UNITED STATES PATENT AND TRADEMARK OFFICE
`____________
`
`BEFORE THE PATENT TRIAL AND APPEAL BOARD
`____________
`
`MYLAN PHARMACEUTICALS, INC., and
`MERCK SHARP & DOHME CORP.,
`Petitioners
`
`v.
`
`GENENTECH, INC. AND CITY OF HOPE
`Patent Owners
`____________
`
`IPR No. 2016-07101
`
`U.S. PATENT NO. 6,331,415
`____________
`
`DECLARATION OF MARC J. SHULMAN IN SUPPORT OF
`PETITIONER’S REPLY
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`1 Case IPR2017-00047 has been joined with this proceeding.
`
`
`
`Mylan v. Genentech
`IPR2016-00710
`Merck Ex. 1091, Pg. 1
`
`

`

`
`
`DECLARATION OF MARC J. SHULMAN
`
`I, Marc J. Shulman, hereby declare and state as follows:
`
`I.
`
`INTRODUCTION
`
`1.
`
`This declaration is submitted on behalf of Merck Sharp & Dohme
`
`Corp. (“Merck”) in IPR No. 2016-00170, regarding U.S. Patent No. 6,331,415,
`
`“Methods of Producing Immunoglobulins, Vectors and Transformed Host Cells for
`
`Use Therein,” owned by Genentech, Inc. and City of Hope (collectively, “Patent
`
`Owners”).
`
`2.
`
`In response to statements made in Patent Owners’ Response (Paper
`
`31) and the opinions expressed by Patent Owners’ expert Dr. John Fiddes (Ex.
`
`2019), I have been asked to provide information on my scientific work related to
`
`recombinant expression of immunoglobulin heavy and light chains in mammalian
`
`cells, which took place in the early 1980s. Specifically, I have been asked to
`
`explain how the work that my colleagues and I performed prior to April 1983
`
`refutes Patent Owners’ and Dr. Fiddes’ arguments regarding the alleged
`
`uncertainties surrounding recombinant expression of antibodies in April 1983 and
`
`their arguments that the “prevailing mindset” in April 1983 was to express only
`
`one exogenous protein per host cell.
`
`II.
`
`BACKGROUND AND EXPERIENCE
`
`3.
`
`I am currently Professor Emeritus in the Department of Immunology
`
`and the Department of Molecular and Medical Genetics, in the Faculty of Medicine
`
`2
`
`Merck Ex. 1091, Pg. 2
`
`

`

`
`
`of the University of Toronto in Toronto, Canada. My curriculum vitae is attached
`
`hereto as Appendix A.
`
`4.
`
`I received a B.A. in Physics from Harvard University in 1962, and a
`
`Ph.D. in Biology from Massachusetts Institute of Technology in 1969.
`
`5.
`
`In 1969-1970, after receiving my Ph.D., I was a postdoctoral fellow at
`
`Massachusetts Institute of Technology and the National Institutes of Health.
`
`During the summers of 1971-1972, I was an investigator at the Marine Biological
`
`Laboratory, Woods Hole, Massachusetts.
`
`6.
`
`In 1972 until 1973, I was a National Science Foundation Fellow at the
`
`Institut de Biologie Moleculaire, L’Universite de Paris, in Paris, France. From
`
`1970-1976, I was also a Staff Fellow, Lab of Molecular Biology at the National
`
`Cancer Institute, National Institutes of Health, in Bethesda, Maryland.
`
`7.
`
`In 1976, I became a Member of the Basel Institute for Immunology, in
`
`Basel, Switzerland, where I remained until 1979.
`
`8.
`
`In 1979, I became Assistant Professor in the Department of Medical
`
`Biophysics at the University of Toronto, in Toronto, Canada. In 1985, I became a
`
`faculty member in the Immunology Department and Medical Genetics (now
`
`Molecular and Medical Genetics) Department and subsequently promoted to
`
`Associate Professor. In 1990, I was promoted to Professor, and in 2007, I retired
`
`as Professor Emeritus.
`
`3
`
`Merck Ex. 1091, Pg. 3
`
`

