🧬 Aging Wiki

2026-05-20

23 min read

log/2026-05-20 — ad-hoc daily entries

Sub-file of log — see parent for index. Holds full content for one-line pointers in log.md.

[2026-05-20] ingest | shin-2025-rehmgb1 (user-requested ad-hoc)

User asked to incorporate https://pubmed.ncbi.nlm.nih.gov/40189139/ — Shin et al. 2025 Metabolism 168:156259, “Propagation of senescent phenotypes by extracellular HMGB1 is dependent on its redox state” (DOI 10.1016/j.metabol.2025.156259). Paper is closed-access (download_status: not_oa in a local paper archive); FWCI 26.05 — top citation percentile. Authorship notable: Jeon Ok Hee lab (Korea University) + Conboy MJ + IM (Berkeley parabiosis lineage) + Wiley CD (UC Davis, MiDAS originator).

Per feedback_paper_impact_broad, framed as a paper that re-frames three atomic substrates beyond the obvious subject page:

  1. The DAMP-class SASP factor hmgb1 (not previously on the wiki as a protein page)
  2. The pattern-recognition receptor rage (not previously on the wiki) as a senescence-propagation node, not just an AGE receptor
  3. The altered-intercellular-communication hallmark — provides the first molecularly explicit mechanism for systemic (not just local-paracrine) senescence spread, which had been a standing no-mechanism for the hallmark
  • added: studies/shin-2025-rehmgb1-paracrine-senescence.md (abstract-only ingest; verified:false; no-fulltext-access)
  • added: molecules/proteins/hmgb1.md (UniProt P09429 confirmed via UniProt API; verified:false)
  • added: molecules/proteins/rage.md (UniProt Q15109 confirmed via UniProt API; verified:false)
  • propagated: processes/sasp.md (new § paragraph under “Paracrine senescence (bystander effect)”; key-proteins frontmatter extended with hmgb1; ^shin2025 footnote added)
  • propagated: hallmarks/altered-intercellular-communication.md (new ReHMGB1 paragraph in “Heterochronic circulating factors”; table row added; ^shin2025 footnote)
  • propagated: hallmarks/chronic-inflammation.md (existing NLRP3/pyroptosis DAMP-release line now wikilinks to hmgb1 + rage + Shin 2025)
  • gaps surfaced:
    • no-fulltext-access on Shin 2025 (closed-access; abstract-only — quantitative n/dose/p-value/tissue-by-tissue claims unverified)
    • Upstream HMGB1 redox-biology primary sources (Lotze 2005, Schiraldi 2012, Venereau, Yang/Antoine) not yet seeded as study pages — flagged on hmgb1
    • Davalos 2013 (HMGB1 nuclear-loss-from-senescent-cells) not seeded as study page
    • SIRT1-HMGB1 acetylation/deacetylation primary sources not seeded
  • forward-queue candidates: (1) Davalos 2013 study page; (2) Schiraldi 2012 study page for HMGB1-CXCL12 heterocomplex biology; (3) when Shin 2025 becomes OA-accessible or a local PDF is procured, run wiki-verifier to flip verified:true and remove banner.

[2026-05-20] verify | shin-2025-rehmgb1 (same-day verification pass)

User provided local PDF (~/Downloads/PIIS0026049525001283.pdf) shortly after the ad-hoc ingest. Read end-to-end (17 pp); all three new pages flipped to verified: true; auto-extraction banners removed; propagation footnotes corrected.

Substantive corrections to the auto-extracted seeds:

  1. OA status wrong. Paper is CC BY-NC-ND open access, not closed. The a local paper archive flagged not_oa (likely OpenAlex lag for Elsevier OA). Banners removed; “closed-access” / “abstract-only” descriptors stripped from footnotes on sasp.md and altered-intercellular-communication.md.

  2. Wiley CD affiliation wrong. Auto-seed said “UC Davis” — paper lists him at Tufts Jean Mayer USDA Human Nutrition Research Center on Aging. Corrected.

  3. Many specifics added (these were gaps in the abstract-only seed, not errors): cell lines (WI-38, BJ, primary renal epi, HSKM), in-vitro dose (20 μg/mL × 72 h), in-vivo dose (5 mg/kg single IV in 3-mo C57BL/6J, 7 dpt), anti-HMGB1 mAb clone 3E8 (BioLegend 651402, 0.1 mg/kg IV), BaCl₂ muscle-injury model details, tissue list (p21 ↑ in GA/TA/liver; heart + kidney ns), pharmacological rescue probes FPS-ZM1 100 nM + Momelotinib 0.3 μM, ~17 min ReHMGB1 serum half-life (Zandarashvili 2013), aged-human-serum ReHMGB1 elevation Supp Fig 6 (70–80 yr vs 40 yr), behavioral readouts (grip strength fully restored to uninjured by anti-HMGB1, p<0.01 vs IgG), GSEA NES values, 1,087 vs 613 DEGs, p15^INK4b^ ReHMGB1-specific bystander marker, SASP secretome profile (IGFBP5/IL15/TGFB1/MMP2/etc).

  4. HMGB1 + RAGE narrative: auto-seed correctly identified the C23/C45/C106 cysteine architecture, the three redox states, and the RAGE → JAK/STAT + NF-κB convergence. Paper confirms; no narrative corrections needed — only specifics layered in.

