ABCG2 Modulators — Pharmacological Levers on the Gut Urate Sink¶

Gut ABCG2 is the apical-membrane efflux transporter that moves urate from blood into the intestinal lumen, accounting for ~30% of daily urate elimination. The engineered-uricase platform's "gut-lumen sink" thesis (see gut-lumen-sink.md) requires the substrate (luminal urate) to be there before the enzyme can act. ABCG2 is the gate that controls substrate supply.

This page maps the pharmacological levers on ABCG2 in the gut: what suppresses it (closing the gate), what induces it (opening the gate), what specifically rescues the most common gout-causing ABCG2 variant (Q141K), and where the existing supplement stack accidentally fights the platform thesis.

The landscape was characterized through PubMed scans on 2026-04-26 with primary-source verification. Evidence tiers are tagged inline.

Two distinct modulation modes — keep them separate¶

Most ABCG2 literature conflates two mechanisms with opposite implications for the platform thesis. Read every claim with this distinction in mind:

Mode	What it does	Implication for gut sink
Functional inhibition	Compound binds existing ABCG2 protein and blocks its pumping action. The pump is present but occupied.	Bad. Even if there is plenty of ABCG2 in the apical membrane, urate cannot be effluxed into the lumen. Studied heavily in oncology because tumor ABCG2 effluxes chemotherapy drugs.
Transcriptional modulation	Compound changes how much ABCG2 protein is produced (via nuclear receptors PPARγ, PXR, AhR, Nrf2, or via NFIB and other transcription factors).	Inducers are good — more pump = more urate efflux capacity. Suppressors are bad — fewer pumps regardless of activity.

The same compound can do both, in opposite directions, dose-dependent. Quercetin is a textbook case: at low μM cytosolic concentrations it is a competitive substrate/inhibitor (functional inhibition), but in some chronic-dosing animal studies it appears to upregulate ABCG2 mRNA (transcriptional induction). Net effect at a given dose is the integrated result, often poorly characterized in the gut-lumen context specifically.

Q141K trafficking rescue is a third, distinct mode discussed in §6 — relevant only to carriers of the Q141K polymorphism but mechanistically different from both functional inhibition and transcriptional induction.

Cross-reference comp-019 2026-05-08: Q141K trafficking rescue interventions (butyrate, fermentable fiber, HDIs) ADD to the gut-lumen uricase sink rather than replacing it. The flux model shows the gut-lumen sink mechanism works ACROSS genotypes — WT/WT shows the LARGEST predicted ΔSUA (−0.83 mg/dL at 25 mg/day mid-dose, 90% CI −1.13 to −0.57), Q141K homozygotes the smallest among typical-genotype patients (−0.50 mg/dL). Mechanism is multiplicative on residual ABCG2 capacity. Q141K-positive patients get a synergy bonus from rescue interventions because the rescue restores some of the residual ABCG2 capacity, which then provides more substrate for the gut-lumen sink to amplify. Q141K + butyrate + gut-lumen uricase is therefore a triple-mechanism intervention, but it is NOT a different mechanism than the WT/WT + gut-lumen uricase pathway — it's the same mechanism with rescued ABCG2 capacity.

The transcriptional regulation map¶

The most authoritative review of ABCG2/BCRP transcriptional regulation across tissues and species is Gorczyca & Aleksunes 2020, Expert Opinion on Drug Metabolism & Toxicology (PMID 32077332). Per that review, the major nuclear-receptor and transcription-factor regulators of ABCG2 are:

Pathway	Inducer signal	Tissue distribution	Gut selectivity
PPARγ	Endogenous: omega-3 metabolites, prostaglandin J2 derivatives. Pharmacologic: pioglitazone, fenofibrate (weak). Microbiome-derived: SCFAs, especially butyrate (Xie et al. 2020).	Adipose-dominant; substantial in gut, liver.	Moderate — also in adipose and liver.
AhR (aryl hydrocarbon receptor)	Indole-3-carbinol & DIM (cruciferous), tryptophan-derived AhR ligands from gut microbes (kynurenine, indole-3-aldehyde from Lactobacillus), some flavonoids.	Highest expression in barrier tissues (gut, skin, lung).	Strong gut enrichment.
Nrf2 / Keap1	Sulforaphane (broccoli sprouts), curcumin (mixed effects — see contradiction note in §8), DMF (drug).	Broad expression.	Moderate — also in liver, kidney, BBB.
PXR	Rifampicin, hyperforin (St. John's wort), some statins.	Gut, liver, also BBB.	Worst tissue selectivity — induces BBB ABCG2 too, which compromises CNS drug penetration.
HNF4α	Bile acids (FXR cross-talk), short-chain fatty acids via complex regulatory cross-talk.	Gut-enterocyte-specific in adult tissue.	Best tissue selectivity for the gut sink thesis.
NFIB (Nuclear Factor I B)	Endogenous expression varies by tissue and genotype (rs28379954).	Broad.	Less characterized; Solbakk 2025 (PMID 40554316) showed NFIB overexpression in Caco-2 enterocytes suppresses ABCG2 by 25–30%.

(All Mechanistic Extrapolation + In Vitro / Animal Model unless specifically tagged as Clinical.)

Key correction relative to first-pass deep dive: The original framing of "butyrate → HDAC inhibition → ABCG2 induction via HNF4α" was incomplete. Per Xie et al. 2020 (Acta Pharmacologica Sinica, PMID 32555444; DOI), butyrate has two independent effects on intestinal transporters:

HDAC inhibition → P-gp (ABCB1) downregulation via NF-κB / p65 — a separate phenomenon affecting drug-efflux pumps, not directly relevant to urate.
PPARγ activation → BCRP / ABCG2 induction — the urate-relevant effect.

