Exon mutations that affect the choice of splice sites used in processing the SV40 late transcripts

Exonic splice regulation imposes strong selection at synonymous sites

What proportion of coding sequence nucleotides have roles in splicing, and how strong is the selection that maintains them? Despite a large body of research into exonic splice regulatory signals, these questions have not been answered. This is because, to our knowledge, previous investigations have not explicitly disentangled the frequency of splice regulatory elements from the strength of the evolutionary constraint under which they evolve. Current data are consistent both with a scenario of weak and diffuse constraint, enveloping large swaths of sequence, as well as with well-defined pockets of strong purifying selection. In the former case, natural selection on exonic splice enhancers (ESEs) might primarily act as a slight modifier of codon usage bias. In the latter, mutations that disrupt ESEs are likely to have large fitness and, potentially, clinical effects. To distinguish between these scenarios, we used several different methods to determine the distribution of selection coefficients for new mutations within ESEs. The analyses converged to suggest that ∼15%-20% of fourfold degenerate sites are part of functional ESEs. Most of these sites are under strong evolutionary constraint. Therefore, exonic splice regulation does not simply impose a weak bias that gently nudges coding sequence evolution in a particular direction. Rather, the selection to preserve these motifs is a strong force that severely constrains the evolution of a substantial proportion of coding nucleotides. Thus synonymous mutations that disrupt ESEs should be considered as a potentially common cause of single-locus genetic disorders.

Complete alternative splicing events are bubbles in splicing graphs

Eukaryotic splicing structures are known to involve a high degree of alternative forms derived from a premature transcript by alternative splicing (AS). With the advent of new sequencing technologies, evidence for new splice forms becomes increasingly available-bit by bit revealing that the true splicing diversity of "AS events" often comprises more than two alternatives and therefore cannot be sufficiently described by pairwise comparisons as conducted in analyzes hitherto. Especially, I emphasize on "complete" AS events which include all hitherto known variants of a splicing variation. Challenges emerge from the richness of data (millions of transcripts) and artifacts introduced during the technical process of obtaining transcript sequences ("noise")-especially when dealing with single-read sequences known as expressed sequence tags (ESTs). Herein, I describe a novel method to efficiently predict AS events in different resolutions ("dimensions") from transcript annotations that allows for combination of fragmented EST data with full-length cDNAs and can cope with large datasets containing noise. At the doorstep of many new splice forms becoming available by novel high-throughput sequencing technologies, the presented method helps to dynamically update AS databases. Applying this method to estimate the real complexity of alternative splicing, I found in human and murine annotations thousands of novel AS events that either have been disregarded or mischaracterized in earlier works. The growth of evidence for such events suggests that the number still keeps climbing. When considering complete events, the majority of exons that are observed as "mutually exclusive" in pairwise comparisons in fact involves at least one other alternative splice form that disagrees with their mutual exclusion. Similar observations also hold for the alternative skipping of two subsequent exons. Results suggest that the systematical analysis of complete AS events on large scale provides subtle insights in the mechanisms that drive (alternative) splicing.

Functional selection of splicing enhancers that stimulate trans-splicing in vitro

The role of exonic sequences in naturally occurring trans-splicing has not been explored in detail. Here, we have identified trans-splicing enhancers through the use of an iterative selection scheme. Several classes of enhancer sequences were identified that led to dramatic increases in trans-splicing efficiency. Two sequence families were investigated in detail. These include motifs containing the element (G/C)GAC(G/C) and also 5' splice site-like sequences. Distinct elements were tested for their ability to function as splicing enhancers and in competition experiments. In addition, discrete trans-acting factors were identified. This work demonstrates that splicing enhancers are able to effect a large increase in trans-splicing efficiency and that the process of exon definition is able to positively enhance trans-splicing even though the reaction itself is independent of the need for the 5' end of U1 snRNA. Due to the presence of internal introns in messages that are trans-spliced, the natural arrangement of 5' splice sites downstream of trans-splicing acceptors may lead to a general promotion of this unusual reaction.

