Appeal No. 2005-1383
Application No. 09/364,847
predictability, only a reasonable expectation of success. In re O’Farrell, 853 F.2d 894,
903, 7 USPQ2d 1673, 1681 (Fed. Cir. 1988).
As to the appellants’ contention that the art does not render the specific enzymes
of claim 2 obvious, we direct attention to the teachings of Peoples. As discussed
above, Peoples discloses that the β-ketothiolase gene is just upstream from acyl-CoA
reductase gene in the bacterial genome (Figure 3, col. 11, line 61-col. 12, line 51). In
addition, Peoples discloses that recombinant plasmids which comprise a β-ketothiolase
gene in the upstream position from an acyl-CoA reductase gene encode catalytically
active enzymes; i.e., enzymes having PHB polymerase activity. See pZT1 and pZT2 in
col. 12, lines 26-51. Accordingly, given the teachings of Peoples with respect to DNA
sequences and recombinant plasmids comprising a β-ketothiolase gene (phbA) and an
acyl-CoA reductase gene (phbB), linked in a tandem arrangement, wherein said genes
encode catalytically active enzymes which act on substrate in successive reactions in a
PHA biosynthetic pathway, and the teachings of Bülow with respect to the construction
of bifunctional enzymes comprising two enzymes which act on a substrate in
succession, we hold that it would have been obvious to one of ordinary skill in the art to
construct a protein fusion consisting of β-ketothiolase (phbA) and acyl-CoA reductase
(phbB).5
Finally, we find the appellants’ argument that “no art has been cited to show that
the fusions of claims 5 and 6, in a host bacterium or plant cell would [have been]
obvious to one of ordinary skill in the art” (Brief, p. 18), to be misdirected.
5 We point out that like claim 1, claim 2 does not require that the protein fusion
have a peptide linker between the two catalytically-active enzymes.
18
Page: Previous 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 Next
Last modified: November 3, 2007