Conservation of an open-reading frame as an element affecting 5' splice site selection

A novel role for nucleolin in splice site selection

Latent 5' splice sites, not normally used, are highly abundant in human introns, but are activated under stress and in cancer, generating thousands of nonsense mRNAs. A previously proposed mechanism to suppress latent splicing was shown to be independent of NMD, with a pivotal role for initiator-tRNA independent of protein translation. To further elucidate this mechanism, we searched for nuclear proteins directly bound to initiator-tRNA. Starting with UV-crosslinking, we identified nucleolin (NCL) interacting directly and specifically with initiator-tRNA in the nucleus, but not in the cytoplasm. Next, we show the association of ini-tRNA and NCL with pre-mRNA. We further show that recovery of suppression of latent splicing by initiator-tRNA complementation is NCL dependent. Finally, upon nucleolin knockdown we show activation of latent splicing in hundreds of coding transcripts having important cellular functions. We thus propose nucleolin, a component of the endogenous spliceosome, through its direct binding to initiator-tRNA and its effect on latent splicing, as the first protein of a nuclear quality control mechanism regulating splice site selection to protect cells from latent splicing that can generate defective mRNAs.

Deep sequencing of pre-translational mRNPs reveals hidden flux through evolutionarily conserved alternative splicing nonsense-mediated decay pathways

BACKGROUND

Alternative splicing, which generates multiple mRNA isoforms from single genes, is crucial for the regulation of eukaryotic gene expression. The flux through competing splicing pathways cannot be determined by traditional RNA-Seq, however, because different mRNA isoforms can have widely differing decay rates. Indeed, some mRNA isoforms with extremely short half-lives, such as those subject to translation-dependent nonsense-mediated decay (AS-NMD), may be completely overlooked in even the most extensive RNA-Seq analyses.

RESULTS

RNA immunoprecipitation in tandem (RIPiT) of exon junction complex components allows for purification of post-splicing mRNA-protein particles (mRNPs) not yet subject to translation (pre-translational mRNPs) and, therefore, translation-dependent mRNA decay. Here we compare exon junction complex RIPiT-Seq to whole cell RNA-Seq data from HEK293 cells. Consistent with expectation, the flux through known AS-NMD pathways is substantially higher than that captured by RNA-Seq. Our RIPiT-Seq also definitively demonstrates that the splicing machinery itself has no ability to detect reading frame. We identify thousands of previously unannotated splicing events; while many can be attributed to splicing noise, others are evolutionarily conserved events that produce new AS-NMD isoforms likely involved in maintenance of protein homeostasis. Several of these occur in genes whose overexpression has been linked to poor cancer prognosis.

CONCLUSIONS

Deep sequencing of RNAs in post-splicing, pre-translational mRNPs provides a means to identify and quantify splicing events without the confounding influence of differential mRNA decay. For many known AS-NMD targets, the nonsense-mediated decay-linked alternative splicing pathway predominates. Exon junction complex RIPiT-Seq also revealed numerous conserved but previously unannotated AS-NMD events.

Structural studies of the endogenous spliceosome - The supraspliceosome

Pre-mRNA splicing is executed in mammalian cell nuclei within a huge (21MDa) and highly dynamic molecular machine - the supraspliceosome - that individually package pre-mRNA transcripts of different sizes and number of introns into complexes of a unique structure, indicating their universal nature. Detailed structural analysis of this huge and complex structure requires a stepwise approach using hybrid methods. Structural studies of the supraspliceosome by room temperature electron tomography, cryo-electron tomography, and scanning transmission electron microscope mass measurements revealed that it is composed of four native spliceosomes, each resembling an in vitro assembled spliceosome, which are connected by the pre-mRNA. It also elucidated the arrangement of the native spliceosomes within the intact supraspliceosome. Native spliceosomes and supraspliceosomes contain all five spliceosomal U snRNPs together with other splicing factors, and are active in splicing. The structure of the native spliceosome, at a resolution of 20Å, was determined by cryo-electron microscopy, and a unique spatial arrangement of the spliceosomal U snRNPs within the native spliceosome emerged from in silico studies. The supraspliceosome also harbor components for all pre-mRNA processing activities. Thus the supraspliceosome - the endogenous spliceosome - is a stand-alone complete macromolecular machine capable of performing splicing, alternative splicing, and encompass all nuclear pre-mRNA processing activities that the pre-mRNA has to undergo before it can exit from the nucleus to the cytoplasm to encode for protein. Further high-resolution cryo-electron microscopy studies of the endogenous spliceosome are required to decipher the regulation of alternative splicing, and elucidate the network of processing activities within it.

