CONTROL NOS. 90/007,542 AND 90/007,859
`
`DOCKET NOS. 22338-10230 AND -10231
`
`Control Nos. 90/007,542; 90/007,859
`
`Attorney Docket No. 22338-10230
`
`Patent
`
`IN THE UNITED STATES PATENT AND TRADEMARK OFFICE
`
`Control Nos.:
`
`Confirmation Nos.:
`
`90/007,542
`90/007,859
`
`7585 (’542)
`6447 (’859)
`
`Filed:
`
`13 May 2005
`23 December 2005
`
`(’542)
`('859)
`
`Patent Owner:
`
`Genentech, Inc. and
`City of Hope
`
`Group An Unit:
`
`3991
`
`Examiner:
`
`B.M. Celsa
`
`For:
`
`Merged Reexaminations of U.S. Patent No. 6,331,415 (Cabilly et al.)
`
`DECLARATION OF STEVEN LANIER MCKNIGHT UNDER 37 C.F.R. § 1.132
`
`1, Steven Lanier McKnight, do hereby declare and state
`
`1.
`
`2.
`
`I am a citizen of the United States and reside in Dallas, Texas. My c.v. is attached as
`Exhibit A.
`
`I have been retained by Genentech and City of Hope to provide my opinion on certain
`issues in the patent reexamination proceedings involving U.S. Patent No. 6,331,415 (“the
`’415 patent”). I am being compensated for my time at a rate of $750.00 per hour.
`
`3.
`
`I have reviewed the following documents in the course of preparing this declaration:
`
`U.S. Patent No. 5,840,545 (“the ’545 patent”);
`
`U.S. Application No. 06/358,414 (“the ’4l4 application”);
`
`The ’415 patent;
`
`U.S. Patent No. 4,816,567 (“the ’567 patent”);
`
`U.S. Patent No. 4,399,216 (“Axel”);
`
`Deacon & Ebringer, BIOCHEMICAL Socnarv TRANSACTIONS 4: 818-820 (1976)
`
`(“Deacon’_’);
`
`European Patent No. 0 044 722 (“Kaplan”);
`
`Ochi er al., NATURE 302: 340-342 (1983) (“Ochi”);
`
`EVIDENCE APPENDIX
`
`GENENTECH 2021
`GEl\'ZY\/IE V. GENENTECH
`lPR2016—00383
`
`PAGE B197
`
`_
`
`GENENTECH 2021
`GENZYME V. GENENTECH
`IPR2016-00383
`
`

`
`CONTROL NOS. 90/007,542 AND 90/007,859
`
`DOCKET NOS. 22338-10230 AND -10231
`
`Control Nos. 90/007,542; 90/007,859
`
`Attorney Docket No. 22338-10230
`
`Patent
`
`-
`
`-
`
`-
`
`-
`
`-
`
`-
`
`~
`
`Oi et al., PROC. NATL. ACAD. Sci. 80: 825-829 (1983) (“Oi”);
`
`Rice & Baltimore, PROC. NATL. ACAD. SCI. 79: 7862-7865 (1982) (“Rjce” ;
`
`Valle et al., NATURE 300: 71-74 (1982) (“Valle l982”);
`
`Valle et aI., NATURE 291: 338-340 (1981) (“Valle 198 I”);
`
`WO 82/03088 (“Dallas”);
`
`The Declaration of Richard Axel filed during prosecution of U.S. Application No.
`08/422,187;
`
`'
`
`Opposition Request of European Patent No. 0120694 filed in the European Patent
`Office on behalf of Genentech, Inc.
`
`4.
`
`I have also reviewed documents associated with the two reexamination proceedings,
`including:
`
`-
`
`-
`
`-
`
`-
`
`-
`
`The PTO Office Action dated February I6, 2007;
`
`The PTO Office Action dated August 16, 2006;
`
`A Request for Ex Parte Reexamination dated December 23, 2005, including
`attachments to that Request;
`
`The Declaration of David Baltimore submitted in connection with the December
`
`23, _2005 Request for Ex Parte Reexamination;
`
`The Declarations of Dr. Rice, Dr. Colman, and Dr. Harris filed with the responses
`of the patent owner to the two office actions
`
`5.