`

`
`
`9.
`
`During my scientific career, I have been awarded over 35 research
`
`grants, and I have published nearly 100 peer-reviewed articles, most of which
`
`pertain to immunoglobulin production and structure/function. (c.f., Appendix A.)
`
`III.
`
`COMPENSATION
`
`10.
`
`I am being compensated for my work on this case at my standard
`
`consulting rate of $600/hour. My compensation is not contingent upon the results
`
`of my analysis or the substance of my testimony. I have no stake in the outcome of
`
`this proceeding or any related litigation or administrative proceedings. I have no
`
`financial interest in Merck, and similarly have no financial interest in the patent at
`
`issue, U.S. Patent 6,331,415 (“the ’415 patent”) or its owners.
`
`IV.
`
`MY INITIAL WORK ON IMMUNOGLOBULINS
`
`11.
`
`I began conducting research in the immunoglobulin (Ig) field in 1976
`
`at the Basel Institute for Immunology, in Basel, Switzerland, where I worked
`
`closely with future Nobel Laureate Dr. Georges Köhler. We published several
`
`papers together including a paper describing our development of a cell line called
`
`Sp2/0. Ex. 1117 (Shulman, et al., 1978).
`
`12. The Sp2/0 cell line makes no immunoglobulin chains, neither the
`
`heavy chain, e.g., µ, nor light chain, e.g., . However, this cell line is capable of
`
`serving as a fusion partner for making hybridomas. Id. Importantly, the Sp2/0 cell
`
`line is also an advantageous recipient cell for expressing vectors encoding
`
`4
`
`Merck Ex. 1091, Pg. 4
`
`

`

`
`
`recombinant immunoglobulin genes, as it provides all the specialized cellular
`
`machinery for transcribing immunoglobulin genes, and for assembling and
`
`processing the immunoglobulin heavy and light chains into functional
`
`immunoglobulin.
`
`13. To elaborate briefly on the preceding point – before the advent of
`
`recombinant DNA and gene transfer technology, cell hybrids were sometimes used
`
`to gain insight into the cellular requirement for expressing tissue-specific genes. In
`
`particular, it had been shown that a hybrid cell could be formed by fusing an Ig-
`
`expressing myeloma cell line with a fibroblast cell line making no Ig. The
`
`interesting result was that this hybrid cell did not produce immunoglobulin. Ex.
`
`1139 (Coffino et al., 1971). This result was a strong indication that
`
`immunoglobulin gene expression required cellular factors that were present in
`
`myeloma cell lines and absent (and presumably suppressed by factors) in
`
`fibroblasts. Accordingly, this result indicated that vectors bearing cloned versions
`
`of the natural genomic immunoglobulin DNA segments would not be expressed in
`
`fibroblast cell lines but would likely be expressed in lymphoid, e.g., hybridoma,
`
`cell lines.
`
`14. Dr. Köhler and I also developed a method of isolating mutant
`
`hybridoma cell lines that were defective in the normal production of either their
`
`immunoglobulin light or heavy chain. Ex. 1118 (Köhler & Shulman 1980). Our
`
`5
`
`Merck Ex. 1091, Pg. 5
`
`

`

`
`
`first application of this technique used the Sp6 cell line that produces IgM()
`
`specific for the hapten, trinitrophenyl (TNP), and yielded regulatory mutants
`
`making either less µ or less  chain and structural mutants that produced an altered
`
`µ chain.
`
`V.
`
`SUCCESSFUL RECOMBINANT EXPRESSION OF
`IMMUNOGLOBULIN LIGHT CHAIN IN A MAMMALIAN CELL
`(“OCHI I”)
`
`15. As noted above, in 1979, I took a position in the Wellesley Hospital in
`
`Toronto in association with the Department of Medical Biophysics of the
`
`University of Toronto. Shortly after arriving in Toronto, I began an extensive
`
`collaboration with Dr. Nobumichi Hozumi, a former colleague at the Basel
`
`Institute who had taken a position in Toronto a year earlier, and Robert Hawley
`
`and Atsuo Ochi, members of Dr. Hozumi’s laboratory. Mr. Hawley undertook to
`
`analyze the  regulatory mutants generated by my work with Dr. Köhler. We
`
`sought to understand the reason that the mutants produced little or no  chain. Mr.
`
`Hawley isolated the functional  gene as a BamHI fragment, and analyzed the 
`
`genes of the mutants. He showed that one mutant (denoted igk14) had lost the
`
`functional  gene, while in two other mutants the  gene had been inactivated by
`
`the insertion of a transposable element. Mr. Hawley also characterized the general
`
`structure of the cloned  gene. In particular, he showed that the  coding regions
`
`on the BamHI segment were flanked at each end by a long stretch of additional
`
`6
`
`Merck Ex. 1091, Pg. 6
`
`

