Skip to content

Medicinal Mushroom Complement Track — Phase 7 Platform Expansion

Why this page exists

Open Enzyme's koji thesis is "engineered A. oryzae secretes therapeutic enzymes via home-fermentable food." The koji chassis is uniquely positioned for that role — decades of ascomycete secretion-biology toolkit (KEX2 protease handling, glucoamylase fusion machinery, CRISPR works, signal-peptide engineering precedent). It is the right engineering chassis.

But comp-014 (medicinal mushroom × chokepoint mapping) surfaced a different category of leverage: medicinal mushrooms are native producers of compounds that don't lend themselves to koji engineering — polysaccharide-peptide hybrids (GLPP), nucleoside analogs (cordycepin), thiol antioxidants (ergothioneine), terpenoid-class secondary metabolites. These don't fit the "express recombinant protein in koji" paradigm; they fit a "cultivate the producer organism, extract the compound" paradigm.

Brian's 2026-05-06 inversion was the right one: not "find a better engineering chassis than koji" (basidiomycete genetics is 5-10× harder than ascomycete; engineering Pleurotus / Lentinula / Hericium / Ganoderma for protein secretion is a multi-year toolkit-building project before first construct expresses) but "are there other home-cultivable fungi whose native compound profile complements what koji produces?" Yes.

This page formalizes the medicinal mushroom complement track as a peer track to the koji engineering track under the broader Open Enzyme platform thesis. Same pattern as engineered-lbp-chassis.md (engineered LBPs as obligate-anaerobe chassis sibling to koji) and sirna-urat1-modality.md (kidney-tropic siRNA as discovery-engine output sibling to fermentable enzymes). All three peer tracks expand Open Enzyme from "engineered enzymes in koji" to "solve gout, every avenue, fully open."

Platform thesis expansion

Track Chassis Engineering effort Therapeutic class Consumption UX
Engineered koji (existing) A. oryzae Heavy — recombinant cassette + secretion engineering Therapeutic enzymes (uricase, lactoferrin, DAF SCR1-4, koji-secreted complement regulators) Shio-koji / amazake / miso (daily food condiments)
Engineered LBPs (peer) F. prausnitzii / Akkermansia Heavy — anaerobe engineering, LBP regulatory path Live therapeutic with sustained colonic activity (butyrate, IL-22, anti-complement) Refrigerated capsule / reconstituted suspension (commercial pharma)
siRNA / kidney-tropic discovery (peer) n/a (synthetic) Heavy — sequence design, conjugate chemistry, delivery Sequence-specific knockdown (URAT1) Subcutaneous injection (commercial pharma)
Medicinal mushroom complement (THIS page) G. lucidum, C. militaris, Pleurotus, Lentinula, Hericium, G. applanatum, Inonotus Light — strain selection + cultivation optimization + extract characterization. NO genetic engineering. Native-compound supplements (GLPP, cordycepin, ergothioneine, eritadenine, erinacines, betulinic acid derivatives) Dried fruiting body / tincture / broth / powder (home cultivation OR consumer supplement)

The fourth track is the lightest engineering effort and the most accessible UX — but it covers a chemistry space koji cannot reach.

What goes on this track vs. the koji track

Routed to the medicinal-mushroom-complement track (per comp-014 Phase 6):

  • GLPP polysaccharide-peptide (Ganoderma lucidum) — ADA + GLUT9 + OAT1 chokepoints, 40.6% UA reduction in HUA mice. Not a koji-engineering target; biosynthesis is mycelium-specific. Native cultivation is the route. MW resolution (grep-verify gate, 2026-05-06): The apparent 520 vs. 37 kDa "discrepancy" is a fractionation-stage difference, not an inconsistency. 520 kDa is the bulk crude polysaccharide-peptide preparation from the Juncao National Engineering Research Center — sister paper PMC11351902 (Zhang 2024) §2.1 explicitly states "the average molecular weight of GLPP was approximately 520 kDa" for the same Juncao prep. 31–42 kDa are the post-DEAE-fractionation sub-fractions — independently verified at 37,121 Da (GL-PP) and 31,130 Da (GL-PP2) per PMID 37852403 and 42,635 Da per PMID 29541200. The Lin 2022 HUA paper itself (PMID 36385640, paywalled body) does not specify which sub-fraction it used; bulk-prep most likely based on sister-paper consistency. SOP-1 SEC-MALS at Tier 3 remains non-negotiable to determine which fraction the HUA mechanism load-bears on.

  • Cordycepin / whole Cordyceps militaris extract — URAT1 chokepoint, animal model SUA 337→203 µmol/L for purified cordycepin. Phase 5b CNKI remediation upgrade (2026-05-20): Xiong 2024 Biotechnology Bulletin tested whole C. militaris water extract in potassium-oxonate + yeast-paste hyperuricemia rats; the best visible dose, 0.5 g/(kgd), reduced serum urate from 281.62 to 93.27 µmol/L and reported URAT1/GLUT9 down, OAT1/ABCG2 up, hepatic XOD inhibition, renal oxidative-stress / inflammatory-marker reduction, and microbiome-diversity effects. The publisher page reports the extract contains 35.86% polysaccharides, 27.05% protein, 0.21% phenolics, and 0.83% cordycepin. This is Animal Model evidence for the whole-extract cultivation route, not purified-cordycepin-alone or human efficacy evidence. Canonical platform route: cultivation, not koji engineering (revised 2026-05-16). C. militaris liquid fermentation is mature; top strain GYS60 (plasma-mutagenesis derivative) hits 7,883 mg/L static liquid (PMID 33463932). Commercial whole-fermentate C. militaris extracts deliver cordycepin with native pentostatin co-protection at the co-evolved ratio at $20–60/month nutraceutical pricing. The food-grade A. oryzae koji-engineering route (Jeennor 2023, PMID 38071331, 564 mg/L/d on glucose) was evaluated and deprioritized* during the 2026-05-15 sweep walkthrough (Item 7) — comp-023 confirmed metabolic-burden feasibility, but koji-engineered cordycepin delivers no novel chokepoint coverage beyond the cultivation route, has an unresolved dose-vs-achievable-titer gap in realistic multi-cassette home-fermentation conditions, and the "endgame strain = full coverage + simple" principle does not justify the engineering complexity for a payload available at the supplement shelf. See koji-endgame-strain.md §3.5 "Cordycepin third-cassette slot — evaluated and deprioritized" and operations/global-lit-scan-p0-remediation-2026-05-20 for the source-read record.

