Ex Parte Short

Board of Patent Appeals and Interferences

Apr 21, 2006

09495052 (B.P.A.I. Apr. 21, 2006)

Administrative Patent Judge Spiegel is presently1
unavailable. A reconstituted panel of Judges Schafer, Gron and
Torczon has been designated to consider the present request for
rehearing. See In re Bose Corp., 772 F.2d 866, 869, 227 USPQ 1,
3-4 (Fed. Cir. 1985)(permissible exercise of discretion).
-1-
The opinion in support of the decision being
entered today is not binding precedent of the Board.
Paper No. 199
Filed by: Trial Section Merits Panel Filed: April 21, 2006
Board of Patent Appeals and Interferences
U.S. Patent and Trademark Office
P.O. Box 1450
Alexandria, VA 22313-1450
Tel: 571-272-9797
Fax: 571-273-0042
UNITED STATES PATENT AND TRADEMARK OFFICE

BEFORE THE BOARD OF PATENT APPEALS
AND INTERFERENCES

JAY M. SHORT
Junior Party
(U.S. Patent 6,479,258)
v.
JUHA PUNNONEN, WILLEM P. C. STEMMER,
ROBERT G. WHALEN and RUSSELL HOWARD
Senior Party
(Application 09/724,869)

Patent Interference No. 105,188

Before: SCHAFER, GRON and TORCZON, Administrative Patent Judges.
GRON, Administrative Patent Judge.1
Interference No. 105,188
Short v. Punnonnen
Punnonen claims benefit of the February 10, 1999,2
filing date of U.S. Application 09/248,716, and the February 11,
1998, filing date of U.S. Provisional Application 60/074,294,
under 35 U.S.C. § 120 for the full scope of the subject matter
defined by Claim 47.
-2-
MEMORANDUM IN SUPPORT OF RECOMMENDATION
Short requested rehearing (Paper No. 188) of Short
Preliminary Motion 10, on June 24, 2004 (Paper No. 60). The
motion was denied by Decision-Motion-Bd.R. 125(a) of the Board of
5 Patent Appeals and Interferences (Board), mailed November 25,
2005 (Paper No. 181). Short’s motion asked for judgment that
Claim 47 of Senior Party Punnonen’s Application 09/724,869, filed
November 28, 2000 (Punnonen Exhibit 1007 (Exh. 1007)), is2
unpatentable under 35 U.S.C. § 103 in view of the combined
10 prior art teachings of Freeman PCT (WO 95/03408, published
February 2, 1995)(Exh. 2040) and Short PCT (WO 91/16427,
published October 31, 1991)(Exh. 2050). In discussing the
teachings of the Freeman and Short PCTs, Short’s motion set the
state of the art and level of skill in the art on February 11,
15 1998, by reference to the Freeman and Short PCTs and Peach, et
al. (Peach), “Both Extracellular Immunoglobin-like Domains of
CD80 Contain Residues Critical for Binding T Cell Surface
Receptors CTLA-4 and CD28,” J. Biol. Chem., Vol. 270, No. 36,
pp. 21181-21187 (1995)(Exh. 2051); Stemmer, Willem P. C.
Interference No. 105,188
Short v. Punnonnen
-3-
(Stemmer), “Searching Sequence Space,” Biotechnology, Vol. 13,
pp. 549-553 (June 1995)(Exh. 2052); Patten, et al. (Patten),
“Applications of DNA Shuffling to Pharmaceuticals and Vaccines,”
Current Opinion in Biotechnology, Vol. 8, pp. 724-733 (1997)
5 (Exh. 2053); and Hoogenboom, et al. (Hoogenboom), “Multi-subunit
proteins on the surface of filamentous phage: methodologies for
displaying antibody (Fab) heavy and light chains,” Nuc. Acid.
Res., Vol. 19, No. 15, pp. 4133-4137 (1991)(Exh. 2054)(Paper
No. 60, pp. 5-13). Punnonen filed a response to Short’s request
10 for rehearing on January 25, 2006 (Paper No. 191).
Short’s request for rehearing (Paper No. 188) alleged that
our decision overlooked and misapprehended teachings of the
Freeman and Short PCTs, the state of the art, the knowledge of
persons skilled in the art, and the full scope of subject matter
15 defined by Claim 47 of Punnonen’s Application 09/724,869 as of
its February 11, 1998, filing date. While Short’s request is no
longer before us, we have reviewed the patentability of
Punnonen’s Claim 47 under 35 U.S.C. § 103 in view of the combined
teachings of the prior art of record.
20 On reconsideration of the prior art of record, we conclude
that Claim 47 of Punnonen’s Application 09/724,869, filed
November 28, 2000, prima facie would have been obvious under
35 U.S.C. § 103 to persons having ordinary skill in the art in
Interference No. 105,188
Short v. Punnonnen
-4-
view of the combined teachings of the Freeman PCT, the Short PCT,
and Stemmer. Accordingly, on remand of Punnonen’s Application
09/724,869, we recommend that further action be taken consistent
with our conclusion.
