1983 Calgene's Luca Comai Files US4535060A — A Glyphosate-Resistant EPSP Synthetase, 13 Years Before Roundup Ready Soy and the First-Generation IP of Herbicide-Tolerant GMOs

Bottom line first

On January 5, 1983, Luca Comai — a plant molecular biologist at the Davis-based startup Calgene LLC — filed US patent US4535060A as sole inventor. The claim covers a culture of cells expressing an EPSPS variant whose glyphosate inhibition is less than about half that of the wild type, generated by in vitro mutation at the aroA structural-gene locus.

Granted on August 13, 1985. The commercial Roundup Ready soybean Monsanto launched in 1996 used a different lineage — soil-bacterium CP4 EPSPS (Agrobacterium sp.), covered by separate patent US5633435A — so Comai's aroA mutant is not the direct ancestor of commercial Roundup Ready. The two are parallel solutions to the same biochemical problem. After Monsanto's full acquisition of Calgene in 1997, this patent passed into the Monsanto portfolio.

Together with Day 14 ep55 — Flavr Savr patent US4801540A (priority 1986, Hiatt and four co-inventors) — this means Calgene as a single company drafted the first-generation IP for herbicide-tolerant GMOs (1983) and for commercial GMO food (1986) within a three-year window. This note reads US4535060A as the entry point of that two-shot, alongside the candidates.tsv corrections it forces.

1. How the candidate was selected (a reproducible pipeline)

[STEP 1] From candidates.tsv FH-001..012, pick a Week 3 closing item;
         FH-008/010/011/012 are the four remaining.
         FH-011 (Roundup-tolerant GMO, US4535060) is priority 12 and pairs
         with Day 14 ep55, so it becomes the note candidate.
[STEP 2] Confirm the DB-listed URL https://patents.google.com/patent/US4535060.
[STEP 3] WebFetch on Google Patents for Claim 1, inventors, assignees, dates.
[STEP 4] Compare DB ("Monsanto Company (now BayerCropScience); filed around 1986")
         against the primary source.
         → Inventor: Luca Comai (sole) — Monsanto attribution is wrong.
         → Original assignee: Calgene LLC; Monsanto succession came after the
           1997 acquisition.
         → Filing date: 1983-01-05 — DB "around 1986" is wrong.
         → Grant date: 1985-08-13.
[STEP 5] Check the Calgene two-shot structure with Day 14 ep55 (US4801540A,
         priority 1986, Calgene assignee).
[STEP 6] Cross-check whether the commercial Roundup Ready soy lineage (1996
         approval) descends from the Comai patent, via secondary sources
         (USDA APHIS petitions, Wikipedia).
         → Commercial soy uses CP4 EPSPS (Agrobacterium sp.), a different
           lineage covered by US5633435A — Comai's patent is not the direct
           ancestor of commercial Roundup Ready.

Selection rationale: (a) it is the entry of Calgene's first-generation IP; (b) within the Day 8–14 DB-correction streak, this case has the right number but wrong inventor and year, a distinct correction pattern; (c) the "Comai mutant vs. CP4 EPSPS" split is a useful design-divergence subject; (d) reading it together with Day 14 ep55 lets one see Calgene's IP architecture in a single session.

2. Claim 1 and the core of the specification

Claim 1 retrieved verbatim from Google Patents:

A culture of cells consisting essentially of cells containing glyphosate resistant 5-enolpyruvyl-3-phosphoshikimate synthetase having at least one mutation originating from in vitro mutation in the aro locus for the structural gene expressing said synthetase, wherein glyphosate inhibition of said synthetase containing said mutation is less than about one-half of the synthetase of the original strain prior to mutation.

Three points matter:

1. The subject is a "culture of cells," not a transgenic plant or commercial cultivar. Commercial-crop coverage is left to follow-on patents.
2. The mutation must originate from "in vitro mutation in the aro locus" — random mutagenesis followed by selection in glyphosate-containing media. aroA is the structural gene for EPSPS.
3. The numerical threshold is "less than about one-half" of wild-type inhibition. This is the patent-range setpoint and the point of contention with later resistance-enzyme patents.

