Sequence requirements for self-splicing of the Tetrahymena thermophila pre-ribosomal RNA.

The sequence requirements for splicing of the Tetrahymena pre-rRNA have been examined by altering the rRNA gene to produce versions that contain insertions and deletions within the intervening sequence (IVS). The altered genes were transcribed and the RNA tested for self-splicing in vitro. A number of insertions (8-54 nucleotides) at three locations had no effect on self-splicing activity. Two of these insertions, located at a site 5 nucleotides preceding the 3'-end of the IVS, did not alter the choice of the 3' splice site. Thus the 3' splice site is not chosen by its distance from a fixed point within the IVS. Analysis of deletions constructed at two sites revealed two structures, a hairpin loop and a stem-loop, that are entirely dispensable for IVS excision in vitro. Three other regions were found to be necessary. The regions that are important for self-splicing are not restricted to the conserved sequence elements that define this class of intervening sequences. The requirement for structures within the IVS for pre-rRNA splicing is in sharp contrast to the very limited role of IVS structure in nuclear pre-mRNA splicing.     

Origin of the phosphate at the ligation junction produced by self-splicing of Tetrahymena thermophila pre-ribosomal RNA

The exons of the self-splicing pre-ribosomal RNA of Tetrahymena thermophila are joined accurately in vitro, even when only 33 nucleotides of the natural 5' exon and 38 nucleotides of the natural 3' exon remain. RNA fingerprint analysis was used to identify the unique ribonuclease T1 oligonucleotide generated by exon ligation. Secondary digests of the ligation junction oligonucleotide with ribonuclease A confirmed the identity of the fragment and demonstrated that the phosphate group that forms the phosphodiester bond at the ligation junction is derived from the 5' position of a uridine nucleotide in the RNA. This observation supports the prediction that the splice junction phosphate is derived from the 3' splice site. These results emphasize the mechanistic similarities of RNA splicing reactions of the group I introns, group II introns and nuclear pre-mRNA introns.     

Alternative secondary structures in the 5' exon affect both forward and reverse self-splicing of the Tetrahymena intervening sequence RNA

The natural splice junction of the Tetrahymena large ribosomal RNA is flanked by hairpins that are phylogenetically conserved. The stem immediately preceding the splice junction involves nucleotides that also base pair with the internal guide sequence of the intervening sequence during splicing. Thus, precursors which contain wild-type exons can form two alternative helices. We have constructed a series of RNAs where the stem-loop in the 5' exon is more or less stable than in the wild-type precursor, and tested them in both forward and reverse self-splicing reactions. The presence of a stable hairpin in ligated exon substrates interferes with the ability of the intervening sequence to integrate at the splice junction. Similarly, the presence of the wild-type hairpin in the 5' exon reduces the rate of splicing 20-fold in short precursors. The data are consistent with a competition between unproductive formation of a hairpin in the 5' exon and productive pairing of the 5' exon with the internal guide sequence. The reduction of splicing by a hairpin that is a normal feature of rRNA structure is surprising; we propose that this attenuation is relieved in the natural splicing environment.     

Activation of a cryptic 5' splice site by U1 snRNA

In the course of analyzing 5' splice site mutations in the second intron of Schizosaccharomyces pombe cdc2, we identified a cryptic 5' junction containing a nonconsensus nucleotide at position +2. An even more unusual feature of this cryptic 5' junction was its pattern of activation. By analyzing the profile of splicing products for an extensive series of cdc2 mutants in the presence and absence of compensatory U1 alleles, we have obtained evidence that the natural 5' splice site participates in activation of the cryptic 5' splice site, and that it does so via base pairing to U1 snRNA. Furthermore, the results of follow-up experiments strongly suggest that base pairing between U1 snRNA and the cryptic 5' junction itself plays a dominant role in its activation. Most remarkably, a mutant U1 can activate the cryptic 5' splice site even in the presence of a wild-type sequence at the natural 5' junction, providing unambiguous evidence that this snRNA redirects splicing via base pairing. Although previous work has demonstrated that U5 and U6 snRNAs can activate cryptic 5' splice sites through base pairing interactions, this is the first example in which U1 snRNA has been implicated in the final selection of a cryptic 5' junction.     

Base pairing between the 3'' exon and an internal guide sequence increases 3'' splice site specificity in the Tetrahymena self-splicing rRNA intron.

