RSAD1 (human, Q9HA92) — curation notes

Part of the RADICAL_SAM project (projects/RADICAL_SAM.md). Literature retrieval was
run with the asta provider (RSAD1-deep-research-asta.md) because the falcon
provider failed with an authentication error; the asta output turned out to be
retrieval noise (database/annotation-methodology papers, none about RSAD1), so the
evidence below was assembled from the cached publications plus targeted PubMed
searches.

Identity and architecture

• UniProt: "Radical S-adenosyl methionine domain-containing protein 1, mitochondrial",
  with AltName: Full=Putative heme chaperone {ECO:0000303|PubMed:29282292}
  [file:human/RSAD1/RSAD1-uniprot.txt].
• Radical SAM core domain 34–270, N-terminal mitochondrial transit peptide 1–17,
  mature chain 18–442; signatures: PF04055 Radical_SAM, PF06969 HemN_C,
  IPR004559 HemW-like, TIGR00539 hemN_rel, PANTHER PTHR13932:SF5
  [file:human/RSAD1/RSAD1-uniprot.txt].
• Family placement: "Belongs to the anaerobic coproporphyrinogen-III oxidase" family,
  HemW subfamily [file:human/RSAD1/RSAD1-uniprot.txt]. This is the key point: HemW
  is a sister subfamily to the catalytic HemN (coproporphyrinogen III dehydrogenase,
  EC 1.3.98.3) proteins, not a HemN itself.
• UniProt function text: "May be a heme chaperone, appears to bind heme." and
  "Binds 1 [4Fe-4S] cluster." [file:human/RSAD1/RSAD1-uniprot.txt]. UniProt explicitly
  records that homologous bacterial proteins lack oxygen-independent
  coproporphyrinogen-III oxidase activity.

HemW is not a coproporphyrinogen oxidase/dehydrogenase

• Lactococcus lactis HemW: "showed that it is devoid of CPDH activity in vivo and in
  vitro" and HemW-like proteins "are usually annotated as CPDHs
  (coproporphyrinogen III dehydrogenases). However, such a function has never been
  demonstrated for a HemW-like protein." The protein instead "covalently bound haem b
  in vitro, with a stoichiometry of one haem per monomer", existed as "a haem-free
  cytosolic form, as well as a haem-containing membrane-associated form", and released
  heme on addition of membranes, leading the authors to "propose a role of HemW in
  haem trafficking" PMID:22142238.
• HemW-like proteins "form a distinct phylogenetic clade that has not previously been
  recognized" PMID:22142238 — i.e. the catalytic HemN activity is a property of the
  sister clade, and the HemW clade is defined partly by its loss (the fourth cysteine
  of the HemN CX3CX2CXC motif is not conserved; see also PMID:34311096, which argues
  mycobacterial "HemN" sequences are mis-annotated HemW proteins).

HemW/RSAD1 as a heme chaperone

• E. coli HemW: "we observed one heme per subunit of HemW", "intact iron-sulfur
  cluster was required for HemW dimerization and HemW-catalyzed" "heme transfer but not
  for stable heme binding"; "Of note, heme that was covalently bound to HemW was
  actively transferred to a" heme-depleted, inactive nitrate reductase, restoring
  activity; interaction partners found were bacterioferritins and the heme-containing
  NarI subunit of nitrate reductase PMID:29282292.
• Human protein: "domain-containing 1 (RSAD1) stably bound heme. In conclusion, our
  findings" "indicate that the radical SAM protein family HemW/RSAD1 is a heme
  chaperone" PMID:29282292. Note the human experiment establishes binding; the
  transfer/insertion experiments were done on the bacterial protein. GOA records only
  GO:0020037 heme binding (IDA) from this paper, which is the right scope.
• Independent biochemistry on the human protein: hRSAD1 "was reconstituted with"
  "a [4Fe-4S]2+ cluster that could be reduced to the [4Fe-4S]+ state", binds PPIX and
  its metal analogues (K_A for Fe(III)-PPIX ≈ 1.6 × 10^6 M⁻¹), and the
  "hRSAD1-[4Fe-4S]-heme complex binds" "oxygen, carbon monoxide, and cyanide"; the
  authors conclude only that hRSAD1 "may" "play a significant role in heme-related
  metabolic processes" PMID:41747499. So: cluster + heme binding confirmed on the
  human protein; no reaction demonstrated.
• Reviews of the human radical SAM set still list RSAD1 among enzymes "of unknown
  function" PMID:27145839; the tetrapyrrole-focused review states that "In bacteria,
  heme is inserted into enzymes by the cytoplasmic heme chaperone HemW, a radical SAM
  enzyme structurally highly related to CgdH" PMID:37101575.