From the abstract:

The patent describes enhanced glyphosate resistance achieved by introducing a mutated aroA gene expressing a modified enzyme in the shikimic acid pathway. This mutated enzyme maintains function "in the presence of significant amounts of glyphosate" while wild-type versions are inhibited.

The shikimate pathway (aromatic amino-acid synthesis: phenylalanine, tyrosine, tryptophan) is present in plants, bacteria, and fungi but not in animals. Glyphosate blocks EPSPS, severs the pathway, and starves the plant of aromatic amino acids — the biochemical basis of Roundup's selective toxicity, since animals lack the pathway entirely.

3. Why it feels uncomfortably close

"Same" rows assert design-ethos continuity; "similar" rows share problem framing while diverging in implementation; "metaphor" rows are cross-era genealogies, technically distinct. The most important correction here is that the direct ancestor of commercial Roundup Ready is not this patent but the CP4 EPSPS line (US5633435A in the 1990s). Conflating the two manufactures a false genealogy.

4. Why Calgene captured first-generation GMO IP — speculation vs. confirmed facts

What primary sources confirm:

• 1983-01-05: US4535060A filed (Calgene; Comai sole inventor)
• 1985-08-13: Granted
• 1986-10-17: US4801540A priority filing (Calgene; Hiatt + four co-inventors; Flavr Savr precursor)
• 1989-01-31: US4801540A granted
• 1994-05-21: FDA clears Flavr Savr (world's first commercial GMO food)
• 1996-05: Monsanto launches Roundup Ready soy (CP4 EPSPS lineage, separate patent)
• 1997: Monsanto fully acquires Calgene; this patent and the rest of Calgene IP transfer

Author's inferences (not directly verified):

• Comai's path to the aroA mutant was open at this exact moment because (a) Steinrücken & Amrhein, 1980, Naturwissenschaften identified EPSPS as glyphosate's target; (b) the E. coli aroA gene was cloned in 1981 (Pittard et al.); (c) glyphosate-resistant E. coli selection systems were established by 1981 (Comai et al., 1983, Science). Three prerequisites converged just before the filing.
• Monsanto's choice of CP4 EPSPS (bacterial origin) over Comai's mutant likely combined (i) avoiding licensing of the Comai claim and (ii) the biochemical advantage that a naturally tolerant bacterial enzyme tolerates higher glyphosate doses than a plant-EPSPS variant. Primary sources: Padgette et al., 1995, Crop Science, and the claim language of US5633435A.
• The 1997 Calgene acquisition was likely targeted at Calgene's full transformation know-how (including the Flavr Savr platform), not this patent in isolation. Primary source: Monsanto's 1997 SEC 10-K.

5. Why this matters for AI Archaeology

The archaeological point is the divergence pattern: the same problem (glyphosate tolerance) gets two solutions in parallel (Comai's mutant vs. CP4 EPSPS), and the later one wins commercially. The 1983 IP holder (Calgene) merges with the 1996 market winner (Monsanto) in 1997 — but the technology that shipped is from a different patent line. A double-layered structure that's easy to flatten in retrospective accounts.

For LLM rereads, the interesting move is Claim 1's numerical threshold "about one-half." A 1980s drafting style: cut the range with a number. Modern GMO claims often use functional definitions (function-defined claims), and numerical-range claims have faced tightening examination since Mayo v. Prometheus (2012) and Alice (2014). A useful contrast for showing how claim drafting differs across eras.

There is also the discrepancy between the popular "Comai 1983 = Roundup Ready ancestor" reading and the primary-source reality. Many general-audience sources (including parts of the Roundup Ready Wikipedia article) blur the two. Separating them by patent number and molecular biology — as this note does — gives a cleaner ancestry for commercial GMO soy IP.

6. Pitfalls (specific to GMO patents)

Pitfall 1: don't conflate "first-generation IP" with "commercial Roundup Ready ancestor" US4535060A is the entry point of first-generation herbicide-tolerant GMO IP, but not the technology Roundup Ready soy actually used in 1996. Commercial soy uses CP4 EPSPS (Agrobacterium sp., separate patent US5633435A). Conflating them creates a false "Comai invented Roundup Ready" lineage. This note flags the divergence explicitly.

