Human CYP1B1 is a 543-residue heme P450 associated principally with the endoplasmic-reticulum membrane. Its oxygenation chemistry uses reducing equivalents delivered by P450 reductase. Recombinant human enzyme directly oxidizes estrogens, retinoids, arachidonic acid and xenobiotics. The physiological contribution of each substrate depends on tissue, substrate availability and competing enzymes; activity in a reconstituted preparation does not establish systemic dominance. Human estrogen 4-hydroxylation is favored over 2-hydroxylation in the directly inspected comparison, whereas rodent regioselectivity differs. The description therefore distinguishes human enzyme chemistry from inherited or cellular-context effects. PMID:15258110(https://pubmed.ncbi.nlm.nih.gov/15258110/), PMID:23821647(https://pubmed.ncbi.nlm.nih.gov/23821647/), PMID:10681376(https://pubmed.ncbi.nlm.nih.gov/10681376/).
A second catalytic mode converts fatty-acid hydroperoxides to oxo products. The original study whose title foregrounds CYP2S1 also directly assays CYP1B1. Its commercial CYP1B1 microsomes contain POR, but the assays receive no added NADPH. This supports the distinct dehydratase function without transferring the paper's CYP2S1-specific anaerobic and reductase-removal controls to CYP1B1. The two core functions describe complementary chemistry and do not imply mutually exclusive products or locations. PMID:21068195(https://pubmed.ncbi.nlm.nih.gov/21068195/), Methods and Table 2.
The human arachidonate source resolves EET positions but groups terminal HETE peaks: 19 and 20 can coelute, as can the assigned 16/17/18 group. It does not establish a separately quantified flux to every terminal product. A different chiral analysis supplies the stereochemical evidence: human CYP1B1 Supersomes predominantly form 8R,9S and 14R,15S EET and produce both 11,12 enantiomers. Its comparator/column procedure is retained as the assignment boundary. PMID:15258110(https://pubmed.ncbi.nlm.nih.gov/15258110/), Table 3; PMID:20972997(https://pubmed.ncbi.nlm.nih.gov/20972997/), Table 2 and Methods.
The official record specifies one displayed 543-aa product and no alternative-product comment. All three official sequence representations inspected agree. The heme ligand is Cys470. Experimental reagents include a human deletion of residues 2–4, polymorphic alleles, and the crystallographic CYP1B1*2 construct replacing residues 1–50 with MAKKTSSKGK, retaining A119S and adding a His4 tag. These are not official splice products. The ANF-bound structure supports the heme/domain architecture without establishing intact-protein localization. UniProt Q16678, PMID:21147782(https://pubmed.ncbi.nlm.nih.gov/21147782/).
Tagged human TEK/CYP1B1 co-immunoprecipitation, including the actual TEK domain comparisons, supports receptor protein tyrosine kinase binding. Six exact interaction records agree with the curated pair. This is an association assay; the data do not establish ligand agonism or an isolated binary interface. The five other retrieved interaction records remain context and do not generate new annotations. PMID:28620713(https://pubmed.ncbi.nlm.nih.gov/28620713/).
Mouse retinal pericyte experiments support context-dependent effects on population expansion, VEGF output, migration, matrix adhesion, apoptosis and endothelial coculture morphogenesis. Increased population does not mean increased DNA-synthesis rate in this experiment. Basal/staurosporine caspase assays and peroxide viability/ROS assays are separate, so the precise oxidative-stress intrinsic-apoptosis route remains unresolved. The reported STAT3 Ser727 result alone does not establish receptor/JAK–STAT signaling. Methods and Results also differ in their peroxide dose. These limits accompany the literal mouse-to-human transfers rather than changing their source identities. PMID:23568032(https://pubmed.ncbi.nlm.nih.gov/23568032/), Methods, Results and Figures 3, 4 and 10.
Endothelial experiments distinguish chronic Cyp1b1 loss, eNOS restoration and acute CYP1B1 rescue. NO bioavailability can improve without restored eNOS protein, and the authors discuss synthesis and bioavailability together. The exact NO-biosynthetic claim remains uncertain despite positive vascular-regulatory evidence. Migration direction differs between endothelial and pericyte contexts. The analysis does not impose a universal direct-sensor requirement or erase positive regulatory annotations because their biochemical intermediate is unresolved. PMID:20032512(https://pubmed.ncbi.nlm.nih.gov/20032512/), Methods, Results and Discussion.
