AIGR Gene Hypothesis Deep Research — S. pombe Rqh1 Four-Way Junction Helicase Activity (GO:0009378)

Target gene: rqh1 (UniProt Q09811), Schizosaccharomyces pombe (strain 972 / ATCC 24843), NCBITaxon:284812
Seed hypothesis: S. pombe Rqh1 directly unwinds four-way Holliday-junction DNA substrates, supporting GO:0009378 four-way junction helicase activity.
Focus type: free_text / function-assignment

Summary

Executive verdict: Partially / weakly supported — supported by orthology inference, not by direct experimental evidence. The claim that Rqh1 possesses four-way junction (Holliday-junction) helicase activity (GO:0009378) rests entirely on phylogenetic propagation from experimentally characterized RecQ orthologues (human BLM, human WRN, and budding-yeast Sgs1), not on any direct biochemical assay of purified Rqh1 protein. The GO annotation of Q09811 for GO:0009378 carries evidence code IBA (ECO:0000318, GO_REF:0000033, PAINT/GO_Central), with the "with/from" field naming human BLM (P54132), human WRN (Q14191), and Arabidopsis RecQ loci — i.e., the assignment is inferred from ancestral reconstruction, not measured in fission yeast.

By contrast, the only experimentally supported molecular-function annotation for Rqh1 is GO:0043138 "3'–5' DNA helicase activity" (IDA:PomBase), which is backed by two independent direct biochemical characterizations of purified Rqh1p: Laursen et al. 2003 (PMID: 12724426) and Ahmad et al. 2002 (PMID: 12478586). Both studies demonstrated generic 3'→5' single-strand-displacement helicase activity on standard duplex/tailed substrates and neither tested a four-way-junction or Holliday-junction substrate. Targeted literature searches did not identify any primary paper demonstrating in-vitro four-way-junction unwinding or branch migration by purified Rqh1. Direct demonstration of Holliday-junction branch migration and double-Holliday-junction (dHJ) dissolution exists only for the orthologues BLM, Sgs1, and WRN.

Supporting the plausibility of the inference, S. pombe in-vivo genetics place Rqh1 firmly in pathways that process branched recombination intermediates — most notably a helicase-activity–dependent suppression of mitotic crossovers (Hope et al. 2007, PMID: 17353272) — and a computed InterPro domain-architecture comparison confirms that Rqh1 conserves the full RecQ core, including the RQC winged-helix domain (IPR018982) and HRDC domain (IPR002121) that mediate structure-specific junction recognition in BLM/WRN/Sgs1. These are, however, biological-process (BP) phenotype inferences and structural/evolutionary inferences respectively — they raise prior confidence in the orthology call but do not constitute a direct molecular-function measurement. Curation lead: retain GO:0009378 as an inferred annotation (IBA/ISS); do not upgrade it to an experimental evidence code; keep GO:0043138 as the directly evidenced molecular function.

Key Findings

Finding 1 — GO:0009378 for Rqh1 is an orthology inference (IBA), not a direct assay

The UniProt/PomBase annotation of Q09811 assigns GO:0009378 "four-way junction helicase activity" with evidence code IBA:GO_Central (phylogenetic inference by ancestral-state reconstruction). The only experimentally supported molecular-function annotation for the same protein is GO:0043138 "3'–5' DNA helicase activity" with evidence code IDA:PomBase, derived from Laursen et al. 2003 (PMID: 12724426), which characterized purified Rqh1 as a 3'→5' helicase and did not test four-way-junction substrates. Targeted PubMed searches combining "Rqh1" with "Holliday junction," "branch migration," and "branched substrate" biochemistry returned no primary paper describing such an assay. The direct four-way-junction / HJ branch-migration and dHJ dissolution activities are demonstrated only for the orthologues human BLM and S. cerevisiae Sgs1.

Laursen et al. 2003 (PMID: 12724426): "We show that Rqh1, the single Schizosaccharomyces pombe homologue, is a 3'-to-5' helicase and exists with Top3 in a high-molecular-weight complex."

On the orthologue BLM (PMID: 16670433): "BLM also promotes branch migration of Holliday junctions." — This is the directly demonstrated orthologue activity that is the basis of the orthology inference to Rqh1.

Interpretation: the four-way-junction MF term is a well-motivated inference, but it is not an experimental fact for the fission-yeast protein.

