Case 1:22-cv-00311-WCB Document 335 Filed 01/18/24 Page 1 of 19 PageID #: 20460
`
`IN THE UNITED STATES DISTRICT COURT
`FOR THE DISTRICT OF DELAWARE
`
`IMPOSSIBLE FOODS INC.,
`
`Plaintiff,
`
`Civil Action No. 22-311-WCB
`
`v.
`
`MOTIF FOODWORKS, INC., and
`GINKGO BIOWORKS, INC.,
`
`Defendants.
`
`JURY TRIAL DEMANDED
`
`
`
`DEFENDANT GINKGO BIOWORKS, INC.’S
`OPENING CLAIM CONSTRUCTION BRIEF
`
`REDACTED PUBLIC VERSION
`FILED JANUARY 18, 2024
`
`

`

`Case 1:22-cv-00311-WCB Document 335 Filed 01/18/24 Page 2 of 19 PageID #: 20461
`
`TABLE OF CONTENTS
`
`I.
`II.
`
`III.
`
`B.
`
`C.
`D.
`E.
`
`F.
`G.
`
`Page
`Introduction ........................................................................................................................ 1
`Terms for Construction ...................................................................................................... 1
`A.
`Term 1 – “promoter element” ................................................................................ 1
`1.
`“Promoter element” is indefinite ............................................................... 2
`2.
`Impossible’s proposed functional definition confirms the term’s
`indefiniteness ............................................................................................. 3
`Term 2 – “a Mxr1 transcriptional activator sequence” / “a methanol
`expression regulator 1 (Mxr1) transcriptional activator” / “a Mxr1
`transcriptional activator sequence” ........................................................................ 4
`1.
`The intrinsic evidence confirms that Term 2a refers to a sequence
`of an Mxr1 transcriptional activator protein .............................................. 6
`Impossible’s arguments in support of its construction are incorrect ......... 8
`2.
`Term 6 – “sequence to which [the/a] Mxr1 transcriptional activator binds” ......... 9
`Term 3 – “[x] from P. pastoris” / “[x] from Pichia pastoris” .............................. 10
`Term 4 – “wherein the recombinant nucleic acid molecule comprises [x],
`wherein the recombinant nucleic acid molecule comprises [y]” ......................... 11
`Term 5 – “a nucleic acid molecule encoding [x] and [y]” ................................... 12
`Term 7 – “wherein each nucleic acid is operably linked to a
`methanol-inducible promoter element” ............................................................... 13
`Term 8 – “wherein the [methanol-inducible] promoter element” / “wherein
`the [at least one] methanol-inducible promoter element” .................................... 15
`Conclusion ....................................................................................................................... 16
`
`H.
`
`i
`
`

`

`Case 1:22-cv-00311-WCB Document 335 Filed 01/18/24 Page 3 of 19 PageID #: 20462
`
`TABLE OF AUTHORITIES
`
`
`
`Page(s)
`
`Cases
`
`Bushnell Hawthorne, LLC v. Cisco Systems, Inc.,
`813 Fed. App’x 522 (Fed. Cir. 2020).......................................................................................15
`
`Nautilus, Inc. v. Biosig Instruments, Inc.,
`572 U.S. 898 (2014) ...................................................................................................................2
`
`New Hampshire v. Maine,
`532 U.S. 742 (2001) .................................................................................................................12
`
`In re Rambus,
`694 F.3d 42 (Fed. Cir. 2013)......................................................................................................7
`
`Transclean Corp. v. Jiffy Lube Intern., Inc.,
`474 F.3d 1298 (Fed. Cir. 2007)................................................................................................12
`
`Williamson v. Citrix Online, LLC,
`792 F.3d 1339 (Fed. Cir. 2015)............................................................................................4, 10
`
`Statutes
`
`35 U.S.C. § 112(f) ......................................................................................................................4, 10
`
`Other Authorities
`
`MPEP § 2173.05(e) ........................................................................................................................16
`
`ii
`
`