  • corrected: studies/shin-2025-rehmgb1-paracrine-senescence.md (full rewrite; verified:true)
  • corrected: molecules/proteins/hmgb1.md (banner removed; verified:true; senescence-propagator section expanded; aged-human-serum elevation + 17-min half-life added)
  • corrected: molecules/proteins/rage.md (banner removed; verified:true; FPS-ZM1 + Momelotinib rescue + Western blot phospho-readouts added)
  • corrected: processes/sasp.md (^shin2025 footnote rewritten)
  • corrected: hallmarks/altered-intercellular-communication.md (^shin2025 footnote rewritten)
  • gaps cleared: no-fulltext-access removed from all four pages
  • forward-queue: Venereau 2012, Schiraldi 2012, Zandarashvili 2013, Davalos 2013, Jeon 2022 Nat Metab, Rebo 2016 — referenced primary sources not yet seeded; high-value because they anchor the upstream HMGB1 redox-biology + heterochronic-blood-exchange literature
  • archive feedback signal: paper marked not_oa when it is in fact CC BY-NC-ND. Possible OpenAlex ingest lag for Elsevier OA. Could surface to a local paper archive FEATURE_REQUESTS as an OA-status-refresh concern; low priority since user-supplied local PDF resolved the verification cleanly

[2026-05-20] update | glucosepane § “Tissue imaging tools”

User Q: “do any studies have photos of high vs low glucosepane dermis?” Term-search on PubMed (glucosepane AND (immunohistochem* OR imaging)) surfaced Li R, Draphoen B et al. 2025 Nanoscale Horizons 10(10):2607–14 (doi:10.1039/d5nh00263j, PMID 40813757) — first published fluorescence imaging of glucosepane in dermis specifically, via FluMag-SELEX aptamer Glu3 in db/db diabetic vs wild-type mouse skin biopsies. Prior verifier pass (2026-05-19) had not surfaced this.

  • added: processes/glucosepane.md § “Tissue imaging tools” — new subsection centralizing Streeter 2020 (antibody / mouse retina) + Li 2025 (aptamer / mouse dermis); explicit gap markers for missing human dermis IHC.

  • updated: processes/glucosepane.md § “Interventions — current” — Spiegel/Revel paragraph trimmed to point to new imaging section + acknowledge Rosenau/Kissmann (Ulm) as the independent aptamer development group, distinct from Spiegel/Yale.

  • updated: processes/glucosepane.md § “Limitations and gaps” — non-invasive measurement bullet refreshed; new bullet on missing human dermis glucosepane imaging.

  • updated: processes/glucosepane.md frontmatter — literature-checked-through: 2026-05-19 → 2026-05-20; verified-scope extended from “14 sources” to “15 sources + Li 2025 abstract-level”.

  • updated: tissues/dermis.md — #gap/needs-replication line on glucosepane in human dermis annotated with partial supersession by Li 2025 mouse data + cross-reference to new imaging-tools section.

  • added: [^li2025] footnote in processes/glucosepane.md.

  • gaps surfaced (new): #gap/needs-human-replication — human dermis aged-vs-young glucosepane IHC not yet published despite reagents now existing; aptamer specificity validation against other AGEs / ECM components flagged by Li 2025 authors as open methodological question.

  • verification status: Li 2025 verified at MEDLINE abstract level via PubMed efetch + RSC HTML for Figure 5 content confirmation (db/db vs WT side-by-side, epidermal-dermal junction marked); full PDF not yet retrieved, attempt via a local paper archive deferred.

  • 2026-05-20 ingest — Remesal et al. 2025 Nature Aging (neuronal FTL1 as cell-autonomous driver of hippocampal cognitive aging in mouse; AAV-shRNA + Syn1-Cre-CRISPR cKO in aged mice rescue NOR + Y-maze; NADH 100 mg/kg IP rescues FTL1-OE deficit; iron redox shift ↑Fe³⁺/Fe²⁺ → impaired ATP synthesis → cognitive decline; DOI 10.1038/s43587-025-00940-z; PMID 40830655; PMC12532579; CC-BY OA). New pages: studies/remesal-2025-ftl1-brain-cognitive-aging.md (verified:true, full PDF read end-to-end; 27 pp + 55 refs) + molecules/proteins/ftl1.md (verified:true; UniProt P02792 human FTL / P29391 mouse Ftl1 confirmed via UniProt REST API; HGNC 3999; NCBI Gene 2512 human / 14325 mouse; Open Targets druggability-tier 3 — Med-Quality Pocket + Structure-with-Ligand, no approved drugs). Propagated to: tissues/brain.md (new “Neuronal iron dysregulation and FTL1” section under Key aging features; hallmark-connection table updated for both loss-of-proteostasis + mitochondrial-dysfunction rows; new ^remesal2025 footnote), hallmarks/mitochondrial-dysfunction.md (new “Iron dyshomeostasis as a route to ATP-synthesis decline in aged hippocampus” paragraph appended to Core mechanism 1 — ETC decline; ^remesal2025 footnote; Related pages list extended with ftl1 + study + brain). Conceptual frames reframed (per feedback_paper_impact_broad): (1) neuron-cell-autonomous cognitive aging via a discrete iron-storage protein — rare in field, most rejuvenation work targets systemic/non-neuronal compartments; (2) iron dyshomeostasis mechanistically distinct from ferroptosis (sub-lethal Fe³⁺-sequestration → ATP-synthesis impairment, opposite redox direction to Wu 2023 AD-brain ferroptotic priming); (3) NADH-direct as a parallel modality to nmn/nr precursors in the NAD-cognitive-aging story — bypasses biosynthetic salvage steps; (4) Villeda lab’s pivot from systemic-rejuvenation factors (GDF11/B2M/CCL11 era) to cell-autonomous neuronal-target gene-therapy lineage (TET2 — Pratt 2022; FTL1 — Remesal 2025). Gaps surfaced: needs-replication (single-lab Villeda); needs-human-replication (mouse-only in vivo); no-mechanism upstream (what drives age-rise of neuronal Ftl1 — transcriptional regulator? IRP-IRE? microglial→neuronal iron handoff?); dose-response-unclear (NADH single dose × 9 days only); not-in-genage (FTL not curated as of 2026-05-20 — aging-gene nomination warranted). NMN/NR compound pages deliberately not edited — Remesal NADH-direct paradigm is cross-referenced from ftl1 but is a parallel modality, not a precursor comparison. ROADMAP not updated per feedback_adhoc_seed_no_roadmap (ad-hoc user-requested ingest). → direct user request (ad-hoc paper-ingest; not part of a numbered round)

