LSM4

LSM4 is a core subunit of the heptameric Lsm1–7 ring that functions in cytoplasmic mRNA decay, P-body assembly, and nuclear pre-mRNA splicing. Its N-terminal Sm fold mediates incorporation into the Lsm ring and is essential for viability, as demonstrated by early embryonic lethality in homozygous knockout mice and yeast complementation studies (PMID:10629062, PMID:11561292, PMID:22615807). The C-terminal low-complexity/RGG domain drives liquid–liquid phase separation and P-body formation through PRMT5-dependent symmetric dimethylation of arginines, and independently contacts SLBP and 3′hExo on the histone mRNP to promote replication-dependent histone mRNA degradation (PMID:27247266, PMID:24255165, PMID:36976747). Symmetric dimethylarginine modifications on LSm4 also mediate binding to the SMN Tudor domain, linking LSm4 to snRNP biogenesis, while an interaction with ICln at the plasma membrane implicates LSm4 in swelling-activated anion channel regulation (PMID:11720283, PMID:19088440).

1. 2000 High

   The question of whether LSm4 is individually essential or redundant with other Lsm proteins was resolved by showing that homozygous Lsm4-knockout mice die shortly after implantation, establishing a non-redundant requirement for LSm4 in early development.

   Evidence Promoter-trap gene targeting and knockout mouse analysis

   PMID:10629062

   Open questions at the time

   • Specific RNA targets or splicing events disrupted in Lsm4-null embryos were not identified
   • Whether lethality reflects loss of splicing, mRNA decay, or both was not resolved
2. 2001 High

   The discovery that LSm4 carries symmetric dimethylarginine (sDMA) modifications and that these are required for Tudor-domain binding to SMN established a post-translational regulatory axis linking LSm4 to snRNP biogenesis.

   Evidence Mass spectrometry, protein sequencing, in vitro methylation with HeLa extracts, SAH inhibition

   PMID:11720283

   Open questions at the time

   • Whether sDMA modification is constitutive or regulated was not determined
   • Functional consequence of disrupting LSm4–SMN interaction on snRNP assembly was not directly tested
3. 2001 Medium

   Yeast complementation with C-terminal truncations showed that the Sm-fold domain alone supports viability but the C-terminal domain is essential for stationary-phase survival, delineating two functionally separable domains.

   Evidence Complementation assays with domain truncation mutants in K. lactis and S. cerevisiae

   PMID:11561292

   Open questions at the time

   • Molecular targets of the C-terminal domain in stationary phase were not identified
   • Whether the C-terminal domain function is conserved in mammals was not addressed
4. 2005 Medium

   Identification of ICln as a physical partner of LSm4 suggested a link between RNA metabolism and cell volume regulation, though the functional significance for LSm4 was initially unclear.

   Evidence Co-immunoprecipitation and NMR structural analysis of ICln PH domain

   PMID:15905169

   Open questions at the time

   • Whether the ICln–LSm4 interaction is direct or bridged was not resolved with purified proteins
   • Functional consequence for RNA processing was not tested
5. 2008 Medium

   FRET and electrophysiology experiments showed LSm4 resides at the plasma membrane under isotonic conditions and dissociates upon swelling, directly modulating IClswell channel kinetics, thereby establishing a non-canonical function for LSm4 in cell volume regulation.

   Evidence FRET, subcellular fractionation, electrophysiology in NIH3T3 and HEK293 cells

   PMID:19088440

   Open questions at the time

   • Mechanism by which LSm4 inhibits IClswell is unknown
   • Whether this plasma membrane pool is distinct from the mRNA decay pool was not addressed
6. 2012 High

   Crystal structures of fission yeast Lsm4 confirmed the conserved Sm fold and revealed a monomer–trimer equilibrium, while RNA binding assays showed isolated Lsm4 does not bind RNA, indicating it functions only in the assembled ring context.

