20091025

Fernandez et al. 2004 MBC-(Lsm2-Lsm7 complex)

In yeast, Lsm2-Lsm8 complex binds and stabilizes the 3′ end of the spliceosomal U6 snRNA,  the Lsm2-8 proteins likely form a heteroheptameric ring;

whereas the Lsm1-Lsm7 complex functions in mRNA decay, Capped mRNA degradation intermediates accumulate in yeast containing mutations in these proteins, indicating that Lsm1-7 complex functions in mRNA decapping. As 3′-shortened mRNAs accumulate in mutant strains, the Lsm1-7 complex may protects 3′ ends of deadelylated mRNAs from nculeases. Immunofluorescence experiments reveal that Lsm1 is mostly cytoplasmic, whereas Lsm7, a component of both complexes, is nuclear and cytoplasmic. Within the cytoplasm, Lsm1 localizes to structures called p-bodies, whch represent sites of mRNA degradation.  Thus, the Lsm2-8 complex may be nuclear, consistent with its role as a U6 snRNP component, whereas the Lsm1-7 complex appears largely cytoplasmic.

Lsm2-Lsm7 associates with snR5, a box H/ACA snoRNA that functions to guide site-specific pseudouridylation of rRNA. snR5, this RNA is a member of the box H/ACA class of snoRNAs that fucntion in pseudouridylation of rRNA. Thre are >20 box H/ACA snoRNAs in yeast, all of which are bound by four core proteins: Gar1, Nhp2, Nop10, and the pseudouridine synthase Cbf5. Approximately half the snR5 RNA in cells i sbound by Lsm proteins. The Lsm complex bound to snR5 is distinct from Lsm2-8 and Lsm1-7 complexes, because neither Lsm1 nor Lsm8 are associated with the snoRNA. Experiments in which binding  of Lsm proteins to snR5 was reconstituted in vitro reveal that the 3′ end of snR5 is required for Lsm protein recognition. We demonstrate that components of the Lsm2-7 complex are prsent in nucleoli. Interestingly, biochemical fractionation and immunoprecipitation experiments suggest that at least some of the Lsm2-7/snR5 complex is distinct from the fraction of the snR5 bound by Gar1 and Nhp2. Consistent with a separate complex, Lsm proteins are not required for the function of snR5 in pseudouridylation.

Spiller et al. 2007 Jounal of cell science-(Lsm2-8 complex)

Nuclear accumulation of Lsm8 requires Kap95.  Production of recombinant human LSM proteins in bacteria, followed by injection of thse prteins into HeLa cells, showed that the pre-assembled LSM2-8 complex localized to the nculeus, whereas LSM injected by itself accumulated in the cytoplasm. These results suggest that LSM8 nuclear  import involves an unidentified nuclear-import signal that is only present when LSM8 interacts with other LSM2-8 subunits.

The three paralogues yeast Lsm proteins, Lsm8, Lsm2, Lsm4 also contain basic C-termini, they might form a nculear-localization signal in a similar fashion. Mutations in yeast Lsm8 are suppressed by overexpression of LSM2 or LSM4.

Under normal physiological conditions, competition between Lsm1 and Lsm8 might provide a link between RNA processing events in the nucleus and mRNA degradation in the cytoplasm.

Lsm proteins are actively imported through the nuclear pore

nup49-313 (a nuclear-pore mutant taht affects protein import), strain ts

In contrast  to the effect of the nup49-313 mutation on Lsm protein localization, the xpo1-1 nuclear-export mutation showed no effect on the localization of Lsm1, Lsm7 or Lsm8. Thus, it seems that the cytoplasmic localization of Lsm1 is not a consequence of nuclear exclusion by continual active export from the nculeus, at least not through this export pathway.

Nuclear localization of Lsm7 requires other Lsm proteins (not Lsm1)

lsm1Δ and lsm6Δ heat sensitive

Depletion of Lsm2 or Lsm4 disrupts Lsm8 localization

Lsm8 by itself does not accumulate in the nucleus, because loss of either of its proposed partners in the lsm2-8 ring results in its delocalization. However, it seems that a complete Lsm2-8 complex is not essential, because lack of Lsm6 has no effect, which might be expected because Lsm6 is a non-essential protein.

Lsm7 and Lsm8 delocalization is not caused by defective splicing

depletion of Lsm2-8 proteins leads to decreased levels of U6 snRNA and an accumulation of per-mRNA.

Lsm8 truncations affect its nuclear localization, Lsm8 needs to interact with other Lsm proteins for its nculear accumulation

Over-production of Lsm1 or Lsm8 has opposing effects on Lsm7 localization

lsm8-1 mutation is synthetic lethal with deletion of LHP1 (which encodes the yeast homolog of La, another U6 RNA-binding protein), and that the requirement for Lhp1 in an lsm8-1 strain can be suppressed by low-copy overexpressiono f LSM2.

The proposed existence of an Lsm2-7 complex that associates with snR5 in thenucleolus is in apparecontradiction with our finding of Lsm8 requirement for nuclear localization. It seems possible that the full Lsm2-8 complex might interact with snR5, but that the Lsm8 epitope tag might be masked in the snR5 RNP.

Pannone et al. 2001 Genetics-(lsm2-8 complex)

A model of Lsm2 and Lsm4 contact Lsm8 in the Lsm2-8 ring.

Deletion of LSM5, LSM6, or LSM7, but not LSM1, are synthetically lethal with deletions of LHP1. Lhp1 acts redundantly with the assembled Lsm2-8 complex to stabilize newly synthesized U6 RNA.