It has been proposed that recognition of the 3' splice site in many group I introns involves base pairing between the start of the 3' exon and a region of the intron known as the internal guide sequence (R. W. Davies, R. B. Waring, J. Ray, T. A. Brown, and C. Scazzocchio, Nature [London] 300:719-724, 1982). We have examined this hypothesis, using the self-splicing rRNA intron from Tetrahymena thermophila. Mutations in the 3' exon that weaken this proposed pairing increased use of a downstream cryptic 3' splice site. Compensatory mutations in the guide sequence that restore this pairing resulted in even stronger selection of the normal 3' splice site. These changes in 3' splice site usage were more pronounced in the background of a mutation (414A) which resulted in an adenine instead of a guanine being the last base of the intron. These results show that the proposed pairing (P10) plays an important role in ensuring that cryptic 3' splice sites are selected against. Surprisingly, the 414A mutation alone did not result in activation of the cryptic 3' splice site.     

Biased exon/intron distribution of cryptic and de novo 3' splice sites

We compiled sequences of previously published aberrant 3′ splice sites (3′ss) that were generated by mutations in human disease genes. Cryptic 3′ss, defined here as those resulting from a mutation of the 3′YAG consensus, were more frequent in exons than in introns. They clustered in ~20 nt region adjacent to authentic 3′ss, suggesting that their under-representation in introns is due to a depletion of AG dinucleotides in the polypyrimidine tract (PPT). In contrast, most aberrant 3′ss that were induced by mutations outside the 3′YAG consensus (designated ‘de novo’) were in introns. The activation of intronic de novo 3′ss was largely due to AG-creating mutations in the PPT. In contrast, exonic de novo 3′ss were more often induced by mutations improving the PPT, branchpoint sequence (BPS) or distant auxiliary signals, rather than by direct AG creation. The Shapiro–Senapathy matrix scores had a good prognostic value for cryptic, but not de novo 3′ss. Finally, AG-creating mutations in the PPT that produced aberrant 3′ss upstream of the predicted BPS in vivo shared a similar ‘BPS-new AG’ distance. Reduction of this distance and/or the strength of the new AG PPT in splicing reporter pre-mRNAs improved utilization of authentic 3′ss, suggesting that AG-creating mutations that are located closer to the BPS and are preceded by weaker PPT may result in less severe splicing defects.     

U5 snRNA interacts with exon sequences at 5' and 3' splice sites.

U5 snRNA is an essential pre-mRNA splicing factor whose function remains enigmatic. Specific mutations in a conserved single-stranded loop sequence in yeast U5 snRNA can activate cleavage of G1----A mutant pre-mRNAs at aberrant 5' splice sites and facilitate processing of dead-end lariat intermediates to mRNA. Activation of aberrant 5' cleavage sites involves base pairing between U5 snRNA and nucleotides upstream of the cleavage site. Processing of dead-end lariat intermediates to mRNA correlates with base pairing between U5 and the first two bases in exon 2. The loop sequence in U5 snRNA may therefore by intimately involved in the transesterification reactions at 5' and 3' splice sites. This pattern of interactions is strikingly reminiscent of exon recognition events in group II self-splicing introns and is consistent with the notion that U5 snRNA may be related to a specific functional domain from a group II-like self-splicing ancestral intron.     

Enhancer elements activate the weak 3' splice site of alpha-tropomyosin exon 2.

We have identified four purine-rich sequences that act as splicing enhancer elements to activate the weak 3' splice site of alpha-tropomyosin exon 2. These elements also activate the splicing of heterologous substrates containing weak 3' splice sites or mutated 5' splice sites. However, they are unique in that they can activate splicing whether they are placed in an upstream or downstream exon, and the two central elements can function regardless of their position relative to one another. The presence of excess RNAs containing these enhancers could effectively inhibit in vitro pre-mRNA splicing reactions in a substrate-dependent manner and, at lower concentrations of competitor RNA, the addition of SR proteins could relieve the inhibition. However, when extracts were depleted by incubation with biotinylated exon 2 RNAs followed by passage over streptavidin agarose, SR proteins were not sufficient to restore splicing. Instead, both SR proteins and fractions containing a 110-kD protein were necessary to rescue splicing. Using gel mobility shift assays, we show that formation of stable enhancer-specific complexes on alpha-tropomyosin exon 2 requires the presence of both SR proteins and the 110-kD protein. By analogy to the doublesex exon enhancer elements in Drosophila, our results suggest that assembly of mammalian exon enhancer complexes requires both SR and non-SR proteins to activate selection of weak splice sites.     