`
`6.
`
`I understand that patentability is evaluated using the perspective of a person of ordinary
`skill in the technical field of the invention just prior to the filing date of the patent (i. e.,
`early April of 1983). A person of ordinary skill in the field of the ’4l5 patent would have
`had a Ph.D. in molecular biology or a comparable scientific discipline, and two to three
`years of practical experience, such as that gained through a post-doctoral appointment or
`comparable assignment.
`I believe I am well-qualified to express an opinion on what a
`person of ordinary skill in the art of the ’4l5 patent would have believed or expected in
`early April of 1983, because at that time I was a person who had a level of experience in
`line with this definition and worked with people who met this definition.
`
`I also
`I understand that the ‘S45 patent issued from an application filed on June 5, I995.
`understand that there were several earlier applications filed between 1982 and 1995
`related to the ’545 patent.
`I understand that the first of these applications was the ’414
`application filed in March of 1982, and that the contents of this application are to be the
`
`EVIDENCE APPENDIX
`
`PAGE 3198
`
`

`
`CONTROL NOS. 90/007,542 AND 90/007,859
`
`DOCKET NOS. 22338-10230 AND -1 0231
`
`Control Nos. 90/007,542; 90/007,859
`
`Attorney Docket No. 22338-10230
`
`Patent
`
`focus of my analysis. In particular, I have been asked to detennine if the ’4l4 application
`describes a host cell that produces two different immunoglobulin chain polypeptides or a
`process where two different polypeptides are expressed in a single host cell.
`I understand
`that the requirements of the host cell and process are outlined in the claims of the ‘S45
`patent.
`
`I also have been asked to determine if there is any description in the ’4l4 application of
`procedures for coexpressing two different polypeptides in a single host cell. Finally, 1
`have been asked to provide my views on the observations of the PTO contained in the
`Final Office Action dated February 17, 2007.
`
`General Observations On The '414 Application
`
`8.
`
`10.
`
`The ’4l4 application describes procedures for making what it calls an rFv binding
`composition, or rFv. An rFv consists of two polypeptides, each with an amino acid
`sequence that corresponds to the variable region sequence of an immunoglobulin chain.
`An “L-rFv” polypeptide contains a variable region sequence from a light chain
`immunoglobulin, and an “H-rFv” contains a variable region sequence from a heavy chain
`immunoglobulin.
`
`The ’4 14 application indicates that an rFv can contain two polypeptides with the same
`amino acid sequence, or with different amino acid sequences. See, p. 3, line 37 to p. 4,
`line 2 (“the L- and H- designations will normally mean light and heavy respectively, but
`in some instances the two chains [of the rFv] may be the same and derived from either
`the light or heavy chain sequences”).
`
`Pages 5 to 18 of the ’4l4 application provide a general description of procedures for
`producing L-rFv and H-rFv polypeptides, and rFv binding compositions. The ’4l4
`application also provides an example of using these procedures on pages 19-42
`(“Example 1”). These procedures can be summarized as follows:
`
`a.
`
`b_.
`
`c.
`
`Produce a hybridoma that makes an antibody with a desired specificity. See, p. 5,
`line 32 to p. 6, line 18.
`
`Prepare a purified whole cell mRNA extract from the hybridoma, and use this to
`prepare a cDNA library using a reverse transcriptase. See, p. 6, line 19 to p. 8,
`line 7. This produces cDNA molecules with sequences that are complementary to
`each of the discrete mRNA sequences (mRNA transcripts) in the mRNA extract.
`
`Amplify the cDNA library. This is done by inserting the cDNA molecules into
`plasmids, transforming a bacterial host cell culture with the plasmids, and
`growing the transformed bacterial host cells under selective pressure (i.e., in the
`presence of an agent that causes bacterial cells that did not incorporate a plasmid
`to die). This produces a collection of bacterial clones, each containing a plasmid
`with one of the cDNA molecules from the cDNA library in it. See, p. 8, line 12 to
`p. 9, line 1.
`
`EVIDENCE APPENDIX
`
`PAGE B199
`
`

`
`CONTROL NOS. 90/007,542 AND 90/007,859
`
`DOCKET NOS. 22338-10230 AND -10231
`
`Control Nos. 90/007,542; 90/007,859
`
`Attorney Docket No. 22338-10230
`
`Patent
`
`d.
