REEXAMINATION CONTROL NOS. 90/007,542 AND 90/007,859
`
`IN THE UNITED STATES PATENT AND TRADEMARK OFFICE
`
`Attorney's Docket No. 2233 8- 1 0230
`
`Patent
`
`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 Art Unit:
`
`3991
`
`Examiner:
`
`B.M. Celsa
`
`For:
`
`Merged Reexaminations of U.S. Patent No. 6,331,415 (Cabilly et al.)
`
`DECLARATION OF DR. TIMOTHY JOHN ROY HARRIS UNDER 37 C.F.R. § 1.132
`
`I, Timothy Harris, do hereby declare and state
`
`1.
`
`2.
`
`3.
`
`4.
`
`5.
`
`I am a citizen of the United Kingdom, and reside in San Diego, California.
`
`I am the same Timothy John Roy Harris who provided a Declaration in Reexamination
`No. 90/007,542 on November 25, 2005 (“First Declaration”).
`
`As I indicated in my First Declaration, I have been retained by Genentech and City of
`Hope to provide my views on certain issues that have been raised in the reexamination of
`U.S. Patent No. 6,331,415 (“the ’415 patent”). I also note thatl have been, and am being,
`compensated for my time at a rate of $500 per hour.
`
`My credentials and experience are essentially as I indicated in paragraphs 1 to 3 of my
`First Declaration.
`I also note that the company of which I was Chief Executive Officer,
`Novasite Pharmaceuticals, recently ceased operations.
`
`For the reasons set forth in my First Declaration, I believe, based on my educational
`training and work experience, I am able to report views that would be representative of a
`person of ordinary skill in the art in early April of 1983 03.9., just prior to April 8, 1983).
`I believe a person of ordinary skill in the field of the ’415 patent claims would have a
`doctorate in molecular biology or a similar scientific discipline, along with about two
`years of post-doctoral experience.
`
`6.
`
`In addition to all of the patents and printed publications I previously reviewed in
`preparing my First Declaration, I reviewed the following publications:
`
`-
`
`Deacon et al., Biochemical Society Transactions, 4:818-20 (1976)
`(“M”);
`SANOFI v. GENENTECH
`|PR2015-01624
`
`
`
`_ 1 _
`
`EXHIBIT 2005
`Mylan v. Genentech
`Mylan V. Genentech
`IPR2016-00710
`Genentech Exhibit 2007
`Genentech Exhibit 2007
`
`IPR2016-00710
`
`

`
`REEXAMINATION CONTROL NOS. 90/007,542 AND 90/007,859
`
`-
`
`-
`
`-
`
`-
`
`«
`
`Valle et al., N_atu1'§, 29l:338-340 (1981) (“Valle 1981”);
`
`Valle et al., %, 30:71-74 (1982) (“ ”);
`
`Dallas, WO 82/03088 (“J”);
`
`Ochi et al., E, 302:340-342 (1983) (“Eli”); and
`
`Oi et al., Proc. Nat’l. Acad. Sci., 80:825—829 (1983) (“Q”).
`
`7.
`
`I also reviewed the following documents (in addition to the materials I identified in
`paragraph 6 of my First Declaration):
`
`-
`
`-
`
`-
`
`A PTO Office Action in Reexamination Nos. 90/007,542 and 90/007,859,
`dated August 16, 2006 (“Second Office Action”);
`
`A PTO Order Granting ex parte reexamination of the ’415 patent, dated
`January 23, 2006 (“Second Reexamination Order”); and
`
`A Request for Ex Parte Reexamination, dated December 23, 2005
`(“Second Request for Reexamination”), including attachments to that
`Request.
`
`In addition, I reviewed relevant literature from that general time period (z'.e., before April
`8, 1983), as I had indicated in paragraph 7 of my First Declaration.
`
`In my First Declaration, I explained why certain scientific findings or observations of the
`Office were inaccurate.