Heat shock activates splicing at latent alternative 5' splice sites in nematodes

Pre-mRNA splicing is essential for the regulation of gene expression in eukaryotes and is fundamental in development and cancer, and involves the selection of a consensus sequence that defines the 5' splice site (5'SS). Human introns harbor multiple sequences that conform to the 5'SS consensus, which are not used under normal growth conditions. Under heat shock conditions, splicing at such intronic latent 5'SSs occurred in thousands of human transcripts, resulting in pre-maturely terminated aberrant proteins. Here we performed a survey of the C. elegans genome, showing that worm's introns contain latent 5'SSs, whose use for splicing would have resulted in pre-maturely terminated mRNAs. Splicing at these latent 5'SSs could not be detected under normal growth conditions, while heat shock activated latent splicing in a number of tested C. elegans transcripts. Two scenarios could account for the lack of latent splicing under normal growth conditions (i) Splicing at latent 5'SSs do occur, but the nonsense mRNAs thus formed are rapidly and efficiently degraded (e.g. by NMD); and (ii) Splicing events at intronic latent 5'SSs are suppressed. Here we support the second scenario, because, nematode smg mutants that are devoid of NMD-essential factors, did not show latent splicing under normal growth conditions. Hence, these experiments together with our previous experiments in mammalian cells, indicate the existence of a nuclear quality control mechanism, termed Suppression Of Splicing (SOS), which discriminates between latent and authentic 5'SSs in an open reading frame dependent manner, and allows splicing only at the latter. Our results show that SOS is an evolutionary conserved mechanism, probably shared by most eukaryotes.

Genome-wide activation of latent donor splice sites in stress and disease

Sequences that conform to the 5′ splice site (5′SS) consensus are highly abundant in mammalian introns. Most of these sequences are preceded by at least one in-frame stop codon; thus, their use for splicing would result in pre-maturely terminated aberrant mRNAs. In normally grown cells, such intronic 5′SSs appear not to be selected for splicing. However, under heat shock conditions aberrant splicing involving such latent 5′SSs occurred in a number of specific gene transcripts. Using a splicing-sensitive microarray, we show here that stress-induced (e.g. heat shock) activation of latent splicing is widespread across the human transcriptome, thus highlighting the possibility that latent splicing may underlie certain diseases. Consistent with this notion, our analyses of data from the Gene Expression Omnibus (GEO) revealed widespread activation of latent splicing in cells grown under hypoxia and in certain cancers such as breast cancer and gliomas. These changes were found in thousands of transcripts representing a wide variety of functional groups; among them are genes involved in cell proliferation and differentiation. The GEO analysis also revealed a set of gene transcripts in oligodendroglioma, in which the level of activation of latent splicing increased with the severity of the disease.