Splicing enhancement in the yeast rp51b intron

Splicing enhancement in higher eukaryotes has been linked to SR proteins, to U1 snRNP, and to communication between splice sites across introns or exons mediated by protein-protein interactions. It has been previously shown that, in yeast, communication mediated by RNA-RNA interactions between the two ends of introns is a basis for splicing enhancement. We designed experiments of randomization-selection to isolate splicing enhancers that would work independently from RNA secondary structures. Surprisingly, one of the two families of sequences selected was essentially composed of 5' splice site variants. We show that this sequence enhances splicing independently of secondary structure, is exportable to heterologous contexts, and works in multiple copies with additive effects. The data argue in favor of an early role for splicing enhancement, possibly coincident with commitment complex formation. Genetic compensation experiments with U1 snRNA mutants suggest that U1 snRNP binding to noncanonical locations is required for splicing enhancement.

Analysis of capsid formation of human polyomavirus JC (Tokyo-1 strain) by a eukaryotic expression system: splicing of late RNAs, translation and nuclear transport of major capsid protein VP1, and capsid assembly

Human polyomavirus JC (JCV) can encode the three capsid proteins VP1, VP2, and VP3, downstream of the agnoprotein in the late region. JCV virions are identified in the nucleus of infected cells. In this study, we have elucidated unique features of JCV capsid formation by using a eukaryotic expression system. Structures of JCV polycistronic late RNAs (M1 to M4 and possibly M5 and M6) generated by alternative splicing were determined. VP1 would be synthesized from M2 RNA, and VP2 and VP3 would be synthesized from M1 RNA. The presence of the open reading frame of the agnoprotein or the leader sequence (nucleotides 275 to 409) can decrease the expression level of VP1. VP1 was efficiently transported to the nucleus in the presence of VP2 and VP3 but distributed both in the cytoplasm and in the nucleus in their absence. Mutation analysis indicated that inefficiency in nuclear transport of VP1 is due to the unique structure in the N-terminal sequence, KRKGERK. Within the nucleus, VP1 was localized discretely and identified as speckles in the presence of VP2 and VP3 but distributed diffusely in their absence. These results suggest that VP1 was efficiently transported to the nucleus and localized in the discrete subnuclear regions, possibly with VP2 and VP3. By electron microscopy, recombinant virus particles were identified in the nucleus, and their intranuclear distribution was consistent with distribution of speckles. This system provides a useful model with which to understand JCV capsid formation and the structures and functions of the JCV capsid proteins.

Multiple distinct splicing enhancers in the protein-coding sequences of a constitutively spliced pre-mRNA

We have identified multiple distinct splicing enhancer elements within protein-coding sequences of the constitutively spliced human beta-globin pre-mRNA. Each of these highly conserved sequences is sufficient to activate the splicing of a heterologous enhancer-dependent pre-mRNA. One of these enhancers is activated by and binds to the SR protein SC35, whereas at least two others are activated by the SR protein SF2/ASF. A single base mutation within another enhancer element inactivates the enhancer but does not change the encoded amino acid. Thus, overlapping protein coding and RNA recognition elements may be coselected during evolution. These studies provide the first direct evidence that SR protein-specific splicing enhancers are located within the coding regions of constitutively spliced pre-mRNAs. We propose that these enhancers function as multisite splicing enhancers to specify 3' splice-site selection.

Novel exonic elements that modulate splicing of the human fibronectin EDA exon

Three exons in the fibronectin primary transcript are alternatively spliced in a tissue- and developmental stage-specific manner. One of these exons, EDA, has been shown previously by others to contain two splicing regulatory elements between 155 and 180 nucleotides downstream of the 3'-splice site: an exon splicing enhancer and a negative element. By transient expression of a chimeric beta-globin/fibronectin EDA intron in COS-7 cells, we have identified two additional exonic splicing regulatory elements. RNA generated by a construct containing the first 120 nucleotides of the fibronectin EDA exon was spliced with an efficiency of approximately 50%. Deletion of most of the fibronectin EDA exon sequences resulted in a 20-fold increase in the amount of spliced RNA, indicative of an exon splicing silencer. Deletion and mutagenesis studies suggest that the fibronectin exon splicing silencer is associated with a conserved RNA secondary structure. In addition, sequences between nucleotides 93 and 118 of the EDA exon contain a non-purine-rich splicing enhancer as demonstrated by its ability to function in a heterologous context.