Mouse trabecular-meshwork/periostin experiments and complementary human TM1 knockdown support matrix, adhesion and redox effects. They do not demonstrate direct CYP1B1 collagen chemistry. Lipid peroxidation is not automatically a resolved lipid-catabolic step. The nine UNDECIDED source objects concern steroid catabolism, NO biosynthesis, oxidative-stress intrinsic apoptosis, lipid catabolism and receptor/JAK–STAT signaling; their individual reasons retain the positive surrounding evidence. PMID:23979599(https://pubmed.ncbi.nlm.nih.gov/23979599/).
Mitochondrial fraction association has been demonstrated after TCDD induction in female mouse tissues with knockout and fraction-marker controls, while carryover is acknowledged. This supports contextual inheritance rather than an unqualified human constitutive-localization claim. The proposed targeting residues do not define an experimentally demonstrated cleavage site. HPA's captured mitochondrial call is uncertain; no HPA image pixels were inspected. The official ER/microsome statements include by-similarity provenance. PMID:23692925(https://pubmed.ncbi.nlm.nih.gov/23692925/), Human Protein Atlas.
The genuine QuickGO response has 118 raw records, the downloaded TSV has 116 unique rows, and the genuine seed has 115 annotation objects. Raw duplicates and a date-only mitochondrial ISS collapse explain these counts. Every seeded source object, qualifier, reference, supporting entity and multiplicity was preserved. No NEW annotation was proposed. The compact evidence guide supplies source URLs, exact used tuples and assay boundaries; the review remains the decision record.
Current PANTHER lookup places Q16678 in PTHR24299:SF11. The older captured official record and committed family context name PTHR24289:SF16; the current PTHR24289 tree lacks the target. The current 585-node tree and 585-record alignment were independently reconstructed, and the target alignment equals all 543 residues. Seven literal IBD nodes lie on its 18-node ancestral path. Raw GAF WITH/FROM is column 8 and aspect column 9, using one-based numbering. Current IBD descendant evidence was expanded separately from the original GOA source list. A target's own experimental evidence and short descendant lists are legitimate phylogenetic grounding, not circularity.
Twelve complete accession queries contain 1,042 donor records. Fifty-one propagation joins plus one binding-partner context join were reconstructed using actual mouse, rat and zebrafish aliases. Same-term records and publication scopes are distinguished from broad family resemblance. For the biological-toxin term, DMBA and 7H-dibenzo[c,g]carbazole are insufficient by themselves because they are combustion-derived xenobiotics. Independent human CYP1B1 conversion of fungal aflatoxin B1 to AFM1 provides the pertinent positive evidence, with the abstract's limits retained. PMID:36442682(https://pubmed.ncbi.nlm.nih.gov/36442682/).
All 213 branches of six ARBA rules were evaluated against the captured official feature set and taxonomy lineage. Each has two matching branches. Four InterPro records match the actual human sequence intervals. Seven literal Rhea source joins match the official UniProt reaction objects; direct Rhea pages returned access failures. Broad true steroid, hormone, primary-alcohol and olefinic classes can be redundant with specific existing metabolic coverage. NADPH is delivered through POR, making the reduced-flavoprotein donor class more precise than direct NAD(P)H donor wording for CYP1B1 oxygenation.
Fourteen Reactome roots, 44 selected raw objects and 29 target paths were reconstructed. WT entity 211021 is in the ER membrane and refers to Q16678 through 52628. Shared CYP catalyst sets do not prove identical substrate flux for every member. The defective reaction 5605147 contains nine disease-variant entities, not alternative products or positive variant activities. The actual target event in the broad endogenous-sterol pathway is estradiol hydroxylation; estrogen metabolism is the informative term. Downstream EET-to-DHET hydrolysis belongs to epoxide hydrolases. No target activity was found in the selected repository GO-CAM index; that absence was not treated as a missing-annotation mandate.