A potential role for initiator-tRNA in pre-mRNA splicing regulation

The translation initiator-tRNA plays a crucial role in the initiation of protein synthesis in both prokaryotic and eukaryotic cells, by employing specific base pairing between its anticodon triplet CAU and the general initiation codon AUG in the mRNA. Here we show that the initiator-tRNA may also act, in a manner that is independent of its role in protein translation, as a pre-mRNA splicing regulator. Specifically, we show that alternative splicing events that are induced by mutations in the translation initiation AUG codon can be suppressed by expressing initiator-tRNA constructs carrying anticodon mutations that compensate for the AUG mutations. These mutated initiator-tRNAs appeared to be uncharged with an amino acid. Our results imply that recognition of the initiation AUG sequence by the anticodon triplet of initiator-tRNA in its unloaded state plays a role in quality control of splicing in the cell nucleus by a yet unresolved mechanism. Identifying the initiator-tRNA as a transacting splicing regulator suggests a novel involvement of this molecule in splicing regulation and provides a critical step toward deciphering this intriguing mechanism.

Cryptic splicing sites are differentially utilized in vivo

It has long been considered that cryptic splice sites are ignored by the splicing machinery in the context of intact genuine splice sites. In the present study, it is shown that cryptic splice sites are utilized in all circumstances, when the authentic site is intact, partially functional or completely abolished. Their use would therefore contribute to a background lack of fidelity in the context of the wild-type sequence. We also found that a mutation at the 5' splice site of beta-globin intron 1 accommodates multiple cryptic splicing pathways, including three previously reported pathways. Focusing on the two major cryptic 5' splice sites within beta-globin exon 1, we show that cryptic splice site selection ex vivo varies depending upon: (a) the cell stage of development during terminal erythroid differentiation; (b) the nature of the mutation at the authentic 5' splice site; and (c) the nature of the promoter. Finally, we found that the two major cryptic 5' splice sites are utilized with differential efficiencies in two siblings sharing the same beta-globin chromosome haplotype in the homozygous state. Collectively, these data suggest that intrinsic, sequence specific factors and cell genetic background factors both contribute to promote a subtle differential use of cryptic splice sites in vivo.

Oriented scanning is the leading mechanism underlying 5' splice site selection in mammals

Splice site selection is a key element of pre-mRNA splicing. Although it is known to involve specific recognition of short consensus sequences by the splicing machinery, the mechanisms by which 5′ splice sites are accurately identified remain controversial and incompletely resolved. The human F7 gene contains in its seventh intron (IVS7) a 37-bp VNTR minisatellite whose first element spans the exon7–IVS7 boundary. As a consequence, the IVS7 authentic donor splice site is followed by several cryptic splice sites identical in sequence, referred to as 5′ pseudo-sites, which normally remain silent. This region, therefore, provides a remarkable model to decipher the mechanism underlying 5′ splice site selection in mammals. We previously suggested a model for splice site selection that, in the presence of consecutive splice consensus sequences, would stimulate exclusively the selection of the most upstream 5′ splice site, rather than repressing the 3′ following pseudo-sites. In the present study, we provide experimental support to this hypothesis by using a mutational approach involving a panel of 50 mutant and wild-type F7 constructs expressed in various cell types. We demonstrate that the F7 IVS7 5′ pseudo-sites are functional, but do not compete with the authentic donor splice site. Moreover, we show that the selection of the 5′ splice site follows a scanning-type mechanism, precluding competition with other functional 5′ pseudo-sites available on immediate sequence context downstream of the activated one. In addition, 5′ pseudo-sites with an increased complementarity to U1snRNA up to 91% do not compete with the identified scanning mechanism. Altogether, these findings, which unveil a cell type–independent 5′−3′-oriented scanning process for accurate recognition of the authentic 5′ splice site, reconciliate apparently contradictory observations by establishing a hierarchy of competitiveness among the determinants involved in 5′ splice site selection.