Xie et al. specifically tested HDAC inhibitors (vorinostat, valproate) and TNF-α / NF-κB pathway disruption. None affected BCRP protein expression. PPARγ antagonist GW9662 abolished butyrate's BCRP induction. Butyrate's effect on the gut urate sink is therefore PPARγ-mediated, not HDAC-mediated, in wild-type ABCG2 carriers.

This refinement matters because it untangles the wild-type induction mechanism from the Q141K rescue mechanism (§6), which IS HDAC-mediated and operates independently. Brian-pattern Q141K-positive gout patients (estimable as ~30–50% of the population per published GWAS) would benefit from butyrate via both mechanisms simultaneously, on different alleles.

Suppressors of intestinal ABCG2¶

What is making the gate leaky in the typical gout patient:

1. Sex-dimorphic intestinal ABCG2 — covered in `androgen-urate-axis.md`¶

[REFRAMED 2026-05-07 per comp-016 evidence-mining.] Earlier framing here described "AR-mediated transcriptional repression of ABCG2 in gut and kidney." The comp-016 17-study primary-literature scan found zero primary studies demonstrating direct androgen-driven suppression of intestinal ABCG2 in vivo. No published ChIP-seq locates a classical androgen response element on the ABCG2 promoter. The closest mechanistic anchor (Jeong 2015 LNCaP, PMID 25615818) is an indirect CREB/CRTC2 axis in prostate cancer cells, not intestinal epithelium. One in vitro study (Klyushova 2023 Caco-2) shows testosterone INCREASES ABCG2 via PXR/FXR — opposite direction to the prior framing. MacLean 2008 (PMID 18378562) found NO sex difference in healthy rat intestinal ABCG2 across all four segments.

Updated framing [refined post-comp-017 full-text re-read 2026-05-07]: Intestinal ABCG2 is sex-dimorphic in a urate-relevant way (Hoque 2020 Nat Commun PMID 32488095 — male Q140K mice show 78% Western-jejunum / 88% combined Western+IHC intestinal ABCG2 protein loss + severe hyperuricemia; female Q140K mice protected), but the mechanistic driver is estradiol POSITIVE on the female side via PI3K/Akt (Yu 2021 Nutr Metab PMID 34144706, In Vitro + Animal Model — at strong-pharmacological E2 tier (100 µM, 5–6 orders above physiological); magnitude at physiological E2 unestablished per comp-017), not androgen NEGATIVE on the male side. The male-female asymptote difference at healthy baseline is empirically near-null (MacLean 2008, replicated by Tubic-Grozdanis 2020) and only emerges under disease state (e.g., Q141K-positive gout patients per Hoque 2020) or strong-pharmacological perturbation. The renal arm of the androgen-urate axis IS partially supported (Hosoyamada/Takiue 2010 PMID 20589576: T → URAT1 mRNA in mouse kidney; NEW per comp-017 full-text re-read: T → URAT1 protein UNCHANGED in non-orchiectomized animals; the actual protein-level renal drivers are Smct1 protein induction + GLUT9 attenuation) — but this is renal, not intestinal.

Klyushova 2023 nuance [post-comp-017 full-text re-read]: ALL three sex hormones (T, E2, P) at all three concentrations (1, 10, 100 µM — lowest active 30–100× above physiological free T) INCREASE Caco-2 ABCG2 via PXR/FXR. This is a xenobiotic-sensor response, not hormone-receptor-specific. The "T INDUCES not suppresses" framing direction is correct; the mechanism is xenobiotic-tier promiscuous induction.

Clinical-cohort anchoring (cohort-level, mechanism-not-isolated-to-intestine): Sakamoto 2018 (PMID 30557349) reports −0.66 mg/dL serum UA at 6 months ADT (n=150 ADT vs 339 surgery); Yahyaoui 2008 PMID 18349066 confirms FtM cross-sex T administration raises serum UA + decreases FEUA over 2 years. These are real and clinically meaningful but consistent with URAT1 mRNA + Smct1 protein + GLUT9 attenuation (renal mechanism) being the dominant transporter axis affected; they do NOT distinguish the intestinal ABCG2 mechanism. (Clinical — observational, not RCT.)

For SERMs (e.g., clomid): the androgen-driver effect on renal transporters propagates as in TRT (clomid raises endogenous T → renal mRNA + Smct1 protein + GLUT9 attenuation); the intestinal ABCG2 arm of the prior framing is not directly supported by primary literature and should be considered open. The H07 hypothesis (hypotheses/H07-clomid-intestinal-er-antagonism.md) proposes clomiphene's intestinal-ER antagonism as the mechanism; sub-claim 2 (clomiphene tissue-specificity at the gut) remains the open core untested by comp-017. (Mechanistic Extrapolation, supported by single-arm clinical observation; intestinal compartment uncertain.)

2. Inflammation / TNFα¶

Ferrer-Picón et al. 2020, Inflammatory Bowel Diseases (DOI, PMID 31211831). Patient-derived intestinal organoids (d-EpOCs) treated with TNFα showed suppressed SLC16A1 (MCT1), ABCG2, and GPR43, mimicking the expression profile of active IBD biopsies. Critically, IBD-derived organoids were not intrinsically less responsive to butyrate — TNFα was the proximal blocker.