Modification of the alternative splicing process of testosterone-repressed prostate message-2 (TRPM-2) gene by protein synthesis inhibitors and heat shock treatment

During the course of the study to examine the effect of cycloheximide on apoptosis-related genes, the variant rat testosterone-repressed prostate message-2 (TRPM-2) mRNA deficient of the exon 5 was found. The putative protein encoded by the variant TRPM-2 mRNA is only constituted from the N-terminal one-third portion of the ordinary TRPM-2 protein. The expression of the variant form was increased dramatically by cycloheximide treatment, while that of the ordinary form was not affected very much. The similar phenomenon was also observed by the use of other types of protein synthesis inhibitors, anisomycin and emetine. The enhancement of expression of the variant was observed in the rat treated with heat shock as well. The variant form was presumably generated by the exon skip mechanism. Systematic analyses of cycloheximide effect on the alternative splicing at various splicing junctions were performed. However, cycloheximide did not exhibit any remarkable effects on other types of alternative splicing, including exon skip in beta A4-amyloid protein precursor (APP) gene, alternative donor selection in Fas antigen gene and alternative acceptor selection in catechol O-methyltransferase (COMT) gene. These results indicated that the induction of exon skip by both protein synthesis inhibition and heat shock treatment occurs in a limited number of genes, if not only in TRPM-2.

Selection of novel exon recognition elements from a pool of random sequences

A 20-nucleotide sequence close to the 3' end of the internal exon of a model two-intron, three-exon pre-mRNA (DUP184 [Z. Dominski and R. Kole, J. Biol. Chem. 269:23590-23596, 1994]) was replaced by a random 20-mer, resulting in a pool of pre-mRNAs which, like the initial DUP184 construct, were spliced in vitro by a pathway leading to predominant skipping of the internal exon. The randomized pre-mRNAs were subjected to a selection protocol, resulting in a pool enriched in pre-mRNAs that efficiently included the internal exon. Isolation and sequencing of a number of clones corresponding to the selected pre-mRNAs showed that two classes of sequences were selected from the initial pool. Most abundant among these were sequences with purine tracts similar to those in the recently identified exon-splicing enhancers while a smaller class included sequences lacking discernible purine tracts within the 20-nucleotide region. Splicing of selected pre-mRNAs showed that the purine tracts vary in their ability to promote exon inclusion and, more important, that sequences lacking purine tracts stimulate inclusion of the internal exon as efficiently as their purine-rich counterparts. The latter result indicates the existence of a novel class of exon recognition sequences or splicing enhancers.

SR proteins promote the first specific recognition of Pre-mRNA and are present together with the U1 small nuclear ribonucleoprotein particle in a general splicing enhancer complex

We show that addition of SR proteins to in vitro splicing extracts results in a significant increase in assembly of the earliest prespliceosomal complex E and a corresponding decrease in assembly of the heterogeneous nuclear ribonucleoprotein (hnRNP) complex H. In addition, SR proteins promote formation of the E5' and E3' complexes that assemble on RNAs containing only 5' and 3' splice sites, respectively. We conclude that SR proteins promote the earliest specific recognition of both the 5' and 3' splice sites and are limiting for this function in HeLa nuclear extracts. Using UV cross-linking, we demonstrate specific, splice site-dependent RNA-protein interactions of SR proteins in the E, E5', and E3' complexes. SR proteins do not UV cross-link in the H complex, and conversely, hnRNP cross-linking is largely excluded from the E-type complexes. We also show that a discrete complex resembling the E5' complex assembles on both purine-rich and non-purine-rich exonic splicing enhancers. This complex, which we have designated the Enhancer complex, contains U1 small nuclear RNP (snRNP) and is associated with different SR protein family members, depending on the sequence of the enhancer. We propose that both downstream 5' splice site enhancers and exonic enhancers function by establishing a network of pre-mRNA-protein and protein-protein interactions involving U1 snRNP, SR proteins, and U2AF that is similar to the interactions that bring the 5' and 3' splice sites together in the E complex.