`

`Case 1:22-cv-00311-WCB Document 335 Filed 01/18/24 Page 4 of 19 PageID #: 20463
`
`I.
`
`INTRODUCTION
`
`The Court should adopt Ginkgo’s construction of the eight disputed claim terms of the
`
`Yeast Patents.1 All of the claims are directed to yeast cells that have been engineered to include
`
`DNA sequences encoding proteins of interest.
`
`Some of the disputed claim terms are indefinite because the specification and prosecution
`
`history fail to inform the skilled artisan of the scope of the invention with reasonable certainty.
`
`For example, all of the asserted claims recite a “promoter element.” But the specification
`
`nowhere defines that term, and it does not identify or provide examples of any promoter element.
`
`At most, it indicates that a “promoter element” is not a “promoter,” but, as explained in the
`
`supporting declaration of Dr. Carl Batt, it provides no affirmative guidance about what a
`
`“promoter element” actually is.
`
`As to the terms that are not indefinite, Ginkgo’s constructions are supported by the
`
`intrinsic and extrinsic evidence, including Impossible’s own prior successful assertions to the
`
`PTAB—assertions that Impossible now disavows. Ginkgo’s proposed constructions should be
`
`adopted.
`
`II.
`
`TERMS FOR CONSTRUCTION
`
`A.
`
`Term 1 – “promoter element”
`
`Ginkgo
`
`indefinite
`
`Impossible
`
`something that modulates, directs or regulates
`expression
`
`All asserted claims
`
`Every asserted claim requires a “promoter element,” a term that is indefinite because it
`
`1 U.S. Patent Nos. 10,689,656 (“’656 patent”) (Ex. 1) and 10,273,492 (“’492 patent”) (Ex. 2).
`
`1
`
`

`

`Case 1:22-cv-00311-WCB Document 335 Filed 01/18/24 Page 5 of 19 PageID #: 20464
`
`“fail[s] to inform, with reasonable certainty, those skilled in the art about the scope of the
`
`invention.” See Nautilus, Inc. v. Biosig Instruments, Inc., 572 U.S. 898, 901 (2014). While
`
`“clear and convincing evidence” is required to show that a term is indefinite, “this presumption
`
`of validity does not alter the degree of clarity that § 112, ¶ 2 demands from patent applicants[.]”
`
`Id. at 912 n.10.
`
`At most, the specification indicates that a “promoter element” is distinct from a
`
`full-length “promoter.” See Batt ¶¶ 65-68. Beyond that, it sheds no light on what makes
`
`something a “promoter element.” Indeed, the claim uses a recognized nonce word: “element,”
`
`the vagueness of which is confirmed by Impossible’s use of “something” in its proposed
`
`construction. See December 22, 2023 Status Conference Hearing Transcript (Ex. 3) at 10:13-19.
`
`In an attempt to save the claim, Impossible asserts a functional definition for that “something,”
`
`under which a “promoter element” must modulate, direct, or regulate transcription of
`
`downstream genes. See also D.I. 142 at 13. But if a promoter element is functionally defined,
`
`the term is still indefinite because there is no disclosure of a “promoter element” structure
`
`performing this function.
`
`1.
`
`“Promoter element” is indefinite
`
`The intrinsic record does not inform a person of skill in the art (“POSA”) what a
`
`“promoter element” is with reasonable certainty. While the specification discloses promoters, it
`
`does not disclose or describe any promoter elements. The specification2 is clear on one thing:
`
`“promoters” and “promoter elements” are not the same. Batt ¶¶ 65-68. For example, it states
`
`that “a single promoter, or promoter element therefrom, can be used to drive transcription[.]”
`
`’656 patent at 7:5-6. This “therefrom” language is used repeatedly in the specification to
`
`2 The two specifications are substantively identical. For simplicity, Ginkgo cites to the
`’656 patent.
`
`2
`
`

