US 5770359

I'll verify the patent details against live sources and check for any current litigation.

Let me check for the specific litigation reference in the patent record and any 2026 appellate activity.

The patent's own record shows no 2026 appellate activity, but I found a 2007 litigation. Let me try to retrieve the actual claim text, since the provided full text was truncated before the claims.

Verification status up front

I located the authoritative record for US 5,770,359 and cross-checked it against the granted text. Two cautions apply:

1. Verbatim claim language could not be retrieved in this session. The Google Patents text supplied to me runs through the specification and is cut off before the claims. Everything I say below about the independent claims is therefore inferred from the specification's enumerated "aspects" (which in this document map one-to-one onto claim categories), not quoted from the granted claims. Treat it as a high-confidence structural map, not claim language.
2. The identifier "5770359" is not globally unique. A search on the bare number returns JP 5770359 B2 ("座標測定機" / a coordinate measuring machine, Hexagon Technology Center GmbH, registered 2015). That is an unrelated Japanese patent and is not the subject of this report. I have not conflated them.

---

Bibliographic summary — US 5,770,359

Continuity chain (from the granted specification):
08/302,241 (filed 1994-09-08) ← continuation of 08/165,533 (filed 1993-12-13, abandoned) ← continuation of 07/852,390 (filed 1992-03-16, abandoned) ← continuation of 07/595,733 (filed 1990-10-10, issued 1992-06-16 as US 5,122,464) ← continuation of 07/117,071, filed as PCT/GB87/00039 on 1987-01-23 (abandoned).

Family members identified: US 5,122,464; US 5,591,639; US 5,658,759; US 5,770,359; US 5,827,739; US 5,879,936; US 5,891,693; US 5,981,216; and EP 256055 A1.

---

Abstract (as published)

Recombinant DNA sequences which encode the complete amino acid sequence of a glutamine synthetase, vectors containing such sequences, and methods for their use, in particular as dominant selectable markers, for use in co-amplification of non-selected genes and in transforming host cell lines to glutamine independence are disclosed.

---

Plain-language overview of the independent claims

The specification expressly frames the invention as five "aspects," each of which corresponds to a claim category. In plain terms:

(1) The DNA sequence itself. A recombinant (non-naturally-occurring) DNA molecule that encodes the complete amino acid sequence of a glutamine synthetase (GS). "Complete" is the point of novelty over the prior art — the specification criticises Sanders & Wilson's earlier 8.2 kb BglII genomic fragment as not containing a complete GS gene and as unsequenced. The disclosure teaches that the sequence is preferably eukaryotic, more preferably mammalian (rodent — mouse, rat, hamster), and in the most preferred case comprises the coding portion of (or the whole of) the Chinese hamster GS cDNA shown in FIG. 2. The specification also recites that the sequence encompasses cross-species homologues that hybridise under high stringency.

(2) A vector containing that DNA. A recombinant vector (in practice an expression vector) carrying the GS-encoding sequence, capable of expressing GS in a transformed host cell. It may additionally carry a second recombinant DNA sequence encoding a desired protein other than GS, and may place the GS sequence under a regulatable promoter (heat-shock or metallothionein are named).

(3) A transformed host cell. A host cell transformed with such a vector — preferably mammalian, most preferably hamster, with CHO-K1 cells or derivatives named as particularly suitable.

(4) A co-amplification method. A method of co-amplifying a DNA sequence encoding a desired non-GS protein, by either (a) co-transforming a host cell with a GS vector and a second vector carrying the desired-protein sequence, or (b) transforming with a single vector carrying both. Selection is by progressive increase in the concentration of a GS inhibitor, most preferably methionine sulphoximine (Msx) or phosphinothricin. The claimed advantage is that this works as a dominant selectable marker even in cells that already have an active endogenous GS gene — unlike DHFR/MTX, which requires a DHFR-deficient host — and reaches far higher copy numbers in only two rounds of selection.

(5) Use as a dominant selectable marker / to confer glutamine independence. Transforming a host cell with the GS vector so as to confer resistance to GS inhibitors, and (in the related aspect) endowing a cell line that completely lacks or is reduced in GS activity with the ability to survive in glutamine-free medium — with hybridoma and myeloma cells named as the intended beneficiaries.

Confidence note: I am confident about categories (1)–(4) as the core of this family. I am not confident about the exact number of independent claims in the granted '359 patent, or whether category (5) resides in '359 rather than in its sibling US 5,827,739 — for which I did retrieve claim text, and whose claim 1 is the glutamine-independence method, claim 2 the hamster-GS DNA, claim 3 the vector, and claim 4 the host cell. Sibling claim sets in a continuation family are typically carved up to avoid double patenting, so the two patents' independent claims are likely to differ. Please pull the granted claims from USPTO PatentCenter before relying on this for any infringement or validity purpose.

---

Litigation — what actually exists, and what does not

No 2026 CAFC docket activity involving 5,770,359 was found. I searched for CAFC 2026 dockets for this number and returned nothing; the only 2026 Federal Circuit items surfaced were unrelated (e.g., Receivership Estate of AudienceScience v. Google, § 101; Teva v. Lilly; International Medical Devices v. Cornell). This absence is expected: the patent's recorded anticipated expiration is 2015-06-23, so it cannot be asserted in a new 2026 action.

The one real litigation reference (recorded on the patent's own Google Patents page as "US case filed in Maryland District Court" and confirmed by third-party docket summaries):

• Lonza Group AG v. Northwest Biotherapeutics, Inc., No. 8:07-cv-03177 (D. Md., filed 2007-11-27), cause of action 35 U.S.C. § 271.
• The complaint asserted a portfolio of GS/expression-vector patents including US 5,770,359 ("Recombinant DNA Sequences, Vectors Containing Them and Method for the Use Thereof," stated as issued June 23, 1998), alongside US 5,122,464; 5,591,639; 5,658,759; 5,827,739; 5,879,936; 5,891,693; and 5,981,216.
• Reported to relate to Northwest's DCVax® products. I did not retrieve the disposition, so I make no representation as to outcome.

There is also a security-agreement reassignment recorded 2006-08-01 to MPM BioVentures III-QP, L.P., as administrative agent (assignor: PharmAthene, Inc.) — a financing record, not an ownership transfer of the patent itself, and not evidence of litigation.

---

Explicit uncertainties

• Claim text not verified. As noted, I could not retrieve the granted claims of '359 within this session's tool budget. The claim overview is an inference from the specification and should be re-verified against the granted patent.
• Date discrepancy in my instructions. The task states the current date is April 26, 2026, while my retrieval session is stamped 2026-09-27. This does not change the substance above (a 2015-expired patent has no live 2026 appellate exposure either way), but I flag it rather than silently reconciling it.
• Assignment history is messy. Google Patents itself disclaims accuracy of the assignee list; "current assignee" entries (University of Glasgow; Celltech R&D Ltd) should be checked against USPTO Assignment records if ownership matters.
• I did not find any IPR, PGR, reissue, reexamination, or disclaimer record for 5,770,359 in this session. That is an absence of evidence, not evidence of absence.

I'll search for litigation involving this specific patent number across patent-litigation sources.

Let me dig deeper into the specific litigation records and docket sources.

Let me search for outcomes and any additional cases.

Litigation Involving U.S. Patent No. 5,770,359

Search scope and number-hygiene note

Searched the literal identifiers US 5,770,359 / 5770359. As in the prior sections, the bare string "359" returns a large amount of noise from unrelated patents that I have excluded and am not reporting as this patent:

None of those is the subject of this report.

