HemW (formerly YggW) is a cytoplasmic heme chaperone of Escherichia coli. It is a radical S-adenosyl-L-methionine (SAM) superfamily protein in the HemW subfamily of the HemN-like (anaerobic coproporphyrinogen III oxidase) family, and the bacterial counterpart of human RSAD1. Like other radical SAM proteins it has a radical SAM core domain whose three cysteines (Cys16, Cys20 and Cys23) and a SAM ligate a [4Fe-4S] cluster, followed by a C-terminal HemN_C domain. Purified HemW binds one heme b per subunit, apparently covalently. Heme binding does not need the cluster, but the intact cluster promotes dimerization and is required for heme transfer. In bacterial two-hybrid assays HemW interacts with the bacterioferritins, which are its likely heme donors, and with NarI, the heme b-containing membrane subunit of the respiratory nitrate reductase NarGHI. Heme-loaded HemW transfers its heme to heme-depleted, inactive NarGHI in membranes from a heme-deficient strain and restores more than half of its nitrate-reducing activity. HemW is not a coproporphyrinogen III dehydrogenase: it cannot replace HemN in a hemN mutant, despite its sequence similarity and historical annotation. A hemW mutant grows only slightly worse under nitrate-respiring conditions, which suggests that a second route of heme delivery exists. HemW binds SAM and cleaves it only at a low, HemN-like background rate, and it is not known whether radical SAM chemistry contributes to heme transfer.
Definition: Binding to heme and delivering it to a specific acceptor protein, typically an apo-hemoprotein, into which the heme is incorporated as a cofactor.
Justification: HemW is the clearest case of a protein whose function is to deliver heme to a client hemoprotein. E. coli HemW passes bound heme to heme-depleted nitrate reductase and restores its activity, in a cluster-dependent way (PMID:29282292). GO has no term for this. QuickGO searches for "heme chaperone", "heme transfer", "heme insertion", "heme delivery", "heme carrier" and "heme incorporation" (checked 2026-10-19) return only heme binding (GO:0020037), membrane transport terms (GO:0015886 heme transport, GO:0015232 heme transmembrane transporter activity) and GO:0017003 protein-heme linkage. GO:0017003 is a process for covalent heme attachment, whereas the NarI heme b is bound non-covalently. GO:0016530 metallochaperone activity is restricted to metal ions, and heme is a metal complex, not an ion. The new term would sit under GO:0140104 molecular carrier activity, beside metallochaperone activity, and would also suit other heme chaperones such as CcmE and PGRMC proteins.
Parent term: molecular carrier activity
Supporting Evidence:
| GO Term | Evidence | Action | Reason |
|---|---|---|---|
| GO:0003824 catalytic activity | IEA GO_REF:0000002 | KEEP AS NON CORE | Summary: Root-level catalytic activity from the radical SAM InterPro signatures. HemW has the radical SAM architecture and a [4Fe-4S] cluster, but no reaction has been established. Its demonstrated activity, transferring bound heme to an apo-hemoprotein, is a carrier (chaperone) activity rather than a catalysed chemical conversion. SAM had no effect on the transfer in vitro (PMID:29282292, full text). Reason: Per the RADICAL_SAM treatment rule for generic rows: where no reaction is known, GO:0003824 is already as precise as the evidence allows, so it is retained but not core. This matches the decision for the human orthologue RSAD1 (genes/human/RSAD1). Supporting Evidence: PMID:29282292 As other radical SAM proteins, HemW contains three cysteines and one SAM coordinating an [4Fe-4S] cluster PMID:37101575 The exact radical SAM-based mechanism of heme transfer remains to be determined |