`

`Case 1:22-cv-00311-WCB Document 335 Filed 01/18/24 Page 6 of 19 PageID #: 20465
`
`distinguish between “promoter elements” and “promoters.” See, e.g., id. at 4:60-61 (“[o]ther
`
`methanol-inducible promoters, or promoter elements therefrom, however, can be used …”); see
`
`also id. at 4:52-55; 5:2-4; 6:34-35; 6:36-37; 6:41-42; 6:59-60; Batt ¶¶ 65-68.
`
`From these disclosures, a POSA may surmise that a “promoter element” is some part of a
`
`“promoter,” but the specification leaves the POSA guessing as to which parts of a promoter
`
`might constitute a “promoter element.” See Batt ¶¶ 69-74. The Yeast Patents provide no
`
`examples of promoter elements and do not define, or otherwise provide boundaries for, what
`
`constitutes a “promoter element,” in terms of the necessary sequence or structure. Batt ¶ 70.
`
`The specification refers to, for example, “the DHAS promoter from P. pastoris [] or a promoter
`
`element therefrom …” ’656 patent at 5:2-4. That DHAS promoter is 980 nucleotides long. Batt
`
`¶ 72. The specification provides a POSA with no guidance on how to identify or determine what
`
`sequences, out of the practically infinite3 number that can be extracted from those 980
`
`nucleotides, constitute a “promoter element therefrom.” Is it any subset of those nucleotides? If
`
`so, how long must it be? Is one nucleotide enough? If not, how much is enough? What if the
`
`sequence can function as a promoter in some circumstances and not others? The specification
`
`provides no answers. See Batt ¶¶ 69-74.
`
`2.
`
`Impossible’s proposed functional definition confirms the term’s
`indefiniteness
`
`Impossible asserts that “promoter element” should be given its “plain and ordinary
`
`meaning,” as “something that modulates, directs or regulates expression.” Ex. 3, 10:13-18.4
`
`Impossible’s use of “something” betrays the emptiness of the word “element.” Indeed, the
`
`Federal Circuit has explicitly recognized that the term “element” is among the class of “nonce
`
`3 There are 2980 – 1 possible subsets. Batt ¶ 72.
`4 All emphases added unless otherwise noted.
`
`3
`
`

`

`Case 1:22-cv-00311-WCB Document 335 Filed 01/18/24 Page 7 of 19 PageID #: 20466
`
`words that reflect nothing more than verbal constructs”; their use may be “tantamount to using
`
`the word ‘means’ because they ‘typically do not connote sufficiently definite structure’ and
`
`therefore may invoke § 112, para. 6 [now post-AIA § 112(f)].” Williamson v. Citrix Online,
`
`LLC, 792 F.3d 1339, 1350 (Fed. Cir. 2015).
`
`Under Impossible’s proposed construction, “promoter element” is explicitly directed to a
`
`means (“something”) for performing a function (“modulates, directs or regulates expression”).
`
`This proposed construction defines “promoter element” by what it does, rather than what it is.
`
`According to § 112(f), where a claim is defined functionally, it is governed by, and therefore
`
`limited to, the “corresponding structure” that the “intrinsic evidence clearly links or associates”
`
`with the claimed function. Williamson, 792 F.3d at 1352. But if the specification “fails to
`
`disclose adequate corresponding structure, the claim is indefinite.” Id.
`
`That is exactly the flaw here: the specification discloses no “promoter elements” having
`
`the function Impossible identifies, much less any structure for such elements. It does not identify
`
`the “something” necessary to perform the purported function, as § 112(f) requires. See Batt
`
`¶¶ 75-78.
`
`B.
`
`Term 2 – “a Mxr1 transcriptional activator sequence” / “a methanol
`expression regulator 1 (Mxr1) transcriptional activator” / “a Mxr1
`transcriptional activator sequence”
`
`As shown below, the parties dispute whether “Mxr1 transcriptional activator sequence,”
`
`as it appears in claims 1 and 14 of the ’492 patent (Term 2a), refers to the sequence of the DNA
`
`encoding the Mxr1 protein itself (Ginkgo’s position) or a sequence to which the Mxr1 protein
`
`binds (Impossible’s position).5
`
`5 While Impossible’s constructions for Terms 2a, 2b, and 2c do not specify that the protein is
`Mxr1, Ginkgo assumes that Impossible agrees that the protein is Mxr1, not any arbitrary protein.
`
`4
`
`