`I also explained why certain comments in the Office Action were
`inconsistent with how a person of ordinary skill in the art would have read certain
`references. In forming these opinions, I evaluated not only what each reference
`individually taught, but whether and how that reference would be considered in
`combination with U.S. Patent No. 4,816,567 (“the ’567 patent”) claims and the other
`references cited by the Office.
`
`Observations on the New Rejections
`
`10.
`
`Beginning on page 22 of the Second Office Action, there is a discussion of the ’567
`patent claims and various references. In this section, the Office identifies two specific
`reasons why claims of the ’41 5 patent are believed to be obvious in view of prior art.
`
`-
`
`-
`
`At page 22, the Office states “(i) One of ordinary skill in the art would have been
`motivated to express, in a single host, light and heavy immunoglobulin chains
`(using one or two vectors) when viewing the reference Cabilly 1 patented
`invention in light of the prior art.” The Office cites the L, 1, Kaplan and
`Dallas references to support this point.
`
`At page 25, the Office states “(ii) The prior art provides further motivation to
`make active antibody with a reasonable expectation of success.” The Office cites
`
`

`
`REEXAMINATION CONTROL NOS. 90/007,542 AND 90/007,859
`
`the Deacon, Valle 1981 and Ochi papers to support this point. I note that the
`Office has also found the information in another paper by Valle (\_lalle 1982) to be
`cumulative to what is taught by the Deacon paper, and information in the Qi paper
`to be cumulative to what is taught by the _OLhi paper.
`
`ll.
`
`12.
`
`I do not believe the Second Office Action accurately portrays what these references
`actually teach. I also do not believe these references would have been considered,
`individually or collectively, in the way the Second Office Action suggests they would
`have by a person of ordinary skill in the art in early April of 1983.
`
`Certainly, by early April of 1983, there was interest within the industry of using
`recombinant DNA technology to produce proteins with known commercial value,
`including fimctional immunoglobulin molecules. However, the state of the art at that
`time and the experiences of those working in the recombinant DNA field, coupled with
`the information in the references cited by the Office, would not have led people to be
`particularly optimistic about achieving this goal, and did not provide any clear direction
`as to how to do so.
`
`Overview ofthe Relevant Technological Field in April of1983
`
`13.
`
`14.
`
`15.
`
`16.
`
`17.
`
`In early April of 1983, the field of genetic engineering was still developing. It was
`nothing like the mature field it is today, over two decades later. A relatively small
`number of proteins had been made by recombinant DNA technology. Almost all of those
`were relatively simple monomeric (i.e., one polypeptide chain) proteins.
`
`In a review article I wrote that was published in April of 1983, I provided a list of
`eukaryotic proteins that had been produced in E. coli using recombinant methods. See
`Harris, Genetic Engineering, 4: 127-85 (1983), attached as Exhibit B to my previous
`declaration, at pages 164 to 169. All but one of these examples concerned production of
`relatively simple monomeric proteins. The exception was insulin, which I reported had
`been produced by individually expressing each of the two chains of the insulin protein in
`different E. coli cell lines, or by expressing “preproinsulin” (a single polypeptide) which
`was enzymatically processed in vitro to form mature insulin. See, pages 137 to 138.
`
`My 1983 review article provides a perspective on the types of recombinant DNA projects
`that had been published by early April of 1983 concerning the expression of recombinant
`proteins in E. coli. I note that all of the examples described in the review article involved
`production of one polypeptide in one transformed host cell.
`
`I was not aware of any published reports as of early April of 1983 documenting
`production of a multimeric protein by independently expressing in a single cell
`recombinant DNA sequences corresponding to the constituent polypeptides of the
`multimeric protein. 1 also was not aware of any published reports at that time of
`production of a multimeric protein of the size (~150 kD) or structural complexity of an
`immunoglobulin tetramer.