`
`e.
`
`f.
`
`g.
`
`h.
`
`Identify colonies of transformed bacterial cells that contain plasmids with cDNA
`encoding either the heavy or the light chain using a nucleotide probe
`corresponding to the constant domain of the heavy or the light chain. Then, select
`these colonies and grow the colonies under selective pressure to produce a
`population of identical copies (clones) of the bacterium with the desired heavy or
`light chain cDNA sequence. See, p. 9, lines l-19.
`
`Extract the cDNA from the individual clone selected by colony hybridization, and
`use it to produce a “tailored” cDNA that encodes the variable region of either the
`heavy or the light chain polypeptide. See, p. 9, line 20 to p. 14, line 15.
`
`Insert the modified cDNA into an expression vector (i. e., a plasmid containing an
`origin of replication, a promoter, and an insertion site), and transform another
`bacterial host (E. coli) with the plasmid. See, p. 14, line 16 to p. 16, line 23.
`
`Express either the light or heavy chain variable region polypeptide by growing a
`transformed bacterial host cell, and then isolate, purify, and renature the
`polypeptide. See, p. 17, line 1 to p. 18, line 14. Repeat the process with the other
`immunoglobulin chain.
`
`Combine the individually produced chains in vitro to form the rFv binding
`composition. See, p. 16, lines 24-28.
`
`1 1.
`
`12.
`
`If these procedures are followed as they are written, individual L-rFv and H-rFv
`polypeptides will be produced in separate cell cultures and these individually prepared
`polypeptides will be isolated, renatured, and combined in a test tube to form an rFv.
`I did
`not find any description of procedures in‘ the ’414 application of a “coexpression”
`strategy (i. e., where two polypeptides with different amino acid sequences would be
`produced in a single transformed cell culture).
`
`All of the processes described in the ’4l4 application relate to bacterial expression
`systems. There are some references to the use of yeast cell cultures to amplify DNA
`sequences, but there are no procedures described in the ’4l4 application for expressing
`proteins in yeast-based systems. There is also no description of using mammalian cell
`lines to produce rFv polypeptides in the ’4l4 application.
`
`The '414 Application Does Not Describe Or Suggest Coexpression OfL-rFv And H-rFv
`Polypeptides In A Single Host Cell
`
`13.
`
`I could find no description in the ’4 14 application of a single host cell that produces two
`different polypeptides, or a process where two different polypeptides are expressed in a
`single host cell. As such, [do not believe there is any description in the ’4 14 application
`of a host cell meeting the requirements of claim 1 of the ’545 patent, or a process meeting
`the requirements of claim 2 of the ’545 patent as these claims have been interpreted by
`the PTO.
`
`EVIDENCE APPENDIX
`
`4
`
`PAGE B200
`
`

`
`CONTROL NOS. 90/007,542 AND 90/007,859
`
`DOCKET NOS. 22338-10230 AND -10231
`
`Control Nos. 90/007,542; 90/007,859
`
`Attorney Docket No. 22338-10230
`
`Patent
`
`14.
`
`The ’4l4 application clearly states that an rFv is to be made by producing the L-rFv and
`H-rFv polypeptides in separate cells and combining them in a test tube afier expression
`and purification. For example, on page 16, lines 24-28, the ’4l4 application plainly
`states:
`
`The resulting construct [i.e., a cDNA insert encoding the L-rFv _o_r H-rFv
`polypeptide in an appropriate expression vector] is then introduced into-an
`appropriate host to provide expression of the heavy 9; light polypeptide
`members of the rFv and the polypeptides isolated. The heavy and light
`polypeptide members of the rFv are then combined in an appropriate ,
`medium to form the rFv. (emphasis added).
`
`15.
`
`16.
`
`17.
`
`18.
`
`This clearly indicates that each of the L-rFv and H-rFv polypeptides will be produced in
`separate cells. The “appropriate medium” being referred to is the test tube environment
`where the two expressed and purified polypeptide chains are finally mixed together afier
`they have been separately produced and isolated. An appropriate medium is not referring
`to a transformed bacterial host cell.
