Phase 5 Deep-Read: AMC-BFE Cordyceps×Astragalus (PMID 41905012, 2026)¶
Citation. Xia X, Song H, Meng Y, Xu C, Niu H, Liu X, Zhang G, Ling J. Biotransformation-derived metabolites from Astragalus membranaceus and Cordyceps militaris alleviate hyperuricemia via multi-target regulation. Bioorg Chem 175:109806 (2026). DOI 10.1016/j.bioorg.2026.109806. PMID 41905012.
⚠️ Access caveat — read this first¶
The full text of this paper is paywalled (Elsevier, no PMC deposit, no preprint mirror found as of 2026-05-06). This deep-read therefore reconstructs the study from:
- The PubMed abstract + MeSH + keywords (full text retrieved).
- The methodological analog from the same group: Zhao X, Li X, Jia X, Gao Y. Multi-omics and compositional analysis of bidirectional solid fermentation products from Cordyceps militaris and Panax quinquefolius L. PMC12805540 (2025). Same Shandong group (Ling J, Zhang G are corresponding on the AMC-BFE paper; Zhao 2025 is the prior bidirectional-fermentation methods paper from the same lab consortium).
- The literature ancestor for the C. militaris × URAT1 axis: Yong T et al. Cordycepin… ameliorates Hyperuricemia through URAT1 in Hyperuricemic Mice. PMC5788910 (2018). The HUA-mouse methodology and URAT1 readout convention this paper inherits.
- Cross-validation against the closest published 2025 SSF cordycepin co-fermentation paper (Zhao 2025).
What this means for downstream use. Every numerical claim in this document that is NOT in the published abstract is marked [METHODS-INFERRED FROM ANALOG] or [NOT IN ABSTRACT]. The 4 chokepoint hits (URAT1 / GLUT9 / ABCG2 / PPARα) are confirmed by the abstract as gene-expression measurements, but the direction, magnitude, p-values, and tissue specificity at the level of individual transcript fold-change cannot be verified without the full text. Treat this deep-read as a provisional read; the canonical numbers must be lifted from the full PDF before any wiki page cites them.
1. Extract characterization¶
What the abstract states explicitly: AMC-BFE = "ethanol extract obtained from bidirectional solid-state fermentation of Astragalus membranaceus and Cordyceps militaris". The naming convention (AMC = Astragalus membranaceus + Cordyceps; BFE = Bidirectional Fermentation Extract) is consistent with the Shandong-Ling-lab pattern for this class of co-fermentation products.
Whole-extract evidence, not isolated compound. The abstract says: "Chemical profiling by LC-MS/MS showed that bidirectional fermentation enriched several bioactive metabolites, including isoflavone aglycones, nucleosides, and polyols." The active component is NOT identified at single-compound resolution — the abstract explicitly hedges: "These compositional differences may be related to the observed urate-lowering activity, although the contribution of individual metabolites requires further investigation."
Inferred chemical classes (from the named families + the parent biology): - Isoflavone aglycones — from Astragalus: calycosin and formononetin (deglycosylated forms of calycosin-7-O-glucoside / ononin / astragaloside). The "aglycone" word matters: bidirectional SSF deglycosylates the parent isoflavones, producing the more bioavailable aglycone forms. This is a known C. militaris β-glucosidase activity. - Nucleosides — almost certainly cordycepin (3'-deoxyadenosine), adenosine, and possibly cordycepic acid. Cordycepin is the canonical C. militaris anti-HUA agent (Yong 2018, PMC5788910 — independently shown to downregulate URAT1). - Polyols — likely mannitol (cordycepic acid) and possibly D-arabinitol/erythritol from the rice substrate processed through fungal metabolism.
Conclusion: AMC-BFE is a chemically-defined fingerprint (LC-MS/MS profile is reported), but mechanistically a whole-extract. No single active component is causally pinned to any single chokepoint. The 4-chokepoint claim is whole-extract evidence; per-chokepoint compound attribution is absent.
2. Fermentation method¶
Confirmed from abstract: "bidirectional solid-state fermentation". Not liquid; not co-culture in a stirred tank. SSF = the C. militaris mycelium colonizes a solid substrate that includes Astragalus.