  • 2026-05-20 verify — studies/geronimo-olvera-2026-apoe2-dna-repair-senescence.md flipped verified:true. PMC HTML (PMC13156074) used as canonical source (CC-BY 4.0 gold OA; paper post-dates a local paper archive OpenAlex snapshot). 4 corrections applied: (1) CRITICAL: NUAT → NNAT (Neuronatin) — gene symbol typo in bulk RNA-seq top-downregulated list; (2) Recombinant APOE2 rescue description expanded: “reduced pγH2AX foci number and 53BP1 foci size” → “reduced pγH2AX foci number and foci size, and 53BP1 foci number and foci size” (PMC confirms both foci number AND size improved for both markers); (3) Rescue experiment concentration and dosing schedule added: 50 ng/mL twice-weekly from day 11; (4) Mouse in-vivo p-value claims standardized: nucleolar-size “p<0.01”, Lamin A/C “p<0.0001”, H3K9me3 “p<0.001–0.0001” → all three corrected to “p<0.05–0.0001 across pairwise comparisons” to reflect the full range reported in the figure legend. Supersession check skipped (paper published 2026-05-08; no meta-analyses possible). Downstream propagation needed on molecules/proteins/apoe.md (NNAT correction in any propagated gene list) and phenotypes/alzheimers-disease.md (same). → direct user request (ad-hoc verification)

  • 2026-05-20 ingest — Gerónimo-Olvera et al. 2026 Aging Cell (APOE2 promotes DNA-repair / resists senescence in human neurons; doi:10.1111/acel.70494; PMID:42103698; PMC13156074; CC-BY OA). New study page studies/geronimo-olvera-2026-apoe2-dna-repair-senescence.md (auto-extracted from PMC; verified:false; archive not yet downloaded). Extended molecules/proteins/apoe.md — added “Neuronal DNA repair and senescence resistance” mechanism section between ε4-AD and “Aging and longevity”; added cellular-senescence + genomic-instability to hallmarks: frontmatter (previously only chronic-inflammation + altered-intercellular-communication); refreshed literature-checked-through: 2026-05-20; updated verified-scope + Limitations & gaps — unsourced resolved for the DNA-repair / senescence axis, needs-human-replication partially resolved (human iPSC neuronal data now available, though tau axis still untested), needs-replication updated (Gerónimo-Olvera 2026 is first isogenic ε2/ε2 multi-system study but single-donor; multi-donor + independent-lab replication still needed). Extended phenotypes/alzheimers-disease.md — APOE2-protective neuronal mechanism added under APOE genetic-risk section + cellular-senescence subsection; new footnote. Key findings ingested: ε4 > ε2 for γH2AX/comet/53BP1/p-ATM; ε2 RNA-seq DEG hubs (BRCA1, CDK1, PLK1, TOP2A, BLM, RAD9B); ε2 preserves Lamin A/C + H3K9me3 under genotoxic stress; ε2 smaller nucleoli + nuclear HMGB1 retention + restrained rRNA (RepEnrich2); ε2 resists p16/CRYAB induction; aged 16-mo APOE2-TR mouse hippocampus (n=3–4) recapitulates lamin/H3K9me3/HMGB1/nucleolar signature; recombinant APOE2 in trans partially rescues APOE4 DDR phenotype; GEO GSE143276. Propagation candidates NOT yet edited: cellular-senescence and genomic-instability hallmark MOCs (synthesis-MOCs that aggregate from atomic pages — propagation occurs implicitly via the updated APOE atomic page; no direct hallmark-page edit warranted unless neuronal-senescence-as-cell-type-specific-mechanism becomes a documented MOC theme). → direct user request (ad-hoc paper-ingest; not part of a numbered round)