Identification and characterization by antisense oligonucleotides of exon and intron sequences required for splicing

Certain thalassemic human beta-globin pre-mRNAs carry mutations that generate aberrant splice sites and/or activate cryptic splice sites, providing a convenient and clinically relevant system to study splice site selection. Antisense 2'-O-methyl oligoribonucleotides were used to block a number of sequences in these pre-mRNAs and were tested for their ability to inhibit splicing in vitro or to affect the ratio between aberrantly and correctly spliced products. By this approach, it was found that (i) up to 19 nucleotides upstream from the branch point adenosine are involved in proper recognition and functioning of the branch point sequence; (ii) whereas at least 25 nucleotides of exon sequences at both 3' and 5' ends are required for splicing, this requirement does not extend past the 5' splice site sequence of the intron; and (iii) improving the 5' splice site of the internal exon to match the consensus sequence strongly decreases the accessibility of the upstream 3' splice site to antisense 2'-O-methyl oligoribonucleotides. This result most likely reflects changes in the strength of interactions near the 3' splice site in response to improvement of the 5' splice site and further supports the existence of communication between these sites across the exon.

SR proteins promote the first specific recognition of Pre-mRNA and are present together with the U1 small nuclear ribonucleoprotein particle in a general splicing enhancer complex

We show that addition of SR proteins to in vitro splicing extracts results in a significant increase in assembly of the earliest prespliceosomal complex E and a corresponding decrease in assembly of the heterogeneous nuclear ribonucleoprotein (hnRNP) complex H. In addition, SR proteins promote formation of the E5' and E3' complexes that assemble on RNAs containing only 5' and 3' splice sites, respectively. We conclude that SR proteins promote the earliest specific recognition of both the 5' and 3' splice sites and are limiting for this function in HeLa nuclear extracts. Using UV cross-linking, we demonstrate specific, splice site-dependent RNA-protein interactions of SR proteins in the E, E5', and E3' complexes. SR proteins do not UV cross-link in the H complex, and conversely, hnRNP cross-linking is largely excluded from the E-type complexes. We also show that a discrete complex resembling the E5' complex assembles on both purine-rich and non-purine-rich exonic splicing enhancers. This complex, which we have designated the Enhancer complex, contains U1 small nuclear RNP (snRNP) and is associated with different SR protein family members, depending on the sequence of the enhancer. We propose that both downstream 5' splice site enhancers and exonic enhancers function by establishing a network of pre-mRNA-protein and protein-protein interactions involving U1 snRNP, SR proteins, and U2AF that is similar to the interactions that bring the 5' and 3' splice sites together in the E complex.

Identification and characterization by antisense oligonucleotides of exon and intron sequences required for splicing

Certain thalassemic human beta-globin pre-mRNAs carry mutations that generate aberrant splice sites and/or activate cryptic splice sites, providing a convenient and clinically relevant system to study splice site selection. Antisense 2'-O-methyl oligoribonucleotides were used to block a number of sequences in these pre-mRNAs and were tested for their ability to inhibit splicing in vitro or to affect the ratio between aberrantly and correctly spliced products. By this approach, it was found that (i) up to 19 nucleotides upstream from the branch point adenosine are involved in proper recognition and functioning of the branch point sequence; (ii) whereas at least 25 nucleotides of exon sequences at both 3' and 5' ends are required for splicing, this requirement does not extend past the 5' splice site sequence of the intron; and (iii) improving the 5' splice site of the internal exon to match the consensus sequence strongly decreases the accessibility of the upstream 3' splice site to antisense 2'-O-methyl oligoribonucleotides. This result most likely reflects changes in the strength of interactions near the 3' splice site in response to improvement of the 5' splice site and further supports the existence of communication between these sites across the exon.

Regulation of alternative pre-mRNA splicing by a novel repeated hexanucleotide element

The alternatively spliced exon EIIIB is regulated in a cell type-specific manner in the rat fibronectin gene. Splicing of EIIIB into fibronectin mRNA is dependent on sequences in the intron immediately downstream of EIIIB. We show that a short, highly repeated TGCATG motif in this intron is important for cell type-specific recognition of EIIIB as an exon. This motif enhances usage of the EIIIB 5' splice site; furthermore, this repeated TGCATG sequence can activate an alternatively spliced exon in the unrelated rat preprotachykinin pre-mRNA. Interestingly, this sequence can also be found within cis-acting elements identified previously in other alternatively spliced genes. This short repeated TGCATG motif is therefore a cell type-specific element that, in addition to controlling fibronectin alternative splicing, may participate in the regulation of other alternative RNA processing events.