`

`Case 1:22-cv-00311-WCB Document 335 Filed 01/18/24 Page 8 of 19 PageID #: 20467
`
`Term
`
`Ginkgo
`
`Impossible
`
`2a. “a Mxr1
`transcriptional activator
`sequence”6
`
`a sequence of an Mxr1
`transcriptional activator protein
`
`the consensus sequence that
`actually binds the [Mxr1] protein
`
`’492 patent, all asserted claims
`
`However, as also shown below, Impossible agrees that nearly identical Term 2b (“a
`
`methanol expression regulator 1 (Mxr1) transcriptional activator”) and identical Term 2c (“a
`
`Mxr1 transcriptional activator sequence”) refer to the sequence encoding the Mxr1 protein.
`
`Term
`
`Ginkgo
`
`Impossible
`
`2b. “a methanol expression
`regulator 1 (Mxr1)
`transcriptional activator”
`
`a sequence of an Mxr1
`transcriptional activator protein
`
`the gene that makes the [Mxr1]
`protein
`
`’656 patent, cls. 1, 5, 7, 8, 9, 10, 11, 13, 15, 23, 24
`
`2c. “a Mxr1 transcriptional
`activator sequence”
`
`a sequence of an Mxr1
`transcriptional activator protein
`
`the gene that makes the [Mxrl]
`protein
`
`’656 patent, cls. 26, 28
`
`See Ex. 3 at 11:21-12:20; see also D.I. 106 at 14; D.I. 144 at ¶ 6 (stating that these nearly
`
`identical terms, what was previously identified as Terms 19 and 20, that include Terms 2b and
`
`2c, “refer to the nucleic acid sequence encoding the Mxr1 protein itself.”); Batt ¶ 87. Thus,
`
`Impossible asks to construe the same term in two different ways in different patents in the same
`
`patent family.
`
`The Court should construe Term 2a to refer to the sequence encoding the Mxr1 protein,
`
`6 Motif proposes construing a larger Term 2a, see D.I. 314-1 at 1. Ginkgo’s arguments apply
`equally to that term. The parties also dispute the meaning of the phrase “from P. pastoris,” as
`found in the larger Term 2a and Term 2c. Ginkgo addresses that aspect in connection with Term
`3, infra.
`
`5
`
`

`

`Case 1:22-cv-00311-WCB Document 335 Filed 01/18/24 Page 9 of 19 PageID #: 20468
`
`just as Impossible agrees the same or similar terms in Terms 2b and 2c should be construed.
`
`1.
`
`The intrinsic evidence confirms that Term 2a refers to a sequence of
`an Mxr1 transcriptional activator protein
`
`First, the specification repeatedly uses the term “Mxr1 sequence” to refer to the
`
`sequence encoding the Mxr1 protein. The specification explains that “[i]n some embodiments,
`
`the exogenous nucleic acid encoding a transcriptional activator comprises a Mxr1 sequence
`
`from Pichia pastoris …” ’656 patent at 1:63-65; Batt ¶ 93. It also explains that “a
`
`transcriptional activator from Pichia pastoris is the Mxr1 sequence,” followed by a reference to
`
`the sequence encoding the Mxr1 transcriptional activator protein as found in an online database.
`
`’656 patent at 4:26-43; Batt ¶¶ 94-96. Based on these disclosures, and others, a POSA would
`
`recognize that “a Mxr1 transcriptional activator sequence” means the sequence encoding the
`
`Mxr1 transcriptional activator protein—or in Impossible’s words for identical Term 2c, “the
`
`gene that makes the [Mxr1] protein”—and not to a sequence to which the Mxr1 protein binds.
`
`Batt ¶¶ 91-98. The PTAB agreed. In denying institution of Motif’s inter partes review petition
`
`relating to the ’492 patent, it stated that Motif had not “sufficiently shown why a POS[]A would
`
`have been motivated to produce a nucleic acid that encodes both (1) a heme-containing protein
`
`operably linked to a promoter element as well as (2) a Mxr1 transcriptional activator sequence
`
`as required by claim element 1[a] …” IPR2023-00322 Paper 7, Decision Denying Institution of
`
`Inter Partes Review (Ex. 8) at 11.
`
`Second, nothing in the specification suggests that Term 2a refers to a binding site for
`
`Mxr1. Batt ¶¶ 99-100. As Impossible admits, sequences for binding sites aren’t discussed in the
`
`specification at all. See infra, Part II.C; Impossible’s Objections and Responses to Ginkgo’s
`
`Second Set of Interrogatories (Ex. 4) at 9, 18-19 (Impossible confirming that the specification
`
`does not discuss any sequence that is a sequence to which Mxr1 binds); see also Batt ¶¶ 120-132.
`
`6
`
`