  • Sanghuangporus vaninii extract — Phase 5b P0-1 remediation upgrade (2026-05-20): open-access Animal Model sources now support S. vaninii as more than a CNKI abstract lead. Hua 2023 Biomedicine & Pharmacotherapy tested 70% ethanol S. vaninii extract (SHEE) in potassium-oxonate + adenine hyperuricemic renal-injury ICR mice at 125/250/500 mg/kg and reported lower blood UA/Cr/BUN/XOD, lower liver ALT/AST, URAT1/GLUT9 down, OAT1/OAT3/ABCG2 up, reduced apoptosis/caspase-3 activity, and LD50 >5,000 mg/kg (Animal Model plus HK-2 cell model; DOI 10.1016/j.biopha.2023.114970). Sun 2022 Nutrients tested 1.5 g/kg S. vaninii in yeast-extract + potassium-oxonate HUA mice and MSU-induced acute gouty arthritis rats, reporting reduced serum UA / serum-liver XOD and reduced ankle swelling / inflammatory markers (Animal Model; DOI 10.3390/nu14204421). This is not human evidence and does not establish direct transporter binding. The 2025 CNKI Journal of Jilin Agricultural University S. vaninii UPLC-Q-TOF-MS record remains quarantined for full-text retrieval because its English/MT allopurinol-comparator phrase is ambiguous; use the open-access 2022/2023 papers as the current promotion-grade anchors. Source-read record: operations/global-lit-scan-p0-remediation-2026-05-20.

  • Phellinus igniarius TFPI — Phase 5b P0-1 remediation upgrade (2026-05-20): Chen 2023 Heliyon tested total flavonoids of P. igniarius (TFPI; 70% ethanol extract, AB-8 resin purified, 66.7% total flavonoids) in potassium-oxonate hyperuricemic / uric-acid-nephropathy ICR mice at 50/150/450 mg/kg. The top dose lowered serum UA from 105.0 to 82.3 micromol/L, Cr from 44.30 to 33.37 micromol/L, and liver XOD from 48.91 to 44.09 U/g prot versus model; HK-2 renal epithelial cells exposed to MSU showed lower LDH/apoptosis/ROS, lower TLR4/NF-kB/TXNIP/NLRP3 and IL-1beta/TNF-alpha, and increased ABCG2 protein (Animal Model plus In Vitro renal-cell model; DOI 10.1016/j.heliyon.2023.e12979). This supports TFPI as a HUA / renal-injury source-read anchor, not human evidence, not direct URAT1 binding, and not a synovial gout-flare model. The 2025 CNKI TFPI record remains quarantined for full-text retrieval because its abstract-level OAT1/ABCG2/URAT1 mRNA-versus-protein directions are not yet interpretable. Source-read record: operations/global-lit-scan-p0-remediation-2026-05-20.

  • The natural Cordyceps ADA-inhibitor pairing — C. militaris natively co-produces pentostatin from the same BGC cluster as cordycepin (Xia 2017, PMID 29056419). Pentostatin is a clinical-grade ADA inhibitor; this means whole-fermentate Cordyceps preparations have a built-in safeguard against cordycepin deamination that purified cordycepin lacks. This reframes the Phase 6 GLPP+cordycepin synergy hypothesis — the native ADA inhibitor is already packaged with cordycepin in fermented C. militaris. The GLPP+cordycepin synergy may be redundant with whole-fermentate Cordyceps, but a purified cordycepin + GLPP combination would still benefit from GLPP's ADA modulation. Wet-lab gate now has a 4-arm question: whole-fermentate Cordyceps vs purified cordycepin vs purified cordycepin + GLPP vs purified cordycepin + pentostatin.

  • Ergothioneine — Phase 7-1c correction: Pleurotus ostreatus is NOT the highest EGT producer; P. citrinopileatus (golden oyster) is 2-3× higher (7.0 vs 2.4 mg/g DW). Apex strain: P. citrinopileatus 303 with two-stage H₂O₂+vit C oxidative stimulus → 641.76 mg/L submerged fermentation (Li 2025). Dietary intake plausibly therapeutic-dose: 50-100g fresh oyster ≈ 12-24 mg EGT, within published RCT-investigational range. OCTN1/SLC22A4 saturates at ~25 mg/day → concentrated extracts have diminishing returns, "daily oyster mushroom in dinner stir-fry" is mechanistically sound. A. oryzae / koji also produces ergothioneine (secondary).

  • Eritadenine (Lentinula edodes) — cardiovascular activity, cholesterol-lowering. Native cultivation route.
  • Erinacines / hericenones (Hericium erinaceus) — NGF-inducing (CNS-relevant). Native cultivation route.
  • Inotodiol / betulinic acid derivatives (Inonotus obliquus) — triterpenoid chemistry overlapping reishi. Native cultivation route.
  • PSK / PSP polysaccharide-protein complexes (Trametes versicolor) — β-glucan immunomodulator, FDA-approved adjuvant in Japan. Native cultivation route.

