Architecture of the U6 snRNP reveals specific recognition of 3′-end processed U6 snRNA

Abstract

The spliceosome removes introns from precursor messenger RNA (pre-mRNA) to produce mature mRNA. Prior to catalysis, spliceosomes are assembled de novo onto pre-mRNA substrates. During this assembly process, U6 small nuclear RNA (snRNA) undergoes extensive structural remodeling. The early stages of this remodeling process are chaperoned by U6 snRNP proteins Prp24 and the Lsm2–8 heteroheptameric ring. We now report a structure of the U6 snRNP from Saccharomyces cerevisiae. The structure reveals protein–protein contacts that position Lsm2–8 in close proximity to the chaperone “active site” of Prp24. The structure also shows how the Lsm2–8 ring specifically recognizes U6 snRNA that has been post-transcriptionally modified at its 3′ end, thereby elucidating the mechanism by which U6 snRNPs selectively recruit 3′ end-processed U6 snRNA into spliceosomes. Additionally, the structure reveals unanticipated homology between the C-terminal regions of Lsm8 and the cytoplasmic Lsm1 protein involved in mRNA decay.

Molecular basis for the distinct cellular functions of the Lsm1-7 and Lsm2-8 complexes

“…Molecular replacement was used to obtain initial phases with initial search templates PDB 4EMG (S. pombe Lsm3) (48), PDB 4EMH (S. pombe Lsm4) (48), PDB 4EMK (S. pombe Lsm5/6/7) (48), and homology models (49,50) of S. pombe Lsm2 and Lsm8 that were constructed from the corresponding S. cerevisiae orthologs (PDB 4C92 and 4M7D, respectively) (11,26). Molecular replacement used a single search model in which the above seven proteins were placed in a fixed orientation relative to one another to resemble the known architecture of the S. cerevisiae Lsm rings (11,25,26). Structure refinement was performed in Phenix.refine (46,51) using secondary structure restraints and TLS parameterization, with iterative rounds of manual model building in Coot (52,53) and additional automated refinement in Phenix.refine.…”

“…Lsm proteins were co-expressed in E. coli and orthogonal affinity tags were used to purify the Lsm1-7 and Lsm2-8 complexes ( Supplementary Figure 1). For Lsm2-8, we demonstrated that the complex can bind to U6 snRNA and further associate with protein Prp24 to form the complete U6 snRNP (25). We wished to determine if the Lsm2-8 ring specifically recognizes the 2′,3′ cyclic phosphate group at the end of U6 snRNA, and if so, how.…”

“…The Lsm2-8 complex shares 6 out of 7 subunits with Lsm1-7, localizes in the nucleus, and binds the 3' ends of U6 and U6 atac snRNAs (25)(26)(27)(28). U6 snRNA is transcribed by RNA polymerase III, which terminates transcription after synthesis of an oligoU tail at the end of U6 snRNA (29).…”

“…cerevisiae Lsm2-8 structure has been determined (25,26) but is unlikely to be representative of most eukaryotes due to significant differences in Lsm8 protein sequence and U6 snRNA posttranscriptional modification (35). Budding yeasts have Usb1 enzymes that open the 2′,3′ cyclic phosphate to produce a 3′ phosphate, and their Lsm8 proteins have evolved a unique Cterminal extension in order to interact with this 3′ phosphate through electrostatic interactions (25,35). In contrast, the Lsm8 proteins of most eukaryotes, from S. pombe to humans, do not have this Lsm8 extension and have U6 snRNAs with terminal 2′,3′ cyclic phosphates.…”

Termination of pre-mRNA splicing requires that the ATPase and RNA unwindase Prp43 acts on the catalytic snRNA U6

“…In vivo, the 3' end of U6 is essential for growth in yeast and plays a critical role in stabilizing U6 snRNA through the recruitment of the La homolog protein Lhp1p and subsequently the Lsm2-8 complex. Unlike Lhp1p, the Lsm complex tolerates the terminal, cyclic 2', 3' phosphate of U6 in metazoans and the terminal 3' monophosphate in yeast, both of which result from 3'-end maturation during U6 biogenesis; whereas Lhp1 binds the UUUOH terminus of all RNA polymerase III transcripts (Wolin and Cedervall 2002), the Lsm complex binds primarily to the GUUUU terminal sequence of U6 (Pannone et al 2001;Achsel et al 1999;Bordonne and Guthrie 1992;Wolff and Bindereif 1995;Zhou et al 2014;Licht et al 2008;Montemayor et al 2018). The 3' end of U6 has additionally been shown in vitro to play an early role in splicing by promoting the annealing of U6 to U4, again through recruitment of the Lsm complex and also through the subsequent recruitment of Prp24p (Ryan et al 2002;Vidal et al 1999;Rader and Guthrie 2002;Licht et al 2008).…”