Lipase Activity in Koji-Derived Products¶

A protocol-and-methods companion to the koji wiki. Specifies how to measure the three native koji enzymes (amylase, protease, lipase) at four tiers of effort, what each tier can and cannot tell you, and how to anchor cheap home-scale assays against bench-grade ground truth so ongoing batch QC stays interpretable.

This page does not relock literature U/g values for wild-type A. oryzae enzymes — those live in digestive-enzyme-optimization.md. It does not relock the GI-survival 2-arm experimental design — that lives in validation-experiments.md §1.18. It is the methods layer that operationalizes the open question called out in koji-home-fermentation.md §"Open questions": quantitative comparison of shio-koji enzyme activity (units/g) vs. commercial PERT (Creon, Zenpep) units per pill — lab-measurable but not yet done.

Framework reference. This page is the canonical first instantiation of the four-tier quantification ladder (kitchen → smartphone → bench → outsourced). The abstract framework definition, operational pattern ("calibrate once at Tier 3, track batches cheap at Tier 1 / Tier 2"), and discipline notes live at quantification-ladder.md. This page covers the koji-enzyme-specific instantiation (lipase / amylase / protease assays + calibration anchors); the sister instantiation for medicinal mushroom extract characterization lives at medicinal-mushroom-extract-sops.md §SOP-6.

1. Decision frame — what to measure and why¶

For the EPI / PERT-comparison clinical question, the three koji enzymes are not equally load-bearing.

Enzyme	Clinical relevance for EPI	Mechanism in food integration	Priority for this protocol
Lipase	Primary bottleneck. Pancreatic insufficiency is dominantly a lipase deficit; Creon dosing is calibrated on USP lipase units.	Hydrolyzes triglycerides → free fatty acids + monoglycerides in the gut. Koji-derived lipase activity per gram is an order of magnitude lower than commercial PERT (Mechanistic Extrapolation from published A. oryzae lipase yields vs. Creon label content; see digestive-enzyme-optimization.md §"Lipase activity").	High — measure first.
Protease	Secondary. Protein digestion has more redundancy (gastric pepsin + brush-border peptidases) than fat digestion.	Different mechanism than gut substitution: pre-marinating protein in shio-koji partially hydrolyzes muscle protein before ingestion, reducing the protease + peptidase load on the gut rather than supplementing it.	High for the marinade-pre-digestion question; low for direct PERT comparison.
Amylase	Low. Salivary + pancreatic amylase + brush-border glucosidases provide redundancy; symptomatic starch malabsorption is uncommon in EPI.	Starch → maltose / glucose during shio-koji ferment + amazake; primary contributor to product sweetness.	Low for clinical comparison; medium for batch QC and fermentation-health monitoring.

What not to bother measuring at this stage: α-glucosidase, glucoamylase, individual peptidase classes (acid vs. neutral protease can be informative but aren't gating), and lactase. These add complexity without changing the EPI / PERT-comparison verdict.

Reference standard for "how much is enough." Creon is labeled in USP lipase / amylase / protease units per capsule (see Creon prescribing information). A standard fat-meal Creon dose is 25,000 USP lipase units (cited in digestive-enzyme-optimization.md §"Creon dosing equivalence"). The OTC analogue benchmark is BoulderBio at 40,000 FIP per capsule (~9,000–10,000 USP equivalent — In Vitro; conversion source in digestive-enzyme-optimization.md §"BoulderBio FIP↔USP conversion"). Any koji-derived measurement should be expressed as a fraction of one of these reference doses to be clinically interpretable.

2. Tier 1 — Kitchen-grade semi-quantitative¶

Equipment ceiling: kitchen scale (0.01 g), pH-buffered water, food-safe thermometer, timer, smartphone camera. Total reagent cost typically <$25 (vendor-dependent; pharmacy-grade indicators + grocery substrates).

