The genetics of nuclear pre-mRNA splicing: a complex story

Optimization of SMN trans-splicing through the analysis of SMN introns

Spinal muscular atrophy (SMA), a neurodegenerative disease, is the leading genetic cause of infantile death and is caused by the loss of survival motor neuron 1 (SMN1). Humans carry a duplicated copy gene, SMN2, which produces very low levels of functional protein due to an alternative splicing event. This splicing difference is the reason that SMN2 cannot prevent SMA development when SMN1 is deleted. SMN2 generates a transcript lacking exon 7 and consequently gives rise to an unstable truncated SMN protein that cannot protect from SMA. To increase full-length SMN protein, we utilize a strategy referred to as trans-splicing. This strategy relies upon pre-mRNA splicing occurring between two separate molecules: (1) the endogenous target RNA and (2) the therapeutic RNA that provides the correct RNA sequence via a trans-splicing event. The initial trans-splicing RNA targeted intron 6 and replaced exon 7 with the SMN1 exon 7 in SMN2 pre-mRNA. To determine the most efficient intron for SMN trans-splicing event, a panel of trans-splicing RNA molecules was constructed. Each trans-splicing RNA molecule targets a specific intron within the SMN2 pre-mRNA and based on the target intron, replaces the downstream exons including exon 7. These constructs were examined by RT-PCR, immunofluorescence, and Western blotting. We have identified intron 3 as the most efficient intron to support trans-splicing in cellular assays. The intron 3 trans-splicing construct targets intron 3 and replaces exons 4-7 and was distinguished based on its ability to produce the highest level of the trans-spliced RNA and full-length SMN protein in SMA patient fibroblasts. The efficiency of the intron 3 construct was further improved by addition of an antisense that blocks the 3' splice site at the intron 4/exon 5 junction. Most importantly, intracerebroventricular injection of the Int3 construct into SMNΔ7 mice elevated the SMN protein levels in the central nervous system. This research demonstrates an alternative platform to correct genetic defects, including SMN expression and examines the molecular basis for trans-splicing.

Rds3p is required for stable U2 snRNP recruitment to the splicing apparatus

Rds3p is a well-conserved 12-kDa protein with five CxxC zinc fingers that has been implicated in the activation of certain drug transport genes and in the pre-mRNA splicing pathway. Here we show that Rds3p resides in the yeast spliceosome and is essential for splicing in vitro. Rds3p purified from yeast stably associates with at least five U2 snRNP proteins, Cus1p, Hsh49p, Hsh155p, Rse1p, and Ist3p/Snu17p, and with the Yra1p RNA export factor. A mutation upstream of the first Rds3p zinc finger causes the conditional release of the putative branchpoint nucleotide binding protein, Ist3p/Snu17p, and weakens Rse1p interaction with the Rds3p complex. The resultant U2 snRNP particle migrates exceptionally slowly in polyacrylamide gels, suggestive of a disorganized structure. U2 snRNPs depleted of Rds3p fail to form stable prespliceosomes, although U2 snRNA stability is not affected. Metabolic depletion of Yra1p blocks cell growth but not splicing, suggesting that Yra1p association with Rds3p relates to Yra1p's role in RNA trafficking. Together these data establish Rds3p as an essential component of the U2 snRNP SF3b complex and suggest a new link between the nuclear processes of pre-mRNA splicing and RNA export.

Fission yeast gene structure and recognition

A database of 210 Schizosaccharomyces pombe DNA sequences (524,794 bp) was extracted from GenBank (release number 81.0) and examined by a number of methods in order to characterize statistical features of these sequences that might serve as signals or constraints for messenger RNA splicing. The statistical information compiled includes splicing signal (donor, acceptor and branch site) profiles, translational initiation start profile, exon/intron length distributions, ORF distribution, CDS size distribution, codon usage table, and 6-tuple distribution. The information content of the various signals are also presented. A rule-based interactive computer program for finding introns called INTRON.PLOT has been developed and was used to successfully analyze 7 newly sequenced genes.

Four yeast spliceosomal proteins (PRP5, PRP9, PRP11, and PRP21) interact to promote U2 snRNP binding to pre-mRNA

We have analyzed the functions of several pre-mRNA processing (PRP) proteins in yeast spliceosome formation. Here, we show that PRP5 (a DEAD box helicase-like protein), PRP9, and PRP11 are each required for the U2 snRNP to bind to the pre-spliceosome during spliceosome assembly in vitro. Genetic analyses of their functions suggest that they and another protein, PRP21, act concertedly and/or interact physically with each other and with the stem-loop IIa of U2 snRNA to bind U2 snRNP to the pre-mRNA. Biochemical complementation experiments also indicate that the PRP9 and PRP11 proteins interact. The PRP9 and PRP11 proteins may be functioning similarly in yeast and mammalian cells. The requirement for ATP and the helicase-like PRP5 protein suggests that these factors might promote a conformational change (involving either the U1 or U2 snRNP) that is required for the association of U2 snRNP with the pre-mRNA.

The Saccharomyces cerevisiae PRP21 gene product is an integral component of the prespliceosome

In Saccharomyces cerevisiae, the prp21 mutation causes accumulation of unspliced pre-mRNA at the nonpermissive temperature. We have cloned the PRP21 gene by complementation of its temperature-sensitive phenotype and found it to be the same as SPP91, an extragenic suppressor of the prp9 mutation previously studied in vivo by Chapon and Legrain [Chapon, C. & Legrain, P. (1992) EMBO J. 11, 3279-3288]. We have analyzed the effects of the prp21 mutation on splicing in vitro and have found that PRP21 is a splicing factor required for prespliceosome assembly. We also have analyzed the interaction of PRP21 with splicing complexes using anti-PRP21 antibodies and found that the RNA components of the prespliceosome--U1 and U2 small nuclear RNA (snRNA) particles and pre-mRNA--are specifically coimmunoprecipitated under splicing conditions in the presence of 0.2 M KCl. At higher KCl concentrations, U1 snRNP dissociates from splicing complexes; nevertheless, U2 snRNA and pre-mRNA are still efficiently immunoprecipitated. Immunoprecipitation of both U1 and U2 snRNA as well as pre-mRNA is ATP-dependent and requires a pre-mRNA capable of supporting prespliceosome assembly. Analysis of the unbound complexes in native gels confirmed that prespliceosomes are specifically immunoprecipitated by anti-PRP21 antibodies. These results demonstrate that PRP21 is an integral component of the prespliceosome and establishes a stable interaction with U2 snRNP and/or pre-mRNA in that complex.