Stays with the koji-engineering track:

  • Recombinant uricase (existing OE thesis — engineered-yeast-uricase-proposal.md, engineered-koji-protocol.md)
  • Lactoferrin (existing OE thesis)
  • DAF SCR1-4 truncated complement regulator (comp-006 / comp-012)
  • DAE (methyl 2,4-dihydroxybenzoate) — small molecule, chemical synthesis preferred over fungal extraction (per comp-014 Phase 6 production-route assessment)
  • Any future therapeutic protein — koji is the chassis when secreted protein is the deliverable

Combined / synergy candidates (per comp-014 Phase 6, refined by Phase 7-1b strain scan):

  • Whole-fermentate Cordyceps (cordycepin + native pentostatin) — Phase 7-1b discovered that C. militaris natively co-produces pentostatin from the same BGC as cordycepin. Whole-fermentate preparations have the ADA-inhibitor safeguard built in. This is the cleanest single-organism medicinal-mushroom-complement product — fermented C. militaris on brown rice (4-8 week home cycle) delivers cordycepin + pentostatin in their natural ratio, no synergy assembly required.
  • GLPP + (purified) cordycepin — only relevant if purified cordycepin is the delivery format. With whole-fermentate Cordyceps, the ADA-inhibitor pairing is already intrinsic. Wet-lab gate (4-arm comparison) now answers: which delivery format wins on dose efficiency + reproducibility.
  • Whole-fermentate Cordyceps + GLPP (two-organism stack) — the cultivation-only, zero-engineering combination: fermented C. militaris (delivering cordycepin + native pentostatin) + G. lucidum-derived GLPP supplement. Targets ADA from two independent biochemical entry points — pentostatin (small-molecule competitive inhibitor) + GLPP (polysaccharide-peptide binding) — while cordycepin is the substrate whose half-life is being protected. This is the fermentable / enzymatically-naive intervention at the ADA chokepoint: requires only cultivating two GRAS fungi, no genetic engineering. Distinct from the whole-fermentate-Cordyceps-alone bullet above because GLPP adds a second, mechanistically orthogonal ADA-blockade path rather than substituting for the native pentostatin. Wet-lab gate: 4-arm ADA half-life assay (per Proposed Experiment in 2026-05-08 sweep) extended with a fifth arm — whole-fermentate Cordyceps + GLPP — to test whether the two-organism combination outperforms either single-organism preparation.
  • ~~Engineered koji cordycepin (cns1+cns2) + GLPP supplement~~ — Deprioritized 2026-05-16. The koji-engineered cordycepin route was removed from active cassette stack; this combination is moot. The cultivation-track equivalent (whole-fermentate Cordyceps + GLPP) above remains the canonical platform path.
  • ~~Engineered koji cordycepin (cns1+cns2) + whole-fermentate C. militaris (native pentostatin)~~ — added 2026-05-15, deprioritized 2026-05-16. This cross-chassis "patch" was an attempt to install pentostatin protection for the koji-engineered cordycepin route. With koji-cordycepin engineering deprioritized (no novel chokepoint coverage + open dose-vs-titer gap + commercially available alternative), the cross-chassis patch is moot. The cultivation route already delivers cordycepin + pentostatin together at the natural ratio. See koji-endgame-strain.md §3.5.
  • Engineered koji uricase + GLPP supplement — koji handles bulk urate degradation in gut lumen; GLPP modulates ADA upstream + GLUT9/OAT1 transporters for renal-side support. Cleanest cross-track synergy (engineering + cultivation).

Substrate engineering as the most-accessible cultivation lever (added 2026-05-19)

Substrate composition is not just a documentation concern — it is a deliberate engineering variable with documented effect sizes from 1.2× (yield aggregate) up to 100× (specific compound profile shifts within a class) and 22× (combined precursor + induction). This finding emerged from the 2026-05-19 substrate-engineering lit scan (logs/substrate-engineering-mushroom-cultivation-lit-scan-2026-05-19.md). The synthesis daemon's prior framing ("substrate accumulation creates a QC documentation discipline") under-claimed the empirical literature by ~10×.

Four mechanisms operate, each with primary-literature anchors:

  1. Passive accumulation — substrate compounds traverse mycelium (plant flavonoids in oak substrate; tree-host polyphenols in I. obliquus conks — Alnus incana conks have 4–30× higher betulinic acid than Betula pendula per Drenkhan 2022 PMC9496626).

  2. Biotransformation — fungal enzymes modify substrate compounds (betulin → betulinic acid in I. obliquus; lentinan biosynthesis from substrate cellulose).

  3. Substrate induction of biosynthetic gene clusters — substrate components act as transcriptional signals. Microcrystalline cellulose 1.5% delivers +85.96% ganoderic acid via HMGR/SQS/LAS upregulation (HMGR up 3.5–4.3×; Hu 2017 PMC5395960). Oleic acid upregulates Cns1/Cns2 in C. militaris, delivering 34× cordycepin difference between A. dichotoma and B. mori insect substrates (Turk 2022 PMC9627333).

  4. Precursor feeding — direct addition of biosynthetic precursors. 12 g/L alanine → 3× cordycepin via Cns2/Cns3 upregulation (Yu 2024 PMC11698586); 2 mM methionine → 1.7–3.1× ergothioneine across multiple species (Lee 2009 PMC3749454).

Critical finding: substrate engineering shifts compound PROFILE, not just yield. Wood-log vs. substitute-substrate G. lucidum produces measurably different triterpenoid spectra — substitute-grown fruiting bodies show 13.5× higher ganosporelactone B and 10× higher ganoderol A, while wood-log fruiting bodies show 2.19× higher total lucidenic acids (Luo 2024 PMC10879320). Hericium minimal vs. complex liquid media shifts erinacine C ↔ erinacine Q ratios by ~100× even when eri gene transcript levels don't change (Doar 2025 PMC11969743) — a post-transcriptional substrate-driven compound-profile shift.