Output type: inter-sample ratios at fixed temperature and pH, not absolute units. Useful for: batch-to-batch comparison, control vs. treatment within a single run, calibrating against a Tier 3 ground truth (see §6). Not useful for: quoting USP units; comparing across labs; clinical claims.

Discipline that makes Tier 1 worth running: - Always include a water blank (zero) and a dissolved Creon capsule (positive reference) in the same run. - Triplicate samples per condition; report median + range. - Hold temperature within ±2°C and pH within ±0.3 across the run — kitchen pH-7 phosphate buffer made from KH₂PO₄ + K₂HPO₄ is fine. - Read endpoints by phone photo at fixed lighting and fixed timing, not by eye.

2.1 Lipase — olive oil + pH indicator¶

Principle. Lipase hydrolyzes olive-oil triglycerides → free fatty acids → pH drops → bromothymol blue (or phenol red) shifts color (blue→yellow for BTB, red→yellow for phenol red). Time-to-shift at fixed substrate load is inversely proportional to lipase activity.

Setup (per sample): - 5 mL emulsified olive oil (homogenized 30 s in 5 mL pH-7 phosphate buffer + 1 drop dish detergent as emulsifier; matches Winkler & Stuckmann 1979 emulsion principle without the p-NPP). - 3 drops bromothymol blue (0.04% in ethanol, pharmacy grade) or phenol red (0.02%). - 1 mL test sample (clarified shio-koji liquid; see §5 for extract prep). - Hold at 37°C in a water bath or yogurt-mode Instant Pot.

Readout. Photograph at t=0, 15, 30, 60, 90, 120 min. Score color shift on a 0–4 scale against the Creon-positive control's progression. Time-to-half-shift is the comparable metric.

Sensitivity floor. This assay reliably detects lipase activity within roughly an order of magnitude of one dissolved Creon capsule's per-mL activity (Mechanistic Extrapolation from indicator pKa range and typical pancreatic lipase specific activity). Below ~5% of Creon per mL the shift is too slow to call against drift.

2.2 Amylase — starch-iodine clearance¶

Principle. α-amylase cleaves starch to maltodextrin / maltose / glucose. Iodine forms a deep blue-black complex with intact amylose and a brown-to-clear color with the cleavage products. Time-to-clear (no blue) at fixed starch load is proportional to amylase activity. Direct kitchen analog of the spot-test stage of the Bernfeld 1955 DNS assay.

Setup (per sample): - 50 mL 1% (w/v) starch solution in pH-6 acetate buffer (pH 6 because A. oryzae α-amylase optimum is acidic-neutral; matches typical assay conditions). - 1 mL test sample. - Hold at 37°C.

Readout. Every 2 min, withdraw 100 µL into a well plate or shot glass containing 2–3 drops Lugol's iodine (pharmacy USP). Score blue intensity by phone photo against a printed calibration card (deep purple-black → red-brown → clear). Time-to-clear is the metric.

Notes. Hishiroku white-koji (A. luchuensis) extract clears more slowly than yellow-koji (A. oryzae) extract at equivalent dose (Mechanistic Extrapolation from published amylase yields by species; see digestive-enzyme-optimization.md §"Amylase activity by strain"). This is expected, not a contamination signal.

2.3 Protease — gelatin liquefaction (or skim-milk-agar plate)¶

Principle. Protease cleaves gelatin (or casein) and liquefies a set gel (or clears an opaque skim-milk-agar lawn). Time-to-liquefy or halo diameter at fixed time = relative activity.

Setup A — Gelatin tube (simplest): - 5 mL liquefied 4% (w/v) gelatin in pH-7 phosphate buffer per test tube; let set at 4°C for 1 h. - Add 0.5 mL test sample on top; hold at 37°C. - Time the inversion-test point (tube can be inverted without flow). Compare against Creon-positive and water-blank tubes.