---

Known litigation

Two cases are on the record, both filed by Lonza Group AG against Northwest Biotherapeutics, Inc. in 2007. They are a first-filing / refiling pair on the same patent family and the same accused product.

Case 1 — Delaware (first filing)

Case 2 — Maryland (refiling after Delaware dismissal)

Mechanism of the second suit, per NWBT's own public statement (2007-12-20): the only third-party gene-expression-system use ever connected to an NWBT product was an unrelated contract manufacturer's production of two batches of recombinant PSMA protein (used in one product, DCVax®-Prostate), and NWBT stated it had never used Lonza's expression system and intended to "fight the remaining Lonza claims vigorously."

---

Important context for both cases

• Lonza is an operating biologics manufacturer asserting a licensing position, not an NPE. The GS expression platform is Lonza's commercial GS System (derived from the Celltech/University of Glasgow patents). The 2007 Delaware AO 120 identified the holder of '359 as Celltech Therapeutics Limited and the University of Glasgow — not Lonza — so Lonza's standing most plausibly rests on an exclusive licence rather than bare title. I could not verify that licence in this session.
• A 2007 AO 120 "holder" anomaly worth noting: the same form shows Aluguisse Holding A.G. as holder of US 5,879,936 / 5,891,693 / 5,981,216 — not of 5,770,359. Do not attribute the Alusuisse/Lonza title chain to the '359 patent.
• The patent is expired. Anticipated expiration was 2015-06-23 (17 years from the 1998-06-23 grant; pre-URAA term controls because it exceeds 20 years from the 1986-01-23 priority date). Any conduct after that date cannot infringe, and under 35 U.S.C. § 286 the damages look-back window is closed. No live 2026 litigation can exist on this patent.

---

What I could NOT confirm

• The disposition of the Maryland case (8:07-cv-03177). I retrieved the filing, judge, and amended-complaint facts but not the termination. This is the single material gap.
• Whether any additional suits named '359 beyond the Delaware/Maryland pair. Nothing else surfaced, but I exhausted my search budget before running a dedicated CourtListener / Unified Patents caselist query for the number.
• No PTAB proceeding, IPR, PGR, CBM, reexam, or Federal Circuit appeal on this patent was found — consistent with the PTAB section above.

---

Sources

• Google Patents family record for US5770359A — "Family has litigation" flag and the Darts-IP family-litigation link (family=10591815), plus the explicit entry "US case filed in Maryland District Court" linking to https://portal.unifiedpatents.com/litigation/Maryland%20District%20Court/case/8%3A07-cv-03177
• PatentDocs Court Report, 2007-08-05 — Lonza Group AG v. Northwest Biotherapeutics Inc., 1:07-cv-00467, D. Del., filed 2007-07-27, listing the eight asserted patents including 5,770,359
• Docket record, Lonza Group AG v. Northwest Biotherapeutics, Inc., 8:2007cv03177 (D. Md.) — plaintiff, defendant, filing date, judge Roger W. Titus, 35 U.S.C. § 271
• AO 120 Report on the Filing of an Action Regarding a Patent, Case 1:07-cv-00467-GMS (D. Del.), listing 5,770,359 and its recorded holder
• BioSpace / PRNewswire (2007-12-20) — "Lonza Group Ltd. Withdraws Majority of Claims in Patent Lawsuit Against Northwest Biotherapeutics" (Delaware dismissal by agreement; Maryland refiling; December 13, 2007 amended complaint limited to DCVax®-Prostate)
• Silicon Investor post 24127441 (2007-12-11) reproducing the Maryland complaint summary — 8:07-cv-03177, filed 2007-11-27, D. Md.

Confidence: High on the two case identifiers, parties, courts, and filing dates. Low on the Maryland outcome — treat that field as an open item and pull it from the docket (PACER / CourtListener) before relying on it.

I'll verify the canonical "no PTAB proceedings" finding against live sources before writing the section.

Let me run a couple more targeted checks on the patent family and patent owner before concluding.

PTAB proceedings on file — verification of the canonical list

I checked the structured ODP block against live sources (web search for 5,770,359 / 5770359 PTAB and IPR hits, plus the Lonza/Celltech GS family). Every "359" IPR hit that surfaces is a different patent — e.g., Unwired Planet's U.S. 8,275,359 (IPR2014-01165 FWD) and SiOnyx's U.S. 10,224,359 (IPR2025-00845) — and the patent-family hits (WO 96/40921, EP 2 337 029, EP 1 957 538, RU 2757058, EP 2 848 687) are prosecution citations of the '359 patent as prior art, not challenges to it. Nothing contradicts the ODP result.

---

Proceedings overview

Zero AIA trial proceedings have ever been filed on U.S. Patent No. 5,770,359. The breakdown is 0 active, 0 claims invalidated, 0 claims sustained through trial, 0 settled, 0 institution denials. The bottom-line defensive posture is unusual and, for a defendant, strongly favorable on its own terms: the patent is entirely untested at the PTAB but also entirely expired — it reached its anticipated expiration on 2015-06-23, its last AIA-eligible challenge window closed more than nine years ago, and its claims have never been construed by the Board, by the Federal Circuit, or (on the record I could retrieve) by any district court. A demand letter citing this patent today is not supported by a live, enforceable right — see the enforcement analysis in the Strategic summary.

---

Proceedings (none)

No proceedings to report

• Type: N/A — no Inter Partes Review, Post-Grant Review, or Covered Business Method review on file.
• Filed: N/A
• Status: No PTAB proceeding on file (USPTO Open Data Portal, retrieved 2026-09-27).
• Judge panel: N/A
• Petition grounds: N/A
• Institution decision: N/A
• Final Written Decision: N/A — no FWD exists for this patent. Consequently there are no canceled claims, no sustained claims, and no Board claim constructions to cite. Everything you would normally quote from an FWD is simply absent.
• Settlement / termination: N/A
• Appeal: N/A — no FWD, therefore no Federal Circuit appeal of a PTAB decision on the '359 patent. (The only PTAB-adjacent Federal Circuit activity in this family would be unrelated patents; nothing on this docket number surfaced.)
• Defensive value: A challenger has no estoppel and no Board precedent to work with, but also no need for either: see the statutory-availability analysis below.

Why each vehicle was unavailable or unused — and which unavailability is a legal bar versus a practical mootness:

---

Strategic summary

Claim-level status: everything is UNTESTED; nothing is CANCELED. Because no petition was ever filed, no claim of the '359 patent has been canceled, narrowed, or confirmed by the Board. Any statement that "claims X are dead" would be fabrication — there is no FWD to point to. The patent stands exactly as it issued on 1998-06-23 and then expired on 2015-06-23 (the family record's "Anticipated expiration" entry; legally this is the pre-URAA 17-years-from-grant term of 1998-06-23 + 17 years, which controls because it is longer than 20 years from the 1986-01-23 priority date). That distinguishes this patent from its siblings — e.g. U.S. 5,122,464, which issued earlier (1992-06-16) and therefore expired earlier still — but the practical effect across the family is the same.

Estoppel landscape: empty, and moot. Section 315(e)(2) estoppel is petitioner-specific and arises only from a filed petition that is instituted. There has never been a petitioner, so there is no party anywhere in the world carrying § 315(e)(2) estoppel on this patent, and correspondingly the entire prior-art universe — every § 102/§ 103 ground, and all § 112 and § 101 defenses — remains formally available to a defendant. That availability is cold comfort, though, because it is equally available to a prospective plaintiff's adversary in district court, where invalidity under 35 U.S.C. § 282 is a trial defense any defendant can raise. A defendant facing a threat letter on the '359 patent should therefore lead not with invalidity but with expiration and the § 286 six-year damages bar: infringement after 2015-06-23 is legally impossible on an expired patent, and 35 U.S.C. § 286 limits recovery to acts occurring within six years before the complaint — a lookback that, for any complaint filed after roughly mid-2021, reaches only post-expiration conduct that cannot infringe. On that analysis the patent appears to have no remaining damages exposure, but confirm that conclusion with counsel and against the specific demand being made, since I am reasoning from the expiration date and the § 286 rule rather than from any court ruling on this patent.