Typically, mammalian genes contain coding sequences (exons) separated by non-coding sequences (introns). Introns are removed during pre-mRNA splicing. The accurate recognition of introns during splicing is essential, as any abnormality in that process will generate abnormal mRNAs that can cause diseases. Understanding the mechanisms of accurate splice site selection is of prime interest to life scientists. Exon–intron borders (splice sites) are defined by short sequences that are poorly conserved. The strength of any splice sequence can be assessed by its degree of homology with a splice site consensus sequence. Within exons and introns, several sequences can match with this consensus as well as or better than the splice sites. Using a system in which a splice site sequence is repeated several times in the intron, the authors showed that linear 5′−3′ search is a leading mechanism underlying splice site selection. This scanning mechanism is cell type–independent, and only the most upstream splice site of all the series is selected, even if splice sites with a better match to the consensus are in the vicinity. These findings reconciliate contradictory observations and establish a hierarchy among the determinants involved in splice site selection.

AUG sequences are required to sustain nonsense-codon-mediated suppression of splicing

More than 90% of human genes are rich in intronic latent 5' splice sites whose utilization in pre-mRNA splicing would introduce in-frame stop codons into the resultant mRNAs. We have therefore hypothesized that suppression of splicing (SOS) at latent 5' splice sites regulates alternative 5' splice site selection in a way that prevents the production of toxic nonsense mRNAs and verified this idea by showing that the removal of such in-frame stop codons is sufficient to activate latent splicing. Splicing control by SOS requires recognition of the mRNA reading frame, presumably recognizing the start codon sequence. Here we show that AUG sequences are indeed essential for SOS. Although protein translation does not seem to be required for SOS, the first AUG is shown here to be necessary but not sufficient. We further show that latent splicing can be elicited upon treatment with pactamycin-a drug known to block translation by its ability to recognize an RNA fold-but not by treatment with other drugs that inhibit translation through other mechanisms. The effect of pactamycin on SOS is dependent neither on steady-state translation nor on the pioneer round of translation. This effect is found for both transfected and endogenous genes, indicating that SOS is a natural mechanism.

Native spliceosomes assemble with pre-mRNA to form supraspliceosomes

Regulation of eukaryotic gene expression is achieved at different levels, which require accurate coordination. Macromolecular assemblies that exist as pre-formed entities can account for such coordination. Processing of pre-mRNA represents one step in this cascade of regulatory events but, moreover, provides explanation for protein versatility. The cellular machine where splicing of pre-mRNA, as well as additional processing events, take place in vivo is termed the supraspliceosome. Here, we show that the supraspliceosome is composed of four active spliceosomes, termed native spliceosomes, connected to each other by the pre-mRNA. Cleavage of pre-mRNA shows that its integrity is essential for the stability of the supraspliceosome. Furthermore, supraspliceosomes can be reconstituted in vitro, from purified native spliceosomes by addition of synthetic pre-mRNAs, providing further support to the supraspliceosome as a preassembled biological complex. The internal setting of the native spliceosomes within the supraspliceosome is most suitable to enable the communication between these structures, which is crucial in order to achieve regulated splicing.

Generation of recombinant influenza A virus without M2 ion-channel protein by introduction of a point mutation at the 5' end of the viral intron

The aim of this study was to inhibit influenza virus M2 protein expression by mutating the splicing signal of the M gene. Mutations were introduced into the GU dinucleotide sequence at the 5'-proximal splicing site of the M gene (corresponding to nt 52-53 of M cRNA). Transfected cells expressing mutated M viral ribonucleoproteins failed to generate M2 mRNA. Interestingly, recombinant viruses with mutations at the dinucleotide sequence were viable, albeit attenuated, in cell culture. These recombinants failed to express M2 mRNA and M2 protein. These observations demonstrated that the GU invariant dinucleotide sequence at the 5'-proximal splicing site of M gene is essential for M2 mRNA synthesis. These results also indicated that the M2 ion-channel protein is critical, but not essential, for virus replication in cell culture. This approach may provide a new way of producing attenuated influenza A virus.

A survey of splice variants of the human hypoxanthine phosphoribosyl transferase and DNA polymerase beta genes: products of alternative or aberrant splicing?