  • 2026-05-20 verify — interventions/lifestyle/protein-intake.md flipped verified:true (partial scope). 6 corrections applied: (1) Morton 2018 FFM CI upper bound corrected 0.51→0.52 (abstract and PMC HTML both confirm 0.52; the 0.51 appeared in wiki body + footnote); (2) Morton 2018 2020 erratum (PMID 32943392) noted — competing-interest correction only, no numerical changes; (3) Mamerow 2014 SKEW intake corrected: “~15 g lunch” → “~16 g lunch” and “~65 g dinner” → “~63 g dinner” (abstract: 16.0±0.5 and 63.4±3.7 g); (4) Levine 2014 body text updated with exact HRs (HR 1.74, 95% CI 1.02–2.97 for all-cause; HR 4.33, 95% CI 1.96–9.56 for cancer; 74% not 75% increase; 4.33-fold not exactly “4-fold”); footnote updated with PMC3988204, exact CIs, and age-group n breakdown (50–65: n=3,039; 66+: n=3,342); (5) Jäger 2017 body text corrected: “2.3–3.1 g/kg lean body mass” → “2.3–3.1 g/kg body weight” (ISSN position stand uses BW not LBM; Helms 2014 uses LBM — distinction documented); (6) Moore 2015 footnote corrected: journal name to full “The Journals of Gerontology. Series A, Biological Sciences and Medical Sciences”; LBM-normalized thresholds added (0.60±0.29 g/kg LBM older vs 0.25±0.13 g/kg LBM young, p<0.01). Additional: Tagawa 2022 (#gap/unsourced resolved) — identified as Tagawa M et al. Sports Medicine Open 2022;8(1):107 (PMID 36057893; doi:10.1186/s40798-022-00508-w); strength-plateau 1.5 g/kg/day (complementary to Morton 2018 FFM-plateau 1.62 g/kg; different endpoint). Supersession check: no 2024–2026 meta-analysis or large RCT contradicts Morton 2018 or Levine 2014. literature-checked-through: 2026-05-20. Verification source: Morton 2018 + Levine 2014 via PMC HTML; all other primaries via PubMed abstract. Full PDFs not locally downloaded. → direct user request (load-bearing clinical protocol verification)

  • 2026-05-20 verify — studies/ottaviani-2023-ppo-flavanol-bioavailability.md flipped verified:true. Full PDF cross-check (Food & Function 14(18):8217–8228). 9 corrections applied: (1) CRITICAL: preregistered: false → true — NCT03526094 confirmed in §2.5; (2) study-design: rct → crossover-controlled — Phase 1 was explicitly non-randomized (fixed order per §2.4); Phase 2 was randomized; (3) All ± values corrected from SD to SEM (paper explicitly states “mean ± SEM” throughout); (4) Mixed-berry Cmax corrected from “~680” to confirmed 659 ± 104 nmol/L (Table 3); (5) Exact p-values added: high-PPO Cmax p = 4 × 10⁻⁵; mixed-berry p = 0.818; AUC₀₋₆ h same p-values; (6) AUC data filled in: capsule 2259 ± 279; banana 272 ± 159; mixed berry 1941 ± 294 nmol/L; (7) Phase 2 exact values filled in: FD Cmax 622 ± 115; FD+BD Cmax 429 ± 80 nmol/L (p = n.s.); AUC reduction 37 ± 6% (p = 0.004); (8) Flavan-3-ol dose confirmed: 541 mg / 638 mg / 484 mg total by condition; 75/88/68 mg (−)-epicatechin; Phase 2: 88 mg (303 µmol); (9) PPO survey units corrected from “nmol O₂/min/g” to KU/100 g edible portion; all 18-product values filled in from Table 4; banana = 3258 ± 71 KU/100 g. Additional items resolved: SREM expansion confirmed (SREMs = structurally related (−)-epicatechin metabolites; 4 main circulating forms named); washout confirmed ≥6 days; age range confirmed 25–60 yr; funding conflict of interest (Mars, Inc.) added to limitations. #gap/needs-pdf-verification tags removed throughout. #gap/needs-pdf-verification on EGCG/gallated catechins replaced with accurate gap tag. “Human RCT” label in significance section corrected to “controlled human study.” Supersession check (PubMed 2023–2026 PPO+epicatechin+bioavailability): no superseding publications found. → direct user request