RNA helicase p68 (DDX5) regulates tau exon 10 splicing by modulating a stem-loop structure at the 5' splice site

Regulation of tau exon 10 splicing plays an important role in tauopathy. One of the cis elements regulating tau alternative splicing is a stem-loop structure at the 5' splice site of tau exon 10. The RNA helicase(s) modulating this stem-loop structure was unknown. We searched for splicing regulators interacting with this stem-loop region using an RNA affinity pulldown-coupled mass spectrometry approach and identified DDX5/RNA helicase p68 as an activator of tau exon 10 splicing. The activity of p68 in stimulating tau exon 10 inclusion is dependent on RBM4, an intronic splicing activator. RNase H cleavage and U1 protection assays suggest that p68 promotes conformational change of the stem-loop structure, thereby increasing the access of U1snRNP to the 5' splice site of tau exon 10. This study reports the first RNA helicase interacting with a stem-loop structure at the splice site and regulating alternative splicing in a helicase-dependent manner. Our work uncovers a previously unknown function of p68 in regulating tau exon 10 splicing. Furthermore, our experiments reveal functional interaction between two splicing activators for tau exon 10, p68 binding at the stem-loop region and RBM4 interacting with the intronic splicing enhancer region.     

Accurate selection of a 5' splice site requires sequences within fibronectin alternative exon B.

Inclusion of fibronectin alternative exon B in mRNA is developmentally regulated. Here we demonstrate that exon B contains two unique purine-rich sequence tracts, PRE1 and PRE2, that are important for proper 5' splice site selection both in vivo and in vitro. Targeted mutations of both PREs decreased the inclusion of exon B in the mRNA by 50% in vivo. Deletion or mutation of the PREs reduced removal of the downstream intron, but not the upstream intron, and induced the activation of cryptic 5' splice sites in vitro. PRE-mediated 5' splice selection activity appears sensitive to position and sequence context. A well characterized exon sequence enhancer that normally acts on the upstream 3' splice site can partially rescue proper exon B 5' splice site selection. In addition, we found that PRE 5' splice selection activity was preserved when exon B was inserted into a heterologous pre-mRNA substrate. Possible roles of these unique activities in modulating exon B splicing are considered.     

Circle reopening in the Tetrahymena ribozyme resembles site-specific hydrolysis at the 3' splice site.

The Tetrahymena intron, after splicing from its flanking exons, can mediate its own circularization. This is followed by site-specific hydrolysis of the phosphodiester bond formed during the circularization reaction. The structural components involved in recognition of this bond for hydrolysis have not been established. We have made base substitutions to the P9.0 pairing and at the 3'-terminal guanosine residue (G414) of the intron to investigate their effects on circle formation and reopening. We have found that disruption of either P9.0 pairing or binding of the terminal nucleotide result in the formation of a large circle, C-413:5E23 from precursor RNA molecules that have undergone hydrolysis at the 3' splice site. This circle is formed at the phosphodiester bond of the 5'-terminal guanosine residue of the upstream exon via nucleophilic attack by the 3'-terminal nucleotide of the intron. The large circle is novel since it can reopen eight bases downstream from the original circularization junction at a site resembling the normal 3' splice site, restoring a guanosine to the 3' terminus and re-establishing P9.0 pairing. The new 3' terminus of the intron is capable of recircularization at any of the three normal wild-type sites. We conclude that both P9.0 and the 3'-terminal guanosine residue are required for the selection of the phosphodiester bond hydrolysed during circle reopening.     

Length changes in the joining segment between domains 5 and 6 of a group II intron inhibit self-splicing and alter 3' splice site selection.

Domain 5 (D5) and domain 6 (D6) are adjacent folded hairpin substructures of self-splicing group II introns that appear to interact within the active ribozyme. Here we describe the effects of changing the length of the 3-nucleotide segment joining D5 to D6 [called J(56)3] on the splicing reactions of intron 5 gamma of the COXI gene of yeast mitochondrial DNA. Shortened variants J(56)0 and J(56)1 were defective in vitro for branching, and the second splicing step was performed inefficiently and inaccurately. The lengthened variant J(56)5 had a milder defect-splicing occurred at a reduced rate but with correct branching and a mostly accurate 3' splice junction choice. Yeast mitochondria were transformed with the J(56)5 allele, and the resulting yeast strain was respiration deficient because of ineffective aI5 gamma splicing. Respiration-competent revertants were recovered, and in one type a single joiner nucleotide was deleted while in the other type a nucleotide of D6 was deleted. Although these revertants still showed partial splicing blocks in vivo and in vitro, including a substantial defect in the second step of splicing, both spliced accurately in vivo. These results establish that a 3-nucleotide J(56) is optimal for this intron, especially for the accuracy of 3' splice junction selection, and indicate that D5 and D6 are probably not coaxially stacked.     