`
`All of the techniques and options in the ’4l4 application for producing L-rFv and H-rFv
`polypeptides are consistent with this approach. For example, page 17, lines 35-38,
`indicates that “[w]here the light or heavy chain is not secreted, the transformed
`microorganisms containing the appropriate ds cDNA for either light or heayy chains are
`grown in liquid culture and cleared lysates prepared.” (emphasis added). This again
`makes clear that each of the L-rFv and H-rFv polypeptides is being produced in a
`separate cell culture.
`
`Similarly, page 18, lines 4-7, indicates that the “eluates from each of the heayy and light
`chain isolations are pooled, followed by treatment to renature the polypeptides to form L-
`rFv and H-rFv respectively.” (emphasis added). These references to multiple isolations
`clearly indicate that separate cell cultures are being used to produce the two different
`polypeptides. A single isolate would be the result of lysing a single host cell that was
`producing both the L-rFv and H-rFv polypeptides.
`
`As such, in my opinion, it is absolutely clear that, if the procedures described in the ’4l4
`application are followed as they are written, each of the polypeptides will be produced in
`separate cells.
`I do not believe any other reading of these sections of the ’4l4 application
`would be rational, logical, or scientifically accurate.
`'
`
`The Procedures Described In The ’414 Application Will Not Yield Genetic Constructs Encoding
`More than One Polypeptide 0r Host Cells That Contain Multiple Plasmids
`
`19.
`
`The procedures in the ’414 application produce a “tailored” cDNA sequence by starting
`with a cDNA obtained from a cDNA library that encodes a full length heavy or full
`length light immunoglobulin chain polypeptide. See, p. 6, lines 19-34. These starting
`cDNA sequences are produced using mRNA transcripts isolated from a hybridoma that is
`producing an antibody with a desired specificity (i. e., a cDNA library is produced
`
`EVIDENCE APPENDIX
`
`5
`
`PAGE B201
`
`

`
`CONTROL NOS. 90/007,542 AND 90/007,859
`
`DOCKET NOS. 22338-10230 AND -10231
`
`Control Nos. 90/007,542; 90/007,859
`
`Attorney Docket No. 22338-10230
`
`Patent
`
`20.
`
`21.
`
`22.
`
`through reverse transcription of the mRNA transcripts in a purified mRNA extract from
`the hybidoma).
`
`By I982, it was well known that the heavy and light chains of an immunoglobulin are
`encoded by separate genes, and that these genes are located on different chromosomes.
`See, e.g., Hood et al., ANN. REV. GENET. 9: 305-353 (1975). When these genes are
`transcribed by the hybridoma cell, discrete mRNA transcripts will be produced — one
`associated with transcription of the light chain gene, and a different one from
`transcription of the heavy chain gene. Because the technique for producing the cDNA
`library makes cDNAs that are complementary to each of these individual mRNA
`sequences or transcripts, none of the cDNAs in the library will contain sequences
`corresponding to both heavy and light immunoglobulin chains.
`
`The amplification procedures described in the ’4 14 application produce copies of the
`cDNAs in the cDNA library. These procedures use simple transformations of bacterial
`cells using a common vector/plasmid. See, e.g., p. 8, lines 11-31; p. 27, lines 14-27.
`When this process is followed as it is described in the application, each plasmid will
`incorporate one cDNA from the cDNA library. None of the plasmids will contain
`sequences from both the heavy and the light chain immunoglobulin chains, so none of the
`amplified cDNA sequences will contain both heavy and light chain sequences.
`
`The “tailored” cDNAs made by the ’4l4 procedures use these amplified cDNA sequences
`as the “starting material” for the tailored cDNA sequence. The individual cDNAs are
`sequenced and subjected to restriction mapping. See, p. 9, lines 20-25 (“these analyses
`insure that the isolated cDNA clones completely encode the variable region and,
`optionally, the leader sequences for the light or heavy chain of the desired
`immunoglobulin.”). This source cDNA from the cDNA library is also used to prepare the
`tailored cDNA sequence encoding the L—rFv or the H-rFv.'
`
`23.