  • 2026-05-20 seed + propagate — interventions/lifestyle/protein-intake.md (new canonical home for dietary-protein dose recommendations). Closes a long-standing citation gap: the wiki carried uncited “1.0–1.5 g/kg” claims on phenotypes/sarcopenia.md, tissues/skeletal-muscle.md, phenotypes/frailty.md, and biomarkers/grip-strength-biomarker.md. 15 DOIs verified via PubMed efetch: Morton 2018 BJSM meta-analysis (10.1136/bjsports-2017-097608; PMID 28698222; n=1,863 across 49 RCTs; FFM-gain plateau at ~1.62 g/kg/day — the load-bearing primary for the 1.6 g/kg deficit floor), Longland 2016 AJCN deficit RCT (10.3945/ajcn.115.119339; PMID 26817506; n=40; 2.4 vs 1.2 g/kg × 40% deficit × 4 wk RT+HIIT; 2.4 g/kg arm +1.2 kg lean / −4.8 kg fat vs 1.2 g/kg arm +0.1 kg lean / −3.5 kg fat), Bauer 2013 PROT-AGE JAMDA (10.1016/j.jamda.2013.05.021; PMID 23867520), Deutz 2014 ESPEN Clin Nutr (10.1016/j.clnu.2014.04.007; PMID 24814383), Jäger 2017 ISSN position stand (10.1186/s12970-017-0177-8; PMID 28642676; 1.4–2.0 g/kg general; 2.3–3.1 g/kg FFM during deficit), Helms 2014 JISSN (10.1186/1550-2783-11-20; PMID 24864135), Moore 2015 J Gerontol meta-regression (10.1093/gerona/glu103; PMID 25056502; older adults ~0.40 g/kg/meal vs ~0.24 young), Cuthbertson 2005 FASEB J anabolic-resistance original (10.1096/fj.04-2640fje; PMID 15596483), Mamerow 2014 J Nutr distribution (10.3945/jn.113.185280; PMID 24477298), Nunes 2022 JCSM age-stratified update (10.1002/jcsm.12922; PMID 35187864), Levine 2014 Cell Metab NHANES (10.1016/j.cmet.2014.02.006; PMID 24606898; high-protein age 50–65 obs-mortality caveat), Antonio 2016a JISSN renal safety (10.1186/s12970-016-0114-2; PMID 26778925), Antonio 2016b J Nutr Metab one-year follow-up (10.1155/2016/9104792; PMID 27807480), Shams-White 2017 AJCN bone safety meta (10.3945/ajcn.116.145110; PMID 28404575), Lim 2021 Nutrients (10.3390/nu13020661; PMID 33670701). R25 recency search: Nunes 2022 surfaced as the most material post-2020 update (does not contradict Morton 2018); Tagawa 2020/2021 flagged unsourced (not findable via PubMed eutils). literature-checked-through: 2026-05-20. New mechanism classes added to frameworks/intervention-classes.md: mps-stimulation, anabolic-protein-support, leucine-sensing. Propagation: (a) phenotypes/sarcopenia.md (2 backlinks: anabolic-resistance mechanism section + adequate-protein recommendation section); (b) tissues/skeletal-muscle.md (intervention table protein-intake row); (c) phenotypes/frailty.md (nutrition section); (d) biomarkers/grip-strength-biomarker.md (protein-intake bullet); (e) interventions/lifestyle/caloric-restriction.md (side-effects section — adds Longland 2016 deficit RCT + explicit IGF-1 tradeoff acknowledgment); (f) biomarkers/igf-1-biomarker.md (Fontana 2008 section — adds explicit cross-link to the tradeoff framing on the new page). All verified: false on the new page pending PDF cross-check. → direct user question (the lean-mass-during-deficit tradeoff is the correct direction over the temporary IGF-1 increase)

  • 2026-05-20 verify — IgG Fc glycosyltransferase protein cluster (4 pages; 7 PDFs read; 6 corrections applied; all 4 flipped verified:true with partial scopes). Pages: molecules/proteins/b4galt1.md, molecules/proteins/st6gal1.md, molecules/proteins/fut8.md, molecules/proteins/mgat3.md. Corrections: (1) ST6GAL1 EC number corrected from erroneous 2.4.99.1 to current IUBMB 2.4.3.1 (seeder had the backwards — UniProt 2.4.3.1 is correct; ExPASy ENZYME confirms 2.4.99.1 is now “Transferred entry: 2.4.3.1” since ~2018 renumbering); (2) FUT8 Wang 2005 KO mortality timing corrected: “die within weeks of birth” → “~70% die within the first 3 days of birth” (per PNAS PDF); strain background added (129SvJ ES cells × B6C3F1 blastocysts); (3) B4GALT1 Landini 2022 attribution corrected: body text and footnote erroneously stated “Landini 2022 identified B4GALT1 as an IgG-specific locus” — PDF confirms Landini 2022’s IgG-specific highlights are ST6GAL1 and MGAT3; FUT8/FUT6 are shared; B4GALT1 attribution reassigned to Wahl 2018; (4) FUT8 Landini 2022 footnote already correct (stated FUT8/FUT6 as shared) — confirmed vs PDF; (5) MGAT3 HGNC corrected 7044→7046 (already applied by seeder per UniProt REST API); (6) Du 2020 (phytoestrogen) tagged no-fulltext-access in body (previously only in footnote). PDFs verified: Shields 2002 (EC50 values confirmed), Anthony 2008 (30 mg/kg, DC-SIGN/FcγRIIb confirmed), Ercan 2017 (n=713 + 4 interventional studies confirmed), Landini 2022 (n=2,020; locus architecture confirmed), Wang 2005 (KO timing + MMP-12/13 + TGF-β1 rescue confirmed), Werner 2024 (B-cell-intrinsic sialylation confirmed), Ruhaak 2010 (n=1,967; longevity association + age trajectory confirmed). Unverifiable: Yamane-Ohnuki 2004 (not_oa), Stanley 2002 (not_oa), Collins 2002 (download failed — 0 candidate URLs), Du 2020 (not_oa), Kaneko 2006 (not_oa), B4GALT1 KO original paper (unlocatable). Supersession check: no superseding meta-analyses or large RCTs found for any of the 4 pages (2023–2026 PubMed). literature-checked-through: 2026-05-20 on all four pages. Archive status updates: Wang 2005 + Werner 2024 + Ruhaak 2010 footnotes updated from “pending” to “downloaded”; Collins 2002 updated to “download failed no-fulltext-access”. → direct user request (R45 precision verify pass)