Errors during the pre-mRNA splicing of metazoan genes can degrade the transmission of genetic information, and have been associated with a variety of human diseases. In order to characterize the mutagenic and pathogenic potential of mis-splicing, we have surveyed and quantified the aberrant splice variants in the human hypoxanthine phosphoribosyl transferase (HPRT) and DNA polymerase beta (POLB) in the presence and the absence of the Nonsense Mediated Decay (NMD) pathway, which removes transcripts with premature termination codons. POLB exhibits a high frequency of splice variants (40-60%), whereas the frequency of HPRT splice variants is considerably lower (approximately 1%). Treatment of cells with emetine to inactivate NMD alters both the spectrum and frequency of splice variants of POLB and HPRT. It is not certain at this point, whether POLB and HPRT splice variants are the result of regulated alternative splicing processes or the result of aberrant splicing, but it appears likely that at least some of the variants are the result of splicing errors. Several mechanisms that may contribute to aberrant splicing are discussed.

Stop codon-mediated suppression of splicing is a novel nuclear scanning mechanism not affected by elements of protein synthesis and NMD

The pre-mRNA splicing machine must frequently discriminate between normal and many potential 5'splice sites that match the consensus sequence but remain latent. Suppression of splicing (SOS) at such latent 5'splice sites is required for the maintenance of an open reading frame, and to ensure that only RNAs that encode for functional proteins will be formed. In this study we show that SOS is a novel mechanism distinct from the known RNA surveillance mechanisms. First, SOS is distinct from nonsense-mediated mRNA decay (NMD) because it is not dependent on translation and is not affected by RNAi-mediated down-regulation of hUpf1 and hUpf2--two key components of the NMD pathway. Second, SOS is distinct from nonsense-associated alternative splicing (NAS), because a mutant of hUpf1, which was shown to abrogate NAS, does not activate latent splicing. Elucidating the mechanism of SOS is pertinent to human disease in view of the large number of human genes that harbor latent splice sites.

Three-Dimensional Structure of the Native Spliceosome by Cryo-Electron Microscopy

Splicing of pre-mRNA occurs in a multicomponent macromolecular machine--the spliceosome. The spliceosome can be assembled in vitro by a stepwise assembly of a number of snRNPs and additional proteins on exogenously added pre-mRNA. In contrast, splicing in vivo occurs in preformed particles where endogenous pre-mRNAs are packaged with all five spliceosomal U snRNPs (penta-snRNP) together with other splicing factors. Here we present a three-dimensional image reconstruction by cryo-electron microscopy of native spliceosomes, derived from cell nuclei, at a resolution of 20 angstroms. The structure revealed an elongated globular particle made up of two distinct subunits connected to each other leaving a tunnel in between. We show here that the larger subunit is a suitable candidate to accommodate the penta-snRNP, and that the tunnel could accommodate the pre-mRNA component of the spliceosome. The features this structure reveals provide new insight into the global architecture of the native splicing machine.

Regulation of splicing: the importance of being translatable

RNA sequences that conform to the consensus sequence of 5' splice sites but are not used for splicing occur frequently in protein coding genes. Mutational analyses have shown that suppression of splicing at such latent sites may be dictated by the necessity to maintain an open reading frame in the mRNA. Here we show that stop codon frequency in introns having latent 5' splice sites is significantly greater than that of introns lacking such sites and significantly greater than the expected occurrence by chance alone. Both observations suggest the occurrence of a general mechanism that recognizes the mRNA reading frame in the context of pre-mRNA.

The effect of nonsense codons on splicing: a genomic analysis

The phenomenon of nonsense-associated altered splicing raises the possibility that the recognition of in-frame nonsense codons is used generally for exon identification during pre-mRNA splicing. However, nonsense codon frequencies in pseudo exons and in regions flanking 5' splice sites are no greater than that expected by chance, arguing against the widespread use of this strategy as a means of rejecting potential splice sites.