`

`
`
`genomic DNA, indicating that the cloned DNA segment was very likely to include
`
`all the regulatory elements, such as were known for other genes. Ex. 1119
`
`(Hawley, et al., 1982).
`
`16. Having isolated the functional  gene, we sought to develop a system
`
`for transferring and expressing the  gene in host cells, and ultimately to make
`
`defined changes in the recombinant  gene that we could test for their effects on 
`
`gene expression. The mutant hybridoma cell line that lacked the functional  gene
`
`was a very advantageous recipient for testing the functionality of the recombinant
`
` gene. That is, if the cloned  gene was functional, transferring the cloned  gene
`
`to the igk14 mutant cell line, would restore  gene expression, and the cells would
`
`then produce and secrete IgM() specific for the original hapten, TNP, which could
`
`then be detected with simple and sensitive assays. In the event, this approach
`
`worked as expected. The results, which are summarized below, are described in a
`
`paper that I co-authored -- “Transfer of a cloned immunoglobulin light-chain gene
`
`to mutant hybridoma cells restores specific antibody production.” Ex. 1021 (Ochi,
`
`et al., 1983 “Ochi I”)).
`
`17. Our plan for this work was to make “stable” transformants, so as to
`
`have a homogeneous population of cells all expressing the transferred gene in the
`
`same manner. Making stable transformants involved inserting the  gene into a
`
`vector with a selectable marker (a gene that conferred drug resistance), selecting
`
`7
`
`Merck Ex. 1091, Pg. 7
`
`

`

`
`
`independently transformed cells that expressed the marker and then screening
`
`those transformants for expression of the  gene. For this purpose, Dr. Hozumi’s
`
`laboratory inserted the  gene into two transfer vectors, pSV2gpt and pSV2neo,
`
`that carried selectable markers that could be used in mammalian cell lines of all
`
`types, thus, creating pSV2gpt- and pSV2neo-, respectively. Ex. 1021 (Ochi I) at
`
`340. The pSV2gpt and pSV2neo vectors were developed in the laboratory of Dr.
`
`Paul Berg. Ex. 1120 (Mulligan & Berg, 1980); Ex. 1004 (Southern & Berg, 1982).
`
`18. As we approached this project we were faced with one significant
`
`uncertainty, namely how to introduce the vectors into the hybridoma cells.
`
`Previous work on DNA transformation had used fibroblast cells as the recipients,
`
`and the conventional wisdom at the time of our work was that the usual
`
`transformation techniques would not work when applied to lymphoid cells, such as
`
`our hybridoma cell lines. Consequently, we adapted an alternative and recently
`
`described method based on protoplast fusion to transfer the  gene to the recipient
`
`hybridoma cells. In this method, which is similar to the process used to produce
`
`hybridomas, the vector is grown in bacterial cells, which are then fused to the
`
`mammalian host cells, thus transferring the expression vector into the recipient
`
`mammalian cells.
`
`19. Dr. Ochi, a post-doctoral fellow in the Hozumi laboratory, then used
`
`protoplast fusion to transfer the pSV2neo- vector to the mutant cell line igk14 that
`
`8
`
`Merck Ex. 1091, Pg. 8
`
`

`

`
`
`lacked the  gene. Ex. 1021 (Ochi I) at 340. By measuring TNP-specific
`
`hemagglutination and TNP plaque formation, Dr. Ochi confirmed that transferring
`
`the recombinant  gene resulted in  chain expression and production of functional
`
`IgM antibodies specific for the antigen TNP. Id. at 340-41.
`
`20. As in the case of the other methods that were used to make stably
`
`transformed mammalian cells, the frequency of drug-resistant cells was low
`
`(~0.0001% of the cells) and the expression of the unselected gene, in our case the 
`
`gene, was variable. This type of variability was the usual outcome for making
`
`stable transformants, and it was attributed in a general way to the fact that
`
`independent transformants each had a different number of vector copies inserted in
`
`both orientations at different chromosomal sites. While this variability was
`
`annoying, we dealt with it by dividing the transformation process into two steps –
`
`first to select the drug-resistant transformants and second to screen those
`
`transformants for their production of the  chain. As reported in Ochi I,
`
`approximately 40% (6/14) of the drug-resistant transformants produced detectable
`
` chain, as measured by TNP-specific hemagglutination, and a smaller but
`
`acceptable fraction produced  chain at near normal levels. In general terms –
`
`frequency of drug-resistant transformants and expression of the unselected  gene
`
`– our system of using protoplast fusion to transfer and express genes of interest in
`
`9
`
`Merck Ex. 1091, Pg. 9
`
`