  • 2026-05-20 seed — IgG Fc glycosyltransferase protein cluster (4 new pages): molecules/proteins/b4galt1.md (UniProt P15291; NCBI Gene 2683; HGNC 924; Ensembl ENSG00000086062; rate-limiting galactosylation enzyme; aging-most-relevant of the four; estrogen→ESR1→B4GALT1 axis documented; druggability-tier 3); molecules/proteins/st6gal1.md (UniProt P15907; NCBI Gene 6480; HGNC 10860; sialyltransferase; DC-SIGN/FcγRIIb mechanism; Werner 2024 B-cell-intrinsic sialylation confirmed; druggability-tier 3); molecules/proteins/fut8.md (UniProt Q9BYC5; NCBI Gene 2530; HGNC 4019; core fucosyltransferase; Yamane-Ohnuki 2004 FUT8-KO CHO cell / POTELLIGENT; obinutuzumab/mogamulizumab; Wang 2005 KO mouse emphysema phenotype; druggability-tier 2); molecules/proteins/mgat3.md (UniProt Q09327; NCBI Gene 4248; HGNC 7046 — corrected from brief’s 7044; bisecting GlcNAc transferase; Stanley 2002 KO mouse; Ruhaak 2010 longevity association; druggability-tier 3). All four verified: false. Canonical-DB IDs confirmed via UniProt REST API 2026-05-20. Key primary sources verified via PubMed efetch: Yamane-Ohnuki 2004 (PMID 15352059), Wang 2005 (PMID 16236725), Stanley 2002 (PMID 12417419), Collins 2002 / Werner 2024 (PMIDs 12213789, 38827747), Ruhaak 2010 (PMID 20830288). Wahl 2018 (PMID 29535710) and Landini 2022 (PMID 35332118) cited on all four pages (archive: downloaded for Landini; pending for Wahl). B4GALT1 KO original primary paper could not be confirmed via PubMed efetch — cited via review (Gudelj 2018) with gap-tag. EC number discrepancy for ST6GAL1 (UniProt API returned 2.4.3.1; standard IUBMB 2.4.99.1) flagged with gap-tag. Implicit stubs resolved: b4galt1, st6gal1, fut8, mgat3 previously listed as implicit stubs in processes/igg-fc-glycosylation.md — now seeded. literature-checked-through: 2026-05-20 on all four pages. → direct user request (four-page GTase batch)

  • 2026-05-20 verify — IgG Fc N-glycosylation process page (processes/igg-fc-glycosylation.md): 8 corrections applied; verified: true (partial scope — 6 PDFs read against local archive; 4 sources abstract-level or not_oa). Key corrections: (1) Shields 2002: “FUT8 knockout CHO cells” → “Lec13 GDP-fucose-biosynthesis-deficient CHO cells” (Lec13 is not a FUT8-KO; FUT8-KO technology is Yamane-Ohnuki 2004); (2) Shields 2002: fold-enhancement framing tightened to “up to 50-fold” with EC50 values; F158 allele claim corrected — both alleles benefit substantially; (3) Anthony 2008: confirmed Science (not PNAS); DC-SIGN mechanistic attribution clarified (DC-SIGN cited in Anthony 2008 body, not Kaneko 2006); dose expressed as 30 mg/kg vs 1,000–2,000 mg/kg; (4) Landini 2022: n corrected from “~5,000+” to n=2,020 IgG GWAS (CROATIA-KORCULA + VIKING, two cohorts); (5) Kaneko 2006: ~5% sialylated fraction % tagged no-fulltext-access (not confirmable from abstract; not_oa); (6) Parekh 1985: “predates clinical diagnosis by years” corrected to “in established RA vs OA vs controls” (cross-sectional study; pre-diagnostic claim not supported by Parekh 1985 specifically); (7) Sex gap “~3 biological-age-years younger” removed from body — unsourced in either Kristic 2014 or Ercan 2017; redirected to glycanage-2017 biomarker page; (8) Duplicate footnote ^kristicrev2018 = ^gudelj2018 (same DOI) — merged to ^gudelj2018; (9) Vinicki 2025 DOI confirmed: 10.1007/s11357-024-01349-z. Supersession check: three 2024-2025 PubMed hits (PMID 38928043 MR bidirectional IgG glycosylation↔senescence; PMID 38542917 MR IgG glycosylation causality; PMID 39363095 Vinicki RCT) — none supersede core mechanism claims; Vinicki 2025 already cited. Unverifiable: Kaneko 2006 (not_oa), Parekh 1985 (download failed), Gudelj 2018 review (pending download), Simurina 2018, Menni 2018 (pending), Nimmerjahn 2023 (not_oa). → direct user request (IgG-glycosylation verification pass)

  • 2026-05-20 seed — IgG Fc N-glycosylation process page: new processes/igg-fc-glycosylation.md (mechanism page for GlycanAge biomarker; hosts enzymology + biology; ~250 lines). Canonical-DB IDs confirmed: FUT8 UniProt Q9BYC5 (via API); B4GALT1 UniProt P15291 (via API). MGAT3 UniProt not confirmed (API returned wrong protein; needs-canonical-id tagged on page). DOIs confirmed via Crossref + PubMed: Kristic 2014 (10.1093/gerona/glt190; downloaded), Shields 2002 (10.1074/jbc.m202069200; pending), Kaneko 2006 (10.1126/science.1129594; not_oa), Anthony 2008 (10.1126/science.1154315; downloaded), Parekh 1985 (10.1038/316452a0; pending), Ercan 2017 (10.1172/jci.insight.89703; pending/gold-OA), Gudelj 2018 (10.1016/j.cellimm.2018.07.009; pending), Nimmerjahn 2023 (10.1038/s41590-023-01544-8; not_oa), Šimurina 2018 (10.1053/j.gastro.2018.01.002; pending), Gaifem 2024 (10.1038/s41590-024-01916-8; downloaded), Menni 2018 (10.1161/CIRCRESAHA.117.312174; pending), Landini 2022 (10.1038/s41467-022-29189-5; pending), Vinicki 2025 (PMID 39363095; DOI unconfirmed — flagged). Implicit stubs created: b4galt1, st6gal1, fut8, mgat3. literature-checked-through: 2026-05-20. → ad-hoc user request