Setup B — Skim-milk-agar plate (more discriminating): - Pour 15 mL skim-milk agar (1% skim-milk powder + 1.5% agar in pH-7 buffer, autoclaved or pressure-cooked) into a sterile petri dish. Set at 4°C. - Punch 6 mm wells with a sterile straw or cork borer. - Add 50 µL test sample per well; hold at 37°C for 24 h. - Photograph; measure the clear halo diameter around each well.

Readout. Halo diameter (mm) at 24 h, or time-to-liquefy in the tube assay. Both are linear-ish in log(activity) over a useful 1–2-log dynamic range.

3. Tier 2 — Smartphone colorimetry¶

Equipment ceiling: Tier 1 plus a smartphone-photometer rig (a 3D-printed cuvette holder or a phone case with a fixed light source and a slot for a 1 cm cuvette; multiple open-source designs, e.g., Adams 2017 PLoS ONE) plus a small reagent kit (DNS reagent, ninhydrin, p-NPP — total ~$50–80 in single-experiment quantities). Output: A405 / A440 / A540 readings translated to relative concentration; with proper standard curves, semi-absolute units.

Output type: approximate U/mL with a 2–4× error bar relative to a calibrated bench instrument; suitable for tracking changes over batch / time / condition with confidence; not suitable for publication-grade absolute claims.

3.1 Lipase — p-NPP with phone OD¶

Principle. Winkler & Stuckmann 1979: p-nitrophenyl palmitate (p-NPP) is hydrolyzed to p-nitrophenol (yellow, A405 max) + palmitate. Rate of A405 increase = lipase activity in U/mL (1 U = 1 µmol p-nitrophenol released per minute at 37°C, pH 8).

Reagent. p-NPP (Sigma N2752 or equivalent), made up fresh at 1 mM in 2-propanol; dilute 1:10 into pH-8 Tris buffer immediately before assay. Reagent ~$30 / 1 g (enough for ~500 assays).

Procedure. 0.9 mL substrate working solution + 0.1 mL test sample in a 1 cm cuvette at 37°C; phone-photograph through the cuvette holder at t = 0, 1, 2, 5, 10 min. Convert RGB to A405 via the phone-photometer calibration curve. Slope (ΔA405/min) → µmol/min via ε(p-nitrophenol, pH 8) = 18,500 M⁻¹cm⁻¹ (well-established).

Calibration. A 5-point standard curve with commercial pancreatic lipase (Sigma L3126) at 0, 0.1, 0.5, 1, 5 U/mL bridges the phone OD to U/mL.

3.2 Amylase — DNS reducing-sugar (Bernfeld 1955)¶

Principle. α-amylase releases reducing-sugar termini from starch; 3,5-dinitrosalicylic acid reagent reacts with reducing sugars to form an orange-red product (A540). 1 U = 1 mg maltose-equivalent released per minute at 37°C, pH 6.

Reagent. DNS reagent: 1% DNS + 30% sodium potassium tartrate + 0.4 N NaOH (Bernfeld 1955 recipe). Premixed kits also available (~$40 / 100 assays).

Procedure. Mix 0.5 mL 1% starch (pH 6 acetate) + 0.5 mL test sample at 37°C; quench at 5 min with 1 mL DNS reagent; boil 5 min; dilute 1:5 with water; phone-photograph A540. Standard curve: maltose 0.1–2 mg/mL.

3.3 Protease — ninhydrin or azocasein with phone OD¶

Two acceptable Tier 2 protease readouts:

Ninhydrin / free amino-N. Hydrolyze 1 mL 1% casein (pH 7) with 0.1 mL test sample at 37°C for 30 min; precipitate residual protein with 2 mL 5% TCA; centrifuge or filter; react supernatant with ninhydrin reagent (heat 100°C × 15 min); read A570. Free amino-N released → relative protease activity. Ninhydrin reagent ~$30 / 100 assays.
Azocasein (Charney & Tomarelli 1947). 0.5 mL 0.5% azocasein (pH 7) + 0.5 mL test sample; 37°C × 30 min; quench with 1 mL 10% TCA; centrifuge; read supernatant A440. Azocasein ~$50 / 1 g (enough for ~100 assays). Higher dynamic range than ninhydrin; closer to bench-grade.