Implication: chronic low-grade inflammation (Hashimoto's, food sensitivities, IBS, active IBD) produces the same gut ABCG2 suppression that high androgens do, by a parallel mechanism. In Vitro + clinical biopsy correlation. Both axes can act simultaneously and additively.

Lactoferrin as a TNFα-mediated ABCG2 rescue candidate. Habib et al. 2023 (PMID 37926296; Animal Model) and older Håversen et al. monocyte/macrophage studies document that oral/parenteral lactoferrin suppresses systemic TNFα. The composed mechanism — koji-derived lactoferrin → ↓ local TNFα drive → relief of TNFα suppression of intestinal ABCG2 → ↑ luminal urate substrate → ↑ effective co-expressed uricase activity — is Speculative (three Animal Model / In Vitro links, no published experiment in this combined geometry), but it positions lactoferrin as a potential substrate-supply synergist for the gut-lumen-sink platform, not merely a parallel NLRP3 modulator. The direct test is the lactoferrin rescue arm in validation-experiments.md §1.14 — a Caco-2 transwell experiment comparing lactoferrin basolateral rescue vs. butyrate PPARγ-mediated rescue at the worst-case (high TNFα) condition. Full mechanistic write-up in lactoferrin.md §4.7. (source: lactoferrin.md, koji-endgame-strain.md)

3. NFIB upregulation¶

Solbakk et al. 2025, Drug Metabolism and Disposition (DOI, PMID 40554316). NFIB overexpression in Caco-2 enterocytes suppressed ABCG2 by 25–30%. In humans, the rs28379954 T>C NFIB variant causes increased clozapine dose requirements consistent with reduced intestinal efflux. In Vitro + pharmacogenomic correlation.

This is a less-studied lever than the others; clinical relevance for urate is unestablished but mechanistically plausible.

4. Statins (mixed)¶

Some statin isoforms suppress ABCG2 expression, contributing to the well-documented modest UA rise on statin therapy (~0.1–0.3 mg/dL). Clinical observation + Mechanistic Extrapolation. Effect varies by statin (rosuvastatin > atorvastatin > pravastatin per limited data).

5. Western diet / low fiber¶

Low fiber → low colonic SCFA production → low PPARγ drive → baseline-suppressed ABCG2. Confounds with the "gout patients eat poorly" stereotype, but the proposed mechanism here is the gut ABCG2 axis specifically, not just purine intake. Mechanistic Extrapolation; clinical evidence in §9.

Inducers of intestinal ABCG2¶

Ranked roughly by evidence + safety + tissue selectivity:

Tier 1 — Strong evidence + good safety¶

Butyrate (via fermentable fiber → colonic SCFA production) - Mechanism: PPARγ activation in enterocytes (Xie et al. 2020; In Vitro + Animal Model) - Anchoring evidence: Animal model — sodium butyrate in HUA mouse model decreased serum UA AND restored intestinal ABCG2 expression (Li et al. 2023, Biomedicine & Pharmacotherapy, DOI, PMID 36948133). Animal Model with direct gout-relevant endpoint. - Practical lever: Fermentable fiber (resistant starch, inulin, GOS, beta-glucan) — not direct butyrate supplementation, which reaches the gut poorly. Aim ≥25–30 g fiber/day, with deliberate fermentable types in the mix. - Effect size in humans: ~0.25 mg/dL UA reduction on DASH diet, up to 0.73 mg/dL in baseline UA ≥8 mg/dL (§9, Juraschek 2021). - Tissue selectivity: good — colonic butyrate is largely consumed by colonocytes before significant systemic distribution.

Sulforaphane (Nrf2) - Mechanism: Nrf2 activation in enterocytes - Anchoring evidence: clinical safety established; broccoli-sprout extracts standardized; animal models show ABCG2 upregulation. In Vitro + Animal Model + clinical safety data. Direct UA endpoint not established. - Practical lever: Standardized broccoli-sprout extract (10–60 mg sulforaphane glucosinolate equivalent/day) or fresh broccoli sprouts (~50 g daily provides similar dose). - Tissue selectivity: moderate — also induces Nrf2 in liver and BBB. The BBB effect is a small concern for someone on lipophilic CNS-active drugs that depend on BBB ABCG2 for safety margins (rare in practice).

Tier 2 — Solid mechanism, modest evidence¶

Alistipes indistinctus / Hippuric acid (PPARγ — microbiome axis) - Mechanism: A. indistinctus produces hippuric acid via aromatic amino acid catabolism. Hippuric acid enhances PPARγ binding to the ABCG2 promoter AND promotes ABCG2 localization to brush border membranes via PDZK1 (a PDZ-domain scaffold that retains ABCG2 at the apical membrane). - Anchoring evidence: A. indistinctus gavage decreased serum urate to baseline in mouse models; A. indistinctus is depleted in hyperuricemia subjects. (Animal Model + Human Observational — Xu et al. 2024, Cell Host & Microbe 32(3):366-381.e9, PMID 38412863; last author: Yan Liu, Sun Yat-sen Univ.) - Practical lever: A. indistinctus is not commercially available as a probiotic. Dietary hippuric acid precursor routes: (1) polyphenol-rich foods (tea, berries, citrus) → gut catabolism → hippuric acid, OR (2) direct dietary benzoic acid → hepatic glycine conjugation → hippuric acid (no bacterial mediation needed). Cranberries are unusually high in natural benzoic acid; cranberry juice reliably elevates urinary hippuric acid (the basis of cranberry-UTI lore). A 4-week unsweetened cranberry juice n=1 (~8 oz/day) before/after panel is the cheapest test of whether the hippuric-acid → ABCG2 mechanism moves serum urate at dietary doses, independent of needing to colonize A. indistinctus. ~$20 cost. (Mechanistic Extrapolation for cranberry route; Animal Model + Human Observational for the hippuric acid mechanism itself.) - OE relevance: This is a second PPARγ → ABCG2 axis operating via a distinct gut microbiome member from the butyrate/PDB pathway — the two axes are additive. A combined prebiotic approach (fiber for PDB → butyrate + polyphenol foods for A. indistinctus → hippuric acid) could achieve higher ABCG2 induction than either alone. See purine-degrading-bacteria.md for the PDB / butyrate connection.