Operational implication for distributed contributors: substrate engineering is the lightest-effort, highest-leverage modality — every load-bearing reagent (methionine, alanine, oleic acid, microcrystalline cellulose, D-galactose, corn steep liquor, casein hydrolysate, nucleosides) is GRAS food-grade and available at consumer pharmacy / grocery retail for $20–50/kg. This compounds with strain selection rather than competing with it. The discipline is target-compound-anchored: a substrate protocol without paired Tier-2/Tier-3 characterization (per medicinal-mushroom-extract-sops.md SOP-6 and the new SOP-7 protocol matrix) is non-falsifiable.

Platform-principle elevation (2026-05-19): this finding promoted from "queued open question" to named Platform Principle 9 in etc/open-source-platform.md. Strain engineering needs academic infrastructure; substrate engineering can be executed from a kitchen. It is the cleanest fit yet between the platform's distributed-accessibility thesis and a primary engineering mechanism.

Falsifiable wet-lab priority surfaced by the lit scan: cordycepin × pentostatin ratio under varied substrate conditions (alanine vs CSLH vs insect substrate vs oleic combinations) — directly extensible from SOP-2 HPLC. Queued as §1.29 in validation-experiments.md (added 2026-05-19). Resolves the whole-fermentate-vs-purified clinical positioning gap that has been open in the wiki.

Scope (Phase 1)

Candidate species

Subset of the comp-014 Phase 5 anchor list, prioritized by: 1. Strength of comp-014 chokepoint-mapping evidence (in vivo > in vitro > predicted) 2. Home-cultivability (kit-scale UX > requires bioreactor > requires specialized substrate) 3. GRAS / pharmacopoeia status (no regulatory novelty needed) 4. Native compound profile diversity (multi-compound producers preferred — broader Open Enzyme value)

Species Common name Top compounds Cultivation UX Regulatory
Ganoderma lucidum reishi / lingzhi GLPP, ganoderic acids (400+), ergosterol; S-GLSP (sporoderm-removed spore powder, distinct NLRP3-axis effect); GLP4 (pentapeptide, direct TBK1-binding, distinct from triterpenoid + polysaccharide fractions) — refinements added 2026-05-19 traditional-name rescan Solid-state on hardwood; commercial mycelium kits widely available; 6-12 months for fruiting body GRAS supplement
Cordyceps militaris cultivated cordyceps cordycepin, polysaccharides, ergosterol peroxide; whole water extract Animal Model HUA signal (Xiong 2024: serum UA 281.62→93.27 µmol/L at 0.5 g/(kgd), URAT1/GLUT9 down, OAT1/ABCG2 up, hepatic XOD down); head-to-head NLRP3 dominance per Wang 2023* (NLRP3-axis primary mechanism for anti-MSU effect, not URAT1 alone — surfaced by 2026-05-19 traditional-name rescan) Liquid fermentation OR solid-state on brown rice; 4-8 weeks home cycle; commercial kits available GRAS supplement
Phellinus igniarius sang huang / 桑黄 total polyphenols (XO + NLRP3 axis); SH-P-1-1 polysaccharide (gut microbiome); PPI polysaccharide (bile acid + hepatic XO); TFPI flavonoids (Animal Model HUA / uric-acid-nephropathy: serum UA 105.0→82.3 micromol/L at 450 mg/kg, Cr 44.30→33.37 micromol/L, liver XOD 48.91→44.09 U/g prot; HK-2 MSU renal-cell TLR4/NLRP3 down, ABCG2 up) — strongest single-species fit for gout-indication coverage across independent papers; 2026-05-19 traditional-name rescan plus 2026-05-20 P0-1 source read Solid-state on hardwood (preferred birch / poplar); commercial mycelium kits available; 6-12 months for fruiting body; cultivated extract matches wild extract per Zhou 2022 — directly addresses wild-supply bottleneck Supplement
Sanghuangporus vaninii yellow-brown sanghuang (modern split from P. baumii) PSH acidic polysaccharide (MW 5.25 × 10⁴ Da); SHEE ethanol extract (Animal Model HUA renal-injury: URAT1/GLUT9 down, OAT1/OAT3/ABCG2 up; LD50 >5,000 mg/kg); whole extract MSU + HUA active Liquid submerged fermentation optimized (Huang 2024); commercial extract available; extract identity/batch reproducibility still SOP-gated Supplement
Inonotus hispidus shaggy bracket whole extract anti-HUA + anti-MSU-arthritis Hardwood substrate; commercially less developed Supplement
Antrodia camphorata niu chang zhi / 牛樟芝 Antcin-H (NLRP3-selective triterpenoid); ACP polysaccharide; Taiwan-endemic Liquid submerged fermentation; specialty supply Supplement (Taiwan-regulated)
P. citrinopileatus golden oyster ergothioneine (highest fungal producer — 7.0 mg/g DW), β-glucans Same straw / sawdust substrate as P. ostreatus; commercially less common than oyster but kits available GRAS food
Pleurotus ostreatus oyster mushroom ergothioneine (2.4 mg/g DW — was originally claimed as highest, corrected by Phase 7-1c scan), lovastatin (190-342 mg/kg DW), pleuran Easiest home cultivation — straw / coffee grounds / sawdust substrate; 4-6 weeks; widely sold consumer kits GRAS food
P. djamor pink oyster β-glucans (43% DW — highest in genus) Same as P. ostreatus GRAS food
P. eryngii king oyster ergothioneine (5.84 mg/g DW Hi-Ergo strain) Same as P. ostreatus GRAS food
Lentinula edodes shiitake lentinan, eritadenine, ergosterol→D2 Log / sawdust block; 6-18 months; widely sold consumer kits GRAS food
Hericium erinaceus lion's mane erinacines, hericenones Sawdust block / log; 4-8 weeks; widely sold consumer kits GRAS food
Trametes versicolor turkey tail / yun zhi PSK, PSP Hardwood log / dowels; 6-18 months for fruiting body Supplement (PSK is approved drug in Japan)
Inonotus obliquus chaga betulinic acid derivatives, inotodiol, melanin Wild-harvest from birch (slow growth); cultivation difficult; commercial mycelium-grown extracts widely available Supplement
Ganoderma applanatum artist's conk DAE (small molecule), polysaccharides Solid-state on hardwood; less commercially developed than G. lucidum Supplement
Aspergillus oryzae (koji) koji ergothioneine (secondary), kojic acid, secreted enzymes Already documented in koji-home-fermentation.md GRAS food

The medicinal-mushroom-complement track is species-additive to comp-014's anchor list, not separate — the same well-studied species, evaluated through a different lens (cultivation feasibility + native-compound consumption rather than engineering-chassis suitability).

