Federal Circuit Affirms Invalidity of Genus Claims to Stevia Production Method Lacking Written Description and Narrower Claims as Combining Natural Phenomena with an Abstract Idea

by Dennis Crouch
In PureCircle USA Inc. v. SweeGen, Inc. , No. 22-1946 (Fed. Cir. Jan. 2, 2024) (non-precedential) [ PureCircle Decision ], the Federal Circuit affirmed a district court summary judgment siding against the patentee, PureCircle.  The court concluded that the stevia production method claims were all invalid, either for: (1) lack of written description for the genus claim or (2) lack of eligibility because the claims are directed to a natural phenomenon combined with an abstract idea.  Judge Dyk authored the opinion of the court that was joined by Judges Schall and Stark affirming the ruling of C.D.Cal Judge James Selna.
The written description portion of the opinion parallels the Supreme Court’s 2023 Amgen decision that focused on the sibling doctrine of enablement. Although non-precedential, the eligibility portion appears important because of the way it combines prohibited eligibility categories and rules that method claims requiring a particular purity level are themselves an abstract idea absent requirement of specific steps on how that purity level is achieved.

PureCircle sued SweeGen for patent infringement back in 2018, asserting U.S. Patent Nos. 9,243,273 and 10,485,257.  The patents here claim a method of making Rebaudioside X. Reb X is a steviol glycoside that occurs naturally in trace amounts in Stevia rebaudiana plants. It is considered one of the best-tasting steviol glycoside sweeteners, but extracting it directly from plants is costly and inefficient. PureCircle’s patents claim methods of producing Reb X by using certain enzymes to add glucose units to rebaudioside D (Reb D), another steviol glycoside occurring in higher amounts in Stevia plants.
Claims 1 and 14 of the 273 patent were treated as representative for the case:
1. A method for making Rebaudioside X comprising a step of converting Rebaudioside D to Rebaudioside X using a UDP-glucosyltransferase, wherein the conversion of Rebaudioside D to Rebaudioside X is at least about 50% complete.
14. The method of claim 1, wherein the UDPglucosyltransferase comprises UGT76G1.
As you can see, claim 1 broadly claims utilizing the functional class of UDP-glucosyltransferase enzymes to transform Reb D into the desired Reb X product at the specified yield. Narrower claim 14 eliminates the genus and instead requires a particular enzyme known as UGT76G1.  This limitation appears to refer to a specific protein enzyme having a particular amino acide sequence.
Lack of Adequate Written Description
The written description requirement is rooted in 35 U.S.C. § 112(a) and requires a patent specification to demonstrate the inventor actually possessed the full scope of the claimed invention at the time of filing. As the Federal Circuit re-established in Ariad Pharmaceuticals, Inc. v. Eli Lilly & Co. , 598 F.3d 1336 (Fed. Cir. 2010) (en banc), the test is whether the specification objectively “reasonably conveys to those skilled in the art that the inventor had possession of the claimed subject matter as of the filing date.” Id . For genus claims that cover multiple embodiments, this generally requires disclosing either a representative number of species within the genus or common structural features shared by genus members, so that a skilled artisan can visualize or recognize the scope of species covered by the claims. Thus, under Ariad , purely functional claiming without correlation to defining structures risks running afoul of the written description requirement.
Here, claim 1 uses functional language to cover a genus of UGT enzymes capable of converting Reb D into Reb X. However, the patents disclosed only a single species – UGT76G1. PureCircle failed to demonstrate that this lone example was representative of the broad genus claimed. Nor did PureCircle establish that there were common structural features linking the members of the genus.  The patentee made four particular arguments that were each rejected by the Federal Circuit:

PureCircle argued that disclosure of a single enzyme species (UGT76G1) could be representative of the entire claimed genus based on prior cases like Invitrogen Corp. v. Clontech Labs., Inc., 429 F.3d 1052 (Fed. Cir. 2005) and Bilstad v. Wakalopulos, 386 F.3d 1116 (Fed. Cir. 2004). However, the court held those cases require adequate “blaze marks” guiding a skilled artisan to visualize the genus, which were lacking here.
PureCircle contended the structure of UGT76G1’s active site was common across the genus, but that commonality was not disclosed in the specification in a way that would permit PHOSITA “to visualize or recognize members of the genus.”
PureCircle asserted that rather than trillions of enzymes covered by genus, homology modeling could narrow the genus down to around 9,000 variants. But the court noted testing so many candidates still indicated lack of adequate written description.  And further, that homology modeling was not disclosed. In the language of Amgen, this was a “research assignment” rather than an invention disclosure.
PureCircle invoked the original claims doctrine to argue those original claims provided support. In particular PureCircle noted that the claims found in the original PCT filing are part of the original disclosure even though later replaced by different claims. On appeal the court agreed with this legal principle but found those claims offered no substantive additional descriptive information beyond the issued claims.

As in other Federal Circuit decisions like AbbVie Deutschland GmbH & Co., KG v. Janssen Biotech, Inc. , 759 F.3d 1285 (Fed. Cir. 2014) and Idenix Pharms. LLC v. Gilead Scis. Inc., 941 F.3d 1149 (Fed. Cir. 2019), PureCircle’s disclosure of a single UGT enzyme did not adequately describe the full scope of the genus claimed. “In short, the one enzyme disclosed in the patents here has not been shown to be typical of the entire genus of UGTs claimed.” Thus, the Federal Circuit affirmed the district court’s finding that the asserted patent claims lacked adequate written description support.
Claim 14 Directed to Patent-Ineligible Subject Matter in the combination of a natural phenomenon and an abstract idea. 
The Federal Circuit then turned to analyzing the eligibility under § 101 of claim 14 of the ‘273 patent. This narrower claim specifically recited the UGT76G1 enzyme rather than the broader genus. The court held that “[t]o the extent that claim 14 claims a ‘method for making Rebaudioside X comprising a step of converting Rebaudioside D to Rebaudioside X using a UDPglucosyltransferase’ it claims a natural phenomenon.” The UGT76G1 enzyme occurs naturally in stevia plants and carries out such conversion in vivo.
Seeking to avoid ineligibility, PureCircle pointed to the requirement in claim 1 that “the conversion of Rebaudioside D to Rebaudioside X is at least about 50% complete.” Natural conversion achieves far lower levels. But the court held that “[t]he problem with PureCircle’s argument is that the 50% completion is itself an abstract idea.” According to the court, the claim merely demanded a result without any limitation on how to produce that result, running afoul of precedents like Interval
Licensing LLC v. AOL, Inc. , 896 F.3d 1335 (Fed. Cir. 2018) where the court wrote that claims that “simply demand[] the production of a desired result . . . without any limitation on how to produce that result” are directed to an abstract idea.
Stan Panikowski of DLA Piper argued for the plaintiffs-appellants PureCircle. He was supported by attorneys Richard Mulloy of DLA Piper and Stuart Pollack of Kilpatrick Townsend. JC Rozendaal of Sterne Kessler argued for the defendants-appellees SweeGen. Supporting Mr. Rozendaal were attorneys Michael Joffre, Anna Phillips, Sasha Rao, Dennies Varughese, and Deirdre Wells, all of Sterne Kessler.