Both methods have a 2–4× spread vs. spectrophotometer readout depending on phone-photometer build quality; serviceable for relative-activity tracking.

4. Tier 3 — Bench enzymology¶

Equipment ceiling: community-college biology lab or undergraduate teaching lab. UV-Vis spectrophotometer (any cuvette-based instrument; even a Spec-20 works for visible-range assays), micropipettes, water bath, centrifuge. No specialized enzymology equipment required.

Output type: publication-grade absolute units (USP, FIP, IU as configured). Comparable to Creon label values directly.

Reagent cost: ~$200 total for one-time per-enzyme stocks (p-NPP, azocasein, DNS, soluble starch, control enzymes from Sigma/Fisher).

The Tier 3 assays are the same chemistry as Tier 2, with the phone-photometer replaced by a calibrated spectrophotometer. The discipline shifts:

Triplicate biological replicates × triplicate technical replicates (n=9 per condition).
Standard curves run same-day with each batch.
Positive controls every plate: pancreatic lipase (Sigma L3126), trypsin (Sigma T1426), α-amylase from A. oryzae (Sigma 10065).
USP unit calibration: USP General Chapter <1601> defines lipase / amylase / protease units against specific substrate / temperature / pH / endpoint conventions. Match those conditions to report USP-equivalent units. Otherwise, report in IU and footnote the assay conditions.

Suggested first run (one-day bench session): 1. Extract preparation from a single shio-koji batch — one extract, three aliquots (§5). 2. Triplicate p-NPP lipase assay vs. dissolved Creon (matched per-mg basis). 3. Triplicate azocasein protease assay vs. dissolved Creon. 4. Triplicate DNS amylase assay vs. dissolved Creon. 5. Output: a single ratio table — koji-extract activity per gram of fresh koji vs. Creon activity per capsule, expressed as percent-of-one-Creon-cap-equivalent per gram koji, separately for lipase / protease / amylase.

This first run is the load-bearing experiment — it converts every subsequent Tier 1 / Tier 2 home assay into a number that means something clinically. Without it, home assays are only inter-sample ratios; with it, home assays become "this batch is 12% of one Creon cap per gram, which is consistent with our Tier 3 calibration of 14% ± 3%."

3.X GI-survival 2-arm design¶

A logical extension of the Tier 3 first-run is the validation-experiments.md §1.18 2-arm GI-survival assay (Koji-S free extract vs. Koji-I whole biomass, both passed through SGF → SIF → assay at each stage). That experiment is fully specified there — the present page provides the assay methodology it depends on. The §1.18 cost estimate of $300–500 includes the assay reagents covered above.

5. Sample preparation — extracting enzymes from finished koji / shio-koji¶

For Tier 1 / 2 / 3, the sample matrix matters. Protocol below is the default for fresh koji rice, shio-koji, and amazake samples; adapt buffer pH for the specific enzyme being measured.

Mass. Weigh 5–10 g fresh sample; record mass.
Homogenize in 4× volume of pH-extraction buffer (pH 7 phosphate for general; pH 5 acetate if measuring acid protease specifically; pH 8 Tris if measuring lipase under USP conditions). Use a stick blender or mortar for ≥1 min.
Cold extract at 4°C for 30 min with occasional inversion.
Clarify by centrifugation (10,000 × g, 10 min, 4°C) or filtration through a folded coffee filter pre-rinsed with extraction buffer.
Aliquot the clear supernatant into 0.5–1 mL portions; assay fresh same day or freeze at −20°C in single-use aliquots (avoid freeze-thaw — koji enzymes lose 20–40% activity per cycle in our Mechanistic Extrapolation from generic soluble-protein literature; verify per batch in §6 calibration).