[METHODS-INFERRED FROM ANALOG] Zhao 2025 (PMC12805540) protocol from the same Shandong consortium, which is almost certainly the protocol AMC-BFE follows with substrate substitution:
- Substrate: rice base + ground botanical (sieved through 65-mesh, ~250 μm). For AMC-BFE, the analogue would be rice + Astragalus membranaceus root powder.
- Inoculum ratio (analog): 30% botanical / 70% rice by total solid weight, optimized in preliminary experiments. AMC-BFE may use a different optimum — not verifiable without full text.
- Strain: C. militaris (the Zhao paper uses strain X6, ITS-confirmed). Spore solution prepared in PDA liquid medium, 25°C, 180 rpm, 48 h.
- Inoculation: ~10 mL spore suspension into solid substrate.
- Conditions: 25°C, dark for 5–7 days until mycelial colonization → switch to 10 h:14 h light:dark cycle, 10°C+ day-night gradient, 60% humidity, 200 lux, total 50 days.
- Sample collection: typical timepoints 7, 15, 20, 25, 30, 40, 50 days. Final material ground and sieved through 65-mesh.
- Extraction: abstract says "ethanol extract" — solvent details not specified.
Bidirectionality clarification: "Bidirectional" SSF in this literature refers to mutual biotransformation — the fungus modifies plant secondary metabolites (deglycosylation, hydroxylation, methylation), AND the plant matrix modifies fungal metabolite output (e.g., the Zhao 2025 paper showed American ginseng addition changes cordycepin yield in C. militaris). It is not sequential fermentation and not co-culture of two organisms; it is a single fungus on a botanical-supplemented solid substrate.
Reproducibility judgment: The abstract confirms the category of method (SSF + bidirectional); the exact AMC-BFE recipe (substrate ratio, day count, extraction solvent ratio, yield) is not in the abstract. Methods section reproducibility cannot be confirmed without the full paper. The analog paper IS reproducible from its published methods, so we have a strong prior that AMC-BFE methods are likewise reproducible — but we cannot verify the AMC-BFE specifics.
3. In vivo mouse data — exact numbers¶
[NOT IN ABSTRACT] Most numerical detail is inaccessible. The abstract confirms qualitative findings only:
| Parameter | Abstract says | Full numerical detail |
|---|---|---|
| Model | Potassium oxonate + adenine HUA mouse | Confirmed |
| Strain | C57BL/6 mice (from MeSH terms) | Confirmed |
| Sex | Male (from MeSH) | Confirmed |
| n per group | — | NOT IN ABSTRACT |
| Dose | — | NOT IN ABSTRACT (MeSH includes "Dose-Response Relationship, Drug" → suggests multi-dose) |
| Route | — | NOT IN ABSTRACT (gavage is conventional for this model) |
| Duration | — | NOT IN ABSTRACT |
| Baseline SUA | — | NOT IN ABSTRACT |
| Treated SUA | — | NOT IN ABSTRACT |
| % reduction | "significantly lowered serum uric acid levels" — direction confirmed, magnitude not stated | NOT IN ABSTRACT |
| p-values | — | NOT IN ABSTRACT |
Confirmed qualitative findings: - SUA significantly reduced ✅ - XOD activity suppressed ✅ - ALT, AST, BUN, serum creatinine ameliorated ✅ - Total cholesterol and triglycerides reduced ✅ - Multi-dose study likely (MeSH "Dose-Response Relationship") ✅
Pattern from analog (Yong 2018, PMC5788910 cordycepin paper): typical n=10 per group, gavage administration, 7-day treatment, baseline SUA in HUA mice ~3.5–6 mg/dL elevation over normal (normal mouse SUA ~1.5 mg/dL → HUA SUA ~5–7 mg/dL), allopurinol positive control at 5 mg/kg, test compound 5–50 mg/kg dose range, % reduction commonly 30–60% at the high dose. AMC-BFE numbers should fall in this range but cannot be confirmed.