`
`There are no steps described in the ’4 14 application where different cDNAs from the
`cDNA library are ligated together before they are inserted into a plasmid for
`
`The process of producing the tailored cDNA sequence also does not create cDNAs that contain both heavy
`'
`and light chain sequences. See, e.g., ’4l4 application at pages 9-I4, 28-39. The first step in that process involves
`sequencing the cDNA clones encoding the light and heavy chain polypeptides. Once the sequence information is in
`hand, an oligomer (a short DNA sequence) is synthesized that will hybridize to a portion of variable region sequence
`in either the heavy or the light chain. The oligomer also has a stop codon at the end of its variable region sequence.
`The oligomer is then incubated with a single strand of amplified cDNA from the cDNA library, and treated with
`enzymes to prepare a double stranded DNA. The '4 14 application refers to this double-stranded DNA as a
`“heteroduplexcd" ds cDNA because it contains two strands of DNA that are not 100% complementary (i. e., the one
`grown from the oligomer contains a stop codon). When this heteroduplexed ds DNA is amplified, it produces
`“homoduplexed” ds DNA (i.e., where the two DNA strands are 100% complementary). This homoduplexed ds
`DNA containing the introduced stop codon is then hybridized with a second synthesized oligomer that contains a
`start codon at the beginning of the variable region sequence. The oligomer is then incubated with a single strand of
`amplified cDNA from the cDNA library and treated with enzymes to prepare a double stranded DNA. This double
`stranded DNA is also a “heteroduplexed” ds cDNA, since it contains two strands of DNA that are not 100%
`complementary (the first strand contains a stop codon, but no start codon). When this heteroduplexed ds DNA is
`amplified, it produces the tailored homoduplexed ds DNA. This homoduplexed ds DNA is then prepared for
`insertion into a plasmid for expression to produce the L-rFv or H-rFv polypeptide.
`
`EVIDENCE APPENDIX
`
`_
`
`PAGE B202
`
`

`
`CONTROL NOS. 90/007,542 AND 90/007,859
`
`DOCKET NOS. 22338-10230 AND -10231
`
`Control Nos. 90/007,542; 90/007,859
`
`Attorney Docket No. 22338-10230
`
`Patent
`
`amplification. As such, none of the steps outlined in the ’4l4 application for preparing
`“tailored” cDNA sequences produce a starting cDNA that contains both heavy and light
`chain sequences.
`
`24.
`
`25.
`
`None of the procedures described in the ’4l4 application insert two different cDNA
`inserts into one genetic construct, either for amplification or expression purposes.
`Instead, all of the procedures and techniques described in the application insert one
`cDNA into each plasmid.
`
`I could find no description in the ’4l4 application of a procedure which produces a
`genetic construct containing either a single cDNA insert encoding two different
`sequences (e.g., encoding both L-rFv and the H-rFv polypeptides), or two different
`cDNA inserts each encoding one of the two polypeptides.
`
`The Techniques Described In The '41 4 Application Clearly Call For Production Oflndividual
`Polypeptides In Separate Bacterial Host Cells
`
`26.
`
`27.
`
`28.
`
`29.
`
`30.
`
`If the transformation procedures described in the application are followed as they are
`written, they will not produce host cells that contain two different plasmids. As such, the
`’4l4 application, in my opinion, cannot be read as describing host cells that produce two
`different polypeptides.
`
`The procedures and techniques described in the ’4l4 application for preparing genetic
`constructs and host cells are all consistent with this one-c1one/one-plasmid/one-
`polypeptide approach.
`
`First, the procedures use hybridization techniques to select transformed bacterial colonies
`and confirm that these colonies contain the cDNA of interest. See, e.g., p. 9, lines 1-9; p.
`12, li.nes 19-25. See also Grunstein & Hogness, PROC. NATL. ACAD. SCI. 72(10): 3961-
`3965 (l975). For example, on page 9, lines 10-13, the ’4 14 application states that “me
`
`host colonies usually bacterial, which have DNA which hflridizes to either the light or
`heafl chain probes are picked and then grown in culture under selective pressure.”
`(emphasis added). This indicates to me that the cells being probed contain only one
`cDNA insert corresponding to either the heavy or the light immunoglobulin chain.
`
`Second, only simple expression vectors and procedures are described in the ’4l4
`application andin its example. See, e.g., p. 14, line 22 to p. 15, line 6; p. 40, lines 3-9.