Finding 2 — In-vivo S. pombe genetics support a helicase-dependent role for Rqh1 in processing recombination / four-way-junction intermediates

rqh1Δ cells show a ~4-fold increase in crossover frequency during mitotic double-strand-break repair, and this crossover suppression is dependent on Rqh1 helicase activity (Hope et al. 2007, PMID: 17353272). Multiple additional studies describe Rqh1 as resolving/disassembling recombination joint molecules — e.g. PMID: 39580958 ("RecQ helicase Rqh1 resolves recombination intermediates") and PMID: 31584934 ("Rqh1-mediated disassembly of joint-molecules"). These are mutant-phenotype (IMP/IGI) inferences consistent with in-vivo action on branched/Holliday-junction-containing intermediates, but they do not directly demonstrate four-way-junction unwinding by the purified enzyme.

Hope et al. 2007 (PMID: 17353272): "This suppression of crossover formation by Rqh1 is dependent on its helicase activity."

Interpretation: the genetics tie Rqh1's helicase function to crossover control — a downstream consequence consistent with junction processing — but map onto a biological-process conclusion rather than a molecular-function assay.

Finding 3 — Two independent direct assays show only generic 3'→5' helicase; the GO:0009378 IBA is propagated from experimental BLM/WRN annotations

QuickGO annotation provenance for Q09811 shows GO:0009378 = IBA (ECO:0000318), GO_REF:0000033 (PAINT), with/from human BLM (UniProtKB P54132), human WRN (UniProtKB Q14191), Arabidopsis RecQ loci (AT1G31360, AT4G35740), PANTHER family PTN000344873 — i.e., propagated from experimentally characterized RecQ orthologues, not from any Rqh1 experiment. The 3'–5' DNA helicase MF (GO:0043138) carries two independent IDA (ECO:0000314) references from PomBase: PMID: 12724426 (Laursen 2003) and PMID: 12478586 (Ahmad, Kaplan & Stewart 2002). Both demonstrate generic 3'→5' DNA helicase activity for purified Rqh1p; neither tested four-way-junction / Holliday-junction substrates. Ahmad 2002 additionally reports that helicase activity of Rqh1p is only partially required for its function.

Ahmad et al. 2002 (PMID: 12478586): "Here we show that, like other RecQ family members, the Rqh1p protein displays 3' to 5' DNA helicase activity."

Ahmad et al. 2002 (PMID: 12478586): "unlike other RecQ family members, the helicase activity of Rqh1p is only partially required for its function."

Interpretation: the experimental record for Rqh1 stops at generic 3'→5' helicase. The four-way-junction annotation is entirely a carry-over from experimentally validated orthologues, and the "partially required" caveat warns against treating any single helicase sub-activity as the sole core function.

Finding 4 — Rqh1 conserves the RecQ RQC winged-helix module shared with BLM/WRN/Sgs1, structurally supporting the four-way-junction IBA inference

A computed InterPro domain-architecture comparison (via UniProt REST) across Rqh1 (Q09811), human BLM (P54132), human WRN (Q14191), human RECQ1 (P46063), and S. cerevisiae Sgs1 (P35187) shows that Rqh1 shares the full RecQ core: RecQ helicase family (IPR004589), P-loop NTPase (IPR027417), Helicase C-like (IPR001650), RecQ Zn-binding (IPR032284), the RQC winged-helix domain (IPR018982), and HRDC (IPR002121) — with BLM, WRN, and Sgs1. The RQC winged-helix is the structural module implicated in structure-specific DNA / Holliday-junction recognition and branch migration in RecQ helicases; allosteric-inhibitor structures of BLM provide "a model for branch migration of Holliday junctions by BLM" (PMID: 33647232).

Interpretation: Rqh1 possesses the machinery known to confer HJ recognition/branch-migration in its orthologues. This raises the prior probability that Rqh1 can act on four-way junctions but is structural/evolutionary evidence — it does not substitute for a direct assay.

Mechanistic Model / Interpretation

The distinction that matters for curation is between direct molecular-function evidence and inference (phylogenetic, phenotypic, or structural). The evidence for Rqh1 stratifies cleanly:

         EVIDENCE FOR Rqh1 four-way-junction helicase (GO:0009378)
  ┌───────────────────────────────────────────────────────────────────────────┐
  │ DIRECT MF ASSAY on purified Rqh1                                            │
  │   • 3'->5' DNA helicase (GO:0043138)  ── IDA  ── PMID 12724426, 12478586    │
  │   • four-way junction unwinding      ── NONE (never assayed)     ✗          │
  ├───────────────────────────────────────────────────────────────────────────┤
  │ IN-VIVO GENETICS (BP-level, phenotype)                                     │
  │   • helicase-dependent crossover suppression ── IMP ── PMID 17353272        │
  │   • resolves/disassembles joint molecules   ── IGI/IMP ─ PMID 39580958,     │
  │                                                          31584934           │
  ├───────────────────────────────────────────────────────────────────────────┤
  │ STRUCTURAL / EVOLUTIONARY (conserved machinery)                            │
  │   • RQC winged-helix (IPR018982) + HRDC (IPR002121) shared w/ BLM/WRN/Sgs1  │
  ├───────────────────────────────────────────────────────────────────────────┤
  │ ORTHOLOGUE DIRECT ASSAYS (the basis of the IBA)                            │
  │   • BLM: HJ branch migration                 ── PMID 16670433               │
  │   • BLM-TopoIIIα-BLAP75: dHJ dissolution     ── PMID 18390547, 24392087     │
  │   • Sgs1/BLM RecQ HJ branch-migration enzyme ── PMID 24984776               │
  └───────────────────────────────────────────────────────────────────────────┘
   │
   ▼
  IBA propagation (GO_Central / PAINT, GO_REF:0000033, ECO:0000318)
  with/from: BLM P54132, WRN Q14191, Arabidopsis RecQ loci
   │
   ▼
  Rqh1 (Q09811) GO:0009378  ── INFERRED, not measured

The coherent narrative is that Rqh1 is the single RecQ helicase of S. pombe, orthologous to Sgs1/BLM/WRN, and functions in a complex with Top3 (topoisomerase III) to process aberrant recombination and replication intermediates during S phase and G2. Its orthologues, when purified, catalyze HJ branch migration and, together with Top3α + BLAP75/RMI, dissolve double Holliday junctions to yield non-crossover products. Rqh1's in-vivo phenotypes (crossover suppression that requires its helicase activity; disassembly of joint molecules; maintenance of circular chromosomes by inhibiting crossovers) are exactly what one would predict if Rqh1 acted on four-way junctions in cells. Conservation of the RQC winged-helix and HRDC modules provides the structural substrate for this activity. What is missing is the one experiment that would convert inference into fact: an in-vitro assay showing purified Rqh1 unwinding or migrating a defined four-way-junction substrate.

GO decision table

Evidence Matrix

Conflicts and Alternatives

• Orthologue carry-over. The strongest alternative interpretation is that GO:0009378 is a database carry-over from BLM/WRN/Sgs1 rather than a fission-yeast fact. QuickGO provenance confirms this directly (IBA, with/from human BLM P54132 and WRN Q14191). Rqh1 is the sole RecQ in S. pombe, so this is not intra-organism paralog confusion, but it is cross-species propagation.
• Helicase-independent functions. Ahmad 2002 (PMID: 12478586) explicitly reports that Rqh1p's helicase activity is "only partially required for its function," and Top3-related genetics (e.g., PMID: 10572171, PMID: 15340008) show that Rqh1/Top3 lethality relationships involve helicase-dependent and independent components. This cautions against over-weighting any single helicase sub-activity as the core function.
• BP vs MF conflation. The in-vivo phenotypes (crossover suppression, joint-molecule disassembly) are readily explained by junction dissolution together with Top3 (a topoisomerase-dependent dissolution) rather than by autonomous four-way-junction unwinding. In the orthologue system, robust dHJ processing requires the BLM–TopoIIIα–BLAP75 complex, not BLM alone — so even the orthologue "activity" is partly complex-dependent. The four-way-junction MF term for Rqh1 alone may therefore be biologically narrower than what the enzyme actually does in its physiological Top3 complex.

Limitations and Knowledge Gaps

1. No direct four-way-junction assay for Rqh1. Checked: PubMed for Rqh1 + Holliday junction / branch migration / branched-substrate biochemistry; PomBase/QuickGO evidence codes. Why it matters: this is the single experiment separating inference from fact. Resolution: purify recombinant Rqh1 (± Top3) and assay on a defined synthetic four-way (X-junction) substrate for unwinding and branch migration.
2. Rqh1-alone vs Rqh1–Top3 complex activity. Checked: complex composition (PMID 12724426) and orthologue dHJ requirements (PMID 24392087, 18390547). Why it matters: the physiological unit is the Rqh1–Top3 dissolvasome; the MF term describes Rqh1 alone. Resolution: compare junction processing by Rqh1 alone vs Rqh1–Top3 (± Rmi1/BLAP75 orthologue) in vitro.
3. Whether the IBA should be annotated as ISS with a specific "with" gene. Checked: QuickGO provenance. Why it matters: curators may prefer an explicit ISS to BLM/Sgs1 over a PAINT IBA. Resolution: curator judgment plus a documented sequence/domain alignment (already supported by Finding 4).
4. Direct correspondence between in-vivo crossover suppression and four-way-junction unwinding. Checked: Hope 2007 genetics. Why it matters: crossover suppression could operate via D-loop dissolution/SDSA rather than HJ migration. Resolution: separation-of-function alleles and substrate-specific in-vitro assays.