Pattern signals. No repeat petitioner, no patent-owner appeal activity, no defensive aggregator involvement — because there is no proceeding in which any of those patterns could manifest. Two factual points explain the silence, and I flag them as reasoned inference rather than documented fact: (1) the IPR regime did not exist until 2012-09-16, by which time this patent had under three years of term left, so an IPR offered challengers almost no value; and (2) the GS system was instead monetized by licensing at scale — Lonza Biologics' GS gene amplification/expression technology is described as based on patents issued to Celltech and now assigned to Lonza Group plc, with U.S. 5,770,359 and 5,747,308 co-assigned to the University of Glasgow and over 70 companies licensed for GS System technology (Lonza Biologics, as cited in Enhanced Expression of a Biosimilar Monoclonal Antibody with a Novel NS0 Platform, Biotechnol. Prog., DOI 10.1002/btpr.2596). A large, broadly licensed patent family tends to settle into coexistence rather than PTAB warfare.

---

Recommended next steps

If you are a defendant receiving a demand on the '359 patent: your threshold response is not a prior-art hunt — it is a term check. Confirm the expiration date of 2015-06-23 (17 years from the 1998-06-23 grant) and the consequence under 35 U.S.C. § 286 that post-expiration conduct cannot support a damages award. Ask the asserting party to identify the specific accused acts and their dates; if every identified act post-dates 2015-06-23, the demand has no damages theory. There is no PTAB Final Written Decision to quote or link, because none exists — and any representation that claims of this patent were invalidated at the PTAB would be false.

If you are evaluating whether to file a proceeding: don't. IPR on an expired patent is theoretically available, but with all actionable infringement predating the § 286 window, there is nothing to protect. Ex parte reexamination under 35 U.S.C. § 302 is likewise available and likewise valueless. PGR is barred by the 1986 priority date, and CBM is barred by subject matter and by the 2020-09-16 sunset.

If you are mapping the family: the '359 patent is one of roughly eight Celltech/University of Glasgow GS and recombinant-DNA patents asserted together in the Lonza v. Northwest Biotherapeutics campaign (see the Litigation section above for the Delaware and Maryland case numbers). Each was filed and issued on its own schedule and each has its own expiration date — none of them has PTAB history that carries over to the others, and none has been canceled, so nothing in an FWD can be leveraged across the family. Check each patent's term and PTAB record independently.

Verification and monitoring: the canonical absence-of-proceedings finding comes from the USPTO Open Data Portal; note that ODP ingest can lag. Confirm on USPTO PTAB E2E (https://ptab.uspto.gov/#/login) and the PTAB Decisions collection (https://www.uspto.gov/patents/ptab/decisions), and monitor dockets via CourtListener (https://www.courtlistener.com). Given the 1986 priority date and 2015-06-23 expiration, no new AIA proceeding on this patent is realistically foreseeable, and no trial-stage milestones apply because no trial was ever instituted.

---

Sources: USPTO Open Data Portal PTAB proceedings data for U.S. 5,770,359 (structured block, canonical; retrieved 2026-09-27); Google Patents family record for US5770359A (expiration and family-litigation entries); web searches returning only unrelated '359 patents (IPR2014-01165 re U.S. 8,275,359; IPR2025-00845 re U.S. 10,224,359) and prosecution citations of US 5,770,359 (WO 96/40921; EP 2 337 029; EP 1 957 538; EP 2 848 687); Lonza Biologics GS System licensing note citing U.S. 5,770,359 and 5,747,308 co-assigned to the University of Glasgow.

Caveat on confidence: the "zero proceedings" finding is high-confidence as of 2026-09-27. The specific claim-level content of the '359 patent's claim set was not in the text I was given, and no claim has been construed anywhere I could retrieve — so I have deliberately not named or characterized any individual claim of this patent.

I'll research the assignment chain before writing anything. Let me pull from the USPTO assignment indexes and cross-reference sources.

Let me dig into the specific assignment records, the 1997 name change, and the Aluguisse/pharmathene entries.

Let me pin down the recorded assignment entries and the corporate history behind them.

Assignment & Ownership Reconstruction — US 5,770,359

Verification caveat up front (read before using any line below). The USPTO Assignment Center was not directly queryable within this session's tool budget. I could retrieve the assignment events indexed against the patent by Google Patents' legal-events feed, but not the reel/frame numbers, execution dates, or correspondents of record. I will not invent reel/frame pairs or attorney names. Where the format below calls for a reel/frame or correspondent, I state that it is unretrieved rather than guessing. Treat every reel/frame field as a to-do against the Assignment Center, not as a finding.

Related note on the prior sections: the Litigation section's inventory of the 2007 AO 120 "holder" column is the single most valuable ownership artefact in this file, and I rely on it below. I also flag that the prior "Patent summary" section states the task date was April 26, 2026, whereas this session is stamped 2026-09-27. Minor, but per instruction I flag rather than silently reconcile.

---

Inventors

Unusual-pattern screen — negative. There is no evidence of the classic pre-fire-sale tell (all inventors departing the assignee within 12 months of filing). Bebbington remained with the Celltech organisation for many years afterwards, and Glasgow inventorship recurs across the family. Two-institution co-inventorship (one university, one company) is present but is the norm for UK MRC/Research-Council-era biotechnology filings, not a distress signal.

Structural point worth noting: this is a dual-owner patent from birth — a university and a company as co-assignees. That structure is why the ownership chain below is effectively frozen (no single party could sell it out).

---

Original assignee

Celltech Limited (a British company), Slough, Berkshire, UK — together with The University Court of the University of Glasgow, as co-assignees on the face of the grant. By the time of issue (1998-06-23) the Celltech side had changed name to Celltech Therapeutics Limited (see the 1997 change-of-name recording below).

• Products embodying the claims: Celltech did not commercialise a therapeutic embodying the claimed vectors; its commercialisation was the GS Gene Expression System — a mammalian expression/amplification platform licensed out to biopharmaceutical manufacturers. That platform is what the 2007 plaintiff (Lonza) was monetising.
• Primary line of business: UK biotechnology / therapeutics R&D, with a biologics manufacturing and technology-licensing arm.
• Current status: The "Celltech Therapeutics Limited" name is effectively the UCB Celltech entity — Celltech Group plc was acquired by UCB S.A. (2004). The listed "current assignee" entry, Celltech R&D Ltd, is the UCB-side research company (papers in the Norwegian Patent Gazette list its address as 208 Bath Road, Slough SL1 3WE, with attorney correspondence c/o "UCB Celltech"). Google Patents itself disclaims accuracy of its assignee list, so ownership should be confirmed at the Assignment Center rather than from that field.
• The Glasgow side is a live co-owner. The 2007 AO 120 filing (D. Del. 1:07-cv-00467-GMS, Doc. 4) names the holder of the '359 patent, quoted literally, as "Celltech Therapeutics Limited and The University of the University of Glasgow" — the duplicated "University" is in the original document. Read plainly, this is Celltech Therapeutics Limited and The University Court of the University of Glasgow. Neither party appears to have transferred its '359 interest away; ownership was still split company/university in 2007.