`These expression vectors contain a single transcription/translation cassette. A plasmid
`with only one such cassette will direct a transformed cell to transcribe and translate only
`one inserted cDNA sequence. This clearly shows that only a single cDNA insert will be
`expressed by each transformed cell.
`
`Third, the example in the ’4l4 application closely tracks this one-plasmid/one-cell
`approach. On page 41, lines 6-15, the ’4l4 application specifies that individual cDNA
`inserts encoding the light or heavy chain variable polypeptides are incorporated into
`separate plasmids, and these separate plasmids are used to transform host cells:
`
`EVIDENCE APPENDIX
`
`PAGE B203
`
`

`
`CONTROL NOS. 90/007,542 AND 90/007,859
`
`DOCKET NOS. 22338-10230 AND -“I 0231
`
`Control Nos. 90/007,542; 90/007,859
`
`Attorney Docket No. 22338-10230
`
`Patent
`
`The “tailored” pGMl is isolated, partially restricted with fist_l and the DNA
`sequences coding for the light and heavy chain variable regions prepared
`above inserted individually into the tailored site to provide two plasmids
`havin DNA se uences codin for the li ht GMIL and heav
`GMI H
`
`chains, in accordance with the procedure described previously for insertion.
`(emphasis added)
`
`31.
`
`This passage clearly indicates that only one cDNA insert, encoding either the L-rFv or
`the I-I-rFv polypeptide, will be inserted into each plasmid. Inmy opinion, this description
`cannot be read as suggesting that a single cDNA sequence encoding both of the
`polypeptides will be inserted into a single plasmid, or that multiple cDNA inserts
`encoding different polypeptides will be inserted into one pGM 1 plasmid.
`
`The Techniques Described In The '414 Application Are Consistent With Only Individual
`Polypeptide Expression
`
`32.
`
`33.
`
`34.
`
`35.
`
`l do not believe someone familiar with basic molecular biology techniques could read the
`‘4 I4 application as describing procedures where a single bacterial host cell is transformed
`with two different plasmids, or where a single host cell is being engineered to produce
`two different polypeptide sequences.
`
`For example, the use ofthe same starting plasmid to produce the pGMlL and the
`pGMlH plasmids suggests that different cell cultures are being produced — one
`transformed with the pGMlL plasmid, and the other with the pGM1H plasmid. This is
`also consistent with the indication in this section of the application that clones containing
`either the pGMl L plasmid or the pGMlH plasmid are to be identified using restriction
`mapping techniques. See, p. 41, lines 13-15.
`
`Restriction mapping techniques compare the enzyme digest of a plasmid extracted from a
`clone to a reference map of the digest of the plasmid produced before the transformation
`step or expected from the genetic engineering process leading to the construction of the
`plasmid. Restriction mapping was, and remains, a standard procedure, and is
`straightforward when applied to comparisons involving a single plasmid. It becomes far
`more complicated when two plasmids are involved, because there will be a mixture of the
`two enzyme digests. This complication makes it inordinately difficult to use restriction
`mapping to confirm the outcome of gene cloning experiments.
`In my opinion, if the
`bacterial colonies under study each contained two different plasmids, the application
`would have mentioned something about how one should perform the restriction mapping
`procedures.
`
`I also do not believe a scientist familiar with molecular biology would read the
`description as indicating that individual bacterial clones are being transformed with two
`plasmids having the same selectable marker. Even assuming one cell incorporated both
`plasmids during the transformation process, the resulting cell colony and culture would
`not retain any of these doubly-transformed cells within a matter of hours. This is because
`the cells in the culture would need only one plasmid to exhibit antibiotic resistance, and
`there is strong evolutionary pressure in the cell culture against maintenance of clones
`
`EVIDENCE APPENDIX
`
`PAGE B204
`
`

`
`CONTROL NOS. 90/007,542 AND 90/007,859
`
`DOCKET NOS. 22338-10230 AND -10231
`
`Control Nos. 90/007,542; 90/007,859
`
`Patent
`,
`Attorney Docket No. 223 3 8- 10230
`
`with multiple plasmids having the same drug resistance marker. These facts would have
`been very familiar to a scientist working in this field in early April of 1983, as well as
`when the ’414 application was written in 1982.
