Guest post by Dr. Jeffrey A. Lefstin, Professor of Law, University of California, Hastings College of Law
Jason’s excellent post on Roslin summarized the case and raised the question of whether the Federal Circuit paid insufficient attention to the inherent differences conveyed by the ‘clone’ limitation of the claims. In this post I address two other issues in Roslin: the Federal Circuit’s interpretation of Ckakrabarty and Funk Brothers, and Roslin’s consistency with Federal Circuit precedent holding that structural or functional differences between natural and synthetic products must be defined in the claims.
Unaltered by the hand of man: Judge Dyk’s opinion in Roslin unfortunately perpetuates the view, now found in the PTO’s Myriad guidelines, that Chakrabarty requires a claimed invention to be “markedly different” from a natural product for patent-eligibility under § 101. Myriad itself imposed no such requirement: the Court found BRCA cDNAs patent-eligible without determining that they were “markedly different” from natural sequences. And though Myriad reiterated the “markedly different” language from Chakrabarty, Chakrabarty’s discussion of ‘products of nature’ was entirely dictum. Only the question of whether living organisms were patent-eligible was before the Court in Chakrabarty; the ‘product of nature’ rejection in the case had not been sustained by the Patent Office Board of Appeals.
The Chakrabarty Court noted the claimed bacteria differed markedly from natural bacteria by way of distinguishing the case from Funk Brothers. As Paul Cole’s recent post notes, the patentee had discovered that certain strains of bacteria could be mixed together without inhibiting their nitrogen-fixing capability. Justice Douglas regarded this compatibility as the unpatentable discovery of a natural phenomenon; the claims were unpatentable because the patentee’s application – a mixed inoculant – was “a simple step” once the underlying discovery was assumed away.
As I discuss in a recent article, Funk was nonetheless very much a patent eligibility case. It reflects Douglas’s view – shared by Justice Stevens in Flook, and Justice Breyer in Mayo – that obvious applications of scientific discoveries or abstract ideas are not patent-eligible “inventions” within the meaning of the statute.
Whether or not we share that view, understanding it shows us that Funk was an ‘inventive application’ case, not a ‘product of nature’ case. Douglas made no reference to the ‘product of nature’ doctrine, nor to the recent cases embodying it. Instead, Douglas emphasized the lack of change in the bacteria to establish that the mixed inoculant was obvious (once the patentee’s discovery was assumed away). Under the old doctrine of ‘aggregation,’ it was not invention to combine old elements where the elements were unchanged, and no new function arose from their combination. It was therefore not inventive for the patentee in Funk to combine old bacteria without changing their structure or function.
But just as many combinations of old elements become patentable when a new function emerges, the mixed inoculants of Funk would have been patentable — even if the bacteria remained unchanged — had the mixed inoculant acquired new functions not performed by its constituent bacteria. If Funk Brothers instead stood for the proposition that a combination is unpatentable if its constituents are ‘unaltered by the hand of man,’ then a very large number of inventions become patent-ineligible. An artificial structure like an arch, formed by piling stones atop each other, would be ineligible unless the stones themselves were altered by the hand of man. Even as ardent a skeptic of the patent system as Justice Douglas would not have gone that far.
Source limitations and expressly claimed distinctions: The second difficulty with Roslin is its demand that the ‘marked differences’ between the natural organism and the invention must be expressly claimed. As Jason discussed, the applicant in Roslin argued that cloned animals differ from their natural counterparts at least in having mitochondrial DNA derived from the egg donor, rather than the animal which donated the somatic nucleus. The Federal Circuit rejected such arguments because neither the difference in mitochondrial DNA, nor any functional consequence of that difference, was recited in the claims.
However, the same argument, albeit in the context of § 102, was before the Federal Circuit in the extensive litigation over Amgen’s recombinant erythropoietin (EPO) patents. Much like Roslin, Amgen had claims to a ‘copy’ of natural product: in that case EPO produced by mammalian cells in culture. While Amgen’s synthetic EPO differed in glycosylation from the natural product, several of the claims in the case recited only the non-natural source of the EPO, not the structural differences. The Federal Circuit recognized that the novelty of the synthetic EPO claims depended on whether synthetic EPO differed from natural EPO. Yet the court found novelty based on the unclaimed structural and functional differences between natural and synthetic EPO, which were demonstrated in part by the specification and prosecution history, and in part by testimony at trial. Amgen Inc. v. F. Hoffman-LaRoche Ltd, 580 F.3d 1340, 1370 (Fed. Cir. 2009). In effect, the court held that the structural and functional differences characterizing the synthetic product were inherent in the source limitations. (The court did not inquire whether all synthetic EPO molecules falling within the scope of the claims would display similar differences in glycosylation.)
Thus under Amgen, a source limitation alone (such as ‘non-naturally occurring’ or ‘purified from mammalian cells grown in culture’) may establish novelty of a product. Assuming the ‘clone’ limitation in Roslin to require derivation from nuclear transfer, then it serves as a source limitation as well. Since Roslin cannot overrule Amgen, we seem to be in a regime where differences between natural and synthetic products may be unclaimed yet confer novelty under § 102, but must be explicitly claimed to establish ‘marked difference’ under § 101. Of course, if Roslin is correct, that doctrinal inconsistency is less significant than the consequence that Amgen-type claims – and perhaps a wider category of product-by-process claims involving natural products – are now ineligible under § 101.
 In Chakrabarty, the Commissioner never suggested that the claimed bacteria were unpatentable under Funk, nor even raised the case. Rather, Chakrabarty argued that if living organisms were not patent-eligible, the Court would have said so in Funk.
 In particular, the General Electric and Marden cases denying patentability to purified tungsten, uranium, and vanadium. The defendant had urged General Electric upon the Funk court in its brief.