Mitochondrial respiratory Complex V (F1Fo ATP synthase) — subunits and F1 assembly; ATP5F1A/ATP5F1B/ATP5F1D/ATP5F1E/ATP5PO/ATP5MC3/ATP5MD + ATPAF2/TMEM70

Mitochondrial respiratory Complex V (F1Fo ATP synthase), grounded to the human subunits ATP5F1A (UniProtKB:P25705), ATP5F1B (P06576), ATP5F1D (P30049), ATP5F1E (P56381), ATP5PO (P48047), ATP5MC3 (P48201), ATP5MD/DAPIT (Q96IX5) and the assembly factors ATPAF2 (Q8N5M1, GO:0044183) and TMEM70 (Q9BUB7). GO molecular-function/BP/location terms were taken from the completed human gene reviews and verified against the local go.db; Reactome reaction ids/titles were verified against the local reactome cache. The complex is modelled as F1 (catalytic head + central stalk) and Fo (membrane/proton-conducting) PROTEIN_COMPLEX nodes plus an assembly node; the catalytic activity (GO:0046933) is carried by the alpha/beta subunits, while the central-stalk/OSCP/DAPIT subunits are structural (GO:0005198). The a-subunit (MT-ATP6/MT-ATP8) and additional peripheral-stalk subunits (ATP5PB/PD/PF, ATP5ME/MF/MG) are mtDNA-encoded or reviewed elsewhere; the current ATP synthase complex term is GO:0045259 (proton-transporting ATP synthase complex). Complex V is driven by the proton gradient built by Complexes I-IV (oxphos.yaml, Complex III and Complex IV modules). Disorders: mitochondrial complex V (ATP synthase) deficiency nuclear types 1-9 — typically neonatal encephalopathy with 3-methylglutaconic aciduria, lactic acidosis and cardiomyopathy (ATPAF2 = MC5DN1, TMEM70 = MC5DN2 the commonest, ATP5F1E = MC5DN3, ATP5F1D = MC5DN5). NOTE: the ATP5MD folder was initially mis-fetched as Q9UI09 (NDUFA12, Complex I) and re-fetched with the correct DAPIT accession Q96IX5. CANDIDATE ASSESSED, NOT MODELLED — ATHENA (PMID:42547818): a 2026 report describes ATHENA, a 170-aa microprotein translated from the lncRNA LINC00528, as an inner-membrane factor that binds the F1 beta and gamma subunits together with ANT and PiC and promotes assembly of the "ATP synthase-associated macromolecular complex", enhancing ATP synthesis and preserving cristae. It is deliberately not added as an assembly annoton here, for three reasons. (i) Grounding: its only protein identifier, UniProtKB:Q8N1L1, is protein-existence level 5 (Uncertain) and carries UniProt's caution "Product of a dubious CDS prediction. May be a non-coding RNA."; adding it as a part would assert coding status the reference proteome does not currently support, and there is no PANTHER family or GO annotation to ground it against. (ii) Boundary: the asserted role is at the level of the ATP synthasome (Complex V plus the ADP/ATP and phosphate carriers, modelled separately in oxphos_adenine_nucleotide_phosphate_carriers.yaml), an association whose existence as a stable entity is itself debated; the report does not separate an effect on F1Fo holoenzyme biogenesis — the role ATPAF2 and TMEM70 occupy in this node — from an effect on carrier association. (iii) Weight: a single primary report, cached abstract-only, with no independent replication, would otherwise sit alongside two assembly factors backed by human disease genetics. Re-evaluate and add if UniProt upgrades Q8N1L1 above PE 5 (or independent proteomic/Ribo-seq evidence establishes the ORF) and if evidence distinguishes a holoenzyme-assembly role from synthasome association; if the effect proves to be on carrier association only, the carriers module is the better home. This candidate is also recorded as a structured, queryable entry in this module's knowledge_gaps block (status OPEN), which carries the reference-level judgments and the branched resolution path in machine-readable form. CANDIDATES ASSESSED, NOT MODELLED — ATPPM1/ATPPM2 (PMID:42809533): complexome profiling of Perkinsus marinus mitochondria found two proteins, Pmar_PMAR021141 (UniProtKB:C5KGI6, 10 kDa) and Pmar_PMAR014495 (UniProtKB:C5KWE2, 19 kDa, thioredoxin-like), co-migrating with the enlarged myzozoan ATP synthase in every complexome dataset and co-purifying with it by size-exclusion chromatography/mass spectrometry; the authors name them ATPPM1-2 and state that they are either novel subunits or late-stage assembly factors. They are not added here, for three reasons. (i) Scope: every part of this module is grounded on a reviewed human accession with a completed gene review, and these are lineage-specific proteins with no human counterpart — they were undetectable even by homology from Toxoplasma. The admission path, if they are confirmed, is a myzozoan node or a separate myzozoan ATP synthase module, not an extra annoton alongside the human subunits. (ii) Category: subunit versus late assembly factor is exactly the distinction this module encodes by placing ATPAF2/TMEM70 in a separate assembly node, and the available data (co-migration plus co-purification) do not decide it, so there is no node to put them in. (iii) Grounding: both are TrEMBL-only, protein existence "Predicted", with no GO annotations and no PANTHER family, so FAMILY-based participant grounding is unavailable; ATPPM1 has no InterPro signature and ATPPM2's only one is the IPR036249 thioredoxin-like homologous superfamily, with UniProt naming it "Monothiol glutaredoxin-4, putative" — a redox assignment with no established connection to ATP synthesis. Re-evaluate if knockdown/structural work assigns them to a sector or to biogenesis. The same paper resolves a distinct point that needs no module change: a Perkinsus ATPTG1 homolog on its own gene (unfused from CytC1) is in complex with ATP synthase, placing the ATPTG1 subunit role before the apicomplexan gene fusion. Recorded as a structured knowledge_gaps entry (status OPEN, BIOLOGY/MF_DARK) with the verbatim provenance.

Connections