Whole-biomass assay (the §1.18 Koji-I arm): lyophilize whole koji, grind to <100 µm powder, resuspend in extraction buffer at the same w/v as the extract for matched comparison. The assay is otherwise identical; the readout difference between whole-biomass and free-extract arms is the cell-wall-encapsulation-protection effect.

Shio-koji liquid as-is (not extracted from solids) is acceptable for direct lipase / amylase / protease assay if the question is "what enzyme activity does the marinade actually deliver to a steak surface" rather than "what's in the koji rice itself." Salt content (typically ~25% w/w in the liquid phase) does suppress some enzymes' activity at standard assay conditions — dilute 1:10 minimum into assay buffer to mitigate, and run salt-matched controls.

6. Calibration — anchoring home tests against bench ground truth¶

The calibration discipline is the bridge that makes Tier 1 home assays clinically meaningful.

One-time calibration run (~half-day bench session): 1. Prepare one well-mixed sample extract (100 mL or more, single source, single time point). 2. Run Tier 3 assays for lipase / protease / amylase → publication-grade U/mL. 3. Run Tier 1 OR Tier 2 assays on aliquots of the same extract → relative-activity scores. 4. Fit a linear regression: bench U/mL = α × (home score) + β. The slope α is the calibration coefficient for this batch / equipment / operator.

Ongoing batch QC (all Tier 1 from then on): - Every batch is assayed at home using the same Tier 1 protocol, scored on the same scale. - Multiply by the calibration coefficient → estimated U/mL → estimated U/g fresh koji. - Report "this batch is X% of one Creon-cap-lipase-equivalent per gram koji" using the Tier 3 PERT comparison.

Recalibrate annually or when any of the following changes: koji starter (yellow ↔ white ↔ black), substrate (rice ↔ rice bran ↔ barley), incubation conditions (>5°C shift in mean temperature or >12 h shift in fermentation duration), home-assay equipment (new phone, new buffer batch).

Interpretive bands (post-calibration, for the lipase axis specifically — the EPI bottleneck):

Lipase activity per gram fresh koji	Clinical interpretation	Action
<1% of one Creon cap	Trace activity.	Shio-koji as a flavor-and-pre-digestion vehicle; not a meaningful PERT contributor.
1–5%	Plausible adjunct effect for moderate-fat meals.	Track in self-experiment-protocol.md — does GI tolerance shift on shio-koji-marinade days?
5–15%	Clinically interesting.	Consider whether it justifies dose reduction of a co-administered PERT cap on shio-koji-marinade meals (clinician discussion).
>15%	Surprising; verify against an outsource (Tier 4) before acting.	High-yield batch + high-lipase strain combination; document conditions, attempt to replicate.

These bands are Mechanistic Extrapolation from PERT dose-response physiology; they are not validated clinical thresholds. Treat as orienting heuristics, not decision rules.

7. Tier 4 — Outsourced contract assay¶

When to use: - One-time benchmarking of a strain × substrate combination before committing to Tier 3 in-house. - Independent verification of a surprising Tier 3 result before publishing or acting clinically. - Regulatory-grade activity reporting if any product-development path opens.

Vendor landscape (food / feed enzyme analytical labs as of 2026, sample-quote basis):

Vendor type	Typical service	Per-enzyme cost (single sample)	Turnaround
Major food/feed contract lab (e.g., Eurofins, Romer, Covance)	USP / FIP / DSC-equivalent units, full QC documentation	~$50–150	1–3 weeks
Specialty enzyme house (Megazyme, Novozymes contract services)	Substrate-specific activity assays (α-amylase, glucoamylase, lipase), often using the vendor's own substrate kits	~$80–200	2–4 weeks
University core fee-for-service	Custom assay design, often cheaper per sample if you're patient	~$30–100	4–8 weeks

Cost ranges are vendor-published 2025–2026 sample quotes; verify per-vendor before committing. (Current as of date on this page; refresh quarterly via the literature-mining queue if needed.)