The cached source flags remain unchanged. Some full originals were recovered separately from abstract-only caches. The primary-access section of the evidence guide distinguishes these cases from unrecovered originals. Positive existing annotations can be retained with explicit curator deference when complete abstracts and independent functional evidence agree; inaccessible paper-specific assertions are not rejected from a title or abstract alone. No direct binding, quantitative flux, physiological dominance or product-specific equivalence is inferred beyond the inspected assay.
The CYP1B1 chemistry consultation independently recovered and read four originals and retained explicit access failures for PMID10426814 and PMID11555828. Its 24 scoped recommendations and product/term reconstruction were reproduced and integrated after the author's own assessment. A separate pericyte consultation read the original text and Figures 3, 4 and 10 and assessed 16 source assertions. These consultations are attributed to independent AI reviewers; the owner personally read the remaining recorded primary scopes and all final judgments. None of these attributions is a human sign-off.
The genuine default Falcon research attempt failed with HTTP403 and the configured fallback failed with HTTP401/quota. No provider-labelled substitute was fabricated. The review was prepared from genuine official/source retrievals, manual primary reading and the attributed consultations. Raw responses, failure bodies and exact audit projections are retained in the immutable preparation packet; the public guide carries concise reproducible locators rather than those response dumps.
Four brief source-exact quotations occur once each in the review YAML, within the aggregate per-source limits. Notes and the guide add no quotation copies. The first science draft is immutable. A forward correction names the actual 7H-dibenzo[c,g]carbazole tested in PMID17166882, and two obsolete delegation clauses were replaced with the completed official Rhea mapping/access scope. No source object, action, product, core or quote changed in those corrections. Ordinary derived status and rendering are recorded separately; independent whole-review acceptance and publication applicability are subsequent gates.
Independent whole scientific reading and saved-source reconstruction supported all 115 annotation decisions and both catalytic core functions. The remaining corrections concern provenance precision: PMID18356043 studies lynestrenol bioactivation and norethindrone metabolism rather than DHEA; its exact CYP1A2 steroid-catabolic donor assay remains unresolved, so UNDECIDED is retained. The interaction guide now orients features by the literal accession: records7–9 have TEK on side A and CYP1B1 on side B. Every raw interaction field is unchanged. The final source checker explicitly verifies that orientation. These corrections follow the earlier two chemical-name and two obsolete-delegation wording corrections; source objects, actions, products, core functions and the four quotation objects remain unchanged.
Human CYP1B1's estrogen chemistry is distinguished by its preference for estradiol 4-hydroxylation in the inspected product-resolved systems. PMID:23821647 directly compares human and other mammalian enzymes and tests the specificity determinant Val395; a rodent hydroxylation ratio cannot substitute for the human result. RHEA:47280 captures the C4 reaction with reduced P450 reductase as donor. A proposed estradiol 4-hydroxylase term now records that chemistry under steroid hydroxylase activity, with a specific ontology question and no invented GO identifier or NEW annotation. The dated QuickGO steroid-hydroxylase closure contains 38 terms including its parent and no C4-specific estrogen member; the separate broad keyword searches are not an exhaustive proof of global term absence.
The compact synthesis now distinguishes estrogen, retinoid and arachidonate substrate contexts. Retinoid oxidation is demonstrated in human CYP1B1 reconstitutions and Supersomes, but does not identify the substrate that links CYP1B1 loss to trabecular-meshwork disease. Arachidonate experiments identify HETE and EET branches; original human table columns and chiral assignments must be kept separate from mouse results and coeluting terminal-HETE groups. Xenobiotic metabolism remains documented in the description and annotations without a core for every substrate. The common reduced-flavoprotein MF in the retinoid and arachidonate units describes distinct substrate contexts, not additional ancestor coverage.
Hydroperoxide dehydration remains a positive secondary activity. Original PMID:21068195 includes actual CYP1B1 microsomal assays, not only CYP2S1 experiments. Those target reactions omit added NADPH but retain POR; their contribution to intact-tissue oxo-lipid flux is unresolved. The existing exact abstract excerpt has moved from the former secondary core to row 110, once, without changing its words. The existing retinoid excerpt has moved into the retinoid core, once. Both other inherited excerpts remain in place.