| GO:0004109 coproporphyrinogen oxidase activity | IEA GO_REF:0000002 | REMOVE | Summary: InterPro2GO mapping from IPR004559 "HemW-like" to coproporphyrinogen oxidase activity. HemW was tested directly and lacks the activity: E. coli hemW expressed in a hemN deletion strain does not restore anaerobic growth, whereas hemN does (PMID:29282292, full text), and UniProt records that HemW has no coproporphyrinogen III dehydrogenase activity in vitro. The L. lactis orthologue also lacks it in vivo and in vitro (PMID:22142238). Besides being wrong for this clade, GO:0004109 names the oxygen-dependent HemF reaction, not even the SAM-dependent reaction of HemN (GO:0051989). Reason: A signature-to-GO mapping contradicted by direct experiment on this protein. The HemW-like signature defines the non-catalytic sister subfamily of HemN. This matches the REMOVE for human RSAD1, triggered by the same InterPro entry. Supporting Evidence: file:ECOLI/hemW/hemW-uniprot.txt Does not have coproporphyrinogen III dehydrogenase activity in vitro (PubMed:29282292). PMID:22142238 showed that it is devoid of CPDH activity in vivo and in vitro file:ECOLI/hemW/hemW-uniprot.txt SIMILARITY: Belongs to the anaerobic coproporphyrinogen-III oxidase family. HemW subfamily. |
| GO:0005737 cytoplasm | IBA GO_REF:0000033 | ACCEPT | Summary: Phylogenetic inference of cytoplasmic location from the HemN/HemW family node. HemW is a soluble protein described as the cytoplasmic heme chaperone (PMID:37101575). It is proposed to move to the membrane to deliver heme to NarI, and the L. lactis orthologue has a heme-free cytosolic form and a heme-loaded membrane-associated form (PMID:22142238). Reason: Location is a family-wide property that the node placement rightly captures, and unlike the enzymatic rows it does not depend on the HemN chemistry. For the bacterial protein the cytoplasm is the right level of detail (for the human orthologue the matching row was refined to mitochondrion). A peripheral membrane-associated pool is likely, but no membrane term is supported directly for E. coli HemW. Supporting Evidence: PMID:37101575 heme is inserted into enzymes by the cytoplasmic heme chaperone HemW file:ECOLI/hemW/hemW-uniprot.txt SUBCELLULAR LOCATION: Cytoplasm {ECO:0000305|PubMed:29282292}. |
| GO:0005737 cytoplasm | IEA GO_REF:0000120 | ACCEPT | Summary: Cytoplasm from UniProt's subcellular-location mapping and the HemW-like signature. Reason: Consistent with the IBA row for this term. Supporting Evidence: file:ECOLI/hemW/hemW-uniprot.txt SUBCELLULAR LOCATION: Cytoplasm |
| GO:0005737 cytoplasm | ISS GO_REF:0000024 | ACCEPT | Summary: Transferred from Lactococcus lactis HemW (Q9CGF7), whose cytosolic form was shown directly (PMID:22142238). Reason: Sound ortholog transfer within the HemW subfamily; consistent with the IBA row. Supporting Evidence: PMID:22142238 In vivo, HemW occurred as a haem-free cytosolic form, as well as a haem-containing membrane-associated form |
| GO:0006779 porphyrin-containing compound biosynthetic process | IBA GO_REF:0000033 | REMOVE | Summary: Phylogenetic inference that HemW takes part in tetrapyrrole biosynthesis. It comes from the same ancestral node, PTN000358281, that gave this row to human RSAD1, and that node's descendant evidence comes from HemN-type biosynthetic enzymes. E. coli HemW performs no biosynthetic step: it cannot replace HemN, has no coproporphyrinogen III dehydrogenase activity, and acts on finished heme b, delivering it to the nitrate reductase (PMID:29282292). Reason: This disputes where the node was placed, not how many donors it has. The PAINT tree itself separates the two clades for the molecular function: coproporphyrinogen dehydrogenase activity (GO:0051989) sits on the HemN-clade node PTN000358335 and reaches E. coli HemN but not HemW. Only the biosynthetic process term was placed deeper, at a node that also covers the HemW subfamily. HemW is outside the clade that performs the biosynthetic step, which was shown for this very protein. Its heme chaperone role acts downstream on the finished cofactor, so it does none of the work of making a porphyrin. This is the same decision and reasoning as for RSAD1 (genes/human/RSAD1). Propagation Review Root cause: PROPAGATION BAD Failure modes: PSEUDO OR SUBACTIVITY LOSS FUNCTIONAL DIVERGENCE Sources checked: PANTHER:PTN000358281 SUPPORTS SOURCE BUT NOT TARGET The node's descendant evidence is HemN-type coproporphyrinogen dehydrogenase activity, which is genuinely biosynthetic in those donors. The process term is placed too deep, because the HemW subfamily descending from this node was shown experimentally (E. coli and L. lactis) to lack that chemistry and to act on finished heme instead. The molecular-function counterpart was placed correctly, at the HemN-only node PTN000358335. Supporting Evidence: file:ECOLI/hemW/hemW-uniprot.txt Does not have coproporphyrinogen III dehydrogenase activity in vitro (PubMed:29282292). PMID:29282292 indicate that the radical SAM protein family HemW/RSAD1 is a heme chaperone PMID:22142238 HemW-like proteins form a distinct phylogenetic clade that has not previously been recognized |
| GO:0006779 porphyrin-containing compound biosynthetic process | IEA GO_REF:0000002 | REMOVE | Summary: The same biosynthetic process claim mapped from the HemW-like InterPro signature. That signature identifies the subfamily shown not to catalyse the coproporphyrinogen step. Reason: Consistent with the IBA row for this term and with RSAD1. HemW performs no step of porphyrin biosynthesis, and the IPR004559 mapping to GO:0006779 should be withdrawn. Supporting Evidence: PMID:22142238 However, such a function has never been demonstrated for a HemW-like protein |
| GO:0006779 porphyrin-containing compound biosynthetic process | ISS GO_REF:0000024 | REMOVE | Summary: Transferred by sequence similarity from L. lactis HemW (Q9CGF7). The donor annotation is an IDA citing PMID:22142238, yet that paper's abstract reports that L. lactis HemW is devoid of coproporphyrinogen dehydrogenase activity, notes that L. lactis cannot synthesize heme, and proposes a role in heme trafficking. Whatever the donor curator meant, the claim cannot be transferred to E. coli HemW, which was itself shown not to perform the biosynthetic step (PMID:29282292). Reason: Contradicted for the target by direct evidence on the target. Heme delivery to hemoproteins happens after heme biosynthesis and is not part of it. The donor annotation looks doubtful too, but its full text was not reviewed here, so the source is flagged as unresolved rather than wrong. Propagation Review Root cause: UNRESOLVED Failure modes: FUNCTIONAL DIVERGENCE Sources checked: UniProtKB:Q9CGF7 UNRESOLVED The L. lactis HemW donor holds an IDA to GO:0006779 citing PMID:22142238, whose abstract reports no coproporphyrinogen dehydrogenase activity in vivo or in vitro, in an organism that cannot make heme. A heme-trafficking reading of the paper would not support a biosynthetic process term. The donor row deserves re-examination against the full text. Supporting Evidence: file:ECOLI/hemW/hemW-uniprot.txt Does not have coproporphyrinogen III dehydrogenase activity in vitro (PubMed:29282292). PMID:22142238 Lactococcus lactis cannot synthesize haem |
| GO:0020037 heme binding | IDA PMID:29282292 The radical SAM protein HemW is a heme chaperone. | ACCEPT | Summary: Purified E. coli HemW binds one heme per subunit. The heme survives SDS and butanone extraction, which suggests covalent or very tight binding, and binding does not depend on the [4Fe-4S] cluster (PMID:29282292). The L. lactis orthologue likewise binds heme b covalently, one per monomer (PMID:22142238). Reason: Experimentally established and the molecular basis of HemW's chaperone function. Core, and the only existing MF row that describes what HemW does. Heme delivery itself has no GO term and is proposed below. Supporting Evidence: PMID:29282292 we observed one heme per subunit of HemW file:ECOLI/hemW/hemW-uniprot.txt Binds one molecule of heme per monomer, possibly covalently. |
