MSH4

MSH4 is a meiosis-specific MutS homolog that acts as a pro-crossover factor by stabilizing recombination intermediates and channeling them toward class I crossover formation with crossover interference (PMID:11454751, PMID:20865162). It localizes to discrete foci along meiotic chromosomes in a manner dependent on the synapsis initiation proteins Zip1 and Zip2, and msh4 loss delays synaptonemal complex assembly, lowers crossover frequency, and abolishes most crossover interference (PMID:11454751); as part of the Msh4-Msh5 (MutSγ) complex it associates in vivo with DSB hotspots, chromosome axes and centromeres, with hotspot and axis binding requiring DSB formation and resection, enhanced by double Holliday junction structures, and dependent on the axis protein Red1 (PMID:34849874). MSH4 functions through a series of physical partners: it binds the MutL components MLH1 and MLH3 via its amino-terminal region, with MSH4 foci giving way to MLH1 foci as pachynema progresses, indicating sequential synapsis then resolution roles (PMID:10928988, PMID:12095912), and it contacts the strand-exchange recombinases RAD51 and DMC1 at early recombination intermediates (PMID:15489243). EXO1 directly binds MSH4 within MutSγ, and this contact is required to activate MutLγ (MLH1-MLH3) endonuclease nicking that resolves joint molecules into crossovers, establishing EXO1 as an integral structural component of the meiotic resolvase machinery (PMID:40319035). Availability of the MSH4-MSH5 heterodimer is modulated by VBP1, which competes with MSH5 for MSH4 binding (PMID:12591739). In humans, a homozygous MSH4 splice-site mutation that deletes the Walker B motif of the ATP-binding domain causes primary ovarian insufficiency, demonstrating that intact MSH4 ATP-binding function is required for human meiosis (PMID:28541421).

1. 2000 High

   Establishing that MSH4 physically links to the MutL machinery answered how a MutS homolog could connect synapsis to crossover resolution, defining an amino-terminal MLH1-binding interface distinct from the ATP/HTH region.

   Evidence Co-immunoprecipitation, domain mapping, and immunolocalization of human MSH4 and hMLH1 on meiotic chromosomes

   PMID:10928988

   Open questions at the time

   • Whether MSH4 acts alone or only as a heterodimer with MSH5 in MLH1 binding not resolved here
   • Functional consequence of the temporal handoff from MSH4 to MLH1 foci not tested
2. 2000 Medium

   Genetic epistasis distinguished MSH4 and EXO1 contributions to meiosis, showing they promote spore viability through separate mechanisms despite both affecting crossovers.

   Evidence Yeast msh4 exo1 double-mutant analysis with crossover frequency and spore viability readouts

   PMID:10855499

   Open questions at the time

   • The basis for synergistic viability loss not mechanistically defined
   • Did not anticipate the later-defined direct EXO1-MSH4 physical/functional partnership
3. 2001 High

   Cytological and epistasis analysis placed Msh4 in the synapsis/crossover pathway, showing its chromosomal localization depends on Zip1/Zip2 and that it governs crossover interference.

   Evidence Immunolocalization with Zip2, genetic epistasis with zip1, and msh4 null phenotyping in yeast

   PMID:11454751

   Open questions at the time

   • Molecular nature of the DNA/intermediate Msh4 stabilizes not defined
   • Mechanism by which Zip1/Zip2 recruit Msh4 unknown
4. 2002 Medium

   Demonstrating MSH4-MLH3 interaction extended the MutS-MutL link to a second MutL subunit, supporting an MSH4-MLH3 module in mammalian recombination.

   Evidence Co-IP from mouse spermatocyte extracts and in vitro interaction with human MLH3 isoforms

   PMID:12095912

   Open questions at the time

   • Single lab, no reciprocal in vivo validation
   • Functional contribution of MLH3 isoform differences not tested
5. 2003 Medium

   Identifying VBP1 as an MSH4 partner that competes with MSH5 introduced a possible regulatory mechanism for controlling MutSγ heterodimer availability.

