial.” That in Judge Rader’s view, was sufficient to reinstate the jury’s verdict of patent validity.