`

`
`
`lymphoid cells produced results that were similar to those obtained with the more
`
`commonly used cell lines and vector expression systems.
`
`21. The variability in gene expression when using the protoplast fusion
`
`technique led us (my colleagues and me) to make two conclusions. On the one
`
`hand, the variability meant that our system was not well suited for analyzing the
`
`features of the  gene that were required for its expression. On the other hand, it
`
`was clear that our system was a reliable method for obtaining stable transformants
`
`that expressed functional  chain in its native form and at near normal levels. In
`
`summary, we concluded that the protoplast fusion technique combined with the
`
`pSV2-neo- vector constituted a very good method for producing the
`
`immunoglobulin  chain and reconstituting functional antigen-specific IgM().
`
`22.
`
`In his criticism of the Ochi I publication, Dr. Fiddes wrote “[d]espite
`
`the commonalities between the host cell and the exogenous light chain gene, Ochi
`
`reported that 10 [sic] of the 14 successfully transformed cell lines failed to regain
`
`any detectable antibody production. …Thus, Ochi reported substantial variability
`
`in the ability of the reintroduced light chain to reconstitute an antibody with the
`
`endogenous heavy chain of the same specificity. Ochi offered no explanation for
`
`these results, and acknowledged that the ‘mechanisms responsible for the
`
`regulation of the expression of rearranged immunoglobulin genes are poorly
`
`understood.’” Ex. 2019 (Fiddes Decl.) at ¶119.
`
`10
`
`Merck Ex. 1091, Pg. 10
`
`

`

`
`
`23. There are two problems with this criticism. First, the quote from the
`
`abstract of Ochi I about poorly understood regulation has been taken out of context
`
`and did not in fact refer to the variability among transformants. Rather this
`
`statement referred to the limited understanding among contemporary molecular
`
`biologists about the cellular factors and regulatory elements that operate in
`
`mammalian cells – what exactly constituted a promoter, which cellular factors
`
`were necessary, etc. Second, as noted above the variability in expression among
`
`transformants was a minor problem where the goal was production of functional
`
`immunoglobulin rather than analysis of gene regulation.
`
`24. Patent Owners have asserted that Ochi I is “evidence that Southern’s
`
`pSV2gpt and pSV2neo vectors were adopted by independent research groups for
`
`single chain immunoglobulin expression prior to the filing date of the ’415 patent.”
`
`Patent Owners then extrapolated from this assertion to argue that because “those
`
`researchers of exceptional skill did not apply Southern [to express heavy and light
`
`chains in the same host cell] there is no reason to believe that a person of ordinary
`
`skill would have.” (Paper 31 Patent Owners’ Response at 60). I do not agree with
`
`this logic. While it is literally true that the cited publications did not explicitly
`
`disclose using the Berg vectors to express more than one gene, that does not mean
`
`that we or other researchers would not have not have thought to do so. Quite the
`
`contrary. My colleagues and I had been working with these very vectors for more
`
`11
`
`Merck Ex. 1091, Pg. 11
`
`

`

`
`
`than a year, and we were, of course, aware that the vectors had multiple cloning
`
`sites. Moreover, as described below, we had already in 1982 made use of the
`
`EcoRI and BamHI sites to insert the µ and  genes into the pSV2neo vector, thus
`
`generating the pSV2neo-µ- expression vector. Furthermore, in his Nobel lecture,
`
`Dr. Berg explained that “[a]dditional DNA segments can also be inserted into the
`
`vector DNA’s at any of several unique restriction sites; consequently, a single
`
`DNA molecule can transduce several genes of interest simultaneously.” Ex. 1069
`
`(Berg Nobel lecture).
`
`25. Dr. Fiddes has opined that if one wanted to make a recombinant
`
`antibody in April 1983, they would have followed a “one-polypeptide-per-host-cell
`
`approach.” Ex. 2019 (Fiddes Decl.) at ¶ 230. This opinion is based on a
`
`misunderstanding of the state of knowledge among workers in the field. First, Dr.
`
`Fiddes’ opinion is contrary to the teachings of Prof. Berg cited above. Moreover, it
`
`is an over-interpretation on Dr. Fiddes’ part to conclude that, because our initial
`
`work involved insertion of only the  gene, we did not think in terms of inserting a
`
`second. Again, and as described below, before Ochi I was published, my
`
`colleagues and I had already expressed both the  gene and the µ gene in a single
`
`cell. More generally, my opinion is that any knowledgeable scientist working in
`
`this area of research would perceive our results as leading directly to the next step
`
`of expressing both heavy and light chains in the same cell.
`
`12
`
`Merck Ex. 1091, Pg. 12
`
`