Ascomycete secondary metabolites — Talaromyces, not Penicillium (identity-corrected 2026-05-19) — chassis-pending discovery note

The track's candidate species above are all basidiomycetes. comp-014 Phase 3 surfaced direct caspase-1 (CASP1) inhibitors at sub-μM potency from "Penicillium" — a chokepoint coverage the basidiomycete corpus lacks. The 2026-05-19 lit scan (logs/food-grade-penicillium-casp1-lit-scan-2026-05-19.md) materially revised the framing:

Identity correction: The original Berkeleyamide-producing strain (Stierle 2008, "P. rubrum" Berkeley Pit isolate) has been reclassified to Talaromyces amestolkiae — a different genus from the cheese-ripening Penicillium species (P. camemberti / P. roqueforti). Berkeleyamides A/D (CASP1 IC50 330/610 nM via comp-014 pChEMBL anchors 6.48/6.21) are Talaromyces chemistry, not Penicillium chemistry. (The reclassification is inferred from secondary sources — Yilmaz 2014 + WebSearch hits; primary-source confirmation via Hoody 2026 [PMC13150583] recommended before any downstream commitment.)

Why the food-grade Penicillium framing was misdirected (three independent reasons):

  1. Wrong genus. P. camemberti / P. roqueforti are in genus Penicillium proper, taxonomically distinct from Talaromyces.
  2. Wrong direction of effect. The closest food-grade Penicillium "anti-inflammatory" candidate — mycophenolic acid — is pro-NLRP3, not anti (Huang 2018 PMC6032679: MPA synergizes with LPS to activate the inflammasome at 5–75 μM). A wet-lab assay seeing "CASP1 modulation" in cheese-strain extracts would risk reading MPA as a positive when it's the opposite direction.
  3. Wrong genome. Domesticated cheese P. roqueforti strains have actively-degraded toxin BGCs (Crequer 2024 PMC11605963 — frameshift in PR-toxin ORF, deletion in mpaC). Substrate-induction cannot unlock what's been mutated out. The 2023 canonical BGC review for P. roqueforti (Chávez PMC10144355) enumerates the entire chemotype (andrastins, MPA, roquefortines, PR-toxin, eremofortins, isofumigaclavines, festuclavine, annullatins) — Berkeleyamide / Berkeleyone are absent from the genome, not just unexpressed.

Corrected platform-relevant path — computational first, wet-lab only if signal:

  1. antiSMASH genome scan ($0, ~3hr compute) of P. roqueforti / P. camemberti / P. rubens for NRPS-PKS hybrid BGCs matching the Berkeleyamide architecture (NRPS amide-bond-forming + meroterpenoid backbone). If no homologs found, the cryptic-Berkeleyamide hypothesis is falsified for cheese strains at zero cost.
  2. Pull Talaromyces amestolkiae BGC if available — Stierle group at U. Montana may have deposited assembly (FAC-NGS data per Cryptic Biosynthesis paper PMC8574098). Reach out via JGI MycoCosm or direct email.
  3. If antiSMASH returns plausible homologs in cheese strains → wet-lab assay becomes platform-relevant. Defensible budget with mycotoxin pre-screen (LC-MS) + CASP1 enzymatic assay + orthogonal cytotoxicity is $5–15K (not the originally-scoped $500–1,000 — that figure assumed direct testing without disambiguation infrastructure).
  4. If antiSMASH returns nothing in cheese strains but T. amestolkiae has a clean BGC → the platform-relevant question shifts entirely to engineering the Berkeleyamide BGC into the koji chassis (A. oryzae heterologous host, which already supports andrastin-type meroterpenoid assembly per Matsuda et al. 2013 cited in PMC5418334). This is a much cleaner play than coaxing cheese strains to express foreign chemistry.

Status: chassis-pending discovery note with computational gates. No wet-lab commitment until antiSMASH signal materializes. The identity correction is the main load-bearing update: any future BGC-mining work must target Talaromyces amestolkiae, not cheese-ripening Penicillium. comp-014 Phase 3's "Penicillium" attribution should also be corrected to "Talaromyces amestolkiae" in any future re-render of that page.

See logs/food-grade-penicillium-casp1-lit-scan-2026-05-19.md for the full lit scan.

Cultivation method comparison (Phase 7 follow-up)

Three home-feasibility levels:

  1. Consumer-kit (oyster, lion's mane, shiitake, reishi, turkey tail) — pre-inoculated grow blocks ($15-50), 4-12 weeks fruiting cycle, no specialized equipment beyond mister bottle + plastic tent.
  2. Mycelium kit (cordyceps, more advanced reishi) — agar plate or grain spawn → fermentation jar / brown rice substrate; 4-8 weeks; requires sterilization (pressure cooker).
  3. Bioreactor (defined-media liquid fermentation for standardized GLPP / cordycepin yield) — not home-feasible at consumer scale; commercial extract supply is the route.