| GO:0042803 protein homodimerization activity | IDA PMID:29282292 The radical SAM protein HemW is a heme chaperone. | KEEP AS NON CORE | Summary: Size-exclusion chromatography of reconstituted HemW resolves monomer and dimer fractions, with more cluster in the dimer, while a cluster-less triple cysteine mutant does not form the dimer (PMID:29282292). The monomer-dimer equilibrium is dynamic, and in the proposed model the cluster-containing dimer is the form that associates with the membrane to deliver heme (PMID:37101575). Reason: Experimentally supported and specific, but it is a structural property linked to the chaperone cycle, not the function itself. Retained as non-core. Supporting Evidence: PMID:29282292 an intact iron-sulfur cluster was required for HemW dimerization file:ECOLI/hemW/hemW-uniprot.txt SUBUNIT: Binding of the [4Fe-4S] cofactor promotes dimerization. |
| GO:0051536 iron-sulfur cluster binding | IEA GO_REF:0000002 | MODIFY | Summary: Generic iron-sulfur cluster binding from the radical SAM signatures. The cluster is a [4Fe-4S] cluster coordinated by three cysteines and a SAM (PMID:29282292). Reason: Per the RADICAL_SAM treatment rule for generic rows, replace with GO:0051539, which the gene already carries by IDA, IBA, ISS and IEA. Proposed replacements: 4 iron, 4 sulfur cluster binding Supporting Evidence: PMID:29282292 As other radical SAM proteins, HemW contains three cysteines and one SAM coordinating an [4Fe-4S] cluster |
| GO:0051539 4 iron, 4 sulfur cluster binding | IBA GO_REF:0000033 | KEEP AS NON CORE | Summary: Phylogenetic inference of [4Fe-4S] cluster binding across the HemN/HemW family, with E. coli HemW's own experimental annotation among the descendant evidence. Here the deep node placement is sound: the cluster-binding motif and cofactor are kept in the HemW clade. Reason: True and as precise as warranted. Per the RADICAL_SAM treatment rule, GO:0051539 is core only where catalysis is established, and no reaction is known for HemW, so this is kept non-core, as for RSAD1. The cluster is nonetheless needed for HemW's chaperone activity: the cluster-less mutant binds heme but cannot transfer it. Supporting Evidence: PMID:29282292 intact iron-sulfur cluster was required for HemW dimerization and HemW-catalyzed heme transfer but not for stable heme binding |
| GO:0051539 4 iron, 4 sulfur cluster binding | IDA PMID:29282292 The radical SAM protein HemW is a heme chaperone. | KEEP AS NON CORE | Summary: Anaerobically purified, reconstituted E. coli HemW contains a [4Fe-4S] cluster ligated by Cys16, Cys20, Cys23 and SAM, characterized by UV-visible and Mossbauer spectroscopy and cyclic voltammetry. The C16S/C20S/C23S triple mutant lacks the cluster (PMID:29282292). Reason: Consistent with the IBA row for this term; non-core while no catalytic reaction is established. Supporting Evidence: PMID:29282292 As other radical SAM proteins, HemW contains three cysteines and one SAM coordinating an [4Fe-4S] cluster |
| GO:0051539 4 iron, 4 sulfur cluster binding | IEA GO_REF:0000002 | KEEP AS NON CORE | Summary: The same [4Fe-4S] cluster binding claim from the HemW-like signature. Reason: Consistent with the IBA and IDA rows for this term. Supporting Evidence: file:ECOLI/hemW/hemW-uniprot.txt Binds 1 [4Fe-4S] cluster. The cluster is coordinated with 3 cysteines and an exchangeable S-adenosyl-L-methionine (Probable). |
| GO:0051539 4 iron, 4 sulfur cluster binding | ISS GO_REF:0000024 | KEEP AS NON CORE | Summary: Transferred from L. lactis HemW, which contains an iron-sulfur cluster whose loss makes the protein monomeric (PMID:22142238). Reason: Sound transfer within the HemW subfamily; consistent with the other rows for this term. Supporting Evidence: PMID:22142238 Recombinantly produced, purified HemW contained an Fe-S (iron-sulfur) cluster and was dimeric |
Loading supporting content…
Download this section (compressed HTML)Q: Should HemW carry a biological process term for its role in nitrate reductase maturation, for example GO:0017004 cytochrome complex assembly, given that its client NarI is a cytochrome b? The in vivo phenotype is slight, so no such term is asserted here.