Alternative splicing of beta-tropomyosin pre-mRNA: multiple cis-elements can contribute to the use of the 5'- and 3'-splice sites of the nonmuscle/smooth muscle exon 6

We previously found that the splicing of exon 5 to exon 6 in the rat beta-TM gene required that exon 6 first be joined to the downstream common exon 8 (Helfman et al., Genes and Dev. 2, 1627-1638, 1988). Pre-mRNAs containing exon 5, intron 5 and exon 6 are not normally spliced in vitro. We have carried out a mutational analysis to determine which sequences in the pre-mRNA contribute to the inability of this precursor to be spliced in vitro. We found that mutations in two regions of the pre-mRNA led to activation of the 3'-splice site of exon 6, without first joining exon 6 to exon 8. First, introduction of a nine nucleotide poly U tract upstream of the 3'-splice site of exon 6 results in the splicing of exon 5 to exon 6 with as little as 35 nucleotides of exon 6. Second, introduction of a consensus 5'-splice site in exon 6 led to splicing of exon 5 to exon 6. Thus, three distinct elements can act independently to activate the use of the 3'-splice site of exon 6: (1) the sequences contained within exon 8 when joined to exon 6, (2) a poly U tract in intron 5, and (3) a consensus 5'-splice site in exon 6. Using biochemical assays, we have determined that these sequence elements interact with distinct cellular factors for 3'-splice site utilization. Although HeLa cell nuclear extracts were able to splice all three types of pre-mRNAs mentioned above, a cytoplasmic S100 fraction supplemented with SR proteins was unable to efficiently splice exon 5 to exon 6 using precursors in which exon 6 was joined to exon 8. We also studied how these elements contribute to alternative splice site selection using precursors containing the mutually exclusive, alternatively spliced cassette comprised of exons 5 through 8. Introduction of the poly U tract upstream of exon 6, and changing the 5'-splice site of exon 6 to a consensus sequence, either alone or in combination, facilitated the use of exon 6 in vitro, such that exon 6 was spliced more efficiently to exon 8. These data show that intron sequences upstream of an exon can contribute to the use of the downstream 5'-splice, and that sequences surrounding exon 6 can contribute to tissue-specific alternative splice site selection.

Identification of a novel exonic mutation at -13 from 5' splice site causing exon skipping in a girl with mitochondrial acetoacetyl-coenzyme A thiolase deficiency

We identified a novel exonic mutation which causes exon skipping in the mitochondrial acetoacetyl-CoA thiolase (T2) gene from a girl with T2 deficiency (GK07). GK07 is a compound heterozygote; the maternal allele has a novel G to T transversion at position 1136 causing Gly379 to Val substitution (G379V) of the T2 precursor. In case of in vivo expression analysis, cells transfected with this mutant cDNA showed no evidence of restored T2 activity. The paternal allele was associated with exon 8 skipping at the cDNA level. At the gene level, a C to T transition causing Gln272 to termination codon (Q272STOP) was identified within exon 8, 13 bp from the 5' splice site of intron 8 in the paternal allele. The mRNA with Q272STOP could not be detected in GK07 fibroblasts, presumably because pre-mRNA with Q272STOP was unstable because of the premature termination. In vivo splicing experiments revealed that the exonic mutation caused partial skipping of exon 8. This substitution was thought to alter the secondary structure of T2 pre-mRNA around exon 8 and thus impede normal splicing. The role of exon sequences in the splicing mechanism is indicated by the exon skipping which occurred with an exonic mutation.

Polypurine sequences within a downstream exon function as a splicing enhancer

We have previously shown that a purine-rich sequence located within exon M2 of the mouse immunoglobulin mu gene functions as a splicing enhancer, as judged by its ability to stimulate splicing of a distant upstream intron. This sequence element has been designated ERS (exon recognition sequence). In this study, we investigated the stimulatory effects of various ERS-like sequences, using the in vitro splicing system with HeLa cell nuclear extracts. Here, we show that purine-rich sequences of several natural exons that have previously been shown to be required for splicing function as a splicing enhancer like the ERS of the immunoglobulin mu gene. Moreover, even synthetic polypurine sequences had stimulatory effects on the upstream splicing. Evaluation of the data obtained from the analyses of both natural and synthetic purine-rich sequences shows that (i) alternating purine sequences can stimulate splicing, while poly(A) or poly(G) sequences cannot, and (ii) the presence of U residues within the polypurine sequence greatly reduces the level of stimulation. Competition experiments strongly suggest that the stimulatory effects of various purine-rich sequences are mediated by the same trans-acting factor(s). We conclude from these results that the purine-rich sequences that we examined in this study also represent examples of ERS. Thus, ERS is considered a general splicing element that is present in various exons and plays an important role in splice site selection.