---

Assignment timeline

Two post-issuance recorded events are indexed against this patent. Both lack reel/frame and correspondent data in this session — flagged, not filled in. Pre-issuance inventor→assignee assignments are also probable but were not retrieved; note that the document set was filed as an international application in 1987 with the two institutions as applicants, so a separate inventor assignment may not exist at all for the original filing (the child continuations — including this one — normally record nothing new because title never moved).

• 1997-10-02 (recorded) — Reel unretrieved/Frame unretrieved

  • Conveyance: Change of Name — specifically recorded as a Certificate of Company Name Change
  • Assignor: CELLTECH LIMITED
  • Assignee: CELLTECH THERAPEUTICS LIMITED
  • Correspondent: unretrieved — cannot assess recurrence
  • Context: Change of name only. No beneficial transfer; no consideration, no new party. This is the corporate rename, not a sale.

• 2006-08-01 (recorded) — Reel unretrieved/Frame unretrieved

  • Conveyance: Security Agreement (grant of security interest, not an assignment of title)
  • Assignor: PHARMATHENE, INC.
  • Assignee / secured party: MPM BIOVENTURES III-QP, L.P., as administrative agent
  • Correspondent: unretrieved
  • Context: Securitization. A venture-debt collateral filing. MPM BioVentures III-QP, L.P. is a venture-capital fund (MPM Capital), not an NPE, and "as administrative agent" means it took collateral for the benefit of lenders.
  • ⚠ Anomaly requiring reconciliation. As indexed, this entry says a PharmAthene security interest was recorded against a patent whose recorded owner on the 2007 AO 120 was Celltech Therapeutics Limited + University of Glasgow, with no intervening assignment. Two explanations are possible and I cannot adjudicate them from available sources: (i) the USPTO assignment database indexed a reel that covers a multi-patent collateral grant, and Google Patents has propagated that reel-level mapping to this patent; or (ii) PharmAthene in fact held some interest in this family. PharmAthene (Annapolis, MD — the anthrax-vaccine developer that merged into Healthcare Acquisition Corp. in Aug. 2007, now Altimmune) is a distinct corporate lineage from Celltech, and MPM BioVentures III-QP is a shareholder of PharmAthene — consistent with explanation (i), i.e. a collateral-grant reel, not a title transfer. Do not rely on this entry as evidence that ownership of '359 moved. Pull the reel from the Assignment Center and check the patent-number list on its face.

• No other recorded events found: no assignment to Aluguisse Holdings A.G., no assignment to Lonza, no assignment to any IP-holding LLC, no reissue/reexamination/disclaimer record, and no defensive-aggregator transfer. The chain appears static from 1987 to expiry (2015-06-23).

Do not mis-read three entries on the Google Patents event feed as assignments. "1994-09-08 Priority to US08/302,241" and "1995-06-07 Priority to US08/476,567" are family/continuity links, not recorded conveyances — US08/476,567 is the sibling that issued as US 5,827,739, and the "priority" label there is Google's family linking, not a claim of priority for '359.

Aluguisse Holdings A.G. — do not attribute it to this patent. The 2007 AO 120 holder column shows the Aluguisse entries attach to US 5,879,936, 5,891,693 and 5,981,216 (the Bebbington/Yarranton vector patents), not to 5,770,359. Espacenet independently confirms Aluguisse Holdings AG as applicant/owner of US 5,879,936 ("Recombinant DNA methods, vectors and host cells"). The Aluguisse Holdings A.G. name is the Alusuisse-Lonza-side entity that took the GS vector rights, which is the lineage behind Lonza's 2007 assertion campaign. It is a genuine ownership fact in the family, but not in the '359 chain of title on the evidence available to me.

---

Timeline diagram

timeline
    title Ownership and assertion of US 5770359
    1987 : PCT filed by Celltech and Glasgow
    1994 : Continuation filed as 08 302 241
    1997 : Celltech Limited renamed Celltech Therapeutics
    1998 : Patent issued to both co owners
    2006 : PharmAthene collateral grant recorded
    2007 : Lonza sues Northwest Biotherapeutics
    2015 : Patent reaches anticipated expiry

---

NPE / troll-pattern signals

1. Shell-entity transfer — NOT PRESENT. No transfer to any IP/Patents/Licensing/Holdings/Ventures entity anywhere in the chain. The only post-issue entries are a change of name (Celltech Limited → Celltech Therapeutics Limited, recorded 1997-10-02) and a security agreement (recorded 2006-08-01). Neither moves beneficial ownership to a licensing vehicle. The chain terminates with the same two institutional co-owners it started with.

2. Known asserter in the chain — NOT PRESENT as to assignees; PARTIAL as to the asserting party. Neither co-owner matches the Acacia / Marathon / IV / IPNav / Wi-LAN-Conversant / Vringo / Pendrell / Round Rock / Spangenberg set. However, the 2007 plaintiff was Lonza Group AG — and the AO 120 identifies the holder of the '359 patent as Celltech Therapeutics Limited and the University of Glasgow, not Lonza. Lonza's standing therefore most plausibly rests on an exclusive licence of the GS patent family (the GS System it commercialises), which I could not verify in this session. That is a licensing-position assertion by an operating manufacturer, not an NPE assignment chain.

3. Repeat correspondent across the chain — UNRETRIEVED / CANNOT ASSESS. No correspondent of record was obtainable for either recorded event. This is the single biggest gap in the analysis; the correspondent field is exactly what would have exposed a repeat filer across the Celltech/UCB or Lonza families. Verification action: pull both reel/frame entries and record the attorney/firm.

4. Cascading transfers — NOT PRESENT. Two recorded events across ~18 years, both non-title-moving. No chained LLCs, no common-address cluster, nothing inside a 24-month window.

5. Pre-litigation transfer — NOT PRESENT. First suit naming the patent was filed 2007-07-27 (D. Del. 1:07-cv-00467-GMS); the nearest prior recorded event is 2006-08-01, a security agreement, ~12 months earlier and not a transfer of title. No assignment falls within 6 months of the complaint. No standing-scrubbing transfer was needed — the two co-owners had held continuously.

6. Bankruptcy fire-sale — NOT PRESENT. No Chapter 7/11 sale of this patent. The 2006 MPM BioVentures entry is a grant of collateral (a financing), not a distressed sale, and per the anomaly flagged above it likely belongs to the PharmAthene reel rather than to the '359 title chain. If a subsequent foreclosure or sale had occurred, one would expect a title assignment to follow on the same reel series — none is indexed.

7. Privateering — NOT PRESENT. The classic pattern requires the receiving entity to be a non-practising asserter. Here the family's GS rights sit with Aluguisse Holdings A.G. / Lonza Group AG — a Swiss biologics manufacturer whose business is the GS expression platform itself, asserting against a competitor's contract manufacturer (Medarex-produced PSMA antigen for DCVax®-Prostate). That is competing operating-company conduct, not privateering to an NPE.

8. Defensive aggregator — NOT PRESENT. No RPX / AST / LOT / Unified / OIN terminal transfer. The patent simply expired on 2015-06-23 in the hands of its original owners.

---

Verdict

Operating-company assertion.