What to send. A homogenized 50 g fresh-koji sample on dry ice, vacuum-sealed, with a one-page chain-of-custody note specifying: sample mass, fermentation start/end timestamps, substrate and starter used, intended assay (lipase / protease / amylase + which units the vendor should report). Most labs accept dry-ice ground shipping; overnight not required for short routes.

8. Unit conventions and cross-mapping¶

Unit	Definition (paraphrased)	Canonical for
USP lipase unit	Amount of enzyme that liberates 1 µmol fatty acid per minute from olive-oil emulsion under USP General Chapter <1601> conditions (37°C, pH ~9, specific substrate prep)	Pancreatic enzyme replacement (Creon, Zenpep, Pancreaze)
FIP unit	Per Fédération Internationale Pharmaceutique standard; substrate, temperature, and pH differ from USP — not directly interchangeable. Approximate empirical conversion: 1 USP lipase unit ≈ 4 FIP lipase units at standard assay conditions (source: digestive-enzyme-optimization.md §"BoulderBio FIP↔USP conversion")	Microbial-derived OTC digestive enzymes (BoulderBio, Now Foods, Pure Encapsulations)
IU (International Unit)	Generic enzymology — 1 µmol substrate transformed per minute at specified conditions	Bench enzymology, research literature
U (Sigma / supplier-specific)	Vendor-defined; always check vendor datasheet for assay conditions before comparing	Commercial assay kits

Always report the unit + the assay conditions. Cross-vendor comparison without specifying conditions is ~order-of-magnitude meaningful only.

9. Common pitfalls¶

No salt-matched control for shio-koji liquid. 25% NaCl suppresses some enzymes' apparent activity 30–60% at standard assay conditions; without a salt-matched blank, you under-report.
Reading kitchen-tier endpoints by eye. Operator drift across a single batch is larger than the inter-batch signal; use phone photos at fixed lighting.
Assuming USP and FIP units interconvert linearly across all enzymes. They don't — the conversion factor is enzyme- and condition-specific. Stay within one unit system per comparison; convert once at the end with a cited conversion factor.
Single-replicate assays. Lipase assays in particular are noisy (emulsion variability, temperature drift); n=3 minimum, n=9 if it's the first time on a new substrate.
Freeze-thawing extracts. Koji enzymes lose meaningful activity per freeze-thaw cycle; aliquot before first freeze; assay one aliquot fresh as a baseline.
Ignoring pH-optimum mismatch. White koji (A. luchuensis) acid protease optimum is pH ~3; reporting its activity at pH 7 (the convention for pancreatic protease) under-reports by 2–3-fold. Run at the strain's optimum if measuring intrinsic capacity; run at pH 7 if measuring expected gut-equivalent activity.
Not running a same-day positive control. Without a reference, a "low" reading is indistinguishable from an instrument or reagent issue. Always include a Sigma standard or a dissolved Creon capsule on every assay day.

10. Hooks into project work¶

Wild-type batch QC (koji-home-fermentation.md): Tier 1 amylase + protease batch QC after every Stage 1 harvest answers "did this batch ferment to a normal enzyme load?" before committing the batch to a 7–14 day Stage 2A shio-koji ferment. Lipase Tier 1 informs whether the resulting shio-koji is worth Tier 3 calibration this round.
Strain comparison (digestive-enzyme-optimization.md §"Yellow vs white vs black"): Tier 3 first-run on yellow koji vs. white koji vs. black koji from matched substrate batches converts the published U/g literature into household-equipment-anchored numbers, which then drive the strain choice for ongoing fermentation.
GI-survival 2-arm assay (validation-experiments.md §1.18): the Tier 3 assays specified here are the readouts §1.18 depends on. The §1.18 estimated cost ($300–500) covers the assay reagents under §3 above plus SGF/SIF reagents and Creon control.
Engineered platform benchmarking (engineered-koji-protocol.md): the Tier 3 wild-type baseline established here is the floor any engineered cassette must outperform. A useful platform metric: engineered-strain U/g per cassette / wild-type U/g per matched-strain control — values >2× justify the engineering complexity; <1.5× suggest substrate or condition optimization (already covered in digestive-enzyme-optimization.md) is the cheaper path.
n=1 self-experiment integration (self-experiment-protocol.md): every shio-koji-marinade-day in the PERT-timing experiment can carry a Tier 1 lipase score for the marinade used that day; correlate against GI-tolerance metrics over the experiment window.