Indole-3-carbinol / DIM (AhR) - Mechanism: AhR activation; gut-AhR is highly active because the receptor evolved partly to sense gut microbial metabolites. - Anchoring evidence: In Vitro + Animal Model. Cruciferous-vegetable consumption epidemiologically associated with lower UA, though mechanism may be multifactorial. - Tissue selectivity: good (gut-enterocyte enrichment of AhR), though hepatic AhR is also active. - Caveat: DIM at high doses (>200 mg/day) has hepatotoxicity case reports. Stay at dietary or moderate-supplement levels.

Probiotic delivery of AhR-active microbes - Mechanism: specific Lactobacillus reuteri strains, L. plantarum, and others produce indole-3-aldehyde (a potent AhR agonist) from dietary tryptophan. - Evidence: Mechanistic Extrapolation. Direct urate endpoint untested for probiotic strain selection. Worth investigating.

Poria cocos / Wolfiporia cocos (茯苓 Fu Ling) — mechanism unidentified, in-vivo evidence robust - Mechanism: Empirically uncharacterized. Could be transcriptional (PPARγ / HNF4α / AhR / Nrf2 — none tested), chaperone-class trafficking rescue (analogous to UDCA / TUDCA per comp-032), or class I HDAC inhibition (no HDAC isoform was tested). The unidentified mechanism is itself the highest-leverage open question. - Anchoring evidence: Sun et al. 2021, Front Pharmacol (PMID 33651969). Hyperuricemic mouse model. Both ethanol and water extracts significantly elevated intestinal ABCG2 mRNA + protein. Water extract effect magnitude exceeded benzbromarone positive control (p < 0.01). Concurrent UA-lowering via XO inhibition + renal protection. Animal Model. - Five computationally predicted bioactives (PubChem CIDs 267, 277, 13824, 15730, 5759) docked to ABCG2; in vitro validation not yet performed. Pachymic acid mentioned only as HPLC reference standard, not implicated as active agent. - Traditional context: Poria cocos is a canonical ingredient of Si Miao San (the gout-indicated TCM formula already surveyed by comp-013) and Wu Ling San (the canonical urinary-tract / fluid-metabolism formula). Per comp-013's inputs/compounds.json, Poria cocos was implicitly included via the Si Miao San source citation but NOT separately spawned as a compound source — exactly the formula-completeness gap flagged in the 2026-05-19 query-framing retrospective audit. - Practical lever: Poria cocos extract is widely available as a TCM supplement (5-15g daily of decoction or 1-3g of standardized extract is the canonical dose range). Mechanism uncertainty makes off-label dosing for ABCG2 induction premature. - Tissue selectivity: Not characterized. - Next move: Caco-2 transwell assay (sister to §1.14 butyrate dose-response arm) — pachymic acid + polyporenic acid C + dehydroeburicoic acid panel against ABCG2 expression + Q141K trafficking. Marginal cost <$2K on existing infrastructure. High platform-relevance: closes the mechanism question on a TCM-grade compound that already has robust in-vivo HUA evidence.

Tier 3 — Pharmacological-grade levers¶

Fenofibrate (PPARα/γ + direct uricosuric effect) - Mechanism: PPARγ activation contributes to gut ABCG2 induction; the larger effect is direct URAT1 inhibition (uricosuric, lowers serum UA via increased renal excretion). - Anchoring evidence: Clinical trials show fenofibrate lowers serum UA ~10–15% in patients with mixed dyslipidemia/hyperuricemia. Clinical Trial. - Caveat: This is a drug, not a supplement. Requires prescription and monitoring (LFTs, creatinine, drug interactions including warfarin).

Pioglitazone (PPARγ) - Mechanism: strong PPARγ agonist. - Anchoring evidence: Clinical Trial for type 2 diabetes; ABCG2 induction is a downstream effect. Modest UA-lowering effect documented. - Caveat: Weight gain, fluid retention, modest bladder cancer signal. Not appropriate for non-diabetic gout patients on cost/risk grounds alone.

Tier 4 — Avoid for primary tissue-selectivity reasons¶

Rifampicin (PXR) - Induces ABCG2 in gut + liver + BBB. The BBB induction compromises the protective drug-efflux function (potentially increases neurotoxic drug penetration). Antibiotic use only — not a chronic urate strategy.

The Q141K rescue mechanism — a separate axis¶

The Q141K (rs2231142, p.Gln141Lys) ABCG2 polymorphism is the single largest genetic risk factor for hyperuricemia and gout (odds ratio ~2–3 across populations). Per Saranko et al. 2013, Biochemical and Biophysical Research Communications (DOI, PMID 23800412):

Q141K has a "mild processing defect" — protein folds imperfectly in the nucleotide-binding domain
Mistrafficked: significant fraction is retained in the aggresome (perinuclear protein-aggregation compartment) instead of reaching the apical membrane
Reduced ATPase activity (~50% of wild-type)
Defect is rescuable by low temperature in cell culture, indicating it is a folding/trafficking defect, not loss of catalytic core. In Vitro.