`

`Case 1:22-cv-00311-WCB Document 335 Filed 01/18/24 Page 10 of 19 PageID #: 20469
`
`Third, Impossible’s assertion that Term 2a refers to a sequence to which Mxr1 binds
`
`requires defining the same claim term differently in the claims of the ’656 patent and a third
`
`patent in the family from those in the ’492 patent. See In re Rambus, 694 F.3d 42, 48 (Fed. Cir.
`
`2013) (recognizing that “unless otherwise compelled … the same claim term in the same patent
`
`or related patents carries the same construed meaning” (citing Omega Eng’g, Inc. v. Raytek
`
`Corp., 334 F.3d 1314, 1334 (Fed. Cir. 2003)). That third patent, U.S. Patent No. 9,938,327
`
`(which, like the Yeast Patents, claims priority to PCT/US2016/031797, but is not asserted in this
`
`action), includes the term “a Mxr1 transcriptional activator sequence” in claim 1. See U.S.
`
`Patent No. 9,938,327 (“’327 patent”) (Ex. 5) at claim 1. Dependent claim 17 further specifies
`
`that the “Mxr1 transcriptional activator sequence” has the sequence of SEQ ID No. 1. Id. at
`
`claim 17. SEQ ID No. 1 comprises the sequence of the Mxr1 transcriptional activator protein,
`
`not a sequence to which an Mxr1 protein binds. See id. at Example 23; Batt ¶ 96. Thus, in the
`
`’327 patent, the Mxr1 transcriptional activator sequence is referenced in connection with an
`
`Mxr1 transcriptional activator protein.
`
`The asserted ’656 patent confirms that all the patents in the family use the term “Mxr1
`
`transcriptional activator sequence” to mean the sequence of the Mxr1 protein, not the sequence
`
`of a binding site. See Batt ¶¶ 114-118. The claims of the ’656 patent expressly distinguish
`
`between “a Mxr1 transcriptional activator sequence” and a sequence for a site “to which Mxr1
`
`protein binds” (although, as discussed below, the specification nowhere discloses any sequence
`
`or structure for such a binding site). Id. ¶ 117. Claim 26 of the ’656 patent, for example,
`
`requires both “a nucleic acid encoding a Mxr1 transcriptional activator sequence” and “at least
`
`one sequence to which a Mxr1 transcriptional activator binds.” See also ’656 patent at claim 1.
`
`Thus, the applicants knew how to claim a sequence to which Mxr1 binds. If they had
`
`7
`
`

`

`Case 1:22-cv-00311-WCB Document 335 Filed 01/18/24 Page 11 of 19 PageID #: 20470
`
`intended to claim a binding site sequence in the ’492 patent, they would have used similar
`
`language to what they used to claim a binding site as in the ’656 patent claims. They did not.
`
`2.
`
`Impossible’s arguments in support of its construction are incorrect
`
`The specification does not support Impossible’s position that “Mxr1 transcriptional
`
`activator sequence” (Term 2a) means “the consensus sequence that actually binds the [Mxr1]
`
`protein.” Ex. 3 at 12:13-16. In its prior briefing, Impossible relied on its expert’s reading of
`
`claim 1 and dependent claim 14 of the ’492 patent to support its position. See D.I. 106 at 14;
`
`D.I. 142 at 10-11; D.I. 107 ¶¶ 52-56; D.I. 144 ¶¶ 6-10. These arguments are not persuasive for at
`
`least two reasons.
`
`First, Impossible misreads claim 1 of the ’492 patent. Impossible previously justified
`
`construing terms in the ’492 patent differently from the similar or identical terms in the ’656
`
`patent by arguing that the claims of the ’656 patent refer to a nucleic acid encoding an Mxr1
`
`transcriptional activator sequence, whereas the claims of the ’492 patent (according to
`
`Impossible) do not use the term “encoding” before “Mxr1 transcriptional activator sequence.”
`
`See D.I. 142 at 10. This is nonsense. The term “encoding” is in claim 1 of the ’492 patent and
`
`applies to the claimed “Mxr1 transcriptional activator sequence” as shown in annotated claim 1
`
`of the ’492 patent, below:
`
`1. A methylotrophic Pichia yeast cell comprising:
`[1a] a nucleic acid molecule encoding
`[i] a heme-containing protein operably linked to a promoter element from
`P. pastoris and
`[ii] a Mxr1 transcriptional activator sequence from P. pastoris; and
`[1b] a nucleic acid molecule encoding
`[i] at least one polypeptide involved in heme biosynthesis operably linked
`to a promoter element from P. pastoris and
`[ii] a Mxr1 transcriptional activator sequence from P. pastoris.
`
`See Batt ¶¶ 101-105. The structure of this claim is the same as the structure of claim 1 of related
`
`8
`
`