Identification of alternative 5'/3' splice sites based on the mechanism of splice site competition

Alternative splicing plays an important role in regulating gene expression. Currently, most efficient methods use expressed sequence tags or microarray analysis for large-scale detection of alternative splicing. However, it is difficult to detect all alternative splice events with them because of their inherent limitations. Previous computational methods for alternative splicing prediction could only predict particular kinds of alternative splice events. Thus, it would be highly desirable to predict alternative 5'/3' splice sites with various splicing levels using genomic sequences alone. Here, we introduce the competition mechanism of splice sites selection into alternative splice site prediction. This approach allows us to predict not only rarely used but also frequently used alternative splice sites. On a dataset extracted from the AltSplice database, our method correctly classified approximately 70% of the splice sites into alternative and constitutive, as well as approximately 80% of the locations of real competitors for alternative splice sites. It outperforms a method which only considers features extracted from the splice sites themselves. Furthermore, this approach can also predict the changes in activation level arising from mutations in flanking cryptic splice sites of a given splice site. Our approach might be useful for studying alternative splicing in both computational and molecular biology.     

Characterization of an authentic intermediate in the self-splicing process of ribosomal precursor RNA in macronuclei of Tetrahymena thermophila.

We have characterized a 1.5 kb RNA species in T. thermophila macronuclei previously found in vivo and including intron sequences linked to the 3' exon. This IVS-3' exon RNA could be detected in gels as a discrete molecule only after denaturation of nuclear RNA. After addition of 32P-GTP, as splicing cofactor in a nuclear in vitro system, the IVS-3' exon RNA was labeled at its 5' terminus, as was the by-product of splicing, the excised IVS RNA. The time course of labeling indicates that the IVS-3' exon RNA acts like a reaction intermediate and specifically a kinetic precursor to IVS RNA. Partial nuclease digestions showed that the IVS-3' exon RNA and the IVS RNA have the same 5' terminal sequence. In addition the IVS-3' exon RNA can release the 15-mer oligonucleotide cleaved off during circularization of IVS RNA under conditions of high temperature. Taken together, the structural, functional, and kinetic properties of the IVS-3' exon RNA strongly suggest that it represents a previously postulated in vivo intermediate in the splicing pathway.     

Probing of the spliceosome with site-specifically derivatized 5' splice site RNA oligonucleotides.

We have developed a site-specific chemical modification technique to incorporate a photoreactive azidophenacyl (APA) group at designated internal positions along the RNA phosphodiester backbone. Using this technique, we have analyzed interactions of the 5' splice site (5'SS) RNA within the spliceosome. Several crosslinked products can be detected within complex B using the derivatized 5'SS RNAs, including U6 snRNA, hPrp8p, and 114-, 90-, 70-, 54-, and 27-kDa proteins. The 5'SS RNAs derivatized at intron positions +4 to +8 crosslink to U6 snRNA, confirming the previously reported pairing interaction between these sequences. hPrp8p and p70 are crosslinked to the 5'SS RNA when the APA is placed within the 5' exon. Finally, a set of unidentified proteins, including p114, p54, and p27, is detected with the 5'SS RNA derivatized at intron positions +4 to +8. Introduction of the bulky APA group near the 5'SS junction (positions -2 to +3) strongly interferes with complex B formation and thus no APA crosslinks are observed at these positions. Together with our earlier observation that hPrp8p crosslinks to the GU dinucleotide at the 5' end of the intron, these results suggest that the inhibitory effect of APA results from steric hindrance of the hPrp8p:5'SS interaction. Unexpectedly, thio-modifications within the region of the 5'SS RNA that is involved in base pairing to U6 snRNA strongly stimulate complex B formation.     

Hierarchy for 5' splice site preference determined in vivo

The relationship between preferences among alternative 5' splice sites and their sequences has been investigated for 37 sequences by assessing their use in splicing relative to the 5' splice site of IVS-2 of rabbit beta-globin. There are strong correlations between the intrinsic strength of a 5' splice site and both the extent to which it resembles the consensus sequence and the calculated stability of its interactions with U1 small nuclear RNA. However, present methods of calculating either of the latter values do not allow predictions to be made of the relative preferences among a small number of sequences.