   Evidence X-ray crystallography, analytical ultracentrifugation, RNA binding assays with S. pombe Lsm4

   PMID:22615807

   Open questions at the time

   • No structure of the full heptameric ring including Lsm4 was obtained
   • C-terminal low-complexity domain was absent from constructs
7. 2013 High

   The C-terminal tail of LSm4 was shown to directly contact SLBP and 3′hExo on histone mRNPs, and mutations disrupting these interactions reduced histone mRNA degradation, defining a substrate-specific role for LSm4 beyond general mRNA decay.

   Evidence Co-immunoprecipitation/pulldown, mutagenesis, histone mRNA decay assays upon DNA synthesis inhibition

   PMID:24255165

   Open questions at the time

   • Whether oligouridylation is required upstream of LSm4 engagement was not addressed
   • Structural basis of the C-tail–SLBP/3′hExo interaction is unknown
8. 2016 High

   Definitive evidence that the LSm4 RGG domain is dispensable for Lsm1–7 assembly, decapping, and mRNA decay but is specifically required for P-body formation established that P-body condensation is mechanistically separable from mRNA decay; PRMT5-mediated sDMA of the RGG domain drives this condensation.

   Evidence RGG deletion rescue, PRMT5 knockdown, fluorescence microscopy, mRNA decay and translation assays in human cells

   PMID:27247266

   Open questions at the time

   • Whether sDMA-dependent P-body formation feeds back to regulate specific mRNAs is unknown
   • The role of the HAT1–RBBP7 interaction identified in this study remains uncharacterized
9. 2016 Medium

   Genetic epistasis in yeast showed that combined loss of Edc3 and the Lsm4 C-terminal Q/N-rich domain reduces mRNA stability and causes nuclear mislocalization of Dcp2, revealing redundant scaffolding roles of Lsm4 and Edc3 in organizing cytoplasmic mRNA decay.

   Evidence Double mutant analysis, mRNA stability assays, Dcp2 localization in S. cerevisiae

   PMID:27543059

   Open questions at the time

   • Whether the yeast Q/N-rich domain and the mammalian RGG domain are functionally analogous remains formally untested
   • Mechanism of Dcp2 nuclear accumulation is unknown
10. 2023 Medium

    In vitro reconstitution demonstrated that purified full-length LSm4 undergoes concentration-dependent liquid–liquid phase separation, directly attributing the condensation-driving activity to LSm4 itself rather than to co-factors.

    Evidence Purified recombinant mCherry-LSm4 LLPS assay, hexanediol and salt perturbation, fluorescence microscopy

    PMID:36976747

    Open questions at the time

    • The specific residues or modifications within the low-complexity domain driving LLPS were not mapped by mutagenesis
    • Behavior of LSm4 within the full Lsm1–7 complex in LLPS was not tested
11. 2025 Medium

    Under hypoxia, reduced tRF-31 binding to LSm4 decreases its ubiquitination and stabilizes the protein, causing nuclear translocation where LSm4 promotes EDN1 pre-mRNA splicing — revealing a tRNA-fragment-based regulatory mechanism governing LSm4 stability and nuclear splicing function.

    Evidence tRF-31 overexpression/knockdown, ubiquitination assays, subcellular fractionation, splicing and proliferation assays in pulmonary artery endothelial cells

    PMID:41354297

    Open questions at the time

    • The E3 ubiquitin ligase targeting LSm4 was not identified
    • Whether tRF-31-mediated regulation applies to other cell types or stress conditions is unknown
    • The specificity of EDN1 as an LSm4 splicing target needs broader transcriptomic validation

• Key open questions include the structural basis of the Lsm4 C-terminal domain interactions with SLBP/3′hExo, the identity of the E3 ligase controlling LSm4 turnover, whether P-body condensation driven by LSm4 feeds back to regulate specific mRNA fates, and the physiological relevance of the plasma membrane LSm4–ICln interaction.
• No structure of the C-terminal RGG/low-complexity domain in complex with any partner
• No genome-wide identification of LSm4-dependent splicing or decay targets
• Relationship between PRMT5-dependent sDMA and tRF-31/ubiquitination regulatory axes is unexplored