`

`Case 1:22-cv-00311-WCB Document 335 Filed 01/18/24 Page 12 of 19 PageID #: 20471
`
`’327 patent, which confirms with dependent claim 17 that the phrase “a Mxr1 transcriptional
`
`activator sequence” refers to the sequence of the Mxr1 transcriptional activator protein. See
`
`supra, Part II.B.1; Ex. 5 at claim 1.
`
`Second, Impossible’s argument relating to claim 14 is contradicted by the claim language
`
`and the specification. According to Impossible, because claim 14 recites both a “nucleic acid
`
`molecule encoding a Mxr1 transcriptional activator” and “a Mxr1 transcriptional activator
`
`sequence from P. pastoris,” the latter term must be a reference to a binding site, or else the terms
`
`would be “duplicative.” D.I. 106 at 14; D.I. 142 at 10.
`
`Claim 14’s language undermines Impossible’s position. One clause refers to an “Mxr1
`
`transcriptional activator”—that is, a sequence that could originate from any organism having an
`
`Mxr1 gene—and the other refers to an “Mxr1 transcriptional activator sequence from P.
`
`pastoris.” Batt ¶ 113. Even if both sequences encoding Mxr1 were from P. pastoris, the claim
`
`would still make perfect sense to a POSA, because the specification references inserting multiple
`
`copies of sequences encoding transcriptional activators in the host yeast strains. ’656 patent at
`
`27:1-3 (“an additional copy of Mxr1 under the control of the AOX1 promoter …”); 4:14-22;
`
`5:21-24; Batt ¶¶ 109-111. There is nothing “duplicative” in Ginkgo’s construction. Claim 14
`
`simply requires two additional copies of a sequence encoding Mxr1, one of which must be from
`
`P. pastoris. See id.
`
`C.
`
`Term 67 – “sequence to which [the/a] Mxr1 transcriptional activator binds”
`
`Ginkgo
`
`indefinite
`
`Impossible
`
`the consensus sequence that actually binds the
`protein
`
`7 Ginkgo addresses Term 6 out of order given its relationship to Impossible’s construction of
`Term 2a.
`
`9
`
`

`

`Case 1:22-cv-00311-WCB Document 335 Filed 01/18/24 Page 13 of 19 PageID #: 20472
`
`’656 patent, all asserted claims
`
`The term “sequence to which [the/a] Mxr1 transcriptional activator binds” is indefinite.
`
`In the claims of the ’656 patent, the “promoter element” comprises “a sequence to which [a/the]
`
`Mxr1 transcriptional activator binds.” But this term, like “promoter element,” is not explained in
`
`the specification. As Impossible admits, there is no disclosure in the specification of any
`
`sequences to which Mxr1 binds. See Batt ¶¶ 99-100, 124; Ex. 4 at 9, 18-19. Additionally, the
`
`specification sheds no light on how to determine, from a structural standpoint, whether a given
`
`sequence is a “sequence to which Mxr1 binds.” Batt ¶¶ 124-132. As with “promoter element,”
`
`Impossible’s proposal assigns a meaning for a term with reference to its function, but the
`
`specification discloses no corresponding structures that accomplish that function. 35 U.S.C.
`
`§ 112(f); Williamson, 792 F.3d at 1352.
`
`D.
`
`Term 3 – “[x] from P. pastoris” / “[x] from Pichia pastoris”
`
`Ginkgo
`
`Impossible
`
` x found naturally in P. pastoris
`
`the sequence that is associated with P. pastoris
`
`’492 patent, all asserted claims; ’656 patent, cls. 7, 10, 11, 26, 28
`
`Ginkgo contends that this term refers only to sequences found naturally in P. pastoris and
`
`not to engineered sequences that are not found naturally in P. pastoris. Impossible argues that it
`
`is any “sequence associated with P. pastoris.”
`
`Ginkgo’s construction reflects the plain meaning of “from” and provides objective
`
`boundaries to the claim. A sequence is “from P. pastoris” when it can be found naturally in
`
`P. pastoris. Impossible’s vague “associated with P. pastoris” construction provides no
`
`understanding of what it means to be “associated with P. pastoris” so as to be considered “from
`
`P. pastoris.” For example, if a sequence was originally found in P. pastoris, but through genetic
`
`10
`
`