Localization

• UniProt: "SUBCELLULAR LOCATION: Mitochondrion" (ECO:0000305), with a predicted
  N-terminal transit peptide [file:human/RSAD1/RSAD1-uniprot.txt].
• MitoCoP high-confidence mitochondrial proteome ("We defined a human mitochondrial
  high-confidence proteome (MitoCoP) of 1,134 protein-coding genes") includes RSAD1
  (GOA HTP row) PMID:34800366.
• The bacterial orthologues are cytoplasmic/membrane-associated [PMID:22142238;
  PMID:37101575]; the eukaryotic protein has relocated to the mitochondrion, so
  GO:0005737 cytoplasm (the IBA transfer from bacterial donors, plus an InterPro
  mapping) is true only in the trivial sense that the mitochondrion is part of the
  cytoplasm, and GO:0005739 is the supported term.

A candidate catalytic activity (2026, not yet in GOA)

• A porcine/mouse picornavirus study reports that RSAD1 "catalyzes methionine
  oxidation of FMDV 3D polymerase and promotes its aggregation and subsequent
  degradation through the autophagy-lysosome pathway", with an in vitro oxidation assay
  using purified GST-RSAD1 vs a cluster-binding mutant (RSAD1-C3A) in SAM/dithionite
  buffer, SAM-domain dependence of the cellular effect, and Rsad1−/− mice being "more
  susceptible to FMDV-, EV71-, or EMCV-triggered death" PMID:42736308.
• This is a single study, mostly in pig cells and mice, on a viral substrate, and the
  "oxidation" chemistry (net two-electron oxidation of a methionine thioether) is
  unusual for a radical SAM enzyme. Treated here as a lead and recorded in
  suggested_questions/suggested_experiments, not asserted as a new molecular
  function or as a defence-response process annotation for the human gene.

Other associations (not annotation-grade)

• Rare-variant burden association of RSAD1 with chronic central serous
  chorioretinopathy in females only, explicitly needing replication PMID:31036833.
• Differential neuronal expression in Alzheimer brain samples PMID:34121475.
• Neither speaks to molecular function; not used in the review.

Decisions taken in the review

• GO:0004109 coproporphyrinogen oxidase activity (IEA, InterPro IPR004559 →
  HemW-like): REMOVE. The mapping is from the HemW-like signature, and HemW is
  precisely the subfamily shown not to have this activity [PMID:22142238;
  PMID:29282292]; UniProt agrees.
• GO:0006779 porphyrin-containing compound biosynthetic process (IBA + IEA):
  REMOVE. The IBD node was placed where coproporphyrinogen dehydrogenase chemistry
  sits (donors are HemN-type bacterial/cyanobacterial proteins); the HemW/RSAD1 clade
  is a sister subfamily that does not perform that step. Heme binding and delivery
  is downstream of tetrapyrrole biosynthesis, so even the chaperone role does not place
  RSAD1 in porphyrin biosynthesis.
• GO:0003824 catalytic activity: KEEP_AS_NON_CORE per the project's treatment
  rule for generic rows — no reaction is established for RSAD1, so the root term is as
  precise as the evidence allows.
• GO:0051536 → MODIFY to GO:0051539; the two GO:0051539 rows →
  KEEP_AS_NON_CORE, since catalysis is not established (project rule), even though
  the cluster is required for heme transfer in the bacterial protein.
• GO:0005737 cytoplasm rows → MODIFY to GO:0005739 mitochondrion.
• No NEW rows. A heme-insertion/chaperone process term would require that the human
  protein itself be shown to perform the insertion; only binding has been shown for
  RSAD1, and the transfer assays are bacterial. The methionine-oxidation activity is a
  single non-human-cell study.