   Evidence Co-IP, colocalization, and yeast three-hybrid competition with a truncated hMSH4 splice variant

   PMID:12591739

   Open questions at the time

   • Physiological relevance of VBP1 competition in meiotic cells not shown
   • Whether VBP1 regulates MSH4 in vivo untested
6. 2004 Medium

   Linking MSH4 to RAD51 and DMC1 positioned the protein at early post-strand-exchange recombination intermediates rather than only at later resolution steps.

   Evidence Biochemical interaction assays and colocalization on mouse meiotic chromosomes

   PMID:15489243

   Open questions at the time

   • Direct versus bridged interaction not distinguished
   • Functional consequence of recombinase contact not tested
7. 2010 High

   Systematic mutagenesis separated MSH4-MSH5 functions, showing the complex promotes both crossover formation and SC assembly and that a crossover threshold is required for viability.

   Evidence Site-directed mutagenesis of 57 conserved residues in yeast with crossover, interference, viability, and SC readouts

   PMID:20865162

   Open questions at the time

   • Biochemical activities of threshold alleles not directly measured
   • Differential sensitivity of Msh5 versus Msh4 not structurally explained
8. 2013 Low

   Homology modeling provided a structural rationale for MSH4-MSH5 heterodimer formation, DNA binding, and the phenotypic classes of point mutants.

   Evidence Computational homology modeling on bacterial MutS templates integrated with prior mutant phenotypes

   PMID:24244354

   Open questions at the time

   • Purely computational with no experimental structural validation in this study
   • Predicted interface residues not directly tested
9. 2017 Medium

   A human disease mutation established that MSH4 ATP-binding function is essential for female meiosis, connecting the gene to primary ovarian insufficiency.

   Evidence Whole-exome sequencing, segregation, and exon trapping confirming exon 17 skipping deleting the Walker B motif

   PMID:28541421

   Open questions at the time

   • ATPase inactivation inferred computationally, not measured
   • Male phenotype of this allele not characterized
10. 2021 High

    Genome-wide ChIP defined where MutSγ binds in vivo and the requirements for that binding, showing dependence on DSB formation, resection, and the axis protein Red1.

    Evidence Msh5 ChIP-seq across spo11Δ, zip3Δ, red1Δ and other meiotic mutants in yeast

    PMID:34849874

    Open questions at the time

    • Msh4-specific binding inferred from Msh5 ChIP
    • How dHJ structures enhance binding mechanistically not resolved
11. 2025 High

    Reconstitution defined EXO1 as a direct MSH4 partner and integral resolvase component, required to activate MutLγ endonuclease independent of EXO1 catalytic activity.

    Evidence In vitro interaction assays, EXO1 point mutagenesis (W371E), and DNA nicking assays with MutSγ-MutLγ

    PMID:40319035

    Open questions at the time

    • Structure of the assembled MutSγ-MutLγ-EXO1 resolvase not determined
    • In vivo requirement of the W371E contact in mammals not tested
12. 2025 Low

    Upstream and pathway-context studies placed MSH4-MSH5 recruitment under HEIP1/HEI10 control and assigned Msh4 anti-MMR activity in the noncrossover/crossover decision.

    Evidence Mouse Heip1 knockout with immunolocalization and Co-IP (preprint); yeast hybrid-zygote genetic epistasis with MMR and HR mutants (preprint)

    PMID:bio_10.1101_2025.02.08.636946 PMID:bio_10.1101_2025.08.25.672081

    Open questions at the time

    • Both findings are unreviewed preprints from single labs
    • Direct biochemical demonstration of Msh4 anti-MMR activity not provided

• How ATP binding/hydrolysis by the MSH4-MSH5 sliding clamp couples intermediate stabilization to crossover designation and interference at the molecular level remains unresolved.
• No experimental structure of the MutSγ-DNA-resolvase assembly
• Mechanistic basis of crossover interference imposed by MSH4 not defined
• Direct measurement of MSH4 ATPase coupling to function lacking