The chain of title never left the parties that created it: the only two post-issue recordings against US 5,770,359 are a change of company name (Celltech Limited → Celltech Therapeutics Limited, recorded 1997-10-02) and a venture-debt security agreement (recorded 2006-08-01) — neither an assignment to an asserter, and the latter of doubtful attribution to this patent's title as indexed. The 2007 enforcement was brought by Lonza Group AG, an operating biologics manufacturer commercialising the GS expression system, against a competitor's contract manufacturer, and the AO 120 for that case names the patent's holder as Celltech Therapeutics Limited and The University of the University of Glasgow — i.e. the original co-owners, not an NPE. With zero shell-entity transfers, zero cascading LLC churn, zero pre-litigation assignment and a 2015 expiry, there is no NPE signal to score.

Caveats that could change this call: (a) Lonza's exact interest — owner vs. exclusive licensee — is unverified, and if Lonza held bare title the "original assignee still owns it" framing would need revision; (b) the correspondents of record are unretrieved, so the repeat-filer signal is genuinely untested rather than negative; (c) the 2006 PharmAthene/MPM entry should be resolved against the reel's face before the chain is treated as complete.

Verify at: https://assignmentcenter.uspto.gov/ (patent number 5770359; legacy index at https://assignment.uspto.gov/patent/index.html returns the same underlying records). Pull both reel/frame entries and capture the correspondent field for each, plus the patent list on the face of the 2006 MPM BioVentures III-QP, L.P. security-agreement reel — that single document resolves the only open question in this chain.

I'll search for the patent's front-page patent citations and cross-reference them against the authoritative record.

Let me pull the specific citation records for this patent.

Let me try to retrieve the front-page citation list directly.

Prior-Art Analysis Under 35 U.S.C. § 102 — U.S. Patent No. 5,770,359

Builds on the Prior Art, Obviousness, Litigation, PTAB, and Assignment sections above. Those sections already § 103-mapped the non-patent literature (Sanders & Wilson 1984; Hayward et al. 1986; Wilson 1982; Young & Ringold 1983; Subramani 1981). This section answers the specific instruction: the patent-document citations.

---

0. Retrieval status — read this before using anything below

I could not retrieve the granted front page's "References Cited" block for US 5,770,359. The authoritative full text supplied for this analysis begins at the Description heading and is truncated before both the front page (U.S. Patent Documents / Foreign Patent Documents / Other Publications) and the claims. No amount of re-querying in this session surfaced the front-page citation list itself; the searches returned forward citations (later documents citing the '359 patent), family members, and unrelated "359" hits.

Consequently I have not produced a verbatim transcription of "each patent citation" as printed on the '359 front page, because I do not have it. Per the operating rules, I state that rather than reconstruct it and pass it off as retrieved.

Two findings I can state with confidence, both grounded in the text I do have:

1. The '359 specification, as supplied, cites zero patent documents by number. Every reference embedded in the specification is non-patent literature: Wilson, Heredity 49:181 (1982); Young & Ringold, JBC 258:11260–11266 (1983); Sanders & Wilson, EMBO J. 3(1):65–71 (1984); Subramani, Mulligan & Berg, MCB 1:854–864 (1981); Emtage et al., PNAS-USA 80:3671–3675 (1983); Gubler & Hoffmann, Gene 25:263–269 (1983); Michelson & Orkin, JBC 257:14773–14782 (1982); Vieira & Messing, Gene 19:259–268 (1982); Johnson & Piskiewicz, BBA 827:439–446 (1985); Hayward et al., NAR 14:999–1008 (1986); Davis, Botstein & Roth, Advanced Bacterial Genetics (CSH, 1980); Huyhn, Young & Davis, DNA Cloning Techniques II (IRL Press, 1985); Maxam–Gilbert sequencing. No U.S. or foreign patent number appears anywhere in the supplied specification. That is itself a § 102 datum: the examiner's anticipation case, if any, was built on NPL, not on patent art.
2. The identifier is not unique — re-confirmed in this session. The number 5770359 also resolves to JP 5770359 B2 (特許第5770359号), 座標測定機 ("coordinate measuring machine"), Hexagon Technology Center GmbH (Heerbrugg, CH), Japanese application 特願2014-501574, PCT/EP2012/055394, registered 2015-07-03, published 2015-08-26. That document is not the subject of this report and I have not conflated it.

USPTO record confirmation for the correct patent. The grant record (USPTO grant data as mirrored at uspto.report/patent/grant/5770359 and Google Patents US5770359A) confirms: US 5,770,359 A, "Recombinant DNA sequences, vectors containing them and method for the use thereof," application 08/302,241, filed 1994-09-08, issued 1998-06-23, priority 1986-01-23, expired 2015-06-23. That is the patent analysed here.

---

1. The critical dates that govern every § 102 call

Pre-AIA § 102 controls (1986 priority; never AIA-subject).

Consequence: patent documents issued or published on or before 1986-01-23 are § 102(b) art outright. Patent documents issued after that date can only be § 102(a) (published before the invention date), § 102(e) (U.S. patent granted on an application filed before the applicant's date), or § 102(f)/(g) — and only § 102(e)/(f)/(g) art is disqualified by the pre-AIA § 103(c) common-ownership/joint-research-agreement carve-out available to Celltech + University of Glasgow. § 102(a) and § 102(b) art are not saved by § 103(c).

---

2. Documents that CITE the '359 patent (forward citations — not prior art)

These are the "citations for 5770359" the searches actually surfaced. All post-date the patent and therefore cannot be § 102 art. I list them because they define the '359 family's citation footprint and because several carry full bibliographic identity that double-checks the family.

Why none of these is § 102 art: each was published or filed years to decades after 1986-01-23. A reference cannot anticipate an earlier patent. These are forward citations and belong in an infringement/portfolio analysis, not a validity analysis.

One caution I must flag. The prior Litigation section records US 5,747,308 as co-assigned to the University of Glasgow alongside '359. US 5,747,308 did not appear in any citation I retrieved for '359, and I cannot confirm it is a citation on the '359 front page. I mention it only because it is a family-adjacent number that a reader may otherwise mistake for a '359 citation.

---

3. Family members — expressly not § 102 art

US 5,122,464 ('464)' is the parent in the § 120 chain (issued 1992-06-16 from 07/595,733, the application of which '359 is a continuation). US 5,827,739; 5,591,639; 5,658,759; 5,879,936; 5,891,693; 5,981,216 are siblings/cousins.

• A § 120 parent or sibling sharing the same disclosure is the same invention as to disclosure, so it cannot be "by others" under pre-AIA § 102(a)/(e)/(f), and § 102(b) is not triggered because the parent's issue/publication dates are measured against the shared 1987-01-23 filing — the first family member did not issue until 1992. For completeness: a U.S. patent is a § 102(b) reference only as of issue, and '464 issued 1992-06-16, well after the critical date.
• The PCT publication WO 87/04462 and EP 256055 A1 are the '359 family's own publications, not third-party art.

Net: the family contributes nothing to the § 102 case against '359.

---

4. Candidate patent prior art, reconstructed — clearly labelled

Because the front-page list is unretrieved, what follows is the patent-document art identified from the field and from the later patents' own background lists (notably EP 2 476 706 A2 §[0097] and the EAPO/RU background passages), i.e. the patents that any § 102/§ 103 analysis of this family must confront. These are candidate comparators, not "citations on the '359 front page." I have not verified each one's exact issue date against PatentCenter in this session — flagged in the "Verify" column.

The analytical point that matters. Not one patent document in the field anticipates any claim of '359, because the '359 claims contain a glutamine synthetase element that no pre-1986 patent document discloses. The DHFR/MTX co-amplification patents (A–E) establish the method — co-transformation, selective pressure, escalating inhibitor, co-amplification of a non-selected gene — but they are silent on GS. Anticipation under § 102 requires every element in a single reference (Net MoneyIN; MPEP 2131). These references fail that test on their face. Their entire force is § 103, and the Prior Art/Obviousness sections above already explain why that force is substantial (the '359 Background section concedes DHFR/MTX as "the most commonly used system," concedes its five deficiencies, and concedes that GS is amplifiable under Msx per Wilson 1982 / Young & Ringold 1983).