Open Enzyme's value-add at the cultivation-method layer is standardization + characterization protocols, not novel cultivation. Consumer-grade reishi grow kits are ubiquitous; what's missing is the protocol that says "this batch contains X mg/g GLPP, Y% triterpene content, Z µg/g ergosterol — verified by standardized extraction + analysis." That's the Open Enzyme contribution to this track.

Production route — sequential cultivation-first default (added 2026-05-06)

For compounds that have both a cultivation route and a koji-engineering route — cordycepin being the canonical case — the platform default is sequential cultivation-first. The koji-engineering route is held as documented contingency, not parallel commit.

Cordycepin example (the worked case):

Route Titer / time Format Maturity Cost to test
C. militaris GYS60 cultivation (PMID 33463932) 7,883 mg/L Liquid submerged fermentation Mature industrial process; established CROs $1–2K outsourced
Engineered A. oryzae cns1+cns2 (Jeennor 2023, PMID 38071331) 564 mg/L/day Solid-state koji One paper, novel $2–4K outsourced; in-house requires separate setup

Why not parallel head-to-head (Principle 6 carve-out): Platform Principle 6 — variant-agnostic empirical head-to-head says default to parallel testing when literature is split AND marginal cost is bounded by shared infrastructure. The cordycepin case fails the second precondition: cultivation requires liquid submerged bioreactor, koji-engineering requires solid-state trays — different fermentation infrastructure, different downstream purification, different QA. Sequential cultivation-first is the right call: GYS60's 7,883 mg/L is mature and reproducible; koji-engineering at 564 mg/L/day (~14× lower per unit time) is one paper. The chassis-coherence appeal of "everything in koji" is real but not load-bearing for a small-molecule supplement — cordycepin's structure doesn't care which organism made it.

Supply check. Even at high-end therapeutic dosing (~500 mg/day pure cordycepin — see Phase 7 follow-up #7 for verification queue), 1 L of GYS60 broth = ~16 daily doses. A modest 1,000-L run = ~44 person-years of supply. Neither route is supply-constrained for this product class; the decision is cost, reproducibility, and CRO availability — not throughput.

⚠️ Reality check — current consumer-grade fruiting-body extracts deliver sub-therapeutic cordycepin doses. Empirically (verified against Real Mushrooms' published HPLC data 2026-05-06), high-quality fruiting-body extracts like Real Mushrooms Cordyceps-M test at 0.1–0.3% cordycepin (~0.4% in recent batches), equating to ~3–4 mg cordycepin per ~1 g serving. That is 25–150× below typical consumer pure-cordycepin nutraceutical doses (100–500 mg/day) and likely orders of magnitude below the dose where animal-model URAT1-inhibition was demonstrated. The implication: today's consumer-grade fruiting-body extracts are valuable as general mushroom adaptogens / β-glucan / immune-layer products, but should not be marketed or recommended as URAT1-inhibition therapy — the dose math doesn't support it. A pure-cordycepin nutraceutical at ~150–500 mg/day is the only product format that delivers a therapeutically-relevant cordycepin dose; that market segment is real but quality-variable and pricier (~$150–300/month). This is the canonical motivating example for Phase 7 follow-up #7 — therapeutic dose grounding pass: without dose-vs-product-content grounding, the platform is at risk of recommending interventions whose actual delivered dose is sub-threshold for the cited mechanism.

When the koji-engineering contingency activates: speculative — none of the trigger conditions are currently load-bearing. (a) Cultivation supply chain disruption / quality issues, (b) scale-up economics later favoring solid-state for some other co-produced compound, or © IP / regulatory positioning making the engineered-organism route strategically valuable. All three are real-but-unlikely triggers. Cordycepin is a small molecule whose therapeutic effect doesn't depend on which organism made it, so chassis-coherence is not load-bearing for this compound. The supply math is fine for the koji route at industrial scale (Jeennor's 564 mg/L/day × 1,000-L bioreactor = ~1,000–3,000 person-days of supply per batch at therapeutic dose), but cultivated GYS60's 14× higher per-unit-time titer + mature CRO ecosystem + classical-use precedent + consumer trust dominate. The contingency is documented for completeness, not as live planning. Default to GYS60 cultivation; revisit only if a specific external trigger surfaces.

The same sequential-first logic applies generally to any compound with both a mature cultivation precedent and a novel-engineering route. Default to the mature route; document the engineering contingency; only switch when supply, cost, or strategic-fit forces it.

Therapeutic UX

Consumption modalities for the cultivation track:

  • Dual decoction (water + ethanol sequential extraction → blended) — captures both polysaccharide and terpenoid fractions. Traditional Chinese medicine standard for reishi. Concentrated to liquid extract or spray-dried to powder.
  • Tincture (ethanol-only extraction) — captures terpenoid + small molecule fractions; not polysaccharides. Better for ganoderic acids, erinacines, betulinic acid derivatives.
  • Hot water tea (decoction only) — captures polysaccharides, water-soluble small molecules. Lentinan from shiitake, GLPP from reishi.
  • Dried powder (whole fruiting body or mycelium → dried → ground) — broad-spectrum but low concentration. Familiar consumption format (capsule or food-incorporation).
  • Co-administration with koji-condiment (the integration point with the koji track) — daily shio-koji + periodic mushroom decoction, or mushroom powder added to koji condiments.

The UX is meaningfully different from the koji track and that's a feature, not a bug — different therapeutic categories may be best served by different formats. Daily koji shio-koji on rice for uricase; weekly reishi tea for GLPP-mediated ADA modulation; oyster mushroom in dinner stir-fry for ergothioneine.