Basseville et al. 2012, Cancer Research (DOI, PMID 22472121) demonstrated:

HDAC inhibitors (vorinostat, romidepsin, others tested) rescue Q141K trafficking from aggresome to plasma membrane
HDIs restore wild-type-equivalent ABCG2 expression and substrate-efflux activity in Q141K cells
Mechanism is via altered microtubule motor protein expression (kinesins/dyneins involved in protein trafficking), not direct chromatin opening at the ABCG2 locus
In Vitro. No human RCT in Q141K-positive gout patients yet.

Implication for the platform thesis:

Butyrate is an HDAC inhibitor at colonic concentrations achievable through fermentable fiber. For Q141K-positive gout patients, butyrate hits two independent targets on ABCG2:

PPARγ-mediated induction of any wild-type allele (most Q141K carriers are heterozygous)
HDI-mediated trafficking rescue of the Q141K variant

This is mechanistically a stronger lever in Q141K carriers than wild-type carriers. Predicted but untested: Q141K + fiber → larger UA reduction than wild-type + fiber.

A pharmacogenomic-stratified RCT of fermentable fiber (or sodium butyrate enteric capsules) on serum UA, stratified by Q141K genotype, would either confirm or falsify this prediction. Cost-effective experiment for the platform's "differential responder" question.

Concentration-gap caveat (added 2026-05-16): The HDAC-rescue mechanism is in vitro-anchored at 1–5 mM butyrate (Basseville 2012). Whether endogenous colonic butyrate flux ever reaches HDAC-inhibitory concentrations at the enterocyte nucleus — after the mucus-layer + epithelial-surface gradient drop from luminal butyrate — is empirically open. Full framing at purine-degrading-bacteria.md §"Concentration gap". Wet-lab resolution: validation-experiments.md §1.14 butyrate dose-response arm — Caco-2 transwell, basolateral butyrate at five concentrations bracketing 0.05–5 mM in both WT and Q141K cells, dual readouts (trafficking + urate efflux). Until the dose-response runs, treat the in vivo Q141K rescue magnitude as mechanistically supported, quantitatively unverified.

Genotype-source framing — clinical, not consumer. The trial must generate its own genotype data at enrollment via a CLIA-certified clinical PCR assay for rs2231142 (Quest, LabCorp, or equivalent single-SNP genotyping; ~$40–80 per patient at clinical-lab pricing). Consumer-grade genotype exports (23andMe, Ancestry, etc.) are explicitly excluded as the data source — not as a personal-preference call but for trial-design rigor: reproducibility, documented assay performance, sample-chain-of-custody, and CLIA-grade QA are all preconditions for a publishable pharmacogenomic stratification. The trial framing is therefore agnostic at recruitment (no presumed prior genotype knowledge); patients are screened, genotyped on-site, and assigned to arms by the trial's own assay.

Population-frequency caveat for trial design. Q141K allele frequency varies substantially by ancestry — published GWAS report ~30–50% in East Asian gout cohorts but only ~10–15% in European-ancestry cohorts (some series lower). The "30–50% of gout patients" figure is correct for an East-Asian-majority cohort but overstates the carrier fraction in a European-American sample. A pharmacogenomic fiber trial should pre-specify population and powering — and, given that Q141K homozygotes are predicted to show the largest effect size, including a homozygous arm (where feasible) gives the cleanest signal-to-noise even at modest n. This framing is not "fiber for everyone with gout" — it's dual-action butyrate as a personalized intervention for Q141K-positive gout patients, with effect size scaling by allele dose.

The clinical-genotyping cost is also self-justifying against the alternative non-stratified design. In a European-ancestry cohort with carrier frequency ~10–15%, a non-stratified RCT looking for the same Q141K-conditional effect size would need roughly 3× the n to power the differential response (the Q141K-positive subset's signal is diluted by the wild-type majority). At ~$40–80/patient, on-site genotyping that lets the trial run with ~120 enriched patients instead of ~360 unstratified is a clear win — and is anyway required to identify Q141K homozygotes (~5–25% of carriers depending on population) for the homozygous arm. See androgen-urate-axis.md for the male-demographic ceiling that interacts with this stratification, and supplements-stack.md for ABCG2-inhibitor counter-indications that should be exclusion criteria at enrollment.

Pharmacological-chaperone route — orthogonal small-molecule rescue (added 2026-05-16)¶

The HDAC/butyrate rescue above is one of two distinct mechanistic strategies for Q141K. The other is a pharmacological chaperone — a small molecule that physically binds the misfolded Q141K protein and stabilizes its native conformation long enough for ER quality control to release it to the apical membrane. The class precedent is overwhelming: CFTR correctors (ivacaftor / tezacaftor / elexacaftor) rescue the ΔF508 CFTR variant via exactly this mechanism, and CFTR is in the same ATP-binding-cassette (ABC) transporter superfamily as ABCG2 (same fold, same general design problem). No published program has applied this chemistry to ABCG2 Q141K specifically.

This route is mechanistically orthogonal to the butyrate / HDAC track: a chaperone restores native folding directly, without needing microbiome-mediated HDAC inhibition or PPARγ-mediated transcriptional induction of a wild-type allele. Implications:

Works in Q141K homozygotes (no wild-type allele to induce — HDI rescue still applies, but PPARγ induction does not).
Independent of gut-microbiome state (no dependence on fiber intake, SCFA production capacity, or PDB colonization).
Daily-pill modality via compounding pharmacy if a hit is found among FDA-approved molecules (off-label / 503A) — see compounding-pharmacy-track.md.
Stacks additively with the HDI rescue (different mechanism, same target outcome).