5 Discussion
1. Claim 47 of Punnonen’s Application 09/724,869
Punnonen’s Claim 47 (Paper No. 181, p. 18) is reproduced
below:
10 Claim 47
A method for obtaining an immunomodulatory
polynucleotide that has an optimized modulatory effect
on an immune response as compared to the response prior to
optimization, or encodes a polypeptide that has an optimized
15 modulatory effect on an immune response as compared to the
response prior to optimization, the method comprising:
a) creating a library of recombinant polynucleotides;
and
20
b) screening the library to identify an optimized
recombinant polynucleotide that has, or encodes a
polypeptide that has, a modulatory effect on the immune
response induced by a vector;
25
wherein the optimized recombinant polynucleotide or the
polypeptide encoded by the recombinant polynucleotide
exhibits an enhanced ability to modulate an immune response
compared to a polynucleotide from which the library was
30 created; wherein said optimized modulatory effect on an
immune response is induced by a genetic vaccine vector,
wherein the optimized recombinant polynucleotide encodes a
co-stimulator selected from B7-1 (CD80) or B7-2 (CD86) and
the screening step involves selecting variants with altered
35 activity through CD28 or CTLA-4, and whereby optimization is
achieved by recursive sequence recombination.
Interference No. 105,188
Short v. Punnonnen
-5-
2. Freeman PCT
First, we revisit the Freeman PCT. On consideration of
its complete disclosure, we fail to see how its recognition
that a “‘soluble B7-2 extracellular region fragment or derivative
5 that binds CTLA4Ig (or CTLA-4 or CD28 or CD28Ig)” need not
“necessarily [be] . . . capable of providing the second
co-stimulation signal required to induce T cells to proliferate
and become functional . . .’ (Paper 181, pg. 133, ll. 11-19)”
establishes that the Freeman PCT does not describe “‘obtaining an
10 optimized polynucleotide encoding a B7-1 or B7-2 co-stimulator
variant having an enhanced ability to modulate an immune response
induced by a genetic vaccine vector and having altered activity
through CD28 or CTLA-4 as compared to a predecessor B7-1 or B7-2
molecule, e.g. wild-type B7-1 or B7-2 . . .’ (Paper 181, pg. 129,
15 11.2-7, and pg. 142, 11. 2-9)”.
The Freeman PCT (Exh. 2040) explicitly states (Exh. 2040,
p. 2, ll. 20-33):
This invention pertains to isolated nucleic acids
20 encoding novel molecules which costimulate T cell
activation. Preferred costimulatory molecules include
antigens on the surface of B lymphocytes, professional
antigen presenting cells . . . which present antigen to
immune cells, and which bind either CTLA4, CD28, both CTLA4
25 and CD28 or other known or as yet undefined receptors on
immune cells. Such costimulatory molecules are referred
to herein as CTLA4/CD28 binding counter receptors or
B lymphocyte antigens, and are capable of providing
costimulation to activated T cells to thereby induce T cell
Interference No. 105,188
Short v. Punnonnen
-6-
proliferation and/or cytokine secretion. Preferred B
lymphocyte antigens include B7-2 and B7-3 and soluble
fragments or derivatives thereof which bind CTLA4 and/or
CD28 and have the ability to inhibit or induce costimulation
5 of immune cells. In one embodiment, an isolated nucleic
acid which encodes a peptide having the activity of the
human B7-2 B lymphocyte antigen is provided. . . . .
In describing further aspects of the invention disclosed, the
10 Freeman PCT teaches (Exh. 2040, p. 6, ll. 10-19, emphasis added):
Macrophages that express a peptide having the activity of a
B lymphocyte antigen, such as the B7-2 antigen, can be used
as antigen presenting cells, which, when pulsed with an
appropriate pathogen-related antigen or tumor antigen,
15 enhance T cell activation and immune stimulation.
Mammalian cells can be transfected with a suitable
expression vector containing a nucleic acid encoding a
peptide having the activity of a novel B lymphocyte antigen,
such as the B7-2 antigen, ex vivo and then introduced into
20 the host mammal, or alternatively, cells can be transfected
with the gene in vivo via gene therapy techniques. For
example, the nucleic acid encoding peptide having B7-2
activity can be transfected alone, or in combination with
nucleic acids encoding other costimulatory molecules.
25 On the other hand, the Freeman PCT acknowledges in the
Summary of the Invention another aspect of its invention
which relates to proteins which block or fail to deliver a
co-stimulatory signal (Exh. 2040, p. 6, l. 24, to p. 13):
The invention also provides methods for inducing both
30 general immunosuppression and antigen-specific tolerance in
a subject by, for example, blocking the functional
interaction of the novel B lymphocyte antigens of the
invention, e.g. B7-2 and B7-3, to their natural ligand(s) on
T cells or other immune system cells, to thereby block co-
35 stimulation through the receptor-ligand pair. In one
embodiment, inhibitory molecules that can be used to block
the interaction of the natural human B7-2 antigen to its
Interference No. 105,188
Short v. Punnonnen
-7-
natural ligands (e.g., CTLA4 and CD28) include a soluble
peptide having B7-2 binding activity but lacking the ability
to costimulate immune cells, antibodies that block the
binding of B7-2 to its ligands and fail to deliver a co-
5 stimulatory signal (so called “blocking antibodies”, such as
blocking anti-B7-2 antibodies), B7-2-Ig fusion proteins,
which can be produced in accordance with the teachings of
the present invention, as well as soluble forms of B7-2
receptors, such as CTLA4Ig or CD28Ig. . . . .
10
Another aspect of the invention features methods for
upregulating immune responses by delivery of a costimulatory
signal to T cells through use of a stimulatory form of B7-2
antigen, which include soluble, multivalent forms of B7-2
15 protein, such as a peptide having B7-2 activity and B7-2
fusion proteins. Delivery of a stimulatory form of B7-2 in
conjunction with antigen may be useful prophylactically to
enhance the efficiency of vaccination against a variety of
pathogens and may also be useful therapeutically to
20 upregulate an immune response against a particular pathogen
during an infection or against a tumor in a tumor-bearing
host.