11. Open questions this protocol enables answering¶

Each of these is currently called out as "lab-measurable but not yet done" in one of the related wiki pages. The methodology above operationalizes them.

What is the per-gram lipase activity of shio-koji from the household's standard fermentation, expressed as percent of one Creon-cap-lipase-equivalent? Tier 3 first-run. Settles whether shio-koji is a meaningful adjunct (>1%) or trace (<1%).
Does shio-koji-marinade pre-digestion measurably reduce the protease + peptidase load on the gut? Tier 3 protease assay on marinated vs. unmarinated protein extracts (not the marinade itself — the meat after marinating). Settles whether the "different mechanism" framing in §1 is empirically supported.
What is the GI-survival fraction of native koji enzymes through SGF → SIF, free-extract vs. whole-biomass? Specified in validation-experiments.md §1.18; methodology operationalized here.
Does A. luchuensis (white koji) have a meaningfully higher acid-protease yield than A. oryzae (yellow koji), and does this matter for shio-koji marinade efficacy at gastric pH? Tier 3 protease assay at pH 3 vs. pH 7 across matched batches.
What is the freeze-thaw stability of koji-derived lipase / protease / amylase? Tier 3 across 0, 1, 2, 5, 10 freeze-thaw cycles. Settles freezer-storage strategy for batched shio-koji production.
What is the salt-tolerance curve of native koji enzymes at shio-koji-relevant NaCl concentrations (0–25% w/v)? Tier 3 across a salt gradient. Informs whether reduced-salt shio-koji variants would deliver meaningfully more enzyme activity to food, or whether the salt concentration is already past the suppression knee.

12. What this protocol does not cover¶

Inhibitor and effector studies (e.g., the effect of fat in the meal matrix on lipase apparent Km). Out of scope; standard enzyme-kinetics protocols apply if pursued.
Identification of which lipase / protease isoforms are active. Requires SDS-PAGE + zymography or LC-MS proteomics; out of scope for a clinical-comparison protocol.
In vivo bioavailability. This protocol measures activity in vitro; the gut does not behave like a buffered cuvette. The §1.18 SGF → SIF bridge in validation-experiments.md is the closest in-scope proxy.
Engineered-strain construct verification. That work belongs in engineered-koji-protocol.md and the per-cassette stability pages (uricase-protease-stability-computational.md, etc.).
Allergenicity, mycotoxin, or contamination assays. Critical for any downstream consumption claim but out of scope for activity quantification. Standard food-safety panels apply.

13. Status¶

Protocol is drafted, not yet field-tested in this repo's household run. The first-run Tier 3 calibration is queued; estimated cost ~$200–400 in reagents at a community-college bench, single-day session. Linked to the Ward 1995 lab-access global landscape work — when the §1.9 dual-cassette feasibility-test partner is identified, the Tier 3 first-run is a low-risk warm-up assay that builds the relationship without committing the partner to the engineered work.

(Evidence level for this page: Mechanistic Extrapolation for the assay-tier mappings, calibration logic, and interpretive bands. In Vitro / Methods for the named canonical assay protocols (Bernfeld 1955, Charney & Tomarelli 1947, Winkler & Stuckmann 1979) and USP unit definitions (General Chapter <1601>). No Clinical Trial or Animal Model claims are made on this page.)