Why this isn't just "buy mushroom supplements"

Consumer-grade mushroom supplements have known quality issues:

  1. Species mis-identification — DNA-barcoded studies of Ganoderma supplements regularly find <50% contain the species labeled.
  2. Compound content uncharacterized — "reishi extract 1000mg" tells you nothing about GLPP, ganoderic acid, ergosterol content. Triterpene content alone varies 100× between products.
  3. Adulteration with inactive carriers — mycelium-on-grain products are mostly grain (β-glucan from grain ≠ β-glucan from mushroom).
  4. Extract method opaque — water-only extracts miss terpenoids; ethanol-only extracts miss polysaccharides; consumers can't tell which they bought.

Open Enzyme's contribution to this track: publish reproducible cultivation + extraction + characterization protocols. Strain selection criteria. HPLC/MS validation methods. Make a comp-014-style triage repeatable for any open-source contributor. The chemistry IS in the public domain — what's missing is the rigor.

Consumer-product caveat — structure-dependent β-glucan NLRP3 directionality (added 2026-05-08)

Mushroom β-glucans are not a monolithic anti-inflammatory class. Their effect on the NLRP3 inflammasome is structure-dependent — different polysaccharide fractions from the same species can activate or inhibit NLRP3 depending on extraction method, branching pattern, and molecular weight.

Per comp-014 Phase 5 (medicinal-mushroom-compound-mapping-computational.md), as captured in modality-chokepoint-matrix.md:

  • G. lucidum exopolysaccharides (EPS) — secreted polysaccharide fractions from liquid-fermentation broth — can activate NLRP3 via the Dectin-1 / Syk pathway. Wrong direction for a gout intervention.
  • G. lucidum spore-powder β-glucans / GLP — fractions from cracked-spore preparations or dual-decoction GLPP-enriched extracts — can inhibit NLRP3 (immune-training / Treg-induction mode of action). Right direction.

Why this matters for consumer products: a generic "reishi extract 1000mg" capsule is opaque about which polysaccharide fraction it contains. EPS-dominant preparations and spore-powder/GLPP-dominant preparations are functionally different products at the NLRP3 axis, and a gout patient can inadvertently worsen inflammation by picking the wrong fraction. The dual-decoction extraction protocol in SOP-1 is specifically designed to enrich for the GLPP fraction, not whole-extract β-glucan; this is part of why "compound content uncharacterized" (#2 above) is load-bearing rather than cosmetic.

Propagation discipline: when reishi / GLPP enters supplements-stack.md as a catalog entry, OR enters gout-action-guide.md as a specific recommendation, this caveat must travel with it. Currently neither catalog mentions reishi specifically, so no propagation-gap exists today — but the manual fresh-stack.py discipline (see synthesis/strategic-reflections/) will need to flag this as a known caveat-with-the-compound entry when promotion eventually fires.

Two species-level corrections from 2026-05-19 traditional-name rescan (logs/mushroom-traditional-name-nlrp3-rescan-2026-05-19.md):

  1. Lentinan (L. edodes) — AIM2 not NLRP3. Earlier comp-014 framing implicitly grouped lentinan in the "mushroom β-glucan NLRP3 modulator" bucket. The 2026-05-19 species-anchored re-read found explicit MSU-arthritis testing of lentinan that came back negative on the NLRP3 axis — shiitake's primary anti-inflammatory contribution travels via the AIM2 inflammasome and cardiovascular eritadenine, not via NLRP3 at MSU-relevant tiers. Lentinan stays on the AIM2 axis for any inflammation framing; it is not a candidate for NLRP3-chokepoint coverage. Falsifying finding — preserve directional accuracy.

  2. PSK (T. versicolor) activates NLRP3 — wrong direction for gout. PSK is an approved drug in Japan (cancer adjuvant) where NLRP3 activation is therapeutically desired. For a gout indication, this is the opposite direction. Consumer "turkey tail" extract products vary widely in PSK content; standardized PSK is contraindicated for gout. Add to the structure-dependent β-glucan directionality discussion above — like G. lucidum EPS, PSK is a "wrong-direction" mushroom polysaccharide for the gout / NLRP3 axis.

Seven Phase 7 follow-ups queued

Same pattern as engineered-lbp-chassis.md and sirna-urat1-modality.md:

  1. Strain selection lit scan — for each top species (G. lucidum, C. militaris, Pleurotus, Lentinula, Hericium, Trametes, Inonotus), identify which commercial / academic strains have characterized compound yields. CNKI + J-STAGE + Korean sources critical here (the multilingual ingestion the comp-014 anchor list was originally drafted for, now applied to cultivation rather than chokepoint mapping). Output: per-species strain table with cultivation precedent + compound yield evidence.

  2. comp-NNN cultivation method × yield study — for ≥3 priority species (start with reishi GLPP + cordyceps cordycepin + oyster ergothioneine), comparative study of liquid fermentation vs solid-state vs fruiting body cultivation, measuring target-compound yield. Open Enzyme runs this as a literature meta-analysis first; wet-lab follow-up only if literature is insufficient or self-experimentation is direct path. The cordycepin yield literature in particular has wide ranges (0.5-5 mg/g dry weight in C. militaris fruiting body) that warrant consolidation.

  3. Extract characterization protocol standardization — write the open-source SOPs for: (a) reishi dual-decoction extraction yielding GLPP-enriched fraction, (b) cordyceps water extraction yielding cordycepin + polysaccharide fraction, © HPLC quantification methods + reference standards. The goal is reproducibility — an Open Enzyme contributor anywhere should be able to follow the SOP and produce a comparable extract.