In the Detailed Description of the Invention, the Freeman
25 PCT teaches:
Costimulatory molecules within the scope of the invention
are referred herein as CTLA4/CD28 ligands (counter-
receptors) or B lymphocyte antigens. Novel B lymphocyte
antigens which provide costimulation to activated T cells
30 to thereby induce T cell proliferation and/or cytokine
secretion include the B7-2 (human and murine) and the B7-3
antigens described and characterized herein.
(Exh. 2040, p. 10, ll. 6-10);
35 [O]ne aspect of this invention pertains to isolated nucleic
acids comprising a nucleotide sequence encoding a novel
costimulatory molecule, such as the B lymphocyte antigen
B7-2, fragments of such nucleic acids, or equivalents
thereof. . . . The term equivalent is intended to include
40 nucleotide sequences encoding functionally equivalent B
lymphocyte antigens or functionally equivalent peptides
having an activity of a novel B lymphocyte antigen, i.e.,
Interference No. 105,188
Short v. Punnonnen
-8-
the ability to bind to the natural ligand(s) of the B
lymphocyte antigen on immune cells, such as CTLA4 and/or
CD28 on T cells, and inhibit (e.g., block) or stimulate
(e.g., enhance) immune cell costimulation. Such nucleic
5 acids are considered equivalents of the human B7-2
nucleotide sequence provided in Figure 8 (SEQ ID NO:1)
and murine B7-2 nucleotide sequence provided in Figure 14
(SEQ ID NO:22) and are within the scope of this invention.
10 (Exh. 2040, p. 10, ll. 24-35);
In one embodiment, the nucleic acid is a cDNA encoding a
peptide having an activity of the B7-2 B lymphocyte antigen.
Preferably, the nucleic acid is a cDNA molecule consisting
15 of a least a portion of a nucleotide sequence encoding human
B7-2 . . . .
(Exh. 2040, p. 10, l. 36, to p. 11, l. 1);
20 A “fragment” of a nucleic acid encoding a novel B
lymphocyte antigen is defined as a nucleotide sequence
having fewer nucleotides than the nucleotide sequence
encoding the entire amino acid sequence of the B lymphocyte
antigen and which encodes a peptide having an activity of
25 the B lymphocyte antigen (i.e., the ability to bind to the
natural ligand(s) of the B lymphocyte antigen on immune
cells, such as CTLA4 and/or CD28 on T cells and either
stimulate or inhibit immune cell costimulation). Thus, a
peptide having B7-2 activity binds CTLA4 and/or CD28 and
30 stimulates or inhibits a T cell mediated immune response, as
evidenced by, for example, cytokine production and/or T cell
proliferation by T cells that have received a primary
activation signal. In one embodiment, the nucleic acid
fragment encodes a peptide of the B7-2 antigen which retains
35 the ability of the antigen to bind CTLA4 and/or CD28 and
deliver a costimulatory signal to T lymphocytes. In another
embodiment, the nucleic acid fragment encodes a peptide
including an extracellular portion of the human B7-2 antigen
(e.g., approximately amino acid residues 24-245 of the
40 sequence provided in Figure 8 (SEQ ID NO:2)) which can be
used to bind CTLA4 and/or CD28 and, in monovalent form,
inhibit costimulation, or in multivalent form, induce or
enhance costimulation.
45 (Exh. 2040, p. 12, ll. 18-33);
Interference No. 105,188
Short v. Punnonnen
-9-
This invention further pertains to expression vectors
containing a nucleic acid encoding at least one peptide
having the activity of a novel B lymphocyte antigen, as
described herein, operably linked to at least one regulatory
5 sequence. . . . .
(Exh. 2040, p. 14, ll. 7-9);
The invention also features methods of producing
10 peptides having an activity of a novel B lymphocyte antigen.
For example, a host cell transfected with a nucleic acid
vector directing expression of a nucleotide sequence
encoding a peptide having activity of the B7-2 protein can
be cultured in a medium under appropriate conditions to
15 allow expression of the peptide to occur. . . . .
(Exh. 2040, p. 14. ll. 29-33);
Transfected cells which express peptides having
20 activity of one or more B lymphocyte antigens (e.g., B7-2,
B7-3) on the surface of the cell are also within the scope
of this invention.
(Exh. 2040, p. 15, ll. 29-31); and
25
[A] peptide having an activity of the B7-2 protein is
defined herein as a peptide having the ability to bind to
the natural ligand(s) of the B7-2 protein on immune cells,
such as CTLA4 and/or CD28 on T cells and either stimulate or
30 inhibit cell costimulation. Thus, a peptide having B7-2
activity binds CTLA4 and/or CD28 and stimulates or inhibits
a T cell mediated immune response (as evidenced by, for
example, cytokine production and/or proliferation by T cells
that have received a primary activation signal). . . .
35 Alternatively, a peptide having both B7-2 binding activity
and the ability to deliver a costimulatory signal to T cells
is used to stimulate or enhance T cell proliferation and/or
cytokine secretion in a subject.
40 (Exh. 2040, p. 16, ll. 22-30; emphasis added).