Polypurine sequences within a downstream exon function as a splicing enhancer

We have previously shown that a purine-rich sequence located within exon M2 of the mouse immunoglobulin mu gene functions as a splicing enhancer, as judged by its ability to stimulate splicing of a distant upstream intron. This sequence element has been designated ERS (exon recognition sequence). In this study, we investigated the stimulatory effects of various ERS-like sequences, using the in vitro splicing system with HeLa cell nuclear extracts. Here, we show that purine-rich sequences of several natural exons that have previously been shown to be required for splicing function as a splicing enhancer like the ERS of the immunoglobulin mu gene. Moreover, even synthetic polypurine sequences had stimulatory effects on the upstream splicing. Evaluation of the data obtained from the analyses of both natural and synthetic purine-rich sequences shows that (i) alternating purine sequences can stimulate splicing, while poly(A) or poly(G) sequences cannot, and (ii) the presence of U residues within the polypurine sequence greatly reduces the level of stimulation. Competition experiments strongly suggest that the stimulatory effects of various purine-rich sequences are mediated by the same trans-acting factor(s). We conclude from these results that the purine-rich sequences that we examined in this study also represent examples of ERS. Thus, ERS is considered a general splicing element that is present in various exons and plays an important role in splice site selection.

cis-elements involved in alternative splicing in the rat beta-tropomyosin gene: the 3'-splice site of the skeletal muscle exon 7 is the major site of blockage in nonmuscle cells

We have been using the rat beta-tropomyosin (beta-TM) gene as a model system to study the mechanism of alternative splicing. The beta-TM gene spans 10 kb with 11 exons and encodes two distinct isoforms, namely skeletal muscle beta-TM and fibroblast TM-1. Exons 1-5, 8, and 9 are common to all mRNAs expressed from this gene. Exons 6 and 11 are used in fibroblasts, as well as in smooth muscle cells, whereas exons 7 and 10 are used exclusively in skeletal muscle cells. Our previous studies localized the critical elements for regulated alternative splicing to sequences within exon 7 and the adjacent upstream intron. We also demonstrated that these sequences function, in part, to regulate splice-site selection in vivo by interacting with cellular factors that block the use of the skeletal muscle exon in nonmuscle cells (1). Here we have further characterized the critical cis-acting elements involved in alternative splice site selection. Our data demonstrate that exon 7 and its flanking intron sequences are sufficient to regulate the suppression of exon 7 in nonmuscle cells when flanked by heterologous exons derived from adenovirus. We have also shown by both in vivo and in vitro assays that the blockage of exon 7 in nonmuscle cells is primarily at its 3'-splice site. A model is presented for regulated alternative splicing in both skeletal muscle and nonmuscle cells.

The cardiac troponin T alternative exon contains a novel purine-rich positive splicing element

We have characterized a novel positive-acting splicing element within the developmentally regulated alternative exon (exon 5) of the cardiac troponin T (cTNT) gene. The exon splicing element (ESE) is internal to the exon portions of the splice sites and is required for splicing to the 3' splice site but not the 5' splice site flanking the exon. Sequence comparisons between cTNT exon 5 and other exons that contain regions required for splicing reveal a common purine-rich motif. Sequence within cTNT exon 5 or a synthetic purine-rich motif facilitates splicing of heterologous alternative and constitutive splice sites in vivo. Interestingly, the ESE is not required for the preferential inclusion of cTNT exon 5 observed in primary skeletal muscle cultures. Our results strongly suggest that the purine-rich ESE serves as a general splicing element that is recognized by the constitutive splicing machinery.

The cardiac troponin T alternative exon contains a novel purine-rich positive splicing element

We have characterized a novel positive-acting splicing element within the developmentally regulated alternative exon (exon 5) of the cardiac troponin T (cTNT) gene. The exon splicing element (ESE) is internal to the exon portions of the splice sites and is required for splicing to the 3' splice site but not the 5' splice site flanking the exon. Sequence comparisons between cTNT exon 5 and other exons that contain regions required for splicing reveal a common purine-rich motif. Sequence within cTNT exon 5 or a synthetic purine-rich motif facilitates splicing of heterologous alternative and constitutive splice sites in vivo. Interestingly, the ESE is not required for the preferential inclusion of cTNT exon 5 observed in primary skeletal muscle cultures. Our results strongly suggest that the purine-rich ESE serves as a general splicing element that is recognized by the constitutive splicing machinery.