  • 2026-05-20 verify — R45 three-paper supersession batch (full PDF cross-check; 3 study pages + 4 entity page extensions): de-decker-2023 flipped verified:true; cordioprev-2024 flipped verified:true; matlack-2025 stays verified:false (preprint; scope documented). Key corrections from PMC HTML → PDF: (1) CRITICAL: De Decker 2023 glycating agent — seeder corrected ‘glycolaldehyde’ to ‘glyceraldehyde dimer’ during seeding; PDF Methods §4.5 explicitly states ‘glycolaldehyde (GA)’; the seeder over-corrected in the wrong direction — ‘glyceraldehyde dimer’ is wrong (glyceraldehyde is 3C C₃H₆O₃; glycolaldehyde is 2C C₂H₄O₂). Corrected back to ‘glycolaldehyde dimer’ across all 5 citing pages (de-decker study + fn3k.md + microbial-amadori-deglycation.md + both footnotes). (2) CRITICAL: CORDIOPREV 2024 MG measurement method — PMC HTML extraction stated ‘LC-MS/MS’; PDF Methods confirms competitive ELISA (OxiSelect Methylglyoxal Competitive ELISA Kit, Cell Biolabs; inter-assay CV 5.8%). Corrected across cordioprev-2024-meddiet-glo1.md + glo1.md + methylglyoxal.md + all footnotes. (3) CORDIOPREV MedDiet baseline MG corrected from imprecise ‘~3.18’ to 3.18 ± 0.07 µg/mL (Table 2). (4) CORDIOPREV GLO1 source confirmed as PBMCs (not ‘whole blood’). (5) CORDIOPREV non-worsening IMT subgroup n confirmed: 408 (166 low-fat + 242 MedDiet). (6) Matlack 2025 five-residue numbering: corrected to specify both numbering systems — ancFN3KRP gain-of-function direction Q179H/M187L/Y253F/Y269F/H293N vs HsFN3K equivalent 5mut H171Q/M179L/F244Y/F260Y/N284H; allosteric hub specified as F244 (Phe) not abstract ‘residue 244’; F244Y alone confirmed insufficient in HsFN3K. (7) Matlack 2025: confirmed no Hill-coefficient or cooperativity analysis — evidence is PSN structural + mutational biochemistry (webPSN, 3 Å cutoff); no Km/Vmax reported. (8) De Decker 2023: AF significance added (p<0.0001 for all three enzymatic arms; p<0.001 for aminoguanidine; ns for PBS). Supersession check: PMID 38719207 (Nutrition Reviews 2025 systematic review; n=2935 from 6 trials) broadly consistent with CORDIOPREV — no supersession. No supersession of De Decker 2023 found (zero competing publications). literature-checked-through confirmed 2026-05-20 on all 7 pages. → direct user request (R45 verify pass)

  • 2026-05-20 seed — R45 three-paper supersession integration (3 new study pages + 4 atomic-page extensions): studies/de-decker-2023-fn3k-faod-skin-combination.md (PMID 37240327; PMC10219073 full-text verified; FN3K+FAOD combination therapy in human skin; n=19; 43% AF reduction); studies/cordioprev-2024-meddiet-glo1.md (PMID 39402581; PMC11475769 full-text verified; 5-yr MedDiet RCT, n=809 CHD patients; MG stabilization + GLO1 upregulation + IMT regression); studies/matlack-2025-fn3k-allostery.md (PMID 40766467; PMC12324494 full-text verified; FN3K allosteric mechanism via APR; PREPRINT). Atomic page extensions: fn3k.md — Supersession candidates section replaced with full “Combination therapy with FAOD” body section + “Sequence/structure/allosteric” section (De Decker 2023 + Matlack 2025; verified-scope updated); glo1.md — CORDIOPREV 2024 subsection added in GLO1 inducers section; extrapolation table row updated from ‘no’ to ‘partial’ for human replication; verified-scope updated; methylglyoxal.md — CORDIOPREV 2024 dietary MG modulation paragraph added in therapeutic section; verified-scope updated; microbial-amadori-deglycation.md — De Decker 2023 combination-therapy section added; verified-scope updated. All study pages verified: false with auto-extraction banner. Matlack 2025 flagged needs-peer-review throughout (bioRxiv preprint). No supersession of existing core claims found — all three papers extend or add combination/dietary/mechanistic evidence to existing framework. Implicit stub: mediterranean-diet (not yet seeded). → direct user request (R45 supersession batch)