4. Per-chokepoint evidence¶
[CONFIRMED FROM ABSTRACT] All four chokepoints are measured at the gene-expression level. No protein-level Western blot or activity assay is referenced in the abstract. No reporter assay for PPARα is referenced.
| Chokepoint | Evidence type | Direction | Tissue | Magnitude / p-value |
|---|---|---|---|---|
| URAT1 (SLC22A12) | mRNA (qRT-PCR implied) | Downregulated ✅ | Kidney ✅ | NOT IN ABSTRACT |
| GLUT9 (SLC2A9) | mRNA | Downregulated ✅ | Kidney ✅ | NOT IN ABSTRACT |
| ABCG2 | mRNA | "Altered expression" — direction not specified in abstract ⚠️ | Liver ✅ (hepatic ABCG2, NOT intestinal) | NOT IN ABSTRACT |
| PPARα | "Activated the hepatic… signaling pathway" | Activated (gene expression downstream targets implied) | Liver | NOT IN ABSTRACT — no reporter, no co-IP, just pathway-level inference from RNA |
Two important hedges that the Phase 4 hit summary may have glossed over:
-
ABCG2 is hepatic, not intestinal. This is unusual. The canonical ABCG2 chokepoint for urate is the intestinal one — ABCG2 in the gut secretes urate into the lumen, and gut ABCG2 dysfunction causes ~⅓ of all hyperuricemia (the "extra-renal underexcretion" phenotype). Hepatic ABCG2 is a different story — it likely affects bile-route urate excretion, which is a minor pathway in mice and humans. The chokepoint mapping to "ABCG2 = the intestinal Open Enzyme target" is therefore weaker than first read.
-
ABCG2 direction is "altered," not "upregulated." This is the abstract author's word — vague. If AMC-BFE downregulated hepatic ABCG2, that's neutral or net-negative for the urate-handling story. If upregulated, that increases bile-route excretion. The Phase 4 hit-list interpretation should be re-examined when the full text is read.
-
PPARα activation is not measured by reporter assay (per abstract). It is "pathway activation" — typically meaning downstream PPARα target gene transcription (acyl-CoA oxidase, CPT1, FGF21, etc.) is upregulated by qRT-PCR. This is suggestive but not direct ligand-binding evidence. The PPARα claim should be tier-2 evidence (gene-expression downstream targets, not direct receptor activation).
5. Compound identification — fractionation?¶
No. The abstract explicitly states: "the contribution of individual metabolites requires further investigation." This is the authors' own caveat that the per-compound mapping is not done. AMC-BFE is whole-extract evidence end-to-end. No bioassay-guided fractionation is referenced. No compound-by-compound URAT1 / GLUT9 / ABCG2 / PPARα assay is referenced.
Implication for Open Enzyme triage: the candidate hits among the LC-MS/MS-identified metabolite classes (cordycepin, calycosin, formononetin, mannitol/cordycepic acid, adenosine) all have prior literature linking them to one or another of these chokepoints. Cordycepin → URAT1 (Yong 2018). Calycosin → PPARα (multiple Chinese-source papers). Mannitol → osmotic / non-specific. The mechanism diversity is plausible and aligns with prior single-compound literature, but the AMC-BFE paper does not deconvolve which compound drives which chokepoint.
6. Other mechanisms — NLRP3 / complement / Lp-PLA2 / HDAC6 / redox?¶
[NOT IN ABSTRACT — full-text check needed]
- NLRP3: not mentioned in abstract or MeSH.
- Complement: not mentioned.
- Lp-PLA2: not mentioned.
- HDAC6: not mentioned.
- PDI / TXNIP / thioredoxin / ergothioneine: not in abstract. However, abstract says "Untargeted metabolomics further identified alterations in gut microbiota-derived metabolites linked to antioxidant pathways." This is a weak signal that the redox/antioxidant axis is touched, but the specific molecules (ergothioneine — C. militaris is a known ergothioneine producer; TXNIP — NLRP3-relevant; thioredoxin — redox poise) are NOT named in the abstract.
Action item for the full-text follow-up: specifically grep the full PDF for: "ergothioneine," "TXNIP," "thioredoxin," "PDI," "NLRP3," "complement," "Lp-PLA2," "HDAC6." If ergothioneine is identified by LC-MS/MS, this is a major plus — C. militaris ergothioneine production is well-documented (PubMed 30+ papers), and ergothioneine is one of the few mammalian-relevant fungal antioxidants with a dedicated transporter (OCTN1/SLC22A4).