`
`As the ’4l5 patent explains, the immunoglobulin tetramer is a large, complex multimeric
`protein made up of four polypeptides: two light chains and two heavy chains. The
`
`-3-
`
`

`
`REEXAMINATION CONTROL NOS. 90/007,542 AND 90/007,859
`
`structure of the tetrameric immunoglobulin molecule is generally maintained by a series
`of disulfide bonds between pairs of cysteine residues and non-covalent interactions
`between the four polypeptides. For example, in an IgG (depicted in Figure 1 of the ’415
`patent, reproduced below), pairs of heavy and light chains are linked through inter-chain
`disulfide bonds, and each pair is linked to the other pair through three disulfide bonds
`formed between cysteine residues within the heavy chains. In addition to these inter-
`chain disulfide bonds, each polypeptide subunit (i. e. , each heavy and each light chain) is
`stabilized by two or four intra-chain disulfide bonds. See, e.g., ’4l5 patent at col. 3, lines
`1 9-3 8.
`
` Fm), fragment
`c fragment
`
`I F
`
`Fun‘ fragment
`
`Fig. 1.
`
`18.
`
`Based on these known structural characteristics of the tetrameric immunoglobulin
`molecule, I believe a person of ordinary skill in the art, in early April of 1983, would
`have expected that the production of an immunoglobulin tetramer using recombinant
`DNA techniques would have been a significantly more challenging undertaking than the
`types of projects described in my review article or the molecules described in Axel et al.,
`U.S. Patent No. 4,399,216 (“Axel”) (i. e., [3—globin) and Rice & Baltimore, Proc. Nat’l.
`
`

`
`REEXAMINATION CONTROL NOS. 90/007,542 AND 90/007,859
`
`Acad. Sci., 79:7862-7865 (1982) (“Rice”) (i.e., a recombinant immunoglobulin light
`chain gene).
`
`The Reasonable Expectations ofa Person of Ordinary Skill in the Art in Early April of] 983
`
`19.
`
`20.
`
`21.
`
`22.
`
`23.
`
`24.
`
`I believe many of the scientific observations in the Second Office Action reflect an
`inaccurate description of the expectations of a person of ordinary skill in the art in early
`April of 1983.
`
`As I previously indicated, in early April of 1983, I was not aware of any literature
`reporting the successful production of a multimeric protein by independently expressing
`in a single host cell recombinant DNA sequences encoding the constituent polypeptides
`of the multimeric protein. I also do not believe a person having ordinary skill at that time
`would have many expectations regarding a project of the scale of the ’4l5 patent process
`based solely on their knowledge of general techniques for producing polypeptides in host
`cells transformed with recombinant DNA sequences.
`
`The Office refers to the transfection experiments conducted in B-lymphoid cell lines in
`the Second Office Action. The Office apparently considers these types of experiments to
`be relevant to the ’4l5 patent claims. In my view, these experiments provide little insight
`into the questions that would have influenced the expectations of a person of ordinary
`skill in the art contemplating production of an immunoglobulin tetramer or a fragment
`derived from it through expression of recombinant DNA sequences encoding the heavy
`and light chains in a single transformed host cell. However, to the extent that the Office
`does consider these experiments, they should also consider how a person of ordinary skill
`would have evaluated them in the context of what else was known about B-cells.
`
`By early April of 1983, there was an extensive amount of literature documenting research
`on how B-lymphocytes produce immunoglobulins. That literature had shown that the
`native processes that govern immunoglobulin production in cells of the B-lymphocyte
`lineage were complicated and involved many variables.
`
`For example, the literature had shown that the processes that govern the assembly and
`expression of immunoglobulin genes were unique compared to other types of genes.
`Immunoglobulin genes are assembled by rearrangement of gene fragments in the B-cell
`incidental to the cell’s development into mature, immunoglobulin secreting plasma B-
`cells. The factors that controlled or influenced the processes of B-cell development as
`well as the assembly and expression of immunoglobulin genes, however, were not
`understood by early April of 1983.
