The review combines one NADPH-dependent ferredoxin-reduction activity with its electron-supply roles in steroid and ubiquinone biosynthesis. All source annotations and seven curated splice products are retained. Reference-only citations identify evidence without claiming a new verbatim quotation was validated.
| Claim | Original evidence and precise scope | Boundary |
|---|---|---|
| Ferredoxin-NADP+ reductase activity and steroid electron supply | PMID:2845396(https://pubmed.ncbi.nlm.nih.gov/2845396/), original PNAS pages 7104–7108, especially page 7104 electron-transfer description and the cDNA/splicing Results. PMID:38885337(https://pubmed.ncbi.nlm.nih.gov/38885337/), Introduction/Figure 1A and Results on patient-derived adrenal-like cells and the Fdxr mutant mouse. | The cloning paper does not compare the two splice enzymes. Steroid output in patient cells and mice is not a purified enzyme-specific assay. |
| Ubiquinone electron supply through FDX2 to COQ6 | PMID:38425362(https://pubmed.ncbi.nlm.nih.gov/38425362/), Results “C5 transformation requires a ferredoxin couple”, Figure 2/Extended Data Figure 3, C1-hydroxylation Results/Figure 4, and Discussion on ancestral reconstruction and short-chain substrates. | Resurrected tetrapod-ancestral FDXR/FDX2, not purified extant human proteins. The measured electron-transfer role supports conserved participation; no stable FDXR complex membership is asserted. |
| Human mitochondrial localization | PMID:29040572(https://pubmed.ncbi.nlm.nih.gov/29040572/), FDXR/MitoTracker Results and Figure 2C–D/Methods. PMID:34800366(https://pubmed.ncbi.nlm.nih.gov/34800366/), original Table S1 sheet B row 343 and Table S2 sheet G row 3234; exact target records. | Positive human organelle evidence, not exclusive localization, membrane topology or separation of every isoform. The MitoCoP protein group contains several FDXR accessions. |
| Submitochondrial source annotations | PMID:9727014(https://pubmed.ncbi.nlm.nih.gov/9727014/), author-original subcellular/submitochondrial Results and Figure 7; the seven Reactome target routes in the database decisions. | Yeast Arh1 inner-membrane fractionation supports the curated ortholog inference. Reactome's matrix compartment and modeled residue span are model assertions, not new human cleavage/topology measurements. |
| Product-dependent targeting | PMID:41511346(https://pubmed.ncbi.nlm.nih.gov/41511346/), Methods 2.4, Results 3.1, Figure 1/Supplemental Figure S1 and Discussion; isoform-source joins. | Entrez/paper 7 = UniProt 4, and Entrez/paper 4 = UniProt 7. Tagged MCF7 localization and mobility do not establish exact endogenous cleavage or a direct DNA-repair/cytosolic-substrate function. |
| Biallelic disease context and correction | PMID:29040572(https://pubmed.ncbi.nlm.nih.gov/29040572/), patient phenotype and fibroblast Results; PMID:29554255(https://pubmed.ncbi.nlm.nih.gov/29554255/), publisher corrigendum. | Clinical context is bounded to mitochondrial disease, optic atrophy and neurological impairment. The corrigendum corrects three Figure 1 genotype labels; no Figure 2 localization correction is stated. |
The direct source guide records original-page access, the abstract-only PMID:1863359 limitation, and exact proteomics workbook joins. Its complete direct decisions bind each review to its full source object and index. The database source guide documents official ontology/mapping records, Reactome graph routes, HPA images and donor/isoform evidence. Its complete database decisions preserve source-specific inference limits.
The phylogeny evidence record explicitly records that literal IBD evidence, target ancestry and MSA were not recovered. Historical release-19 PTHR11938 placement is not silently joined to current PTHR48467. Target self-WITH is expected descendant grounding and is not treated as circularity. The current family label is not converted into a new FDXR activity.
The main notes journal, unchanged direct journal and unchanged database journal preserve synthesis and source-access history. Genuine provider authentication/quota failures produced no research report; this guide is manually authored evidence documentation. Normal caches and machine records are immutable, and discovery-only preserved caches are not presented as independently read support for the functional synthesis.
The commit-pinned production GO-CAM explicitly models FDXR reductase activity as part of ubiquinone biosynthesis and directly positively regulating FDX2 electron-transfer activity. The direct guide explains the preserved upstream source qualifier, the two probable evidence-code CHECK requests, and the actual typed ontology result behind the broad-metabolism non-core decision. No molecular source fields or model bytes are rewritten.
Shared PAINT, yeast-donor, Reactome-coordinate and isoform limits are retained once in the database guide rather than repeated in every reason. It also records UniProt's tentative Long-isoform inactivity and distinguishes that inference from the tagged 4/7 localization study. FAD remains part of the reductase mechanism, not an added binding-function claim.