We find that the Freeman PCT alone describes procedures for
obtaining polynucleotides encoding active B7-2 co-stimulator
Interference No. 105,188
Short v. Punnonnen
-10-
variants having an enhanced ability to modulate an immune
response induced by a genetic vaccine vector and having altered
activity through CTLA4 and/or CD28 as compared to a predecessor
B7-2 molecule, e.g., natural or wild-type B7-2. The new
5 polynucleotides obtained in accordance with the procedures
described in the Freeman PCT may encode proteins which upregulate
immune responses by either inhibiting delivery of a costimulatory
signal to T cells or enhancing delivery of a costimulatory signal
to T cells. The altered B7-2 antigens include soluble,
10 multivalent forms of B7-2. While the reference teaches that
enhanced delivery of stimulatory B7-2 may be useful to improve
the efficiency of vaccination against a variety of pathogens, it
also teaches that delivery of nonstimulatory B7-2 may be useful
to construct B7-2 “knock out” animals. In either case, the
15 procedures described by the Freeman PCT are designed to produce
polynucleotides that encode peptides having an altered modulatory
effect on an immune response as compared to the response prior to
alteration.
Whether or not each of the methods expressly described by
20 the Freeman PCT optimizes the modulatory effect of the encoded
protein is indeterminable from the teaching of the Freeman PCT.
We find, however, that all the methods described in the Freeman
PCT were designed to improve the modulatory effect of
Interference No. 105,188
Short v. Punnonnen
-11-
the encoded protein on an immune response as compared to the
response prior to alteration. The procedures described in the
Freeman PCT for obtaining new polynucleotides comprise the steps
of: a) creating a library comprising recombinant polynucleotides,
5 and b) screening the library to identify polynucleotides which
encode B7-1 and/or B7-2 variants having altered modulatory
effects on an immune response, whether the effects are enhanced
immune responses or inhibited immune responses to a vaccine
vector. For a description of construction of a cDNA library,
10 see the Freeman PCT at page 18, l. 27, to page 19, l. 20 (II.
Isolation of mRNA and Construction of cDNA Library); page 70,
l. 11, to page 72, l. 4 (Example 4: Cloning, Sequencing and
Expression of the B7-2 Antigen, A. Construction of cDNA Library);
and page 78, l. 17, page 80, l. 6 (Example 6, Cloning and
15 Sequencing of the Murine B7-2 Antigen, A. Construction of cDNA
Library). On consideration of more detailed instructions in the
Freeman PCT, we also find that the inventive procedures described
therein include recursive screenings for DNA encoding altered,
active B7-2 antigens (Exh. 2040, III. Transfection of Host Cells
20 and Screening for Novel B Lymphocyte Activation Antigens, p. 19,
l. 23, to p. 20, l. 6; emphasis added):
The thus prepared cDNA library is then used to clone
the gene of interest by expression cloning techniques.
. . . .
Interference No. 105,188
Short v. Punnonnen
-12-
According to one embodiment, plasmid DNA is introduced
into a simian COS cell line . . . by known methods of
transfection . . . and allowed to replicate and express cDNA
inserts. The transfectants expressing B7-1 antigen are
5 depleted with an anti-B7-1 monoclonal antibody . . . and
anti-murine IgG and IgM coated immunomagnetic beads.
Transfectants expressing human B7-2 antigen can be
positively selected by reacting the transfectants with the
fusion proteins CTLA4Ig and CD28Ig, followed by panning with
10 anti-human Ig antibody coated plates. Although human
CTLA4Ig and CD28Ig fusion protiens were used in the examples
described herein, given the cross-species reactivity between
B7-1 and, for example murine B7-1, it can be expected that
other fusion proteins reactive with another cross-reactive
15 species could be used. After panning, episomal DNA is
recovered from the panned cells and transformed into a
competent bacterial host . . . . Plasmid DNA is
subsequently reintroduced into COS cells and the cycle of
expression and panning repeated at least two times. After
20 the final cycle, plasmid DNA is prepared from individual
colonies, transfected into COS cells and analyzed for
expression of novel B lymphocyte antigens by indirect
immunofluorescence with, for example, CTLA4Ig and CD28Ig.
25 Example 4 (Exh. 2040, B. Cloning Procedure, p. 72, ll. 6-32) and
Example 6 (Exh. 2040, B. Cloning Procedure, p. 80, ll. 8-34)
explain how the recursive screening steps of the procedures
described in the Freeman PCT select polynucleotide sequences
which encode novel human or murine B7-2 variants having markedly
30 enhanced or inhibited immune responses to a vaccine vector.
Focusing on the detailed teachings of the Freeman PCT, we
find that one method described in this primary reference for
obtaining an immunomodulatory polynucleotide that encodes a B7-2
antigen which has an altered modulatory effect on an immune
35 response induced by a genetic vaccine vector as compared to the
Interference No. 105,188
Short v. Punnonnen
-13-
response to the unaltered antigen comprises: a) creating a
library of DNA encoding a natural B7-2 antigen and DNA altered to
encode a B7-2 variant having reduced activity or enhanced
inactivity; b) inducing homologous recombination between the
5 natural DNA and the altered DNA comprising a DNA library; and
c) screening all new DNA resulting from homologous recombination
of natural DNA and altered DNA comprising the DNA library for
expression of B7-2 variants which reduce the animal’s immune
response to a vaccine as compared to its natural immune response.
10 See the following excerpts from the Freeman PCT (Exh. 2040,
p. 22, l. 17, to p. 23, l. 17; emphasis added):
A transgene is a DNA which is integrated into the genome of
a cell from which a transgenic animal develops. In one
embodiment, murine B7-2 cDNA or an appropriate sequence
15 thereof can be used to generate transgenic animals that
contain cells which express B7-2 protein. Methods for
generating transgenic animals, particularly animals such as
mice, have become conventional in the art and are described,
for example, in U.S. Patent Nos. 4,736,866 and 4,870,009.