`

`
`
`VI.
`
`WORK IN OCHI I LED TO THE SUCCESSFUL RECOMBINANT
`PRODUCTION OF FUNCTIONAL IMMUNOGLOBULIN VIA
`HEAVY & LIGHT CHAINS IN A SINGLE CELL (“OCHI II”)
`
`26. Around the same time that Dr. Ochi, Dr. Hozumi and I had begun our
`
`work on expression of the recombinant  gene, Dr. Köhler’s laboratory cloned the
`
`functional µ gene and identified a cell line, denoted igm10, that lacked that gene.
`
`In early 1982, Dr. Köhler provided us with the recombinant µ gene, which had
`
`been cloned as an EcoRI segment. The cloned µ gene was also flanked by
`
`extended non-coding DNA segments, indicating that, as in the case of the cloned 
`
`gene, the cloned µ gene was very likely to include the ordinary regulatory elements
`
`needed for its expression in lymphoid cells.
`
`27. Our easy success in developing a system for expressing the  gene led
`
`us immediately to plan how we could express the µ heavy chain gene and then both
`
`genes together with the goal of making functional immunoglobulins of different
`
`types. We of course recognized that the pSV2 vector system provided two unique
`
`restriction sites – BamHI and EcoRI – that were suitable for inserting the light and
`
`heavy chain genes, respectively. We anticipated using two methods. One method
`
`would take advantage of the fact that the two vectors used different and
`
`independently selectable markers, which would allow us first to transfer and select
`
`cells expressing the  gene and then in a second step transfer the other vector
`
`bearing the µ heavy chain into the -expressing cells. This was in fact the method
`
`13
`
`Merck Ex. 1091, Pg. 13
`
`

`

`
`
`that we designed in late 1982 for the production of chimeric mouse/human IgM.
`
`Ex. 1148 (Boulianne, et al., 1984). The other method would take advantage of the
`
`fact that each vector provided two convenient cloning sites which would allow us
`
`to insert both the heavy and light chain genes into the same vector. This vector
`
`bearing both Ig genes could then be transferred into a cell line like Sp2/0, which
`
`would then be expected to produce and combine both chains into functional
`
`immunoglobulin. We used this second method of inserting both genes in the same
`
`vector in our work on the expression of the µ heavy chain gene. This work, which
`
`also proceeded very smoothly was reported in a second paper that I co-authored
`
`along with Drs. Hozumi and Ochi. Ex. 1040 (Ochi et al., 1983 Functional
`
`immunoglobulin M production after transfection of cloned immunoglobulin heavy
`
`and light chain genes into lymphoid cells (“Ochi II”)).
`
`28. To summarize the work reported in Ochi II, on receiving the µ gene
`
`from Dr. Köhler, researchers in Dr. Hozumi’s lab then inserted it into two vectors:
`
`pSV2neo and pSV2neo-, to create vectors that are denoted as pSV2neo-µ and
`
`pSV2neo-µ-, respectively. Ex. 1040 (Ochi II) at 6351-52. Following the same
`
`approach that we used in the case of the  gene, we used the pSV2neo-µ vector to
`
`confirm that the recombinant µ gene was functional by introducing it into the
`
`igm10 cell line lacking the µ heavy chain gene identified by Dr. Köhler. Id.
`
`14
`
`Merck Ex. 1091, Pg. 14
`
`