Primary chemistry assays and table locators gives public originals, DOI/section/table coordinates, the decisive human numerical rows and their units. These are authored paraphrases and target-specific data transcriptions, not fabricated cache quotations. Publication caches remain unchanged. Six references now have concise findings; four include short exact cached anchors, counted together with all retained quotes. The guide's self-verification label is removed: it is navigation and interpretation, while the cited experiments provide the evidence.
The sole annotation action change is row 40, from MODIFY to TOO_GENERAL, because accepted estrogen metabolism already entails the broad steroid process. Its positive experimental scope remains intact. Thirteen monooxygenase refinements now explicitly point to an already represented donor-class term rather than imply new activities. All nine UNDECIDED judgments remain: steroid/lipid catabolism, NO formation versus availability, oxidative-stress apoptotic-route specificity and receptor/JAK–STAT route specificity are unresolved at their individual source scopes. Neither accessible full text nor a desired completion status resolves those distinctions. All 115 imported source objects and the displayed product remain unchanged.
The pericyte study PMID:23568032 contains positive evidence involving intrinsic apoptotic machinery: Cyp1b1 loss lowers Bax and Bim expression and basal/staurosporine caspase-3/7 activity. Its Discussion proposes Bax/Bim loss as a contributor, alongside increased Akt phosphorylation. These observations are weighed explicitly rather than dismissed because the readouts occur in different figures. Nevertheless, the knockout also increases DHE-detected oxidative stress and reduces peroxide viability. NAC lowers the DHE signal; the reported experiment does not show rescue of apoptosis or establish that oxidative stress initiated a Bax/Bim-dependent death pathway. The exact GO:0008631 transfer therefore remains unresolved. PMID:23979599 independently measures increased stress-induced caspases in mouse trabecular-meshwork cells, but those endpoints likewise do not identify the initiating route.
The STAT3 result has a different resolution problem. Figure 10 and Discussion in PMID:23568032 connect Cyp1b1 loss with reduced STAT3 Ser727 phosphorylation and VEGF production, without establishing receptor/JAK dependence. The two GO:0046427 assertions are now MARK_AS_OVER_ANNOTATED for pathway specificity beyond these measurements. Ser727 is not treated as proof that all JAK signaling is absent. The positive VEGF-production annotations remain intact.
Hydroperoxy-icosatetraenoate dehydration remains directly measured human CYP1B1 chemistry in PMID:21068195. Its action is now KEEP_AS_NON_CORE, consistent with its secondary status and unresolved intact-tissue flux. The actual microsomes retained POR and received no added NADPH; the existing exact excerpt and numerical source note remain unchanged. This change does not reject the enzyme assay.
The three accepted estrogen, retinoid and arachidonate cores remain intact. Steroid hydroxylase activity continues to represent the independently demonstrated C4 reaction while its specific term proposal is unresolved; the accepted C2 activity alone does not represent the C4 product. This differs from broad steroid metabolism, already entailed by accepted estrogen metabolism. Current saved GO definitions explicitly place GO:0016712 beneath GO:0004497 by is_a, so the existing donor-class refinements retain their wording. The NO synthesis-versus-bioavailability uncertainty remains; a candidate positive-regulation term does not itself resolve that experimental distinction.
The primary source table note remains a compact guide to original experiments and target cells, not independent experimental evidence. The two numerical annotation reasons now identify that locator explicitly without adding a self-supporting quote from authored commentary. Relevant primary locators are PMID:23568032, Methods apoptosis/viability and ROS assays, Results Figures 3, 4, 9 and 10 and the corresponding Discussion; PMID:23979599, Methods apoptosis assay and Figure 4C; and PMID:21068195, eicosanoid-metabolism Methods and Table 2. No new quotations, source objects, products, core units or reference identifiers are introduced. Prior notes and history remain unchanged.
This checkpoint supersedes earlier action-count summaries without rewriting those historical notes. PMID:20032512 establishes positive endothelial NO regulation, including chronic eNOS changes and acute restoration of NO availability without restoration of eNOS. Its DAF-FM assay and discussion retain synthesis versus bioavailability alternatives. Rows 36/37 now mark the biosynthetic-process assertion as over-annotated because the study does not establish a CYP1B1-performed chemical, structural or cofactor step in NO formation. This is not a claim that every biosynthetic participant must be an NO synthase or that CYP1B1 cannot affect NO synthesis. No replacement regulatory annotation is added.