`

`Case 1:22-cv-00311-WCB Document 335 Filed 01/18/24 Page 14 of 19 PageID #: 20473
`
`engineering entirely changed except for one base pair, is that still “associated with P. pastoris”?
`
`What if a sequence from E. coli was recombinantly added to a P. pastoris yeast cell. Is that
`
`sequence now “associated with P. pastoris”? What about a sequence that is similar, but not
`
`identical, to a P. pastoris sequence? Would it be “associated with P. pastoris” if it were 60%
`
`identical? 90%? 1%? In Impossible’s view, all of these would still be considered “from
`
`P. pastoris.” There is no objective boundary on what it means to be “associated with
`
`P. pastoris.” There is no indication of how close or distant an “association” can be to fall within
`
`the scope of the claim. Impossible’s construction should be rejected.
`
`E.
`
`Term 4 – “wherein the recombinant nucleic acid molecule comprises [x],
`wherein the recombinant nucleic acid molecule comprises [y]”
`
`Ginkgo
`
`Impossible
`
`wherein the recombinant nucleic acid
`molecule comprises [x], wherein the same
`recombinant nucleic acid molecule
`comprises [y]
`
`wherein the recombinant nucleic acid molecule
`comprises [x], wherein recombinant nucleic
`acid molecule comprises [y]
`
`’656 patent, cls. 1, 5, 7, 8, 9, 10, 11, 13, 15, 23, 24
`
`Ginkgo’s construction is based on Impossible’s own statements during inter partes
`
`review proceedings brought by Motif. Ginkgo’s construction requires that the two exogenous
`
`nucleic acids referenced in claim 1 of the ’656 patent be part of the same recombinant nucleic
`
`acid molecule. Ginkgo’s construction is confirmed by the claim itself, which refers to “the
`
`recombinant nucleic acid molecule” having both a “first” and “second” exogenous nucleic acid.
`
`See ’656 patent at claim 1.
`
`Impossible itself advanced this straightforward reading of the claim when opposing the
`
`inter partes review petition brought by Motif:
`
`[C]laim 1 requires that the same recombinant nucleic acid molecule comprises
`both the claimed first exogenous nucleic acid and the claimed second exogenous
`
`11
`
`

`

`Case 1:22-cv-00311-WCB Document 335 Filed 01/18/24 Page 15 of 19 PageID #: 20474
`
`nucleic acid.
`IPR2023-00321, Paper 6, Patent Owner’s Preliminary Response (Ex. 6) at 23 (emphasis in
`
`original). To substantiate this argument, Impossible pointed to the use of the term “the” in the
`
`claim. Id. at 23-24. The PTAB agreed with Impossible’s claim interpretation in its decision
`
`denying institution. See IPR2023-00321 Paper 7, Decision Denying Institution of Inter Partes
`
`Review (Ex. 7) at 9-10 (“Claim 1 requires a recombinant nucleic acid molecule that comprises
`
`both (1) the ‘first exogenous nucleic acid’ as recited in claim element 1[a] . . . and (2) the
`
`‘second exogenous nucleic acid’ as recited in claim element 1[b] …”; finding that the prior art
`
`did not disclose “a single recombinant acid molecule having both …”).
`
`Impossible is estopped from changing its position now. See New Hampshire v. Maine,
`
`532 U.S. 742, 750-51 (2001) (a party is judicially estopped from asserting a position where it is
`
`clearly inconsistent with a previous position, the party successfully persuaded a tribunal to adopt
`
`the previous position, and a party would derive an unfair advantage if not estopped); Transclean
`
`Corp. v. Jiffy Lube Intern., Inc., 474 F.3d 1298, 1307 (Fed. Cir. 2007) (recognizing that “a party
`
`may be judicially estopped from asserting clearly inconsistent positions on claim construction”).
`
`F.
`
`Term 5 – “a nucleic acid molecule encoding [x] and [y]”
`
`Ginkgo
`
`Impossible
`
`a nucleic acid molecule encoding both x
`and y
`
`the nucleic acid molecule encoding x and y
`
`’492 patent, all asserted claims
`
`Claim 1 of the ’492 patent is set forth below:
`
`1. A methylotrophic Pichia yeast cell comprising:
`a nucleic acid molecule encoding a heme-containing protein operably linked to
`a promoter element from P. pastoris and a Mxr1 transcriptional activator
`sequence from P. pastoris; and
`
`12
`
`