`

`
`
`29. We then tested the expression vector pSV2neo-µ- encoding both µ
`
`and . In the first transfections of the vector pSV2neo-µ-, we used the myeloma
`
`cell line X63, which itself produces a  heavy chain and a  light chain, with the
`
`result that the recipient cells secreted a mixture of IgM and IgG. Id. at 6352. We
`
`subsequently transfected SV2neo-µ- into the hybridoma Sp2/0, as this cell line
`
`produces no Ig chains itself, but retains the machinery for producing Ig chains
`
`from genomic Ig vectors. Id. When the pSV2neo-µ- vector was transferred into
`
`Sp2/0, the recipient cells produced IgM that behaved like the TNP-specific IgM of
`
`the original Sp6 hybridoma.
`
`30.
`
`I have described the basis on which my colleagues and I had
`
`reasonable expectations that transferring the recombinant µ and  genes into a
`
`single host cell would yield native light and heavy chains and functional IgM. And
`
`indeed, the above-described results thoroughly confirmed these expectations.
`
`These results also directly refute Patent Owners’ and Dr. Fiddes’ argument that
`
`there was a prevailing mindset in April 1983 that researchers would have
`
`expressed only one exogenous polypeptide per host cell. See Paper 31 Patent
`
`Owners’ Response at 37-38, 11; Ex. 2019 (Fiddes Decl.) at ¶ 87. As evidenced by
`
`our contemporaneous research and demonstrated results, any notion that the
`
`“prevailing mindset,” in 1983 would have been to express the heavy and light
`
`chains is simply wrong. Nature had already provided the perfect example of how
`
`15
`
`Merck Ex. 1091, Pg. 15
`
`

`

`
`
`to make functional immunoglobulin – namely to express the heavy chain and light
`
`chain genes in the same cell. Moreover, as detailed above, the technology existed
`
`for transferring two genes into the same cell. Under these circumstances, no one
`
`needed further advice or instruction on the most efficacious route to
`
`immunoglobulin production.
`
`31.
`
`In addition, there was also specific biochemical and physiological
`
`evidence that militated against expressing the heavy and light chain genes in
`
`separate cells. First, mammalian cells do not secrete a normal heavy chain except
`
`in the presence of a light chain that can combine with that heavy chain. Moreover,
`
`unpaired heavy chains are often insoluble. It would be a daunting task to purify
`
`heavy chain away from the other contents of mammalian cells.
`
`32. Second, solubilizing, renaturing, and combining the isolated chains
`
`are poorly defined techniques at best, with no assurance of actually yielding a
`
`product in which the heavy and light chains have assembled normally and created
`
`the true binding site. In contrast, producing Ig by expressing both heavy and light
`
`chains in the same host cell, e.g., Sp2/0, ensured that the normal protein would be
`
`secreted as soluble, active material that could be purified by well-established
`
`methods. Because of this evidence, my colleagues and I never even considered
`
`producing the Ig light and heavy chain genes in separate host cells and then
`
`combining them in vitro to make functional Ig.
`
`16
`
`Merck Ex. 1091, Pg. 16
`
`

`

`
`
`33.
`
` As I explained earlier, nature itself provided an excellent example of
`
`how to produce immunoglobulin from recombinant DNA. Presented with this
`
`example, any person with an appropriate scientific background, i.e., a so-called
`
`person of ordinary skill in the art or “POSA” as described by Dr. Fiddes (Ex. 2019
`
`(Fiddes Decl.) at ¶¶ 147-149), would know that transferring the heavy and light
`
`chain genes into the same cell was a very likely route to success.. In other words, a
`
`person of ordinary skill in the art would have had a reasonable expectation of
`
`success in seeking to produce an immunoglobulin by using recombinant DNA
`
`techniques to express both the heavy and light chain genes in a single host cell.
`
`34. Regarding the skill level of such POSA, I agree with Dr. Fiddes’
`
`definition regarding the knowledge he attributes to that hypothetical person for the
`
`purposes of this declaration. Indeed, my colleagues and I were relative beginners
`
`by most measures and certainly would not be considered as having exceptional
`
`skill beyond the level described for a POSA at the time. That is, Dr. Hozumi and I
`
`had only recently begun our careers as independent investigators and were working
`
`under relatively austere conditions supported by the first renewals of our
`
`government grants. Mr. Hawley was a beginning graduate student. Dr. Ochi had
`
`been trained as a dentist, and this work on immunoglobulin gene expression was
`
`his first serious experimental undertaking.
`
`17
`
`Merck Ex. 1091, Pg. 17
`
`