The full chassis-pending entry is at chassis-pending-interventions.md §7 "Pharmacological chaperones for ABCG2 Q141K folding rescue". The cheapest first move — AlphaFold Q141K structure + virtual screen of FDA-approved molecules — was completed as comp-032 2026-05-16: GREEN. All four CFTR-corrector positive controls ranked in the top 11% of a 134-molecule curated library; the FDA-approved drug surface is NOT empty for ABCG2 Q141K chaperone candidates. Top wet-lab-priority candidates (ranked by composite score):

Lumacaftor (Tier 2, CFTR corrector) — strongest mechanistic prior; same ABC superfamily; on-patent for CF (Vertex), navigate patent landscape for off-label 503A
Tafamidis (Tier 2, TTR tetramer stabilizer) — aromatic-acid stabilizer at hydrophobic interface; misfolded-state selective
Ursodiol / UDCA (Tier 1, bile acid chaperone) — broad ER-stress chaperone via ATF6/Hsp70; F508del-CFTR rescue precedent; off-patent USP/NF monograph
Diflunisal (Tier 1, lowest-friction first call — off-patent NSAID with USP/NF monograph + off-label ATTR-stabilization precedent) — anionic at pH 7.4, strongest electrostatic match for Q141K +1 pocket
TUDCA (Tier 2, bile acid chaperone) — CNS-penetrant; F508del-CFTR + ALS-clinical-trial precedent

The Q141K folding-rescue thesis stands; pivot to AI-aided novel binder design (RFdiffusion) is NOT triggered. Next move: per-hit cell-based Q141K trafficking-rescue assay on the top 3-5 candidates in a Caco-2 Q141K-transfected line (sister to the §1.14 butyrate dose-response arm — same Caco-2 infrastructure, marginal cost ~$500–1,500 for the chaperone-class screen on top of the existing factorial). Compounding-pharmacy partner conversation can lead with diflunisal as the Tier-1 off-patent low-friction first call.

Tissue selectivity matters¶

ABCG2 is expressed at multiple barrier sites with different physiological roles:

Tissue	ABCG2 role	Inducing it does what?
Intestinal apical membrane	Effluxes urate, drugs, xenobiotics into gut lumen	Good for urate platform
Renal proximal tubule	Effluxes urate from cell into urine	Good for urate (parallel renal sink)
Blood-brain barrier	Effluxes drugs out of brain (protective)	Bad — reduces CNS penetration of e.g. statins, some psychotropics, some antibiotics
Placenta	Effluxes drugs from fetal circulation	Clinically irrelevant for adult gout
Hepatic canalicular membrane	Effluxes substrates into bile	Mixed — affects drug clearance kinetics

Pick gut-selective inducers, not pan-tissue ones. The HNF4α and PPARγ axes are relatively gut-enriched (PPARγ also adipose/liver but minimal BBB). PXR is the worst — gut/liver/BBB simultaneously, so rifampicin is a no-go for chronic use.

A patient on chronic CNS-active medication (e.g., SSRIs, antipsychotics, anti-epileptics) should be cautious about pan-tissue ABCG2 inducers — measurable drug-level changes can result.

The supplements-stack contradiction¶

Several compounds in supplements-stack.md are functional ABCG2 inhibitors at typical supplement doses, even though they were added for anti-NLRP3 or anti-inflammatory reasons. For male gout patients on TRT or supraphysiological clomid (the highest-leverage demographic), the stack may be working pharmacologically against the platform thesis.

Documented functional ABCG2 inhibitors at supplement-relevant doses:

Compound	Inhibition tier	Source
Curcumin	Established BCRP/ABCG2 inhibitor in vitro (Ki ~5–10 μM). At 500–1000 mg supplement doses, gut-lumen concentrations easily reach this.	Pharmacology literature, multiple in vitro studies
Quercetin	Substrate/inhibitor at low μM (functional); transcriptional upregulation in chronic dosing (mixed). Net effect on gut sink: probably negative acutely.	Pharmacology + nutritional biochemistry literature
EGCG	Functional BCRP inhibitor in pharmacology assays. Yu et al. 2024 (Food Funct, PMID 38757391) showed mouse PO-induced hyperuricemic model net-favorable effect on ABCG2/URAT1/GLUT9 expression at the tissue level — direction opposite to the in vitro inhibition story. Net clinical effect on gut sink: unresolved.	Mixed: pharmacology in vitro vs. animal model in vivo
Genistein / soy isoflavones	Established BCRP substrate-inhibitor. Dietary intake from natto/miso is much smaller than supplement doses.	Pharmacology literature

Risk-tier stratification (propagated from supplements-stack.md §"Stack-level contradictions" 2026-04-27; updated framing 2026-05-05):

User profile	ABCG2 status	Risk tier	Practical implication
Q141K homozygote + androgen-suppressed (TRT / SERM / AAS) + high-dose flavonoid (>500 mg quercetin OR >600 mg EGCG OR >500 mg curcumin)	Triple-hit suppressed	Highest concern	Gut sink may be functionally closed during dose window. Pause inhibitor flavonoids; favor inducers (sulforaphane, fermentable fiber → butyrate, §"Inducers of intestinal ABCG2" above).
Q141K heterozygote OR androgen-dominant (high-T, no SERM) + supplement-grade flavonoid	One axis suppressed + acute pharmacological inhibition	High concern	Meaningful gut-sink narrowing during the dose window. Time inhibitor flavonoids away from urate spikes (post-fructose meals, peri-flare). Acceptable with UA monitoring.
Wild-type ABCG2 + supplement-grade flavonoid	Pharmacological inhibition only	Moderate concern	Net effect dose- and chronicity-dependent. Watch UA trajectory after introduction; down-titrate if UA rises.
Any genotype + dietary-level flavonoid (onions, tea, turmeric, fermented soy at normal food portions)	Sub-Ki gut concentrations	Minimal concern	No restriction. Food-level intake is unlikely to be clinically significant for the gut sink.