20 Typically, particular cells would be targeted for B7-2
transgene incorporation with tissue specific enhancers,
which could result in T cell costimulation and enhanced T
cell proliferation and autoimmunity. Transgenic animals
that include a copy of a B7-2 transgene introduced into the
25 germ line of the animal at an embryonic stage can be used
to examine the effect of increased B7 expression. Such
animals can be used to examine the effect of increased B7
expression. Such animals can be used as tester animals for
reagents thought to confer protection from, for example,
30 autoimmune disease. In accordance with this facet of the
invention, an animal is treated with the reagent and a
reduced incidence of the disease, compared to untreated
animals bearing the transgene, would indicate a potential
therapeutic intervention for the disease.
Interference No. 105,188
Short v. Punnonnen
-14-
Alternatively, the non-human homologues of B7-2 can be
used to construct a B7-2 “knock out” animal which has a
defective or altered B7-2 gene as a result of homologous
recombination between the endogenous B7-2 gene and altered
5 B7-2 genomic DNA introduced into an embryonic cell of the
animal. For example, murine B7-2 cDNA can be used to clone
genomic B7-2 in accordance with established techniques. A
portion of the genomic B7-2 DNA (e.g., such as an exon which
encodes an extracellular domain) can be deleted or replaced
10 by another gene, such as a gene encoding a selectable marker
which can be used to monitor integration. Typically,
several kilobases of unaltered flanking DNA (both at the
5' and 3' ends) are included in the vector (see e.g.,
Thomas, K.R. and Capecchi, M. R. (1987) Cell 51:503 for a
15 description of homologous recombination vectors). The
vector is introduced into an embryonic stem cell line (e.g.,
by electroporation) and cells in which the induced DNA has
homologously recombined with the endogenous DNA are selected
(see e.g., Li, E. et al. (1992) Cell 69:915). The selected
20 cells are then injected into a blastocyst of an animal
(e.g., mouse) to form aggregation chimeras (se e.g.,
Bradley, A. in Teratocarcinomas and Embryonic Stem Cells: A
Practical Approach, E.J. Robertson, ed. (IRL, Oxford, 1987)
pp. 113-152). The chimeric embryo can then be implanted
25 into a suitable pseudopregnant female foster animal and the
embryo brought to term to create a “knock out” animal.
Progeny harbouring the homologously recombined DNA in their
germ cells can be identified by standard techniques and used
to breed animals in which all cells of the animal contain
30 the homologously recombined DNA. Knockout animals can be
characterized for their ability to accept grafts, reject
tumors and defend against infectious diseases and can be
used in the study of basic immunobiology.
We have reviewed the findings at page 142, first paragraph,
35 of our prior decision in light of the reference’s teaching as
a whole (Paper No. 181). We now find that the Freeman PCT
prima facie describes a method for obtaining an optimized
polynucleotide encoding a B7-2 co-stimulator variant having an
enhanced ability to modulate an immune response induced by a
Interference No. 105,188
Short v. Punnonnen
-15-
genetic vaccine vector and having altered activity through CTLA4
or CD28 as compared to a predecessor B7-2 molecule, e.g., wild-
type B7-2. However, the reference also teaches that optimization
may be achieved by recursive activity whereby (Exh. 2040,
5 pp. 19-20, III. Transfection of Host Cells and Screening for
Novel B Lymphocyte Activation Antigens) a prepared cDNA library
is used to clone the B7-2 gene of interest, plasmid DNA is
introduced into a simian COS cell line by known methods of
transfection and allowed to replicate and express the cDNA
10 inserts, transfectants expressing human B7-2 antigen are selected
by reacting the transfectants with the fusion proteins CTLA4Ig,
CD28Ig and other cross-reactive proteins and panning, episomal
DNA is recovered from the panned cells and transformed into
E. coli, plasmid DNA is reintroduced into COS cells and the
15 expression and panning cycle is repeated at least two times, and
the COS cells of the final repeat are analyzed for expression of
novel B lymphocyte antigens using CTLA4Ig and CD28Ig.
Moreover, we find explicit teaching in the Freeman PCT that
optimization may be achieved by homologous recombination of DNA
20 comprising a library of DNA including endogenous and altered B7-2
DNA (Exh. 2040, p. 22, l. 17, to p. 23, l. 17). Accordingly,
while our previous finding that Short failed to establish that
the Freeman PCT explicitly describes recursive sequence
Interference No. 105,188
Short v. Punnonnen
-16-
recombination appears to have been correct, we did not consider
the more general teachings in the Freeman PCT to obtain an
immunomodulatory polynucleotide having an optimized modulatory
effect on an immune response as compared to the response prior to
5 optimization by methods whereby optimization is achieved by
homologous sequence recombination or selecting and recycling DNA
exhibiting altered antigenic activity.
3. Short PCT
We now revisit the Short PCT. We consider the teaching of
10 the Short PCT anew in light of our finding herein that the
teaching of the Freeman PCT is much more pertinent to the subject
matter defined by Punnonen’s Claim 47 and much less deficient
than we previously had found.