`

`
`
`VII. CONCLUSION
`
`35.
`
`In summary, the work that my colleagues and I performed in 1982 on
`
`expression of the  gene, and published in Ochi I (March, 1983) demonstrated that
`
`all the techniques necessary for expressing immunoglobulin genes for producing
`
`functional immunoglobulin were at hand. This state of knowledge allowed us to
`
`foresee with confidence that we could apply these same techniques first to test and
`
`express the µ heavy chain gene and then to express both heavy and light chains
`
`together. That is, our expectation from our work on  gene expression was that it
`
`would be straightforward to express both the heavy and light chains in a single cell,
`
`using either a single vector or two vectors, and obtain functional immunoglobulin.
`
`Our subsequent work over the course of 1982 and early 1983 in which we did in
`
`fact express both genes in the same cell confirmed these expectations and was
`
`published in Ochi II.
`
`
`
`
`
`18
`
`Merck Ex. 1091, Pg. 18
`
`

`

`Executed this 7th day oprril 20l 7.
`
`I declare under penalty of perjury that
`
`the foregoing is true and correct.
`
`.
`
`36.
`
`I declare that all statements made herein of my own knowledge are
`
`true and that all statements made on inl'omiation and belief are believed to be true,
`
`statements and the like so made are punishable by fine or imprisomnent, or both,
`
`under Section 1001 of Title 18 ofthe United States Code.
`
`and further that these statements were made with the knowledge that willful false
`
`- Marc J. Shulmam
`
`Merck Ex. 1091, Pg. 19
`
`

`

`
`
`
`
`
`
`
`
`
`
`EXHIBIT A
`
`EXHIBIT A
`
`Merck Ex. 1091, Pg. 20
`
`Merck Ex. 1091, Pg. 20
`
`

`

`
`
`
`
`Marc J. Shulman Jan 2017
`
`
`
`Curricula Vitae
`
`MARC J. SHULMAN, Ph.D.
`
`11 Hyatt Road
`Woods Hole MA 02543 USA
`
`Tel: 508 444 8892
`
`marc.shulman@gmail.com
`
`
`
`U.S., Canada
`
`
`Mailing Address:
`
`
`Telephone
`
`E-mail
`
`Citizenship:
`
`Current Positions:
`1990 -present
`
`
`
`
`Associate, Marine Biological Laboratory
`2007-present
`Woods Hole, MA
`
`
`
`Previous Professional Experience
`1985-1990
`Assistant/Associate Professor
`
`Department of Immunology and
`
`Department of Molecular and Medical Genetics
`
`University of Toronto
`
`l979 - 1985
`
`
`
`l976 - l979
`
`
`l970 - l976
`
`
`l972 - l973
`
`
`l97l - l972
`(summers)
`
`l969 - l970
`
`Education:
`
`Professor (Emeritus, 2007)
`Department of Immunology and
`Department of Molecular and Medical Genetics
`University of Toronto
`
`Associate of the Arthritis Society of Canada
`Assistant Professor, Dept. of
`Medical Biophysics, University of Toronto
`
`Member, Basel Institute for Immunology
`Basel, Switzerland
`
`Staff Fellow, Lab of Molecular Biology
`National Cancer Institute, NIH, Bethesda, MD
`
`NSF Fellow, Institut de Biologie Moleculaire
`L'Universite de Paris
`
`Investigator, Marine Biological Lab
`Woods Hole, Mass. USA
`
`Postdoctoral Fellow at M.I.T. and NIH
`
`Ph.D. Biology; Massachusetts Institute of Technology, 1969
`
`
`
`
`
`Merck Ex. 1091, Pg. 21
`
`

`

`
`
`
`
`Languages:
`
`
`
`Marc J. Shulman Jan 2017
`
`
`
`B.A. Physics; Harvard, 1962
`
`English; moderate facility in Spanish, French and German
`
`
`
`
`
`Merck Ex. 1091, Pg. 22
`
`