  • 2026-05-20 verify — natural-AGE-clearance enzyme cluster (5 pages + 1 extension verified): fn3k.md, glo1.md, microbial-amadori-deglycation.md, methylglyoxal.md, carboxymethyl-lysine.md flipped verified:true; age-crosslink-breakers.md verified-scope extended. Key corrections: (1) Campbell 2012 n corrected 96→26 (carboxymethyl-lysine.md); (2) Genuth 2015 complication-specific AGE attribution corrected — neuropathy=MG-H1+furosine, retinopathy=glucosepane+furosine, nephropathy=furosine (methylglyoxal.md + carboxymethyl-lysine.md); (3) Monnier 2015 n corrected from ~466 to endpoint-specific subsamples CAC=187/IMT=127/cardiac-MRI=142; (4) Sanghvi 2019 MAFK→MAFG corrected (fn3k.md); NRF2 glycated-residue list (K462/K472/K487/R499/K543/K554/R569) + quantitative glycation levels added (~29% glycated; 40–50% peptide cleavage decline; >50% destabilization in KEAP1-WT); (5) De Bruyne 2021 model expanded from “isolated human lenses” to multi-species ex-vivo/in-vivo (equine/porcine/murine/human) with specific functional data; (6) Hirokawa 2003 FAOD subclass classification corrected from “free-amino-acid-specific vs peptide-permissive” to “alpha-glycated (Class A) vs epsilon-glycated (Class B)”; (7) Samanta 2024 citation updated Brain 2024→Brain 2025;148:262-275; quantitative MG accumulation data added from full PDF (6.2→26.2 nM synaptic mitochondria 6→30 mo; 60-80% MG/AGE reduction in GLO1-Tg; LTP/mEPSC rescue confirmed); (8) Rabbani 2021 design confirmed as crossover RCT; specific doses (90+120 mg/day) flagged unconfirmed from abstract. Unverifiable: Thornalley 2003 confirmed not_oa; Prevenzano 2022, Cely 2026, Frischmann 2005 (closed), Hirokawa/Jeong papers abstract-only. Supersession check: CORDIOPREV trial (PMID 39402581; 2024) new human RCT data on MedDiet→GloxI upregulation in CHD patients — complementary, not superseding. No supersession of any core claim found. → direct ad-hoc user request

  • 2026-05-20 ingest — Mijakovac 2026 medRxiv preprint (IgG glycome + mortality + interventions): new studies/mijakovac-2026-igg-glycome-mortality.md; extended biomarkers/glycanage-2017.md (2026 update section: mortality HR data + three-intervention reversal; comparison table mortality column updated; limitation #2 updated to “preliminary”); extended interventions/blood-product/plasma-exchange.md (new IgG-glycome TPE section + Fuentealba 2025 cross-link). Key data from PDF (fetched 2026-05-20): mortality HR 1.10 (1.08–1.12) unadjusted → 1.05 (1.03–1.06) fully adjusted (GCKD n=4,827, 840 deaths, 8.5 yr); partial replication Vis cohort n=796 HR 1.03; HRT −0.12 glycan yr/mo (n=19, P=5.76e-8); TPE −0.4 glycan yr/mo (n=9, P=0.0245); CR significant deceleration at 11 kg loss (P=0.0017, DIOGENES n=680). All preprint content #gap/needs-peer-review. Implicit stubs: hormone-replacement-therapy, fuentealba-2025-plasma-exchange-multi-omics. literature-checked-through updated to 2026-05-20 on both extended pages. → ad-hoc user request

  • 2026-05-20 seed — natural-AGE-clearance enzyme cluster + key AGE molecules (5 new pages + 1 extension): new molecules/proteins/fn3k.md (FN3K fructosamine-3-kinase; Tier 2 deglycation enzyme; UniProt Q9H479; NCBI Gene 64122; key papers: Delpierre 2000 PMID:11016445, Szwergold 2001 PMID:11522682, Veiga da-Cunha 2006 doi:10.1042/BJ20060684, Sanghvi 2019 doi:10.1016/j.cell.2019.07.031, De Bruyne 2021 doi:10.3390/ijms22083841); new processes/microbial-amadori-deglycation.md (amadoriase/FAOX/FAOD class; fungal/bacterial Amadori oxidases; key paper: Delanghe 2024 doi:10.3390/ijms25094779 — expanded substrate scope including CML, abstract-only); new molecules/proteins/glo1.md (GLO1 glyoxalase-1; Tier 1 dicarbonyl detoxification; UniProt Q04760; NCBI Gene 2739; key papers: Thornalley 2003, Rabbani 2016 doi:10.1007/s10719-016-9705-z, Prevenzano 2022 doi:10.15252/embr.202152990, Samanta 2024 doi:10.1093/brain/awae229 — locally available PDF); new molecules/compounds/methylglyoxal.md (MG; PubChem CID 880; ChEMBL CHEMBL170721; key AGE precursor; Genuth 2015 doi:10.2337/db14-0215; ACTION II 1999); new processes/carboxymethyl-lysine.md (CML; canonical non-crosslink RAGE ligand; Ahmed 1986 PMID:3082871; Campbell 2012 doi:10.1371/journal.pone.0049813; Monnier 2015 doi:10.1186/s12933-015-0266-4); extended interventions/pharmacological/age-crosslink-breakers.md with three-tier natural defense framework section + new cross-references + literature-checked-through updated to 2026-05-20. All content verified: false / abstract-level (GLO1 Samanta 2024 PDF locally available). Implicit stubs created: mg-h1, aminoguanidine, carnosine, benfotiamine, pyridoxamine, fn3k-rp (FN3K-related protein). → direct ad-hoc user request

Graph View

Backlinks