The Short PCT teaches (Exh. 2050, p. 33, l. 35, to p. 35,
15 l. 17; emphasis added):
The present invention also provides a novel method for
screening variants of a parental clone or clones. If the
parental clone or clones contain two nucleotide sequences
that, when expressed together, create a phenotype, then such
20 nucleotide sequences can be altered to create populations of
variants of such nucleotide sequences. If the two variant
populations are coexpressed in a random fashion (that
is with no correlation between the specific alterations
made in the two different nucleotide sequences), then a
25 combinatorial collection of such nucleotide sequence
variants has been created. Such combinatorial collections
may be screened for the presence of phenotypes that are
unlike the parental clone or clones. Generally, the method
combines the following elements:
30
Interference No. 105,188
Short v. Punnonnen
-17-
1. Replicating a clone containing a nucleotide
sequence under conditions that allow mutations to occur.
2. Replicating a second clone containing a second
nucleotide sequence under conditions that allow mutations to
5 occur.
3. Randomly combining and co-expressing the two
mutated populations of nucleotide sequences.
4. Screening clones containing combinations of the
mutated nucleotide sequences for phenotypes that were not
10 present in either parent clone.
Alternatively, the methods combine the following
elements:
15 1. Replicating at least portions of two nucleotide
sequences contained within a single clone under conditions
that allow mutations to occur in either nucleotide sequence.
2. Allowing recombination events between the two
nucleotide sequence populations to reassociate mutant
20 nucleotide sequences to form new pairs of the two sequences
that were not paired in the original mutated, replicated
population.
3. Screening clones containing combinations of
nucleotide sequences for phenotypes that were not present in
25 the parent clone or in the mutant replicas of the parent
clone.
For example, assume a parent clone containing two
nucleotide sequences A and B is replicated under mutating
30 conditions such that variant clones are formed:
Parent: A/B
Variant 1: A1/B
Variant 2: A/B1
35 Variant 3: A2/B1
Variant 4: A/B2
Variant 5: A3/B
However, within this mutated population, the combinations
40 A1/B2, A2/B, A2/B2, A3/B1, and A3/B2, do not occur. If the
mutant population (including non-mutated parent clones) is
allowed to recombine sequences A and B and their variants,
then combinations such as A1/B2, A2/B etc. can be created.
Such new combinations may express a desired phenotype that
45 was not present in the parental or variant population.
Interference No. 105,188
Short v. Punnonnen
-18-
We find that the hereinabove transcribed paragraphs of the
Short PCT describe methods for obtaining polynucleotides having
an optimized phenotype comprising: a) creating a library of
recombinant polynucleotides; and b) screening the library to
5 identify those polynucleotides having a variant phenotype,
whereby optimization is achieved by recursive sequence
recombination. The Short PCT adds (Exh. 2050, p. 40, ll. 5-18):
In some cases, it is desirable to bias the repertoire
for a preselected activity, such as by using as a source of
10 nucleic acid cells (source cells) from vertebrates in any
one of various stages of age, health and immune response.
For example, repeated immunization of a healthy animal prior
to collecting rearranged B cells results in obtaining a
repertoire enriched for gentic material producing a ligand
15 binding polypeptide of high affinity.
Accordingly, the foundation for our previous finding that
“the Short PCT fails to remedy the deficiencies of the Freeman
PCT [because] . . . Short has not shown that the Short PCT
20 describes using recursive sequence recombination to create
recombinant polynucleotide libraries as required by Punnonnon
claim 47" (Paper No. 181, p. 142, third full paragraph) did not
take into account the full scope and content of the teachings of
the Freeman PCT.
25 4. Stemmer
We find that, as of the date Punnonen’s involved application
was first filed, recursive sequence recombination was a procedure
Interference No. 105,188
Short v. Punnonnen
-19-
well-known in the pertinent art for creating polynucleotide
variants encoding polypeptides eliciting an optimized immune
response in an expression host as compared to the immune
response elicited by the polypeptides encoded by the library of
5 polynucleotides from which the novel polynucleotide variants were
recombined. This finding is consistent with the prior art
teaching of the publication of Punnonen’s co-inventor Stemmer of
record. Willem P. C. Stemmer (Stemmer), “Searching Sequence
Space,” Biotechnology, Vol. 13, pp. 549-553 (June 1995)
10 (Exh. 2052), teaches that recursive sequence recombination is a
procedure well-known for rapidly creating the best polynucleotide
sequences for a specific task (Paper No. 60, p. 6, para. 12).
The 1995 Stemmer publication teaches:
Experiments conducted to date suggest that benchtop
15 in vitro evolution techniques are capable of creating
libraries as large as molecules. Researchers can mimic
natural evolution by searching these libraries for the best
candidates for a specific task: Repeated round[s] . . . of
selection and amplification of candidates has already been
20 shown to produce novel molecules capable of binding with
equal or higher affinity than their natural counterpart.
(Exh. 2052, p. 549, col. 1, third para.);
25 The problem with point mutagenesis strategies can be traced
to the fact that they introduce random “noise” into a
message at every cycle. If the noise level is too high
–relative to the library size and the selection stringency-
the message will gradually become nonsensical. This problem
30 has generated a great deal of speculation about how to
create “biased” libraries that attempt to use similar,
rather than random, amino acid substitutions. The optimum
Interference No. 105,188
Short v. Punnonnen
This resembles the procedure used in the Freeman PCT3
(Exh. 2040, pp. 22-23, bridging para.).
-20-
level of random point mutations introduced at each cycle
would be much lower if existing, already proven, point
mutations or combinations could be permuted by
recombination.