7. Author affiliation, Chinese-source evidence¶
Corresponding authors: - Guoying Zhang, School of Pharmacy, Shandong University of Traditional Chinese Medicine, Jinan 250355, China. - Jianya Ling, School of Pharmacy + State Key Laboratory of Microbial Technology, Shandong University, Qingdao 266237, China.
All authors Chinese-affiliated. The paper is published in Bioorganic Chemistry (Elsevier, English-language journal; SCImago Q1; impact factor in 2026 not yet final but historically ~5). This is a Chinese-research, English-publication paper — exactly the class of research the umbrella-CLAUDE.md global-multilingual rule says we should NOT discount. The Shandong U TCM + Shandong U State Key Lab Microbial Tech consortium has a track record of bidirectional SSF papers (Zhao 2025 Cordyceps×Panax is the same consortium's prior work).
Chinese-source cross-check: the related-Chinese-literature on Astragalus + C. militaris co-fermentation in CNKI / WanFang would likely show 5–15 prior Chinese-language papers from the same lab and adjacent labs. None checked here yet — but the existence of the lineage suggests this is a mature methodological line, not a one-off finding. Confidence in the methods reproducibility is therefore higher than a first-paper-from-a-new-group claim of similar magnitude.
Summary of what's verified vs. what needs the full PDF¶
| Claim | Status |
|---|---|
| AMC-BFE reduces SUA in HUA mice | ✅ Confirmed (abstract) |
| AMC-BFE inhibits XOD activity | ✅ Confirmed (abstract) |
| Renal URAT1 downregulated | ✅ Confirmed direction (abstract); magnitude unverified |
| Renal GLUT9 downregulated | ✅ Confirmed direction; magnitude unverified |
| Hepatic ABCG2 modulated | ⚠️ Direction NOT specified in abstract |
| Hepatic PPARα pathway activated | ✅ Pathway-level, not direct ligand binding |
| Liver/kidney function markers improved | ✅ Confirmed |
| Whole-extract, no compound-level deconvolution | ✅ Confirmed by author hedge |
| Gut microbiota antioxidant metabolites altered | ✅ Confirmed (abstract) |
| Bidirectional SSF chemistry shifted (isoflavone aglycones, nucleosides, polyols enriched) | ✅ Confirmed (abstract) |
| Exact mouse n, dose, % SUA reduction, p-values, fold-changes | ❌ Need full PDF |
| NLRP3, ergothioneine, TXNIP, complement, HDAC6, Lp-PLA2 | ❌ Need full PDF (not in abstract) |
| Toxicity / body-weight / safety data | ❌ Need full PDF |
Citation hygiene¶
Per umbrella CLAUDE.md and the "Pre-commit grep-verify gate" rule in Open Enzyme/CLAUDE.md §4, no number from this paper should propagate to wiki pages until the full PDF is retrieved and grep-verified. The four-chokepoint mechanism diversity claim (the basis for triaging this as the highest-priority Phase 6 hit) holds at the qualitative level from the abstract. Quantitative claims (e.g., "URAT1 downregulated 0.4×", "% SUA reduction = X%") are placeholder-only until full-text verification.
Recommended action for Phase 6:
1. Retrieve full PDF (institutional access, interlibrary loan, or contact corresponding author Ling J at lingjian-ya@sdu.edu.cn).
2. Grep PDF for: load-bearing numbers per the table above + the redox/NLRP3/ergothioneine list.
3. Update this deep-read in place; remove [NOT IN ABSTRACT] markers as numbers are verified.
4. Then and only then, propagate the canonical AMC-BFE numbers to wiki/comp-014-medicinal-mushroom-compound-mapping.md and wiki/synthesis.md.
References¶
PubMed: Xia 2026, PMID 41905012, DOI 10.1016/j.bioorg.2026.109806.
Methodological analog from same Shandong consortium: Zhao et al 2025, PMC12805540, https://citations.gxl.ai/papers/PMC12805540 — bidirectional SSF C. militaris × Panax quinquefolius L., same lab pattern (rice substrate, 65-mesh sieve, 25°C, 50-day SSF).
URAT1-axis prior art: Yong et al 2018, PMC5788910, https://citations.gxl.ai/papers/PMC5788910 — cordycepin downregulates URAT1 in HUA mice (independent confirmation that the C. militaris → URAT1 axis is real, regardless of fermentation context).