Stratification matters because a blanket "avoid quercetin" message undermines compliance for the largest cohort (wild-type genotype, dietary intake) where the risk is essentially zero. The clinically meaningful signal concentrates in androgen-suppressed Q141K-positive readers at supplement-grade doses.

Practical inference for high-T or Q141K-positive patients: avoid high-dose curcumin and quercetin acutely when the gut sink matters most (post-meal urate spikes, fructose challenges, etc.). Dietary-level intake of these compounds (turmeric in food, onions, tea) is unlikely to be problematic; supplement-grade doses are the concern.

This also surfaces a research-level open question: how much of the "non-responder" rate in nutraceutical gout RCTs is explained by ABCG2-inhibitor co-supplementation rather than by per-compound efficacy failure?

Human clinical evidence — what the RCTs actually show¶

Dose-anchoring is essential because in vitro and animal effect sizes don't always translate.

Non-CKD adults — fiber/DASH effect is real but modest¶

Juraschek et al. 2021, Arthritis & Rheumatology (DOI, PMID 33615722): secondary analysis of the DASH feeding study, n=327 adults with mild-to-moderate hypertension.

DASH diet (high fiber + low-fat dairy): mean serum UA reduction 0.25 mg/dL vs. control (p=0.004)
Fruit-and-Vegetables diet alone: 0.17 mg/dL reduction (p=0.051, borderline)
Effect dose-dependent on baseline UA severity:
UA <5: 0.08 mg/dL reduction
UA 5–5.9: 0.12
UA 6–6.9: 0.42
UA 7–7.9: 0.44
UA ≥8: 0.73 mg/dL reduction (P-trend = 0.04)

Clinical Trial. This is gold-standard evidence. The effect is small but reproducible and dose-dependent on baseline severity, which is exactly the pattern expected from a mechanism that opens the gut sink against a variable starting load.

CKD/dialysis — mixed; meta-analysis says null for UA specifically¶

He et al. 2021, European Journal of Nutrition (DOI, PMID 34491388): randomized crossover trial of inulin-type prebiotics in 16 peritoneal dialysis patients, 12 weeks.

Serum UA reduced ~10% in prebiotic phase vs. placebo (p=0.047)
Mechanism attributed to enhanced fecal UA degradation (microbial uricolysis), not increased renal excretion
Microbiota changes: enrichment of purine-degrading species (Anaerostipes caccae, Clostridium species)

Clinical Trial (small, single-center).

Khosroshahi et al. 2019, Nutrition & Metabolism (DOI, PMID 30911321): RCT of resistant starch in 44 maintenance hemodialysis patients, 8 weeks. Reduced creatinine and uric acid (p<0.05). Reduced p-cresol. Smaller study, supportive.

But Wathanavasin et al. 2025, Toxins (DOI, PMID 39998074) — meta-analysis of 21 RCTs and 700 CKD patients on dietary fiber (6–50 g/day, ≥4 weeks):

Significant reductions: p-cresyl sulfate, indoxyl sulfate, BUN, IL-6, TNFα
No significant reduction in serum uric acid (or TMAO, hs-CRP)

Clinical Trial — Meta-Analysis. Across the broader CKD literature, fiber lowers other uremic toxins and inflammation but the UA signal does not survive aggregation.

Reconciliation: in advanced CKD, reduced renal clearance dominates the UA balance. Even if gut ABCG2 is upregulated, the systemic UA pool is dominated by impaired filtration. The Juraschek/DASH effect was in non-CKD adults — the gut-sink lever has room to matter when renal function is intact. The platform thesis applies to non-CKD gout patients first; CKD is a confounding context.

Engineering implications¶

For the engineered koji (uricase + lactoferrin endgame strain)¶

Three potential additions to the koji platform that would couple substrate-supply (ABCG2 induction) with substrate-degradation (uricase):

Glucoraphanin co-production. A. oryzae can be cultured on cruciferous substrates (or engineered to produce glucoraphanin from glucose precursors). Glucoraphanin is the sulforaphane precursor; gut myrosinase from cruciferous-resident bacteria converts it to active sulforaphane. Co-delivery of uricase + glucoraphanin would pair "degrade urate in lumen" with "induce more urate transport into lumen" in a single product. Speculative — synthetic biology feasibility not yet assessed for this specific coupling.
Resistant-starch-rich substrate. The koji rice itself is the substrate. Engineering or selecting rice strains with higher resistant-starch content → more colonic butyrate → more PPARγ drive on ABCG2. The protein/enzyme delivery vehicle becomes a fiber-delivery vehicle simultaneously. Mechanistically plausible; needs quantification of incremental SCFA yield from engineered vs. wild-type substrate.
Tryptophan-pathway probiotic co-strain. L. reuteri or similar AhR-agonist-producing strain co-formulated with the engineered S. boulardii or S. cerevisiae uricase strain. Each contributes a different mechanism to the gut sink. Speculative — strain compatibility under fermentation conditions and survival post-ingestion not yet established.