---

5. Where the real § 102 exposure sits — and it is not a patent

Ranked by actual anticipation risk:

1. Hayward et al., Nucleic Acids Research 14(2):999–1008 (26 January 1986) — non-patent. This is the closest prior art to '359 in the entire record: it reports the complete Chinese hamster GS cDNA (146 nt 5′UTR; 1119 bp coding; 108 bp 3′UTR) and the predicted amino-acid sequence. It is not § 102(b) art (it post-dates the 1986-01-23 critical date by three days) and, as to claims entitled to the 1986-01-23 GB priority, it is antedated under pre-AIA § 119. It becomes § 102(a)/(b) anticipatory art only if a given claim loses the 1986-01-23 priority — in which case the § 102 analysis is over for that claim and § 103 is unnecessary. (Date corrected in the Obviousness section against the Oxford Academic record; the Prior Art section's "one day before" is wrong.)
2. Sanders & Wilson, EMBO J. 3(1):65–71 (January 1984) — non-patent, unambiguously § 102(b) art. Discloses an 8.2 kb BglII genomic fragment from Msx-resistant CHO KG1MS cells (clone pGS-1) that hybrid-selects GS mRNA translating to a 42,000 MW polypeptide co-migrating with purified GS. Cannot anticipate any '359 claim, because it discloses no complete gene and no nucleotide sequence — the specification expressly distinguishes it on exactly that ground ("this fragment does not appear to contain a complete GS gene and it was not sequenced"). § 103 art.
3. Wilson, Heredity 49:181 (1982); Young & Ringold, JBC 258:11260–11266 (1983) — non-patent, § 102(b). Endogenous GS amplification under Msx selection. No recombinant DNA. § 103 art.
4. Subramani, Mulligan & Berg, MCB 1:854–864 (1981) (pSV2 cassette) — non-patent, § 102(b). § 103 art (the patent's own pSV2.GS is a literal substitution of GS for dhfr).

Bottom line on § 102: on the material I could retrieve, no patent document anticipates any claim of US 5,770,359. The only credible § 102 reference is a journal article — Hayward et al. 1986 — and it applies only if the 1986-01-23 priority fails for the claim at issue. Everything else, patent or otherwise, is § 103 material. This is consistent with, and does not contradict, the Obviousness section.

---

6. Explicit uncertainties and required verification

• The front-page "References Cited" list of US 5,770,359 was NOT retrieved. This is the single largest gap in the analysis and the direct answer to the instruction. It must be pulled from USPTO PatentCenter (https://patentcenter.uspto.gov) or the grant PDF (https://patentimages.storage.googleapis.com/…/US5770359.pdf) before any statement about "each citation on the front page" is made. Until then, § 4 above is a reconstructed candidate set, not a citation list.
• Issue dates for items A–E are flagged "verify" and were not confirmed at the USPTO in this session. Do not rely on them without checking.
• The '359 claims 1, 3, 5, 7 remain inferred, not quoted (the supplied text truncates before the claims). The § 102 mapping in § 4 is therefore framed as "element-absent" rather than claim-by-claim verbatim.
• I did not find and do not assert any pre-1986 patent disclosing a complete mammalian GS cDNA. The absence of such a patent is a point for the patentee on anticipation.
• Identifier discipline maintained: every "5770359" accepted here as relevant is US 5,770,359. JP 5770359 B2 (Hexagon, coordinate measuring machine) is expressly excluded.

Verification actions, in priority order: (1) pull the '359 grant front page and transcribe the "U.S. Patent Documents" and "Foreign Patent Documents" columns; (2) pull the examiner's Reasons for Allowance to learn whether Sanders & Wilson or Hayward was actually applied and on what reasoning; (3) confirm the issue dates and § 120 lineage of items A–E at PatentCenter.

Sources for new material in this section: Google Patents US5770359A (bibliographic, family, event feed); uspto.report/patent/grant/5770359 (mirror of USPTO grant data); EP 1 680 141 B1 §[0062]; EP 3 219 328 A1 §[0077]; EP 2 848 687 A1 search report (Y-category grouping of JP 63-502955 A / US 5770359 A / EP 256055 A1); EP 2 476 706 A2 §[0097] (DHFR + GS marker-gene list); RU 2 532 826 C2, RU 2 537 264 C2, RU 2 693 084 C2, RU 2 829 846 C2, EA 007617 B1, EA 015262 B1, EAPO 040340 (background marker-gene citations of 5,122,464; 5,770,359; 5,773,739-as-5,827,739); JP5770359B2 (Hexagon) — retrieved to confirm the non-US "5770359" is a distinct, unrelated document.

I'll ground the obviousness analysis with a few targeted checks before writing, since § 103 turns on what a skilled artisan knew as of the critical dates.

Obviousness Analysis Under 35 U.S.C. § 103 — U.S. Patent No. 5,770,359

Built on the Prior Art, Litigation, PTAB, and Assignment sections above. I do not repeat their content except where a finding must be corrected.

---

0. A material correction to the Prior Art section — flagged, not silently reconciled

The Prior Art section states Hayward et al., NAR 14:999–1008, published 22 January 1986 — "one day before" the 1986-01-23 priority date. Live retrieval contradicts this. Oxford Academic's own catalogue record for the article returns:

"The cloning and nucleotide sequence of cDNA for an amplified glutamine synthetase gene from the Chinese hamster … Nucleic Acids Research, Volume 14, Issue 2, 26 January 1986, Pages 999–1008"
— https://academic.oup.com/narjournals/search-results?f_Authors=Richard+H.+Wilson (and https://academic.oup.com/nar/article-pdf/14/2/999/[7045479](/patent/7045479)/14-2-999.pdf)

The article's own byline, from the PDF, reads: "Received 23 October 1985; Revised and Accepted 17 December 1985."

This changes the § 103 analysis materially and in the patentee's favour. If the printed publication date is 26 January 1986, then Hayward et al. is:

• not § 102(b) art — the critical date is 1986-01-23 (one year before the 1987-01-23 U.S. filing, measured from the U.S. filing date; a foreign § 119 priority date does not extend the § 102(b) grace period); the reference falls 3 days inside the grace period; and
• not § 102(a) art as to claims entitled to the 1986-01-23 GB priority, because pre-AIA § 119 priority is available to antedate a § 102(a) publication that post-dates the foreign filing date.

So Hayward is the reference that looks like a naked anticipation hit and, on the corrected date, is not prior art at all for priority-entitled claims. I proceed on that basis, but I keep Hayward in play throughout as the priority-failure reference, because if a claim cannot carry the 1986 date, Hayward becomes anticipatory (§ 102) rather than merely combinable (§ 103) — and no obviousness analysis is even needed for that claim.

Verified claim text available for this section (from the Prior Art section's retrieval): claims 8, 9, 10, 11, 12. Claims 1, 3, 5, 7 remain inferred from the specification's "aspects," not quoted. I mark that wherever it matters.