Pitfall 2: candidates.tsv "Monsanto Company; filed around 1986" is wrong on multiple counts

• Inventor: actually Luca Comai (sole). Monsanto is a successor assignee post-1997 acquisition; the original assignee is Calgene LLC.
• Filing year: actually 1983-01-05. The "around 1986" date matches Calgene's Flavr Savr precursor (FH-009), a different lineage.

This is a notable correction case in the Day 8–14 streak.

Pitfall 3: avoid the GMO=safe / GMO=unsafe dichotomy This note covers the patent specification and IP structure. GMO-food safety is adjudicated through different processes (FDA, USDA APHIS, EFSA reviews); patent issuance is not a safety certification. This note does not litigate that debate.

Pitfall 4: caveats around "low animal toxicity" of the shikimate pathway The standard explanation that glyphosate is low-toxicity to animals because they lack the shikimate pathway has limits: (a) animal gut microbiota do carry the pathway; (b) long-term effects beyond acute toxicity remain in scientific dispute (IARC 2015 classification "Group 2A — probably carcinogenic to humans"; EFSA and EPA assessments diverge). This note does not adjudicate that debate either; it limits itself to the patent's biochemical design.

To be precise

Confirmed facts:

• US4535060A "Inhibition resistant 5-enolpyruvyl-3-phosphoshikimate synthetase, production and use"; sole inventor Luca Comai; original assignee Calgene LLC (Davis, California). Claim 1, abstract, inventors, and assignees retrieved from Google Patents (https://patents.google.com/patent/US4535060A/en).
• Filed 1983-01-05; granted 1985-08-13.
• Current assignees Monsanto Co. and H&Q Ventures III A CA LP per Google Patents Assignment tab — succession after the 1997 Calgene acquisition.
• The "about one-half" numerical threshold and the "in vitro mutation in the aro locus" limitation are quoted verbatim from Claim 1.

Author's interpretation:

• The "commercial Roundup Ready ancestor is not this patent" reading is built from a comparison with US5633435A (CP4 EPSPS); Calgene/Monsanto have not, as far as I have seen, drawn that line publicly in those terms.
• The "Calgene two-shot (1983 GMO crops origin / 1986 commercial GMO food origin)" framing is editorial.
• The "two parallel solutions (Comai mutant vs. CP4 EPSPS)" framing is editorial; it is not part of any standard patent-history taxonomy.

Metaphor / analogy:

• "aroA mutant ↔ CP4 EPSPS" is a metaphor; the designs are technically distinct.
• "Calgene LLC ↔ Bayer Crop Science" is a metaphor for the bio-startup consolidation arc.

Open items:

• Forward-citation count for this patent and its links to subsequent GMO patents.
• License value and IP appraisal at the 1997 Monsanto acquisition.
• Verbatim Claim 1 of US5633435A (CP4 EPSPS) and the overlap with the Comai claim.
• Verbatim text of Padgette et al., 1995, Crop Science.
• USDA APHIS Roundup Ready soy approval document (1994 Petition).
• Calgene's 1997 SEC 10-K disclosures.
• The full structure of Monsanto's Roundup Ready IP portfolio.

Where the comparison breaks:

• The 1983 composition-and-use claim and the post-2010s genome-editing regulatory frame (post-Mayo / post-Alice patent-eligibility debate) sit in fundamentally different regimes.
• Calling this "first-generation herbicide-tolerant GMO IP" is a cross-era summary; Comai's contemporaneous intent was much narrower.
• The cause of glyphosate-resistant superweed spread cannot be attributed to this patent alone; it is a property of agricultural practice at scale.

References:

• Original patent: US4535060 on Google Patents
• Food & Health Patents Note #4: Aspartame US3492131A
• Food & Health Patents Memo #6: Flavr Savr US4801540A
• Food & Health Patents Memo #7: Margarine US146012A
• Wikipedia: Genetically modified crops (https://en.wikipedia.org/wiki/Genetically_modified_crops)
• Wikipedia: Calgene (https://en.wikipedia.org/wiki/Calgene)