`I
`
`In my opinion, ifthe authors ofthe ’4l4 application were intending to produce bacterial
`clones containing two different plasmids, they would have included a very clear
`description of a procedure that could be used to produce and to maintain a stable culture
`of these bacterial double transformants. The absence of that description indicates to me
`that they did not intend to make double transformants.
`
`The ’4l4 application also describes procedures which use a single antibiotic resistance
`gene.
`In particular, Example 1 uses the same pGMl plasmid to produce both expression
`plasmids. The pGMl plasmid contains a single antibiotic resistance gene. See, Miozzari
`& Yanofsky, J. BACTERIOL. 133(3): 1457-1466 (1978). Antibiotic resistance genes, or
`marker genes, allow scientists to differentiate bacterial cells that have successfully
`incorporated a plasmid, from cells that have not, by growing the culture in the presence
`of the antibiotic. Those cells that do not contain and express the antibiotic resistance
`gene in the plasmid die, thereby producing a culture which only contains bacterial clones
`that have successfully incorporated the plasmid.
`
`The use ofthe single-marker pGMl-based plasmid also would not make sense ifthe ’4l4
`application was describing procedures for transforming bacterial clones with two
`different plasmids. For example, culturing the transformed cells in the presence of the
`marker antibiotic would not differentiate clones that successfully incorporated only the
`pGM1L plasmid from those that successfully incorporated only the pGM1l-1 plasmid, or
`from clones that incorporated both plasmids. This would also make it impossible to use
`the antibiotic as a source of selective pressure to prepare and maintain a homogenous
`culture of clones that maintained both plasmids.
`
`Considering all of these points, I believe the ’414 application can only be read as
`describing procedures that produce only one polypeptide in one host cell (i.e., either the
`L-rFv polypeptide or the H-rFv polypeptide).
`
`36.
`
`37.
`
`38.
`
`39.
`
`There Is No Description OfHost Cells Meeting The Requirements OfClaim 1 Of The ’545
`Patent In The '4 14 Application
`
`40.
`
`As written, claim 1 seems to cover a scenario where the two polypeptides of the rFv have
`an identical amino acid sequence. Specifically, the claim states “[a] host cell which
`expresses a recombinant double-chain antibody fragment (rFv) comprising two
`polypeptide chains having substantially the same amino acid seguence of at least a
`portion ofgig variable region .
`. .ofa mammalian immunoglobulin .
`.
`. .” (emphasis
`added). This reading seems to be the scenario described on pages 2-3 of ’4l4
`application, which indicates that the L-rFv and H-rFv polypeptides can actually have the
`same sequence. See, p. 3, line 32 to p. 4, line 2. A host cell meeting the requirements of
`claim 1 read in this way would only have to be transformed with one plasmid containing
`one cDNA insert encoding either the L-rFv or H-rFv polypeptide.
`
`EVIDENCE APPENDIX
`
`PAGE B205
`
`

`
`_ CONTROL NOS. 90/007,542 AND 90/007,859
`
`DOCKET NOS. 22338-10230 AND -10231
`
`Control Nos. 90/007,542; 90/007,859
`
`Attorney Docket No. 22338-10230
`
`Patent
`
`41.
`
`42.
`
`Despite this, the PTO states that claim 1 of the ’545 patent defines a host cell that
`produces two different polypeptides. This means that the PTO is reading this claim to
`require the host cell to be transformed with cDNA sequences encoding two different
`polypeptides, and that the host cell express those two sequences.
`
`In my opinion, as I explained above, a scientifically correct reading of the description in
`the ’4l4 application makes it clear that it is not describing at any point a host cell that
`will produce both heavy and light immunoglobulin chain variable region polypeptides.
`Instead, the ’4l4 disclosure very clearly describes procedures in which each variable
`region polypeptide is produced in a separate host cell culture and is then isolated,
`renatured, and combined in a test tube to form an rFv.
`
`43.
`
`-H
`
`In the February 16, 2007 Office Action, the PTO identified a number of sections of the
`’545 patent that it believes are describing a host cell that produces both heavy and light
`chain variable region polypeptides.
`
`44.