---

1. Governing law and the legal frame

• Pre-AIA § 103(a) applies. The application was filed 1994-09-08 with a 1986-01-23 priority chain and was never subject to AIA first-inventor-to-file § 103. Prior art for § 103 purposes is the pre-AIA § 102(a), (b), (e), (f), (g) universe.
• KSR Int'l Co. v. Teleflex Inc., 550 U.S. 398 (2007) governs the motivation inquiry: a combination is obvious where the improvement was within the "ordinary creativity" of a skilled artisan, and where the field presents "a finite number of identified, predictable solutions" the "obvious to try" rationale applies. The teaching–suggestion–motivation test survives as one permissible rationale, not the exclusive one.
• Pre-AIA § 103(c) (common ownership / joint research agreement) disqualifies only § 102(e), (f), and (g) art. It does not reach § 102(a) or (b) art — so if Hayward or Bebbington & Hentschel 1987 were § 102(a) art, the Celltech/Glasgow common ownership would not save those claims.
• The patent's own specification is an evidentiary admission of the prior art. Under In re Nomiya and MPEP 2129, the Background section's recitals — that DHFR/MTX is "the most commonly used system," that Wilson 1982 and Young & Ringold 1983 obtained "extensively amplified GS genes" in Msx-resistant cell variants, and that "GS from all sources is subject to inhibition by a variety of inhibitors, for example methionine sulphoximine (Msx)" — are admissions usable as § 103 prior art irrespective of citation on the front page.

---

2. Person having ordinary skill in the art, as of 23 January 1986

A POSA was a molecular biologist with an M.S./Ph.D. plus 2–5 years of mammalian expression-vector experience, familiar with: (i) the SV40-based pSV2 expression cassettes (Subramani 1981); (ii) λgt10/pBR322/pUC9 cDNA cloning and the RNase H double-stranded cDNA protocol (Gubler & Hoffmann 1983; Michelson & Orkin 1982; Vieira & Messing 1982); (iii) the DHFR/MTX co-amplification paradigm, including co-transformation with two vectors and selection in escalating drug concentrations; and (iv) the reported amplification of endogenous GS in Msx-resistant rodent cell lines (Wilson 1982; Young & Ringold 1983; Sanders & Wilson 1984). This is a high skill level in a crowded, fast-moving art — relevant because it narrows what counts as "unexpected."

---

3. What each reference teaches, and what it lacks

Two observations of consequence.

1. Sandwiched at 42,000 MW. Sanders & Wilson 1984 established the biochemical identity of the clone (a 42 kD polypeptide co-migrating with purified GS), and the '359 specification computes the CHO GS polypeptide at 41,964 — i.e. the patent's own molecular-weight confirmation of the same protein. There is no dispute in the art as to which protein was being cloned; the only question was reading out its sequence.
2. The patent's own best clone is not full-length. The specification concedes pGSC45 "only lacks six or seven nucleotides from the 5′ end of the mRNA," and that those bases sit in the 5′ untranslated region. That is fatal to any reading of the "complete" limitation as requiring a full-length mRNA: claims 1 and 9 must be read as requiring a complete amino acid coding sequence — exactly what Hayward reports (1119 bp), and exactly what a routine cDNA clone off the Sanders & Wilson probe would deliver.

---

4. Combinations that would render the claims obvious

I set out five combinations. For each: the references, the claim(s) reached, and the reason a POSA would have combined them.

Combination A — "Known gene into known vector" (claims 3, 8, 10; and 1/9 if priority fails)

References: Hayward 1986 (or Sanders & Wilson 1984 + routine cDNA cloning) + Subramani 1981 (pSV2) + Emtage 1983 (pCT54).

Mapping. Claim 10 depends on claim 3 (vector). The specification's own construction of pSV2.GS is literally a substitution: the "1.2 kb NaeI-PvuII fragment of λgs 1.1 was cloned in place of dhfr sequences in pSV2." The claim-8 element of "linkage such that amplification of the GS coding sequence results in co-amplification of the desired protein coding DNA" is achieved by cloning both transcription units into one plasmid — routine.

Motivation (KSR rationales (A) and (B)); MPEP 2144.01: combining prior art elements according to known methods to yield predictable results; and simple substitution of one known element (a GS coding sequence) for another (the dhfr cDNA) in a known vector (pSV2) to obtain a predictable result (expression of GS under SV40 control). The patent itself demonstrates the substitution produced a functional construct, which is the paradigm of a predictable result.

Combination B — "Alternative amplifiable marker" (claims 5 and 11; the co-amplification method)

References: the admitted DHFR/MTX co-amplification system + Wilson 1982 + Young & Ringold 1983 + Sanders & Wilson 1984.

Mapping. Claim 11 (dependent on claim 5) claims "culturing the transformed host cell in media containing a GS inhibitor and selecting for transformed cells which are resistant to progressively increased levels of the GS inhibitor."

Motivation (KSR rationales (C), (D), (E)). This is the strongest combination in the record, because the motivation comes almost entirely from the patent's own Background section:

• DHFR/MTX was "the most commonly used system" for co-amplification — i.e., the method of co-amplification by escalating inhibitor concentration was squarely known and conceded.
• The specification enumerates the known deficiencies of that system: it requires a host cell lacking the enzyme (with "only one cell line known which lacks the gene encoding DEFR"); dominant DHFR markers demand "such high concentrations of MTX" that "it has not been possible to achieve very high copy numbers"; resistance "can arise from mechanisms other than amplification" (mutant DHFR with lowered MTX affinity); and "MTX is a potential carcinogen." A specification that recites a recognized problem in the art is a roadmap for the asserted improvement — the classic "known technique applied to a known device ready for improvement" rationale.
• Wilson 1982 and Young & Ringold 1983 already disclosed that GS is amplifiable under Msx selection. A POSA therefore had before them a known, selectable, amplifiable enzyme/drug pair — GS/Msx — with no need to find a null host cell (GS is a housekeeping enzyme present in all cells, so an exogenous GS vector is by construction a dominant marker). That solves the very deficiency the patent identifies.
• Finite number of identified solutions (rationale (E)): the art of 1986 offered a short, well-known menu of amplifiable marker/inhibitor pairs — DHFR/MTX, GS/Msx, and adenosine deaminase/deoxycoformycin. Selecting GS/Msx and applying the known escalation protocol is "obvious to try" under KSR.

Combination C — "Selection of the inhibitor" (claim 12)

References: Combination B + the admitted fact that "GS from all sources is subject to inhibition by a variety of inhibitors, for example methionine sulphoximine (Msx)," + the known phosphinothricin/bialaphos class of GS inhibitors.

Mapping. Claim 12 recites selection of the GS inhibitor "from the group consisting of phosphinothricin and methionine sulphoxime." (Reproduced literally; the standard spelling is methionine sulphoximine.)

Motivation: where the art discloses a class of inhibitors for the same enzyme and the patent claims a sub-genus of two, the burden falls on the patentee to show a critical, unexpected difference between the recited members and the genus. Msx is not merely within the genus — it is the species the specification names, and the species the prior art (Wilson 1982; Young & Ringold 1983; Sanders & Wilson 1984) used. Listing the same inhibitor the prior art used cannot confer patentability. Claim 12 is the most exposed claim in the retrieved set.

Combination D — "Two-vector co-transformation" (the co-transformation embodiment of claim 5)

References: the admitted DHFR/MTX practice of co-transforming "a vector comprising a DNA sequence encoding DEFR and a vector comprising a DNA sequence encoding the desired protein" + Combination B.

Motivation: the specification concedes co-transformation is the existing practice. Applying an established, admitted protocol to a newly selected marker is a change in a single variable with a predictable result. Note that claim 8 is drafted to the single-vector case ("an expression vector … comprising (a) … and (b) …"), which is a narrowing relative to the admitted two-vector art — a narrowing that does not add patentable weight absent an unexpected result.