`
`For example, as Drs. Rice and Baltimore explained in the introduction of their 1982
`PNAS paper:
`
`B-cell differentiation proceeds from the “pre-B” lymphocyte, which
`synthesizes pl immunoglobulin (Ig) heavy chains but no light chains, to the
`mature B lymphocyte, which synthesizes both heavy and light chains and
`expresses surface lg, and finally to the lg-secreting plasma cell (1-5). The
`availability of transformed cell analogs has allowed biochemical
`
`-5-
`
`

`
`REEXAMINATION CONTROL NOS. 90/007,542 AND 90/007,859
`
`25.
`
`26.
`
`27.
`
`characterization of these stages of cellular differentiation (6-11). Recently
`such studies have contributed greatly to our understanding of the structure
`of Ig gene segments and the joining of these segments to produce a
`functionally rearranged Ig gene (12-17).
`
`Although much is now known about Ig gene structure, relatively little is
`known about the molecular mechanisms that control Ig gene expression.
`
`It was also known then that a variety of factors affect the ability of B-lymphocytes to
`produce and secrete immunoglobulin tetramers. Some of these insights came from the
`study of lymphocyte cell lines, such as hybridomas and myeloma cells lines. For
`example, several groups had reported that mutant hybridoma cell lines that produce
`excess heavy chain oflen would die, leading to the view that the presence of excess free
`heavy chain polypeptides in these cells was toxic to the cells. See, Wilde & Milstein,
`Eur. J. Irnmunol., l0:462-467 (1980); Kohler, Proc. Nat’l. Acad. Sci., 772197-2199
`(1980) (attached as Exhibits A and B, respectively). This would have caused a person of
`ordinary skill to question whether unbalanced or uncontrolled production of heavy chain
`and light chain polypeptides in a transformed host cell would be toxic to the cell.
`
`Similarly, it was assumed that mature B-lymphocytes had unique features or attributes
`that gave these cells the specialization required to produce, properly fold, assemble and
`secrete immunoglobulins. This was supported by findings in the literature. For example,
`Wabl and Steinberg reported on the existence of a protein, called “BiP,” which they
`reported bound to free heavy chain in pre-B cells. See Wabl & Steinberg, Proc. Nat’l.
`Acad. Sci., 7916976-6978 (Nov. 1982) (attached as Exhibit C). They suggested this
`“helper” protein might be involved in the assembly and expression of immunoglobulin
`genes or the production of the immunoglobulin molecule.
`
`Thus, research that had been done by early April of 1983 had shown that a number of
`independent but interrelated factors could affect the successfiil production of the
`immunoglobulin by the B-lymphocyte including:
`
`(i)
`
`(ii)
`
`(iii)
`
`(iv)
`
`The timing and levels of expression of messenger RNA from the native
`immunoglobulin genes in the B-lymphocytes,
`
`The amount of heavy and light chain polypeptides present in the cell at various
`times and locations (i.e., the “stoichiometry” of polypeptides in the cellular
`environment where the immunoglobulin tetramer might be formed),
`
`The developmental state of the B-lymphocyte (e.g., whether it had gained the
`capacity through its development to express the immunoglobulin genes at
`appropriate levels, or could process the gene expression products to form the
`immunoglobulin tetramer), and
`
`The presence of agents in B-lymphocytes that facilitated proper assembly and
`secretion of the tetrameric immunoglobulin molecule (so-called “helper”
`proteins).
`
`

`
`REEXAMINATION CONTROL NOS. 90/007,542 AND 90/007,859
`
`28.
`
`The picture all this work painted was that the expression of immunoglobulin genes in B-
`lymphocytes, and the natural processes governing production and secretion of tetrameric
`immunoglobulin molecules by these cells, was complicated and influenced by many
`interrelated variables. To the extent that a person of ordinary skill in the art turned to the
`work involving B-lymphocytes for guidance, that person would have reached the
`conclusion that successful co-expression of recombinant heavy and recombinant light
`chain DNA sequences in a single host cell, and assembly of immunoglobulin tetramers
`following co-expression of the introduced sequences, would have been dependent on
`many interrelated factors. Such a person would not have expected the task to have been
`as straightforward or predictable as the Office suggests it was in early April of 1983.