Q: HemW binds SAM (shown with [14C]SAM) but no SAM-dependent reaction has been shown. Should GO:1904047 S-adenosyl-L-methionine binding be asserted for it, as it is for the catalytic radical SAM enzymes?
Q: Can the InterPro2GO mappings of IPR004559 (HemW-like) to GO:0004109 and GO:0006779 be withdrawn, and the PAINT annotation of GO:0006779 be moved from PTN000358281 to the HemN-clade node PTN000358335, where coproporphyrinogen dehydrogenase activity already sits?
Q: Is the L. lactis HemW (Q9CGF7) IDA to porphyrin-containing compound biosynthetic process (PMID:22142238) intended, given that the paper reports no coproporphyrinogen dehydrogenase activity?
Suggested experts: Marc Solioz, Dieter Jahn
Experiment: Compare heme b content and specific activity of NarGHI purified from wild-type and hemW strains, and from strains that also lack the bacterioferritins, grown anaerobically on glycerol with nitrate.
Hypothesis: HemW is the main heme delivery route to NarI under nitrate respiration in vivo.
Type: genetics with biochemical phenotyping
Experiment: Map the heme attachment site of heme-loaded HemW by proteolysis and LC-MS/MS of heme-containing peptides, then test whether mutating that residue blocks binding or only transfer.
Hypothesis: The HemW-bound heme is covalently attached through a specific residue.
Type: mass spectrometry / mutagenesis
What is not known — curated, literature-grounded statements of the open unknowns (the inverse of core functions).
Gap: How the [4Fe-4S] cluster enables heme transfer, and whether SAM cleavage or radical chemistry takes part, is unknown. HemW binds SAM and cleaves it only at a low background rate, and SAM did not affect heme transfer in vitro.
OPEN BIOLOGY MF_DARK
What is known: Established: the cluster is ligated by three cysteines and a SAM; it is not needed for heme binding but is needed for dimerization and for heme transfer to NarGHI.
Significance: Determines whether HemW is a radical SAM enzyme in any functional sense or a radical SAM-fold protein that uses its cluster structurally, which also decides how its human orthologue RSAD1 should be annotated.
What would resolve it: Quantify 5'-deoxyadenosine formation by reduced, heme-loaded HemW in the presence of apo-NarGHI membranes, and compare transfer by SAM-site and cluster-ligand variants.
Provenance (the field's own admissions):
Gap: The client range of HemW beyond NarGHI, and the identity of the backup heme-delivery route that lets a hemW mutant grow almost normally, are unknown.
OPEN BIOLOGY BP_DARK
What is known: Established: HemW interacts with NarI and the bacterioferritins and restores heme to NarGHI in vitro. A hemW mutant shows only a slight growth defect on glycerol with nitrate.
Significance: Without a clear in vivo phenotype, HemW's place in respiratory hemoprotein maturation, and therefore its biological process, cannot be fixed.
What would resolve it: Measure heme content and activity of nitrate reductase, cytochrome bo3 and bd oxidases, and other b-type hemoproteins in hemW single and combined mutants, and identify HemW partners by in vivo crosslinking.
Provenance (the field's own admissions):
Loading supporting content…
Download this section (compressed HTML)Loading supporting content…
Download this section (compressed HTML)