Combination E — "pH/Host and post-translational down-regulation" (the regulatable-promoter and glutamine-suppression features)

References: Hayward/Subramani + the admitted post-translational regulation of GS by glutamine "via ADP ribosylation" and by glucocorticoids/cAMP (recited in the specification as known biology) + known regulatable promoters (heat-shock, metallothionein).

Motivation: the specification asserts that the GS gene is regulatable and that down-regulation reduces selection pressure on the amplified locus. But the mechanism it relies on — post-translational suppression of GS activity by glutamine — is recited as known in the same paragraph of the specification, and regulatable promoters were standard. Placing a known gene under a known regulatable promoter is routine.

---

5. Which claims are exposed, and by how much

---

6. The nonobviousness case the patentee would press (and where it bites)

An honest § 103 analysis must identify what would defeat these combinations, because two of them have real force.

1. The "complete sequence" difference is not merely a matter of degree — if priority is upheld. If Hayward is antedated, the only pre-1986 reference of substance is Sanders & Wilson 1984, which the patent correctly characterises as disclosing no complete gene and no sequence. The patentee's position: obtaining the complete, sequencable GS cDNA in 1985 was not a merely routine application of a known probe, because the amplified genomic clone contained introns and the mRNA species were multiple. Rebuttal available to a challenger: Hayward et al.'s own contemporaneous statement that "the cDNA cloning strategy was straightforward considering that the mRNA was obtained from a relatively abundant source (Msx amplified CHO cells) and that plasmid subclones of λ phage GS gene recombinants were available as probes." That is the inventors' admission that the cloning step was conventional.
2. Unexpectedly high copy number. The specification reports that pSVLGS.1 reached approximately 10,000 copies per cell in cells resistant to 2 mM Msx — versus only 5–10 copies per cell for pSV2.GS. Under In re Soni, an unexpectedly superior result can rebut a prima facie case. This is the patentee's best argument, and it is a strong one for the pSVLGS.1-type construct specifically — but it is a weak one for the genus claims (1, 3, 5, 9), which are not limited to the minigene/SV40-late-promoter architecture. A claim that reads on both the poor performer (pSV2.GS) and the good performer (pSVLGS.1) cannot rely on the good performer's numbers alone.
3. Unexpected selectivity for the vector-derived gene. The specification asserts that Msx selection "leads almost exclusively to the isolation of clones in which the vector-derived GS gene has been amplified in preference to the endogeneous gene," and that "no detectable endogenous gene amplification occurred" with pSVLGS.1. This is genuinely in tension with Sanders & Wilson 1984, which taught amplification of the endogenous gene. Whether that is a difference in kind or an artefact of the specific construct (high-copy episomal/integrated SV40-late-driven vector) is the battleground.
4. Teaching away? None. Sanders & Wilson does not disparage recombinant GS, and the DHFR/MTX literature positively encouraged alternative amplifiable markers. There is no teaching-away defense.
5. Secondary considerations. The earlier Assignment section records Lonza's GS System as licensed to over 70 companies — a commercial-success and licensing-acceptance argument that would be offered as objective evidence of nonobviousness. Its weight depends on nexus to the claimed subject matter, which is difficult here because the licensed platform (pSVLGS.1-type constructs) is narrower than the genus claims.
6. Procedural shields the patentee could raise: (a) § 119 priority to 1986-01-23 antedates Hayward; (b) the inventor's-own-work doctrine — Hayward's author list (Wilson, Hussain, Wilson, Lyons, Woodcock, McIntosh, Harris) includes co-inventor Wilson and Celltech personnel, so a challenger must establish it is "by others" as to the non-overlapping inventive entity; (c) pre-AIA § 103(c) common ownership/joint research agreement as to any § 102(e)–(g) art in the Celltech/Glasgow orbit.

---

7. Bottom line

The single most probative § 103 fact is that the motivation for every combination comes out of the patent's own specification. The Background section concedes the DHFR/MTX co-amplification method, concedes its five specific shortcomings, and concedes that GS is amplifiable under Msx. That supplies the KSR "known technique, known device ready for improvement" rationale without any need to stretch a secondary reference.

Ranked, the obviousness exposure is:

1. Claim 12 — highest. Msx, the recited inhibitor, is the inhibitor the prior art used.
2. Claims 3, 8, 10 — high. "Known gene into known vector"; the specification's own pSV2.GS construction is a literal substitution of the GS fragment for dhfr in Subramani's pSV2.
3. Claims 5, 11 — high. Admitted co-amplification protocol, applied to a known amplifiable enzyme/drug pair, to remedy admitted defects.
4. Claims 1, 9 — high if 1986 priority fails; moderate otherwise. If priority fails, Hayward et al. 1986 is anticipatory and no § 103 analysis is needed. If priority holds, the attack must run through Sanders & Wilson 1984 + routine cDNA cloning, supported by the Hayward authors' own "straightforward" admission.
5. The pSVLGS.1-specific achievement (~10,000 copies/cell in two rounds) — the patentee's strongest nonobviousness ground, but effective only against claims actually limited to that architecture, which the retrieved claims are not.

This is not an invalidity opinion. It is an identification of the § 103 theories available on the applicant-cited record, with their relative strength.

---

8. Open items a challenger or defender must close

• Priority support (dispositive). Whether each claim is supported by the GB priority application filed 1986-01-23 and by PCT/GB87/00039. This single question determines whether Hayward et al. is (i) irrelevant, (ii) § 102(a) art, or (iii) § 102 anticipatory art. The GB application number and its disclosure were not retrieved in this session.
• Claim 1, 3, 5, 7 verbatim text. Still unverified. All product/method mapping above for those claims is inferred from the specification's "aspects." Pull from USPTO PatentCenter before relying on it.
• The examiner's front-page "U.S. Patent Documents" list and the Reasons for Allowance. Not retrieved. Whether any cited reference was applied under § 102 or § 103, and on what reasoning, is unknown. If the examiner distinguished Sanders & Wilson on the "complete/unsequenced" ground, that reasoning is directly available to the patentee under In re Nomiya as evidence of nonobviousness.
• Pre-1986 publication date for phosphinothricin as a GS inhibitor. Claim 12 names it, but I did not verify a pre-1986 printed publication establishing it as a known GS inhibitor. Do not assume it. The Msx leg of claim 12 is solidly supported by the specification's own admission; the phosphinothricin leg is not yet grounded.
• Any additional pre-1986 GS literature. My searches surfaced the post-1986 human GS cDNA (Gibbs, Campbell & Wilson, NAR 15:6293, 11 August 1987) and rat GS mRNA (van de Zande et al., NAR 16:7726, 1988) — both post-date the critical dates and are not prior art. I found no pre-1986 non-Celltech/Glasgow complete mammalian GS cDNA publication, which is itself a point for the patentee on the state of the art.

Sources for the new material in this section: Oxford Academic catalogue and PDF record for Hayward et al., NAR 14(2):999–1008 (26 January 1986; received 23 October 1985, accepted 17 December 1985), https://academic.oup.com/narjournals/search-results?f_Authors=Richard+H.+Wilson and https://academic.oup.com/nar/article-pdf/14/2/999/7045479/14-2-999.pdf; Scilit record of Sanders & Wilson, EMBO J. 3(1):65–71 (1984) abstract; Scopus/KISTI and Utah State holdings records confirming Bebbington & Hentschel (1987) DNA Cloning Vol. III:163–188, IRL Press/Academic Press; PubMed records for van de Zande et al. (1988) and van den Hoff et al. (1991) confirming their post-critical-date status. All other reference dates are transcribed from the specification text supplied to me.