Sub-bullet (added 2026-05-08): the SOP-6 framework in medicinal-mushroom-extract-sops.md is the canonical standardization scaffold — kitchen → smartphone → bench → outsourced tiers, calibrate-once at Tier-3 / track-batches at Tier-2 discipline, per-compound method specs. Next operational step: a first-execution demonstration batch — commercial C. militaris grow kit (~$50-150) cultivated through Tier 1 yield + Tier 2 EGT colorimetry done at home (Ellman's reagent, well-established), with one outsourced Tier 3 HPLC reference run for cordycepin (~$200-400 CRO). Scope: cordycepin + EGT only; GLPP deferred until Tier 3 SEC-MALS access (load-bearing assay per Pass 3 caveat) is sourced. Cordycepin Tier 2 diazo-coupling stays speculative and is covered by follow-up #7's primary-literature verification pass before being committed to the SOP. Outcome: de-stubbed SOP-2 / SOP-3 with first-batch numbers; reproducibility benchmark for any open-source contributor's later batches. Currently unscheduled — needs a free Brian-weekend or a contributor with a grow setup.

  1. GLPP + cordycepin synergy wet-lab gate — comp-014 Phase 6 flagged this as the cleanest synergy pair in the breadth pass. Study design: PO HUA mouse model, 4 arms (control / GLPP alone / cordycepin alone / combination), measure SUA + cordycepin PK + ADA activity. Critical question: does ADA inhibition co-treatment extend cordycepin half-life enough to lower the daily dose meaningfully? This is a focused, falsifiable experiment.

  2. Hypothesis card H06 — medicinal-mushroom-complement track viability. Specifies: what's the kill criterion? When would we abandon this track? Default: if (a) compound yield variability across cultivation batches exceeds 50% even with standardized protocols AND (b) standardized extracts cannot replicate the published in vivo effect sizes within 2× — that's pipeline failure regardless of mechanism truth. Falsifiable.

  3. Comparative chassis/platform matrix for gout — extend modality-chokepoint-matrix.md with a new row "Native-compound mushroom complement" populated with the Phase 7 candidates × chokepoint mapping. Forces explicit comparison: koji-uricase vs reishi-GLPP at the gut-lumen / urate-axis cells. Doesn't pre-determine which wins; makes the comparison tractable for case-by-case decisions.

  4. Therapeutic dose grounding pass (added 2026-05-06) — for each load-bearing compound in the track (cordycepin, GLPP, ergothioneine, eritadenine, erinacines, PSK, inotodiol, astilbin), grep-verify the human therapeutic dose range from primary clinical / supplement-trial literature under CLAUDE.md Rule 4 pre-commit grep-verify gate. The mushroom track scope page currently discusses production yields (mg/L, mg/g DW) without a dose-context anchor — without that anchor, "GYS60 hits 7,883 mg/L" is meaningless to a supplement-stack decision. Output: per-compound dose-grounding table (typical supplement range / clinical-trial range / mechanism-derived effective range / load-bearing-confidence tier). This should land before any wet-lab-gated Phase 2 follow-up depends on compound dose assumptions. Cross-applies to the TCM compound triage compounds — same gap.

Sub-bullet (added 2026-05-06): while doing the per-compound primary-source dive for dose grounding, also note any validated colorimetric-assay precedents at smartphone-tier (~Tier 2) sensitivity. Specifically — Pass 2 proposed diazo-coupling for cordycepin (per medicinal-mushroom-extract-sops.md SOP-6) but cordycepin lacks the phenolic/amine motifs typical for diazo chemistry, so the proposal needs primary-literature verification before committing to SOP. Ellman's reagent for EGT is well-established and needs no additional literature check. GLPP Tier 2 phenol-sulfuric is total-polysaccharide (not GLPP-specific) — already known; SEC-MALS at Tier 3 is the load-bearing assay. Closing this sub-bullet alongside #7 keeps the literature-search budget bounded — one primary-source pass per compound covers both dose and assay-precedent questions.

Regulatory clarity (compared to the koji track)

The medicinal-mushroom-complement track has simpler regulatory positioning than engineered-organism tracks:

  • All species are GRAS food (Pleurotus, Lentinula, Hericium) or established supplement-grade (Ganoderma, Cordyceps, Trametes, Inonotus)
  • No genetic modification → no GMO regulatory burden
  • Traditional-use precedent is decades to centuries (TCM, Kampo, Korean medicine, Western mycology)
  • Existing supplement industry framework (DSHEA in US, equivalent elsewhere) — not a novel regulatory category

This is a real platform advantage, not just a cultivation-feasibility advantage. The koji-engineered tracks (uricase, lactoferrin, DAF SCR1-4) need GRAS-pathway-certification or equivalent for any therapeutic claim. The medicinal-mushroom track is already there.

Caveat: dosing claims and efficacy claims are still regulated. "Reishi extract supports urate metabolism" is supplement-language-acceptable; "reishi extract reduces serum uric acid by 40.6%" is structure-function-claim regulated under DSHEA + equivalent. Open Enzyme contributes the reproducibility and characterization layer; therapeutic claims remain user-side / clinician-side responsibility.

Cross-references

Maintenance

  • When a new candidate species lands (via comp-NNN extension, new wiki research, or a new compound-pharmacology paper): add to candidate species table; mark its top compounds + cultivation UX + regulatory status; consider for Phase 7 cultivation comp-NNN.
  • When a Phase 7 follow-up resolves: promote from "queued" to canonical wiki content (e.g., the strain-selection lit scan output becomes a dedicated page per top species).
  • When the koji track adds a new compound: re-evaluate whether the medicinal-mushroom track has a complementary compound at the same chokepoint (synergy pair candidate).
  • The track is a peer to koji, not a replacement. Phase 7 deliverables expand the platform without competing with the koji-engineering thesis.

Status: Phase 1 (scope) committed 2026-05-06. Phases 2-6 are the queued follow-ups above. Brian's call on which phase to fire first.

CTO-actionable items (no scientific expertise required): tracked operationally in operations/todos.md §"Phase 7 medicinal-mushroom-complement track" — order golden oyster grow kit, email Lin Zhanxi 林占熺 lab for Juncao GLPP SOP, open GitHub Issue for Phase 5b multilingual deep-dive contributions, frame comp-014 as recruiting material for the open collaborator roles.