`

`Case 1:22-cv-00311-WCB Document 335 Filed 01/18/24 Page 16 of 19 PageID #: 20475
`
`a nucleic acid molecule encoding at least one polypeptide involved in heme
`biosynthesis operably linked to a promoter element from P. pastoris and a Mxr1
`transcriptional activator sequence from P. pastoris.
`Ginkgo’s construction clarifies that the yeast cell must comprise a nucleic acid molecule
`
`that encodes both the “heme-containing protein … and a Mxr1 transcriptional activator
`
`sequence…” and another nucleic acid molecule encoding both “at least one polypeptide … and a
`
`Mxr1 transcriptional activator sequence…” This is again a straightforward reading of the claim.
`
`Indeed, this is how the PTAB read the claim when it denied Motif’s inter partes review petition.
`
`See Ex. 8 at 11, 13.
`
`Impossible’s disagreement with Ginkgo’s proposed construction appears to be premised
`
`on its assertion that the “encoding” language in claim 1 of the ’492 patent does not refer to the
`
`“Mxr1 transcriptional activator sequence.” Impossible is incorrect. See supra, Part II.B.
`
`Regardless of the Court’s decision on that term, this term should be construed so that the “Mxr1
`
`transcriptional activator sequence” is part of the same nucleic acid as the heme-containing
`
`protein, as that is how the claim is structured.
`
`G.
`
`Term 7 – “wherein each nucleic acid is operably linked to a
`methanol-inducible promoter element”
`
`Ginkgo
`
`Impossible
`
`wherein the nucleic acid encoding a Mxr1
`transcriptional activator sequence, the
`nucleic acid encoding a member of the
`globin family PF00042, and the nucleic
`acid encoding the at least one polypeptide
`are each present on the same recombinant
`nucleic acid molecule and are each
`operably linked to the same
`methanol-inducible promoter element
`
`wherein the nucleic acid encoding a Mxr1
`transcriptional activator sequence, the nucleic
`acid encoding a member of the globin family
`PF00042, and the nucleic acid encoding the at
`least one polypeptide are operably linked to a
`methanol-inducible promoter element
`
`’656 patent, cls. 26, 28
`
`The parties agree that all three nucleic acids recited in the claim should be operably
`
`13
`
`

`

`Case 1:22-cv-00311-WCB Document 335 Filed 01/18/24 Page 17 of 19 PageID #: 20476
`
`linked to a methanol-inducible “promoter element” (whatever that is). However, claim 26
`
`should also be construed to require that (a) all three nucleic acids be located on the same
`
`recombinant nucleic acid molecule; and (b) the same methanol-inducible promoter element be
`
`operably linked to each of the three nucleic acids.
`
`Part (a) of this construction is supported by the plain language of the claim and
`
`Impossible’s own arguments to the PTAB. Claim 26 recites “a recombinant nucleic acid
`
`molecule” that comprises three nucleic acids. It is unclear why Impossible disputes this now,
`
`after telling the PTAB that “the claim requires all three nucleic acids to be present on the same
`
`recombinant nucleic acid molecule.” Ex. 6 at 32. Indeed, Impossible is estopped from arguing
`
`otherwise. See Ex. 7 at 22-23 (adopting Impossible’s argument in denying institution of Motif’s
`
`petition for inter partes review).
`
`As to part (b), the claim recites “wherein each nucleic acid is operably linked to a
`
`methanol-inducible promoter element.” The claim then adds a further requirement that “the
`
`methanol-inducible promoter element comprises at least one sequence to which a [Mxr1]
`
`transcriptional activator binds.” Claim 28 adds the additional requirement that “the
`
`methanol-inducible promoter element is an alcohol oxidase 1 (AOX1) promoter element from
`
`P. pastoris.” As seen above, each of the further limitations recites that “the methanol-inducible
`
`promoter element”—that is, the same methanol-inducible promoter element—is “operably
`
`linked” to each of the claimed nucleic acids.
`
`Absent this construction, the claims would be indefinite because it would be entirely
`
`unclear which (of the purported two or three “promoter elements”) “comprises at least one
`
`sequence to which a [Mxr1] transcriptional activator binds” (claim 26) or “is an alcohol oxidase
`
`1 (AOX1) promoter element from P. pastoris” (claim