`
`The ’5 6 7 Patent Claims
`
`29.
`
`30.
`
`31.
`
`32.
`
`33.
`
`34.
`
`There are three significant differences between the claims of the ’41 5 patent and the
`claims of the ’567 patent.
`
`First, the ’567 patent claims require that a host cell be transformed with a DNA sequence
`encoding only one immunoglobulin polypeptide, either a light or a heavy chain. The
`’4l5 patent claims, however, require that a single host be transformed with two DNA
`sequences, one encoding at least the variable domain of a heavy chain polypeptide, and
`one encoding at least the variable domain of a light chain polypeptide.
`
`Second, the ’567 patent claims do not require the production of heavy gd light
`immunoglobulin chains as separate molecules in a single transformed host cell, which is
`required by the ’41 5 patent claims. The ’567 claims only refer to the production of either
`a heavy g a light chimeric immunoglobulin chain polypeptide.
`
`Third, unlike the ’41 5 patent claims, the ’567 patent claims do not refer to, or require, the
`assembly of multiple immunoglobulin heavy and light chain polypeptides into an
`immunoglobulin molecule or immunologically active fragment. Instead, the claims only
`refer to the isolation of an individual heavy or light chimeric immunoglobulin chain
`polypeptide after it has been produced by the cell.
`
`These are significant distinctions. As I explained above, the challenges of achieving co-
`expression of two eukaryotic genes in the same cell were different in character than the
`challenge of expressing only one eukaryotic gene. A person of skill in the art would have
`expected the approach of expressing recombinant heavy and light chain DNA sequences
`in the same cell to present a distinct challenge.
`
`The Second Office Action also addresses some of the language used in the ’567 patent
`claims and what that language would have meant to a person of ordinary skill.
`Specifically, at page 33, the Office appears to suggest that a person of ordinary skill in
`the art would have read the phrase “specificity for a particular known antigen” in the ’567
`claims as indicating that one should perform the ’567 process only to produce a fully
`assembled and functional immunoglobulin molecule. This is not how a person of
`ordinary skill in the art would have read that phrase in early April of 1983. As I
`explained in paragraphs 11 to 14 of my First Declaration, a person of ordinary skill in the
`
`

`
`REEXAMINATION CONTROL NOS. 90/007,542 AND 90/007,859
`
`art in early April of 1983 would have understood this phrase in the ’567 patent claims as
`referring to the amino acid sequences found within the variable domain of the individual
`light or heavy chain being produced as it is these sequences that determine the specificity
`of antibody. Such a person would not have read these passages in the claims as being a
`requirement that a fimctional, antigen-binding immunoglobulin molecule containing
`heavy and light chains be produced.
`
`The Axel Patent
`
`35.
`
`At page 23, the Second Office Action states:
`
`Axel describes the process as particularly suited for transformation of DNA
`into eukaryotic cells for making antibodies (see col. 3, lines 31-36). Axel
`discloses and claims the expression of antibodies in mammalian host cells as
`intact (assembled) proteins. See Axel: abstract; col. 5, lines 3-7 and 24-28;
`patent claims 1, 7, 22-24, 28 and 29.
`
`36.
`
`The Second Office Action (page 34) also states:
`
`Additionally, the Axel reference suggests expressing two immunoglobulin
`chains in a single cell, since Axel discloses and claims (e.g. claim 7) DNA
`(i.e. DNAI) encoding an antibody that necessarily possesses both light and
`heavy immunoglobulin chains. In this respect, the Axel reference clearly
`encompasses one or more genes which encode one or more proteins: e.g.