5
(Exh. 2052, p. 550, col. 2, paragraph bridging cols. 1-2);
Since recombination can occur even with “naked” DNA,
“it would seem that it would be easy to employ recombination
10 techniques in order to evolve complex sequences-or even
whole genomes-at the benchtop. But in reality this has not
been the case. To the surprise of most researchers,
recombination methods for mutagenesis, selection, and
replication have been extremely difficult to develop in the
15 laboratory.
(Exh. 2052, pp. 550-551, bridging para.);
So far, a technique called sexual PCR, or DNA
20 shuffling, comes closest to mimicking natural recombination
by allowing the in vitro homologous recombination of DNA.
(Exh. 2052, p. 551, col. 1, first full para.);
25 This method offers practical and theoretical advantages
over existing recursive mutagenesis methods, such as error-
prone PCR or recursive oligonucleotide directed mutagenesis.
By recombining point mutations and wild-type sequences,
followed by selection for function, sexual PCR will[3]
30 rapidly fine tune the mutation load present in different
parts of a protein. Over several cycles, those areas that
tolerate a high mutation load accumulate more diversity than
those areas that are less permissive to modification.
35 (Exh. 2052, p. 551, cols. 1-2, bridging para.);
Sexual PCR’s advantages for searching sequence space
can be further explained by understanding how it employs
poolwise and pairwise recombination strategies.
40 Recombination by sexual PCR is, in principal, poolwise-more
than two parental sequences can contribute to form the
Interference No. 105,188
Short v. Punnonnen
-21-
chimeric molecule. This is a very efficient method of
combining useful mutations from many different parental
sequences through multiple crossovers. Sexual PCR’s high
recombination rate takes advantage of this by recombining a
5 large number of mutations in a minimum number of selection
cycles.
(Exh. 2052, p. 551, col. 2, first full para.); and
10 The advantage of sexual PCR is that it uses a very low
rate of random mutation . . . . The longer the sequence,
the lower the rate needs to be. Recombination and selection
quickly yield functional blocks containing combinations of
compatible point mutations. At a larger scale, these
15 functional building blocks are permutated and the best
combinations are selected. The low point mutation rate and
the homology requirement for recombination make it a process
that is mostly gentle and conservative, but also capable of
big sequence jumps.
20
(Exh. 2052, p. 551, col. 2, second to last full para.).
Regarding a promising alternative technique called exon
shuffling, Stemmer states (Exh. 2052, p. 552, cols. 1-2, bridging
25 para.):
Exon shuffling may also be used to obtain random
peptide libraries of very large effective sizes. Without
homologous introns, the degree of homology between peptides
selected from a random library would be insufficient for
30 recombination. . . . . Recombination inside the exon is
unlikely because of their random sequence origin and much
shorter length. The recombined exons are cloned back into
the phage, and a new library of 10 is obtained and8
screened. This process is repeated until no further
35 improvement occurs.
5. Prima facie obviousness
We have considered the combined teachings of the Freeman and
Short PCTs in light of Stemmer’s description of the knowledge and
Interference No. 105,188
Short v. Punnonnen
-22-
skill of a person having ordinary skill in the art at the
pertinent time. We conclude that the method defined by Claim 47
of Punnonen’s involved application prima facie would have been
obvious to persons having ordinary skill in the art as of the
5 effective filing date of Punnonen’s application.
Prior art disclosures should be considered for everything
they would have taught a person having ordinary skill in the art
when the issue of patentability under 35 U.S.C. § 103 in view of
that prior art is raised. “The description of . . . ‘specific
10 preferences in connection with . . . [the] generic’ is
determinative in an analysis of anticipation under 35 U.S.C.
§ 102. . . . But in a section 103 inquiry, ‘the fact that a
specific [embodiment] is taught to be preferred is not
controlling, since all disclosures of the prior art, including
15 unpreferred embodiments, must be considered.’ In re Lamberti,
545 F.2d 747, 750, 192 USPQ 278, 280 (CCPA 1976).” Merck & Co.
Inc. V. Biocraft Laboratories Inc., 874 F.2d 804, 807, 10 USPQ2d
1843, 1846 (Fed. Cir. 1989). In re Fracalossi, 681 F.2d 792, 794
n.1, 215 USPQ 569, 570 n.1 (CCPA 1982), teaches:
20 It is axiomatic that a reference must be considered in its
entirety, and it is well established that the disclosure of
a reference is not limited to specific working examples
contained therein. In re Lamberti, 545 F.2d 747, 750,
192 USPQ 278, 280 (CCPA 1976).
25
Interference No. 105,188
Short v. Punnonnen
-23-
In re Boe, 355 F.2d 961, 965, 148 USPQ 507, 510 (CCPA 1966),
instructs, “All of the disclosures in a reference must be
evaluated for what they fairly teach one of ordinary skill in the
art.” Similarly, In re Bode, 550 F.2d 656, 661, 193 USPQ 12, 17
5 (CCPA 1977), directs, “A reference must be evaluated for all it
teaches and is not limited to its specific embodiments.” “[A]
prior art reference is relevant for all that it teaches to those
of ordinary skill in the art.” In re Fritch, 972 F.2d 1260,
1264, 23 USPQ2d 1780, 1782 (Fed. Cir. 1982). Summarizing, In re
10 Heck, 699 F.2d 1331, 1332-33, 216 USPQ 1038, 1039 (Fed. Cir.
1983), instructs:
As the Court of Customs and Patent Appeals said in a section
103 case (In re Lemelson, 397 F.2d 1006, 1009, 158 USPQ 275,
277 ([CCPA] 1968)):
15
“The use of patents as references is not limited to
what the patentees describe as their own inventions or
to the problems with which they are concerned. They
are part of the literature of the art, relevant for all
20 they contain.”