The other four challenged rows remain unresolved for distinct reasons. PMID:23568032 supplies positive Bax/Bim/Akt and caspase context, and PMID:23979599 supplies peroxide-induced caspase evidence, but the precise oxidative intrinsic-apoptosis route remains uncertain (rows 53/54). NAC lowered the ROS readout; no apoptotic rescue by NAC is inferred. Lipid peroxidation and oxidative-stress phenotypes do not identify a lipid-breakdown step, while the full PMID:19005183 original remains inaccessible (rows 58/59). These are source-specific uncertainties, not an access-only definition of UNDECIDED or a rejection of the positive mouse phenotypes. The existing receptor/JAK over-annotation judgment concerned untested defining receptor/JAK elements and does not mechanically determine these other pathway judgments.
The steroid-catabolic IBD at PTN001210461 was independently reconstructed on the actual human path (585 family nodes, 18 target-path nodes); its P05177 donor and PMID:18356043 IMP are real. Row 33 remains uncertain because the precise inherited catabolic scope and original donor assay are not resolved, not because of a phylogenetic error or donor count. Positive human estrogen hydroxylation remains accepted.
Three integrated catalytic cores remain: estrogen hydroxylation, retinoid oxidation and arachidonate oxygenation. PMID:23821647 directly distinguishes the human C4 and C2 estradiol products, with C4 favored in the stated preparation. GO:0101021 is already represented by six accepted objects (five EXP and one IEA); the C4 term proposal remains separate. The arachidonate unit includes both hydroxylation and epoxidation, with the specific EET branches retained in existing annotations. No extra parent/child units or unsupported schema slot are introduced merely to increase machine-readable term counts.
Current disposition: 52 ACCEPT, 32 KEEP_AS_NON_CORE, 17 MODIFY, 5 UNDECIDED, 4 MARK_AS_OVER_ANNOTATED and 5 TOO_GENERAL (115 total). The remaining UNDECIDED rows are 33, 53, 54, 58 and 59. The submitted PR body already reported seven before this follow-up; its alleged stale nine-count was not present. All 115 source objects, 49 reference identifiers/titles, official product objects, eight existing quotation objects, three core objects and previous note bytes are preserved. No NEW annotations, new publication cache, source edit or history rewrite is part of this finite follow-up.
The undated opening sections record the initial review checkpoint. In particular, the paragraph under “Contextual and unresolved functions” that lists nine UNDECIDED source objects is historical, not the current disposition. At approved commit 4c4360695, the 115 assessments comprise 52 ACCEPT, 32 KEEP_AS_NON_CORE, 17 MODIFY, 5 UNDECIDED, 4 MARK_AS_OVER_ANNOTATED and 5 TOO_GENERAL. The five unresolved objects concern steroid catabolism (row 33), oxidative-stress intrinsic apoptosis (rows 53/54) and lipid catabolism (rows 58/59). The two NO-biosynthesis objects (rows 36/37) and two receptor/JAK–STAT objects (rows 92/93) are MARK_AS_OVER_ANNOTATED, with the positive surrounding biology retained. Row numbers here use the same zero-based convention as the preceding checkpoints.
The preceding Follow-up 3 sentence about the “submitted PR body” referred only to that separate artifact, which reported seven UNDECIDED objects before that follow-up; it did not deny the historical nine-object paragraph in these notes. The PR body captured at 11:00 UTC on 2026-10-16 already reports the current five UNDECIDED and four MARK_AS_OVER_ANNOTATED objects. This appendix clarifies the checkpoints without revising prior notes, annotations or scientific conclusions.
Rows: GO:0005739 IBA (is_active_in, PTN001210461), IEA GO_REF:0000120 (Q64429 + SL-0173),
ISS GO_REF:0000024 (Q64429; GOA has two ISS rows, dated 2014 and 2024, reviewed as one).
No sub-compartment rows. All KEEP_AS_NON_CORE; kept, with dated reasons.