`x “. .. DNA which includes a gene or genes coding for desired proteinaceous
`materials...” (Abstract lines 1-4, with emphasis). Accordingly, although
`Axel lacks an antibody example, Axel nonetheless suggests recombinant
`antibody production in a suitable host (e.g. eukaryote).
`
`37.
`
`38.
`
`39.
`
`These observations are factually incorrect, and inconsistent with how a person of
`ordinary skill would have read the Axel patent in early April of 1983.
`
`I addressed the Axil patent in my First Declaration at paragraphs 20 to 30. I explained
`that the ALI patent literally describes a process that uses only tlo recombinant DNA
`sequences. “DNA I” encodes a desired “proteinaceous material not associated with a
`selectable phenotype” that is to be isolated from the transfonned host cell. “DNA II”
`encodes a selectable marker which, when expressed by the transformed cell line, allows a
`scientist to add an agent into the cell culture where the transformed cells are growing to
`select out those that have been successfiilly transformed with DNA 11. The polypeptide
`encoded by DNA II is not recovered from the host cell. I found no examples or other
`information in the Axiel patent describing a process in which a “DNA III” was included,
`or in which a DNA I was constructed encoding more than one polypeptide chain. See,
`e. g., paragraph 25 of my First Declaration.
`
`The E patent cotransformation experiments only involve one instance in which the
`DNA I is a mammalian gene -- rabbit B-globin protein. This is a relatively simple
`monomeric polypeptide with a molecular weight of about 16 kD, unlike the large (~l 50
`kD) complex immunoglobulin tetramer. I believe the examples in the Axjel patent would
`
`-3-
`
`

`
`REEXAMINATION CONTROL NOS. 90/007,542 AND 90/007,859
`
`40.
`
`41.
`
`42.
`
`43.
`
`44.
`
`have provided little or no insight into the expression of an immunoglobulin tetramer in a
`single cell, consistent with my opinion that the Axel patent is addressing production of
`one polypeptide by one transformed host cell.
`
`1 also pointed out that the Axel patent disclosure does not contain any suggestions
`regarding the desirability of expressing both heavy and light immunoglobulin chains in a
`single mammalian cell line. See, paragraphs 27 and 28.
`
`The passages of the Axel patent referred to in the Second Office Action do not change
`my opinion. For example, the Office points to column 3, lines 31 to 36 of the Axel
`patent, which reads:
`
`The process of this invention is particularly suited for the insertion into
`eukaryotic cells of DNA which codes for proteinaceous materials which are
`not associated with a selectable phenotype such as interferon protein,
`insulin, growth hormones clotting factors, viral antigens, antibodies and
`certain enzymes.
`
`Contrary to what the Second Office Action states, this passage does not suggest that the
`Axjel patent “DNA I + DNA II” method is particularly suited for making assembled
`antibodies any more than it suggests that the method is “particularly suited” for making
`any other type of polypeptide listed along with “antibodies.” Instead, in my opinion, the
`reference to “interferon protein, insulin, growth hormones clotting factors, viral antigens,
`antibodies and certain enzymes” would have been viewed by a person of ordinary skill in
`the art as simply being a laundry list of types of proteins having economic value at the
`time the Axel patent was filed. I believe a person of ordinary skill would view these and
`other references throughout the patent to “antibodies” as simply identifying an antibody
`polypeptide (i. e., a heavy or a light chain polypeptide) as a type of polypeptide that can
`be produced by the Axel process.
`
`The Second Office Action also refers to claim 7 of the Ail patent. Claim 7 is consistent
`with my opinion that the Axjel patent is not making any suggestion particularly relevant to
`production of tetrameric antibodies. This is because the claim is referring in the
`alternative to “a viral antigen” and to “an antibody.” Viral antigens do not have a
`characteristic or uniform physical structure like a tetrameric immunoglobulin. I do not
`believe a person of ordinary skill in the art would have referred to these classes of
`polypeptides in the alternative if they were intending to convey a particular observation
`about the multimeric nature of the antibody molecule.
