COA4 (Q9NYJ1) — research notes

Human cytochrome c oxidase assembly factor 4 homolog, mitochondrial.
Synonyms: CHCHD8, E2IG2. HGNC:24604. 87 aa. Chromosome 11.

Why this gene, now

COA4 was one of the ~100 "orphan" mitochondrial proteins in MitoCarta3.0. UniProt still
carries only FUNCTION: Putative COX assembly factor. {ECO:0000250} — i.e. by similarity,
with no experimental function statement. The 2026 MitoMatch paper
(doi:10.1038/s41467-026-77112-z) supplies the
missing direct physical interaction (COA4–COX11) and the first human loss-of-function
characterization, converting a long-standing genetic inference into experimentally grounded
human data. See projects/MITO_INTERACTOME.md.

Structure and import

Yeast Coa4 — the prior evidence chain

Discovery (2009, as Cmc3). Systematic survey of the 14 yeast twin Cx9C proteins; most
deletions impair respiratory chain assembly or stability [PMID:19703468 — abstract only in cache].

Renaming and first characterization (2010). Isolated as an allele-specific suppressor of a
Shy1 Leigh-syndrome mutant; renamed Coa4. Key phenotypes PMID:20624914:
- "Cells lacking Coa4 are depressed in CcO activity but show no impairment in Cox1 maturation or
formation of the Shy1-stabilized Cox1 assembly intermediate" — so Coa4 acts downstream of
Cox1 synthesis and hemylation.
- "Cells lacking Coa4 resemble shy1 Δ cells in exhibiting a reduced mitochondrial copper content."
- "Coa4 may likewise be part of the Cu(I) routing pathway."
- Respiratory function restored by CYC1 (cytochrome c) overexpression.

ROS confound (2013). coa4Δ respiratory growth is rescued by DTT/GSH/ascorbate without
restoring CcO assembly — the growth defect is largely H₂O₂ from partially assembled CcO
intermediates, not the assembly defect itself PMID:23198688. Important caveat when reading respiratory-growth rescue as evidence of restored assembly.

Genetic placement in the copper pathway (2022). Targeted suppressor screen PMID:35666203:
- COX11 overexpression restores Cox1 abundance, CcO assembly and respiration in coa4Δ.
- The rescue requires the copper-coordinating cysteines of Cox11 — so it is restored copper
delivery, not ROS suppression: "cysteine mutants of Cox11 that are incapable of binding copper,
failed to rescue the respiratory defect".
- Coa4 and Cox11 abundance are reciprocally regulated in mitochondria.
- coa4Δ cells have reduced cellular copper; copper supplementation partially rescues.
- Human COA4 complements yeast coa4Δ — "we demonstrate that human COA4 can replace the
function of yeast Coa4 indicating its evolutionarily conserved role". This is what licenses
transferring the yeast mechanism to the human protein.
- Directionality: "only Cox11 can rescue coa4Δ and not the other way around ... suggests that
Coa4 acts upstream of Cox11 in the copper delivery pathway".
- Explicit negative result: "Our initial attempts to detect protein:protein interaction
between Cox11 and Coa4 via coimmunoprecipitation/mass spectrometry experiments were not
successful (data not shown)." The 2026 paper closes exactly this gap.

MitoMatch (2026) — what is new

Swaminathan et al., The predicted interactome of the human mitochondrial proteome,
Nat Commun 2026 (in press), doi:10.1038/s41467-026-77112-z.
No PMID assigned yet, so it is cited and cached by DOI as
publications/DOI_10.1038_s41467-026-77112-z.md (full text retrieved via OpenAlex under
CC-BY), which means its supporting quotes are machine-verifiable like any PMID reference.

Prediction: AlphaFold-Multimer predicts an evolutionarily conserved COA4–COX11 interaction, with
IMS-localized COA4 contacting the IMS-facing domain of COX11 — topologically coherent.

Experimental validation and human phenotyping:

Experiment Result
Yeast Coa4-V5 co-IP/MS, DSSO-crosslinked mitochondria (n=3) Cox11 recovered; also IMS phosphatase Ptc5
COX11-FLAG IP from 293T mitochondria (n=3) Recovers COA4-V5, plus COX1 and COX2
Reciprocal COA4-V5 IP (n=3) Recovers COX11-FLAG; not COX1 or COX2
CRISPR COA4-KO, MCH58 fibroblasts, 2 clones COA4 absent
COX11 immunoblot in KO (n=3) Striking reduction in COX11 abundance
ICP-MS of KO mitochondria (n=3) Reduced Cu; Fe, Zn, Mn unaffected
BCS copper-chelator titration (n=3) COX1 loss more pronounced in KO than WT
BN-PAGE/western (n=2) Drastic, specific loss of complex IV–containing supercomplexes
Seahorse OCR (n=3) Reduced respiration

The crosslinking step is plausibly why 2026 succeeded where 2022 failed — consistent with a
transient or low-affinity contact.

Independent support for COA4–COX11 that predates the 2026 paper

UniProt Q9NYJ1 records INTERACTION: Q9NYJ1; Q9Y6N1: COX11; NbExp=2; IntAct=EBI-22303661, EBI-2963275, sourced from BioPlex PMID:33961781. This is also the origin of the GOA
IPI GO:0005515 protein binding row with WITH/FROM UniProtKB:Q9Y6N1. So the human COA4–COX11
interaction had affinity-purification support before MitoMatch; the 2026 work adds reciprocal
targeted co-IP, cross-species conservation, and functional consequence.

Annotation assessment

Well supported and core
- GO:0033617 mitochondrial respiratory chain complex IV assembly — previously IBA/IEA only.
Now backed by human KO data (complex IV supercomplex loss, reduced OCR, COX1 sensitized to
copper limitation) plus two decades of yeast genetics. This is the core BP.
- GO:0005758 mitochondrial intermembrane space — twin CX9C/MIA substrate, yeast localization
data, InterPro IPR039870. Core CC.

Uninformative but not wrong
- GO:0005515 protein binding (IPI, PMID:33961781, with COX11). Per repository guidelines this
term is uninformative regardless of evidence quality. The real content — a specific,
conserved, functionally consequential contact with COX11 — is carried by the complex IV
assembly annotation and by core_functions. Mark over-annotated, keep the partner recorded in
supporting_entities.

Redundant generalization
- GO:0005739 mitochondrion (×3: IEA, HTP, IDA). Correct but subsumed by the IMS annotation.

Candidate not currently annotated
- GO:0006878 intracellular copper ion homeostasis. Reduced mitochondrial/cellular copper is
reproducible across yeast coa4Δ [PMID:20624914, PMID:35666203] and human COA4-KO (2026, with
Fe/Zn/Mn controls). But it may be a downstream consequence of failed CcO metallation
rather than a separate role — the 2022 authors themselves frame it as one of "2 distinct but
closely related roles", and comparable copper deficiency is seen in SCO1/SCO2/COX10/COX15
patient fibroblasts, where it is attributed to CTR1 turnover/mislocalization. Left out of
core_functions; raised in suggested_questions instead.

Molecular function: still unknown. No catalytic activity is established, and metallochaperone
activity is positively excluded. The best current statement is an accessory/regulatory role
supporting COX11-mediated CuB metallation of COX1 — which the ontology does not currently express
well for this case. Asserting a specific MF term here would over-reach; core_functions therefore
carries the BP + CC with a mechanistic description rather than an invented MF.

Open questions