`

`
`
`
`
`Marc J. Shulman Jan 2017
`
`
`
`TEACHING EXPERIENCE
`
`A. IMMUNOLOGY
`University of Toronto
`
`l984 - 1985 "Cellular and Molecular Aspects of B-cell Differentiation" MBP l00lL
`
`l985 - l987 "Introductory Immunology". MPL 334Y (12 lectures).
`
`l987 - l988 Immunology for medical students.
`
`l987 - 2000 "Recent Advances in Molecular/Cellular Immunology" IMM 1016H, Imm l0l7H (two
`lectures per series)
`
`1996 - 2002 “Student seminar series" Imm1019
`
`2000 -2001 “Advanced Immunobiology” IMM430 (two lectures)
`
`
`
`Other Institutions
`University of Health Sciences of Antigua:
`
`2000
`Immunology for medical students
`
`St. Matthew’s Medical University (Belize)
`
`2001
`Immunology For medical students
`
`B. MOLECULAR & CELLULAR BIOLOGY
`University of Toronto
`
`l982 - l984 "Cell Biology for Physical Scientists" MBP l022H
`
`l982 - l983 "Recent Advances in Molecular Genetics" (Ad hoc course for the clinical fellows of
`the Rheumatic Disease Unit)
`
`1989 - 1990 "Principles of Microbial and Molecular Genetics" MGB 310 (four lectures)
`
`1990 - 1991 "Eukaryotic Genetics" MGB 410 (four lectures)
`
`
`
`Other Institutions
`National Institute of Health (Bethesda):
`
`1970
`Biology of Temperate Bacteriophage
`
`1974
`DNA replication, recombination, and repair
`
`C. MATHEMATICS
`Federal City College (Washington, D.C.):
`
`1971-1972 Algebra II
`
`
`
`
`
`Merck Ex. 1091, Pg. 23
`
`

`

`
`
`
`
`Marc J. Shulman Jan 2017
`
`
`
`
`1991-1993
`
`1997-2002
`
`2000-2002
`
`
`ADMINISTRATIVE EXPERIENCE
`
`Chair, Department of Immunology
`
`Graduate Coordinator, Department of Immunology
`
`Chair, Student Awards Committee, Division of Life Sciences
`
`
`
`
`
`Merck Ex. 1091, Pg. 24
`
`

`

`
`
`
`
`Marc J. Shulman Jan 2017
`
`
`
`Major Scientific Accomplishments
`
`Biotechnology
`Hybridoma technology
`
`Shulman, M.J., Wilde, C.D. & Kohler, G. (l978). A better cell line for making hybridomas
`
`secreting specific monoclonal antibodies. Nature 276:l69.
`
`This work developed the Sp2/0 cell line, which is now widely used to make specific
`hybridomas which produce only the Ig chains donated by the spleen cell fusion partner.
`DNA transfection.
`
`
`Ochi, A., Hawley, R.G., Shulman, M.J. & Hozumi, N. (l983). Transfer of a cloned
`immunoglobulin light chain gene to mutant hybridoma cells restores specific antibody production.
`Nature 302:340.
`
`This paper reported a technique for transfecting DNA into hybridoma cells. It is notable
`because the conventional wisdom of the day insisted that only adherent cells could be transfected.
`
`Antibody engineering
`Boulianne, G.L., Hozumi, N. & Shulman, M.J. (l984). Production of functional chimeric
`
`mouse/human IgM. Nature 3l2:643.
`
`This paper described a method for preparing chimeric mouse V region/human C region
`immunoglobulin, which is superior to mouse Ig in human therapy. The concept underlies the
`methods which are generally used to produce specific Ig for human therapy.
`
`Genetic recombination/Gene targeting
`Mammalian cells
`
`Baker, M.D., Pennell, N., Bosnoyan, L. & Shulman, M.J. (l988). Homologous
`
`recombination can restore normal immunoglobulin production in a mutant hybridoma cell line.
`Proc. Nat. Acad. Sci. USA 85:6432-6436.
`
`This paper reported that, contrary to the conventional wisdom in the '80's, homologous
`recombination can occur between transfected and chromosomal genes. Second, this technology
`allowed us to test rigorously for expression activating elements in the normal chromosomal context
`.
`Microbial cells Shulman, M.J., Hallick, L., Echols, H. & Signer, E. (l970). Properties of
`
`recombination-deficient mutants of bacteriophage lambda. J. Mol. Biol. 52:50l.
`
`This work proved that the phage  exonuclease and  protein are required for homologous
`recombination.
`
`Somatic cell molecular genetics
`Kohler, G. & Shulman, M.J., (1980) Immunoglobulin M mutants. Eur. J. Immunol. 10:467.
`
`
`This paper showed for the first time that it was feasible to select mutant cell lines in which
`one could identify the corresponding mutations