EME1 (OsEME1) — Oryza sativa subsp. japonica — research notes

UniProt: Q0J9J6 | Gene: EME1 / OsEME1 | Locus: Os04g0648700, LOC_Os04g55500
"Crossover junction endonuclease EME1" / "Essential meiotic endonuclease 1"; EC 3.1.22.-
Family: EME1/MMS4 family; PANTHER PTHR21077:SF5 (CROSSOVER JUNCTION ENDONUCLEASE MMS4)

1. What EME1/MUS81 is — conserved biology

EME1 (Eme1 in fission yeast; Mms4 in budding yeast; EME1/EME2 in humans) is the
non-catalytic regulatory subunit of a heterodimeric, structure-specific DNA
endonuclease whose catalytic subunit is MUS81. EME1 itself lacks endonuclease
activity in yeast and human; the interaction with MUS81 stimulates and regulates
MUS81's endonucleolytic activity.

PMID:40333587
PMID:40333587

MUS81-EME1 is one of three structure-selective endonuclease systems
(MUS81-EME1/MMS4, SLX1-SLX4, GEN1/Yen1) that resolve branched DNA
intermediates ("Holliday junction resolvases" in the broad sense). MUS81-EME1
has substrate preference for nicked Holliday junctions, 3'-flaps, D-loops and
replication-fork structures
— i.e. branched substrates with a 5'-end at the
branch — and resolves intact HJs only with reduced efficiency.

PMID:24008669
PMID:24008669

UniProt FUNCTION (Q0J9J6, by similarity): "Interacts with MUS81 to form a DNA
structure-specific endonuclease with substrate preference for branched DNA
structures with a 5'-end at the branch nick. Typical substrates include 3'-flap
structures, D-loops, replication forks, nicked Holliday junctions and also
intact Holliday junctions with a reduced efficiency. May be required in mitosis
for the processing of stalled or collapsed replication fork intermediates.
Plays a role in DNA repair and in genotoxic stress-induced homologous
recombination (HR) in somatic cells. Mediates a subset of meiotic recombination
events that are insensitive to crossover interference."

This dual mitotic/meiotic role is key: MUS81-EME1 functions in BOTH (a) somatic
(mitotic) replication-fork rescue and DSB repair AND (b) meiotic crossover
formation (the interference-insensitive / "class II" CO pathway). It is NOT a
meiosis-restricted enzyme.

DNA-damage-responsive regulation

MUS81-EME1 activity is cell-cycle and DNA-damage regulated. In fission yeast,
DNA-damage-induced activation of Mus81-Eme1 requires Cdc2(CDK1)- and
Rad3(ATR)-dependent phosphorylation of Eme1; the complex is activated at G2/M.
PMID:23584455
PMID:40333587

In meiosis, MUS81-EME1 (with Yen1/GEN1 and SLX1-SLX4) must be kept inactive
during prophase I to allow proper CO patterning; precocious activation disrupts
crossover distribution.
PMID:29920281

2. OsEME1 — the primary rice paper (Du et al. 2025, Plant Biotechnol J)

PMID:40333587 / DOI:10.1111/pbi.70101 — "ESSENTIAL MEIOTIC ENDONUCLEASE 1 is
required for chloroplast development and DNA repair in rice." This is the
definitive functional characterization of OsEME1 (Q0J9J6).

Key findings:

3. Rice OsMUS81 — meiosis (Mu et al. 2022, New Phytol)

PMID:36495065 / DOI:10.1111/nph.18668 — "MUS81 is required for atypical
recombination intermediate resolution but not crossover designation in rice."

So the rice MUS81-EME1 complex has a meiotic role — but it is a minor/atypical-
intermediate-resolution role, NOT a meiosis-restricted "essential meiotic
endonuclease" role. The "Essential meiotic endonuclease" name is historical
(from S. pombe Eme1) and is somewhat misleading for the plant ortholog.

4. Other plant context

5. Assessment of the IEP annotations from PMID:12869764

PMID:12869764 (Kikuchi et al. 2003, Science) is "Collection, mapping, and
annotation of over 28,000 cDNA clones from japonica rice." It is a large-scale
full-length cDNA sequencing/annotation consortium paper. Its abstract describes
ONLY cDNA collection, sequencing, mapping to genomic DNA, and InterPro-based
function assignment. There is NO gene-specific expression-profiling experiment
for EME1 and NO experiment testing response to UV, gamma radiation or high
light.

The three IEP annotations (response to UV GO:0009411, response to high light
intensity GO:0009644, response to gamma radiation GO:0010332) all cite
PMID:12869764 as the IEP (Inferred from Expression Pattern) reference. This is
almost certainly a curation artefact: the rice FL-cDNA consortium built cDNA
libraries from many tissues and stress conditions, and a clone for this gene may
have come from a stress-treated library (UV-/gamma-/high-light-treated tissue).
The presence of a cDNA clone in a stress library is library-of-origin
information, NOT an expression-response measurement, and certainly not evidence
of biological "involvement in" a stress response. IEP from this kind of source
is not appropriate functional evidence. Expression ≠ function.

Note: high-light sensitivity of OsEME1 IS genuinely demonstrated — but in
PMID:40333587 (oseme1 mutants hypersensitive to high light), as a downstream
consequence of a genome-maintenance defect, not a primary "response to high
light" function. The IEP annotations remain unsupported by their cited
reference.

6. Summary of SPKW-retired annotation assessment

The six retired SPKW (GO_REF:0000043) annotations were keyword-to-GO mappings:
- meiotic cell cycle (GO:0051321) <- KW "Meiosis" — OVER-SPECIFIC. EME1/MUS81 is
not meiosis-restricted; its dominant role in rice is somatic DSB/replication
repair. Removal justified.
- cell division (GO:0051301) <- KW "Cell division" — weakly supported, generic;
EME1 acts in cell-cycle-coupled repair but "cell division" per se is not its
function. Removal justified (over-annotation).
- DNA recombination (GO:0006310) <- KW "DNA recombination" — correct in essence
(EME1 resolves HR intermediates) but generic; better captured by HR-repair /
resolution terms. Removal acceptable; correct biology retained by current
IBA/IEA terms.
- endonuclease activity (GO:0004519) <- KW "Endonuclease/Nuclease" — GENUINELY
CORRECT. EME1 IS an endonuclease (EC 3.1.22.-), demonstrated biochemically in
rice (PMID:40333587). Current GOA has NO catalytic MF term — only "DNA
binding". Removing endonuclease activity LOST a correct, core annotation. This
removal was NOT justified; a NEW catalytic MF term should be added.
- hydrolase activity (GO:0016787) <- KW "Hydrolase" — correct but vague parent
of endonuclease activity. Removal of the vague term is acceptable IF a
specific endonuclease MF is present (it is not — see above).
- metal ion binding (GO:0046872) <- KW "Metal-binding" — correct (Mg2+/Ca2+
cofactor) but the keyword mapping is generic. MUS81-EME1 nuclease chemistry is
Mg2+-dependent. Removal of the generic term is acceptable; magnesium ion
binding (GO:0000287) would be the accurate specific term if retained.