Discriminating Tests

1. In-vitro four-way-junction unwinding/branch-migration assay on purified recombinant Rqh1 using a defined synthetic X-junction (fluorescently or radiolabeled), ± ATP, ± Mg²⁺. A positive result would directly justify GO:0009378 at IDA; a negative would confirm the term is inference-only for Rqh1.
2. Rqh1 vs Rqh1–Top3(±Rmi1) comparison on double-Holliday-junction and mobile-D-loop substrates, mirroring the BLM–TopoIIIα–BLAP75 experiments (PMID 24392087, 18390547), to determine whether junction processing is an autonomous Rqh1 activity or a complex property.
3. Separation-of-function helicase-dead and junction-binding mutants (e.g., Walker-A K→I, analogous to rad12-K547I, PMID 10572171) assayed both in vitro (substrate specificity) and in vivo (crossover frequency), to map the crossover-suppression phenotype onto a specific molecular activity.
4. Structure-guided comparison of the Rqh1 RQC winged-helix against the BLM branch-migration model (PMID 33647232) via AlphaFold/homology modeling to predict junction-binding residues, then test them experimentally.

Proposed Follow-up Actions (Curation Leads — require curator verification)

• Retain GO:0009378 "four-way junction helicase activity" as an inferred annotation (IBA, ECO:0000318, or ISS to BLM/Sgs1). Do not upgrade to an experimental evidence code (IDA/EXP) — no direct Rqh1 assay exists.
• Retain GO:0043138 "3'–5' DNA helicase activity" (IDA) as the primary, directly evidenced molecular function, backed by two independent references (PMID 12724426, 12478586).
• Candidate references to verify:
• PMID 12478586 — snippet: "Here we show that, like other RecQ family members, the Rqh1p protein displays 3' to 5' DNA helicase activity." (confirms MF is 3'→5', not four-way-junction)
• PMID 12478586 — snippet: "unlike other RecQ family members, the helicase activity of Rqh1p is only partially required for its function." (caveat on core-function weighting)
• PMID 17353272 — snippet: "This suppression of crossover formation by Rqh1 is dependent on its helicase activity." (BP-level support)
• PMID 16670433 — snippet: "BLM also promotes branch migration of Holliday junctions." (orthologue basis of IBA)
• Suggested question for curators: should the four-way-junction MF be re-expressed as an ISS to a named orthologue (BLM P54132 / Sgs1 P35187) rather than a PAINT IBA, given the strong domain-architecture conservation documented here?
• Suggested experiment: direct in-vitro four-way-junction assay on purified Rqh1 (± Top3), which would resolve the inference-vs-fact gap in a single experiment.

Evidence Base (selected literature)

• PMID: 12724426 — Laursen et al. 2003. Direct assay: Rqh1 is a 3'→5' helicase in a Top3 complex (basis of GO:0043138 IDA). No HJ substrate tested.
• PMID: 12478586 — Ahmad, Kaplan & Stewart 2002. Second independent 3'→5' helicase IDA; notes helicase activity only partially required.
• PMID: 17353272 — Hope et al. 2007. Helicase-dependent crossover suppression in vivo (BP support).
• PMID: 39580958 / PMID: 31584934 — Rqh1 resolves/disassembles recombination joint molecules (BP support).
• PMID: 16670433 — BLM promotes HJ branch migration (orthologue direct evidence).
• PMID: 24392087, PMID: 18390547, PMID: 24984776 — BLM/Sgs1 + Top3(+BLAP75/Rmi1) dHJ dissolution and HJ processing (orthologue basis of the conserved-function inference).
• PMID: 33647232 — BLM structures providing a model for HJ branch migration (structural basis).
• PMID: 32085395, PMID: 23161011 — Reviews of yeast/human RecQ helicase function (orientation only).

Report prepared for AI Gene Review hypothesis-level curation. All molecular-function claims are separated into direct Rqh1 assays vs orthologue-derived inference. Computed provenance: QuickGO evidence-code lookup for Q09811 and InterPro/UniProt-REST domain-architecture comparison across Rqh1/BLM/WRN/RECQ1/Sgs1.

Artifacts

• OpenScientist final report
• OpenScientist final report