`
`First, on page 20 of the Office Action, the PTO states that the ’545 patent describes:
`
`a “host cell” transformed with a single genetic construct (e.g. including
`pBR322; see e.g. Moore at col. 5, lines 32-35 and col. 7, lines 39-50) .
`.
`encoding variable light and heavy chains .
`. .. (emphasis in original).
`
`.
`
`45.
`
`I disagree. There is no description in these sections of the patent of a host cell has been
`transformed with a single genetic construct that encodes two different polypeptide
`sequences.
`
`-
`
`-
`
`Col. 5, lines 32-34 states that “a wide variety of vectors may be employed for
`amplification or expression of the ds cDNA to produce the light and heavy chains
`of the immunoglobulin.” This simply indicates that many vectors were available
`in 1982 for amplification and expression of cDNA sequences.
`It does not suggest
`that a single cDNA encoding two immunoglobulin chains should be expressed in
`a single host cell.
`
`In addition, this section of the patent (i. e., col. 5, lines 35-47) describes
`amplification of cDNA in the cDNA library, not expression of cDNA. This is
`done by incorporating the ds cDNA from the cDNA library into plasmids and
`transforming a host to incorporate each plasmid. The statements in this section
`make it clear to me that individual cDNAs are being inserted into individual
`plasmids for amplification. For example, col. 5, lines 33-36, makes references to
`vectors that contain “Q appropriate restriction site” (emphasis added), which
`indicates that each plasmid will incorporate only one cDNA insert from the cDNA
`library. And, col. 5, lines 36-37 states that “[t]he ds cDNA obtained from the
`reverse transcription of the mRNA” is being used. This indicates that a single ds
`‘cDNA is inserted into each vector, not multiple distinct ds cDNAs, and each
`cDNA being amplified will encode only one of the two immunoglobulin chains.
`As I explained earlier, there will be no cDNA molecules in the cDNA library
`
`EVIDENCE APPENDIX
`
`10
`
`w
`
`,
`
`PAGE B206
`
`

`
`CONTROL NOS. 90/007,542 AND 90/007,859
`
`DOCKET NOS. 22338-10230 AND -10231
`
`Control Nos. 90/007,542; 90/007,859
`
`Attorney Docket No. 22338-10230
`
`Patent
`
`produced by reverse transcription of the mRNA extract that contain sequences
`from both the heavy and light immunoglobulin chains.
`
`Col. 7, lines 39-50 is describing amplification procedures. The first sentence of
`this section (lines 39-4|) indicates that the vectors will have a single ds cDNA
`insert encoding only one polypeptide (i.e., “the vector which is employed
`provides for amplification and convenient isolation of transformants having Q
`variable region coding sequence insert.”). The references to “hybrid plasmid” and
`mismatched sequences refers to double-stranded DNA, where one strand contains
`the “native” cDNA produced from the cDNA library (and, ultimately, from the
`mRNA extract) and the other strand contains a cDNA grown from the synthetic
`strand that incorporates a stop codon at the end of the variable region coding
`sequence. When transformants containing a plasmid with this mismatched ds
`cDNA divides, one cell will contain a copy of the native sequence and the other
`cell will contain a copy of the “site mutated” sequence. Both of these sequences
`still encode only a heavy chain sequence or a light chain sequence, not both.
`Also, these transformed cells will not produce the polypeptides because the
`plasmid containing the cDNA insert does not contain regulatory elements that will
`direct the host cell to transcribe and translate (i.e., express) the inserted cDNA.
`
`46.
`
`Next, the PTO, at pages 20-21 , states that the ’545 patent describes “a ‘host cell’
`transformed with .
`.
`. two separate constructs comprising DNA (e.g. ds cDNA derived
`from a hybridoma as in instant claim I4: see Moore patent claim 2) encoding variable
`light and heavy chains [E.g. see Moore patent claim 1; col. 10, lines l-5; col. 23, lines 35-
`45 (pBR322); and col. 24, lines 50-60 (pGMlL and pGM1H); col. ll, lines 5-12] .
`.
`. .”
`(emphasis in original). Again, the PTO is mistaken in its interpretation of these sections
`of the patent.
`
`As I explained above, there is no description in the original ’4l4 application of a
`process having the ste