For an engineered LBP peer track (commercial-pharmaceutical chassis)¶

The butyrate dual-action documented above (PPARγ → WT ABCG2 induction + class-I HDAC inhibition → Q141K trafficking rescue) is the cleanest genotype-agnostic mechanism in this entire page. Most ABCG2-relevant interventions hit either WT or Q141K, not both. Butyrate hits both simultaneously — but its bioavailability problem (rapid small-intestinal absorption before reaching the colon) makes oral butyrate dosing impractical at clinically meaningful colonic concentrations.

An engineered colonically-resident butyrate producer solves the bioavailability problem at the dose-frequency level. Faecalibacterium prausnitzii, a strict-anaerobe colonic resident already producing butyrate natively at 3–5% gut abundance, is the obvious chassis. Engineered overproduction would put butyrate where it needs to be (the colonic crypt, against the epithelium) on a continuous basis after a single quarterly capsule rather than requiring constant oral dosing.

This is fundamentally a commercial-pharmaceutical track, not a koji-style home-fermentation track — obligate anaerobes require anaerobic bioreactor manufacturing, cold-chain stabilization, and the FDA Live Biotherapeutic Product (LBP) regulatory path. See engineered-lbp-chassis.md for the full peer-track scope, regulatory discussion, and queued Phase 2 follow-ups (lit scans on engineering toolkit, commercial landscape, FDA LBP path; comp-008 expression feasibility analysis; falsification card H02). The two tracks (koji + LBP) serve complementary patient populations: koji for the broader market and prevention, LBP for the high-severity / Q141K-genotype / refractory subset where pharmaceutical-grade durability justifies the cost-and-distribution overhead.

For the supplements stack catalog¶

Per-compound stack-interaction fields including ABCG2 status should be a required field for any new compound considered for supplements-stack.md. The four functional inhibitors documented above are the current contradiction set; new entries that meet that pharmacology profile should carry a counter-indication note for high-T or Q141K-positive readers.

Open research questions¶

Q141K × fiber differential response. Predicted but untested: Q141K-positive gout patients should show larger UA reduction with fermentable fiber than wild-type-positive patients, because butyrate hits both PPARγ-induction and HDI-rescue. A pharmacogenomic-stratified RCT (n ~120, baseline UA ≥7 mg/dL, stratified Q141K hetero/wild-type by on-site CLIA-grade rs2231142 PCR genotyping at enrollment — see §6 for genotype-source rationale, 12-week inulin or equivalent) would resolve this. Estimated cost: $150K, 6 months. High platform-relevance.
Tissue-selective PPARγ agonists. Pharmacology question: are there PPARγ agonists with gut-enrichment selectivity (sparing adipose, liver, BBB)? Selective PPARγ modulators (SPPARMs) are an active drug-development area; some have differential tissue activity. Worth a desk review.
Engineered koji + glucoraphanin co-production. Synthetic biology desk review: is co-expression of glucoraphanin biosynthetic pathway feasible in A. oryzae? The plant pathway involves several enzymes; whether a fungal host can support it is open.
EGCG net effect on the gut sink. Yu 2024 (PMID 38757391) shows favorable in vivo phenotype despite EGCG being a known functional ABCG2 inhibitor in vitro. Resolution requires direct in vivo measurement of gut ABCG2 protein/function before and after EGCG dosing in a relevant model.
Inflammation-suppression overlap with androgen suppression. TNFα and androgens act independently on ABCG2 expression. For patients with both high T and chronic inflammation (e.g., Hashimoto's), is the suppression additive or saturating? Animal model question; can be tested with TNFα + DHT co-treatment vs. either alone.
Stack-level audit of nutraceutical gout RCT non-responders. Hypothesis: a measurable fraction of the ~30% non-responder rate in fiber/polyphenol gout RCTs is explained by ABCG2-inhibitor co-supplementation in those participants. Tractable as a meta-analytic question if the RCTs reported concomitant supplement use (often they do).

Provenance and citation tier¶

All claims tagged with evidence level: Clinical Trial / Animal Model / In Vitro / Mechanistic Extrapolation. Primary sources for the most actionable claims:

Transcriptional regulation map: Gorczyca & Aleksunes 2020 (DOI, PMID 32077332) — review of TF-mediated BCRP regulation across tissues.
Microbiome-transporter axis: Rzeczycki et al. 2025 (DOI, PMID 41465322) — recent review specifically on gut microbiota → intestinal drug transporters via SCFAs / bile acids / indole metabolites.
PPARγ as the wild-type ABCG2 induction mechanism: Xie et al. 2020 (DOI, PMID 32555444) — primary study, rat in vivo + Caco-2 in vitro.
Q141K trafficking defect: Saranko et al. 2013 (DOI, PMID 23800412).
HDI rescue of Q141K: Basseville et al. 2012 (DOI, PMID 22472121).
TNFα suppression of intestinal ABCG2: Ferrer-Picón et al. 2020 (DOI, PMID 31211831).
Animal model gout efficacy: Li et al. 2023 (DOI, PMID 36948133).
Human RCT — DASH/fiber: Juraschek et al. 2021 (DOI, PMID 33615722).
Human RCT — inulin in CKD (positive): He et al. 2021 (DOI, PMID 34491388).
Meta-analysis — fiber in CKD (UA null): Wathanavasin et al. 2025 (DOI, PMID 39998074).
NFIB regulation: Solbakk et al. 2025 (DOI, PMID 40554316).

Information retrieved via PubMed on 2026-04-26.