`
`In my opinion, the Second Office Action is incorrect when it suggests that these various
`references to the word “antibody” or “antibodies” mean that the gel patent is
`specifically describing methods for transforming a single mammalian cell line with
`recombinant DNA sequences encoding both the heavy and the light immunoglobulin
`chains. The Axel patent certainly does not provide any specific guidance or suggestion
`how to do so. See, paragraphs 27 and 28 of my First Declaration. It also does not
`provide an indirect suggestion to do so because the i DNA I plus DNA II method
`contemplates a one protein-one transformed host cell process.
`
`

`
`REEXAMINATION CONTROL NOS. 90/007,542 AND 90/007,859
`
`45.
`
`46.
`
`47.
`
`48.
`
`The passage in the ;Ax_<:l patent Abstract cited in the Second Office Action as referring to
`“gene or genes” does not change my opinion. This passage is referring to the concept of
`inserting multiple copies of the same gene coding for a single polypeptide of interest in
`order to enhance the yields of the desired polypeptide. It is not referring to insertion of
`multiple distinct genes coding for different proteins, such as distinct genes coding for the
`heavy and light chains of an immunoglobulin molecule.
`
`The Q patent explains that the point of introducing multiple copies of DNA I is to
`increase the number of copies of DNA I that can be expressed by the cell, by which
`process “multiple copies of proteinaceous or other desired molecules can be produced
`within eukaryotic cells.” Axel at colurrm 7, lines 32-34; see also colunm 14, lines 16
`through 18 (“[t]he number of rabbit globin genes integrated into these transformants is
`variable” — note that there is only one rabbit B-globin gene being referred to in this
`passage). The body of the gel patent explains two approaches for obtaining eukaryotic
`host cells with multiple copies of a gene encoding a desired recombinant protein; namely:
`(i) inserting multiple copies of the same gene coding for a single polypeptide of interest,
`or (ii) inserting a construct that will amplify with increasing selection agent (e.g.,
`methotrexate) concentrations. See, column 6, line 44 to column 7, line 45 .
`
`I also disagree with the statement in the Second Office Action that the Q patent
`“discloses and claims the expression of antibodies in mammalian host cells as intact
`(assembled) proteins.” (Emphasis added.) Certainly, the mere use of the word “antibody”
`does not convey this suggestion. In addition, after a thorough review of the Axtel patent, I
`have been unable to locate any description concerning “intact” or “assembled”
`antibodies. In my opinion, if the Axjel patent were describing techniques particularly
`suited for expressing recombinant DNA sequences encoding light and heavy chain
`polypeptides in a single host cell, and their subsequent assembly into an immunoglobulin
`tetramer, there would have been some discussion in the patent about how assembled
`antibodies consist of multiple discrete polypeptides. There is none.
`
`Finally, I do not believe the Q patent would have suggested to a person of ordinary
`skill in the art in early April of 1983 that they modify a process of producing one
`immunoglobulin chain in one cell (i. e., the ’567 patent process) by producing both heavy
`and light immunoglobulin chains in one host cell (i.e., the ’4l5 patent process). As I
`explained above, the Ax_el patent is concerned with transformation of a eukaryotic cell
`with a DNA encoding a single recombinant protein (or multiple copies of that single
`DNA) in one transformed host cell.
`
`The Rice Paper
`
`49.
`
`The Second Office Action also provides an inaccurate description of what the Rice paper
`would have taught to a person of ordinary skill in the art in early April of 1983. At
`page 23, the Second Office Action states:
`
`Rice inserted the light chain gene into a plasmid, transfected the cells, and
`then examined the polypeptides as well as the RNA produced by the cells
`(see pages 7863-7864 and Figures 2 and 3). Lastly, since the cells were
`
`-10-
`
`

`
`REEXAMINATION CONTROL NOS. 90/007,542 AND 90/007,859
`
`producing both immunoglobulin light and heavy chain, the cells were
`examined for the ability to assemble the chains into IgG molecules, leading
`to the observation that “[e]ssentia1ly all of the K chain produced in the K-2
`cells appear to be assembled into IgG2b” (see page 7864 and Abstract
`penultimate sentence). Thus, Rice demonstrates the successful expression
`of both heavy and light chains in a host with snbseg Ltent assembly into
`immunoglobulins. (Emphasis added.)