We previously focused on the preferred embodiments of both
the Freeman PCT and the Short PCT and their shortcomings relative
to the specific subject matter defined by Punnonen’s Claim 47.
25 However, Stemmer’s view of the state of the art (Exh. 2052) is
entitled to considerable weight because it is the unbiased view
of one of Punnonen’s co-inventors and it was published long
Interference No. 105,188
Short v. Punnonnen
-24-
before the effective filing dates of both Punnonen’s involved
application and Short’s involved patent.
We conclude that the method defined by Claim 47 of
Punnonen’s involved application prima facie would have been
5 obvious under 35 U.S.C. § 103 to a person having ordinary skill
in the art on the effective filing date of Punnonnen’s involved
application in view of the combined teachings of the Freeman PCT,
the Short PCT, and Stemmer. Our conclusion newly emphasizes
Stemmer’s view of the knowledge, skill and state of the art at
10 the pertinent time. We emphasize our view that the Freeman PCT
describes, and/or reasonably would have suggested, all aspects of
a method encompassed by Claim 47 of Punnonen’s involved
application to persons having ordinary skill in the art but for
the limitation “whereby optimization is achieved by recursive
15 sequence recombination” (Claim 47, Paper No. 181, p. 18).
However, as we have indicated above, we find in the Freeman PCT
suggestions to optimize the modulatory effect on an immune
response sought by the methods taught in the Freeman PCT by use
of mutation procedures, including homologous recombination and
20 point or sequence mutations, followed by screening, recycling
and/or repeating the procedures for optimum results. In light of
the knowledge and skill in the art at the pertinent time, as
evident from the disclosures of the Short PCT and Stemmer, the
Interference No. 105,188
Short v. Punnonnen
-25-
use of recursive sequence recombination and/or other recursive
mutation techniques for creating novel DNA using a recombinant
DNA library would have been obvious to achieve optimum results.
Persons having ordinary skill in the art would have required no
5 more instruction and/or motivation than is provided by the prior
art relied upon to make and use the method of Punnonen’s
Claim 47.
Where, as here, the general conditions of a claimed process
are disclosed in the prior art, it is not inventive to optimize
10 by routine experimentation. In re Aller, 220 F.2d 454, 456,
105 USPQ 233, 235 (CCPA 1955). Here, the evidence shows that
persons having ordinary skill in the art understood that optimum
results could be achieved by recursive homologous recombination
at the pertinent time without undue experimentation. Stemmer
15 acknowledges that it was within the ordinary skill of the artisan
to optimize the modulatory effect on an immune response of a
protein encoded by recombinant DNA by recursive sequence
recombination. Moreover, the obviousness of the method of
Claim 47 of Punnonen’s involved application is kindled by
20 specific intructions in the Short PCT to repeat recombination
procedures to create a desired phenotype (Exh. 2050, p. 33,
l. 35, to p. 35, l. 17). The Short PCT directs persons skilled
in the art to utilize recursive sequence mutations, including
Interference No. 105,188
Short v. Punnonnen
-26-
recursive sequence recombinations, to produce novel DNA which
encodes a polypeptide having desirably enhanced properties.
Accordingly, we conclude that the combined teachings of the
Freeman PCT, the Short PCT, and Stemmer reasonably would have led
5 persons having ordinary skill in the art to optimize the
modulatory effect of B7-2 variants encoded by novel DNA produced
by homologous recombination by: a) creating a library of
recombinant DNA encoding B7-2, b) promoting sequence
recombination in the recombinant DNA library, c) screening the
10 library for novel recombinant DNA which exhibits an enhanced
ability to modulate an immune response induced by a genetic
vaccine vector, and optimizing the enhanced ability of the novel
recombinant DNA to modulate the immune response by recursive DNA
sequence recombination, and reasonably expect success.
15 Subject matter defined by Punnonen’s Claim 47 appears to be
unpatentable under 35 U.S.C. § 103 as having been prima facie
obvious to a person having ordinary skill in the art in view of
the combined teachings of the Freeman PCT, the Short PCT, and
Stemmer. Therefore, we recommend that further examination and
20 subsequent action relating to the patentability of the subject
Interference No. 105,188
Short v. Punnonnen
-27-
matter defined by Punnonen’s Claim 47 proceed consistent with the
views expressed in this memorandum.
5
10
RICHARD E. SCHAFER )
Administrative Patent Judge )
)
)
15 ) BOARD OF PATENT
) APPEALS AND
TEDDY S. GRON ) INTERFERENCES
Administrative Patent Judge )
)
20 )
)
)
RICHARD TORCZON
Administrative Patent Judge
25
30
35
40
Interference No. 105,188
Short v. Punnonnen
-28-
cc (via overnight mail):
Attorney for Short:
5 Jane M. Love Ph.D., Esq.
Caren K. Khoo, Esq.
WILMER CUTLER PICKERING HALE and DORR LLP.
399 Park Avenue
New York, New York 10022
10 Tel: (212) 937-7233
Fax: (212) 937-7300
Attorney for Punnonen:
15
R. Danny Huntington, Esq.
Sharon E. Crane, Ph.D., Esq.
BINGHAM McCUTCHEN LLP.
2020 K Street, NW
20 Washington, DC 20006-1806
Tel: (202) 373-6000
Fax: (202) 373-6367
25