`
`50.
`
`It then states, at page 35, that:
`
`Further, the Rice reference clearly teaches to one of ordinary skill in the art
`that an exogenous immunoglobulin light chain assembles with a heavy chain
`endogenously produced by the cell even though both chains possess
`di/ferent antigen specificity. Thus, in light of this teaching it would be
`reasonable for one of ordinm skill in the art to expect that expressing a
`light and heayy chain of the same antigen specificity 1e.g. derived from a
`known antibody) in a competent host would result in assembly of a
`functional antibody. (Underlined emphasis added.)
`
`51.
`
`52.
`
`53.
`
`54.
`
`This description of the R paper and what it would have suggested to a person of
`ordinary skill in the art in early April of 1983 is factually inaccurate and inconsistent with
`how I believe a person of ordinary skill in the art would have actually interpreted the
`paper at that time.
`
`As I explained in my First Declaration (paragraphs 33 to 39), the _Rii paper
`acknowledges that the factors responsible for control of immunoglobulin gene expression
`in lymphoid cells were not known at the time of the publication. See, page 7862
`(“relatively little is known about the molecular mechanisms that control Ig gene
`expression”) The R paper (at pages 7864-5) also states that:
`
`The major result of these studies is the demonstration that a fimctional K
`gene can be introduced into a lymphoid cell line in which it will be
`continuously expressed. This opens the possibility of examining control and
`rearrangement mechanisms in lymphoid cells by using inserted genetic
`elements.
`
`Based on this, a person of ordinary skill in the art would view the Rice paper as teaching
`a technique that could be useful in designing experiments to deduce the mechanisms that
`control expression of endogenous immunoglobulin genes.
`
`Even with respect to the work they actually performed, the authors of the Rice paper were
`careful to point out what they did not know about their experimental system. In Rice, the
`authors did not insert a well-defined DNA sequence encoding only the immunoglobulin
`light chain into the cell. Instead, they inserted a piece of genomic DNA that contained
`several uncharacterized sequences beyond the sequence of the light chain. These
`included the intervening sequence between the sequences coding for the variable (Vk),
`joining (Jk) and constant (Ck) regions, and about 1-1.5 kb of DNA on either side of the
`
`-11-
`
`

`
`REEXAMINATION CONTROL NOS. 90/007,542 AND 90/007,859
`
`light chain sequence. The authors noted that “any of this extra DNA could be involved in
`promoter and control fimctions,” suggesting that more than just the DNA sequence of the
`light chain is required for expression. (See, page 7865.) They also noted that the
`rearranged light chain gene “apparently” used its own promoter to control expression,
`rather than the standard promoter selected and inserted by the authors of the @ paper.
`The promoter that caused expression was not identified or characterized. (See, page
`7865)
`
`The Rice authors also indicated that it was “an open question” as to whether the
`endogenous heavy chain expression controlled expression of the recombinant light chain,
`and suggest further investigation. (See, page 7865.)
`
`These quotes are consistent with my opinion that a person of ordinary skill in the art in
`early April of 1983 would not have characterized the @ paper as the Office does;
`namely, that it generally established that one could express recombinant DNA sequences
`encoding heavy £1 light immunoglobulin chains in a single host cell, and that these
`chains would be expected to form immunoglobulin tetramers.
`
`I do not believe a person of ordinary skill in the art would have extended the results
`observed in Q to expression of DNA sequences encoding both immunoglobulin chains
`in any type