In vivo recognition of a vertebrate mini-exon as an exon-intron-exon unit

Very small vertebrate exons are problematic for RNA splicing because of the proximity of their 3' and 5' splice sites. In this study, we investigated the recognition of a constitutive 7-nucleotide mini-exon from the troponin I gene that resides quite close to the adjacent upstream exon. The mini-exon failed to be included in spliced RNA when placed in a heterologous gene unless accompanied by the upstream exon. The requirement for the upstream exon disappeared when the mini-exon was internally expanded, suggesting that the splice sites bordering the mini-exon are compatible with those of other constitutive vertebrate exons and that the small size of the exon impaired inclusion. Mutation of the 5' splice site of the natural upstream exon did not result in either exon skipping or activation of a cryptic 5' splice site, the normal vertebrate phenotypes for such mutants. Instead, a spliced RNA accumulated that still contained the upstream intron. In vitro, the mini-exon failed to assemble into spliceosome complexes unless either internally expanded or accompanied by the upstream exon. Thus, impaired usage of the mini-exon in vivo was accompanied by impaired recognition in vitro, and recognition of the mini-exon was facilitated by the presence of the upstream exon in vivo and in vitro. Cumulatively, the atypical in vivo and in vitro properties of the troponin exons suggest a mechanism for the recognition of this mini-exon in which initial recognition of an exon-intron-exon unit is followed by subsequent recognition of the intron.

In vivo recognition of a vertebrate mini-exon as an exon-intron-exon unit

Very small vertebrate exons are problematic for RNA splicing because of the proximity of their 3' and 5' splice sites. In this study, we investigated the recognition of a constitutive 7-nucleotide mini-exon from the troponin I gene that resides quite close to the adjacent upstream exon. The mini-exon failed to be included in spliced RNA when placed in a heterologous gene unless accompanied by the upstream exon. The requirement for the upstream exon disappeared when the mini-exon was internally expanded, suggesting that the splice sites bordering the mini-exon are compatible with those of other constitutive vertebrate exons and that the small size of the exon impaired inclusion. Mutation of the 5' splice site of the natural upstream exon did not result in either exon skipping or activation of a cryptic 5' splice site, the normal vertebrate phenotypes for such mutants. Instead, a spliced RNA accumulated that still contained the upstream intron. In vitro, the mini-exon failed to assemble into spliceosome complexes unless either internally expanded or accompanied by the upstream exon. Thus, impaired usage of the mini-exon in vivo was accompanied by impaired recognition in vitro, and recognition of the mini-exon was facilitated by the presence of the upstream exon in vivo and in vitro. Cumulatively, the atypical in vivo and in vitro properties of the troponin exons suggest a mechanism for the recognition of this mini-exon in which initial recognition of an exon-intron-exon unit is followed by subsequent recognition of the intron.

The role of exon sequences in splice site selection

Using mouse immunoglobulin mu (IgM) pre-mRNA as the model substrate for in vitro splicing, we have explored the role of exon sequences in splicing. We have found that deletion of the 5' portion of exon M2 of the IgM gene abolishes the splicing of its immediately upstream intron. Splicing was restored when a purine-rich sequence found within the deleted region was reinserted into the deletion construct. This M2 exon sequence was able to stimulate the splicing of a heterologous intron of the Drosophila doublesex pre-mRNA that contains a suboptimal 3' splice site sequence. These results show that the IgM M2 exon sequence functions as a splicing enhancer. We found that the assembly of the early splicing complex is stimulated by the M2 exon sequence. In vitro competition experiments show that this stimulatory effect is mediated by the interaction of some trans-acting factors. Our results suggest that the U1 snRNP is one such factor. We propose that recognition of an enhancer exon sequence by the components of splicing machinery plays a vital role in the selection of splice sites, not only for the IgM pre-mRNA but for other pre-mRNAs. We designate such a sequence as exon recognition sequence (ERS).

Occurrence of a 2-bp (AT) deletion allele and a nonsense (G-to-T) mutant allele at the E2 (DBT) locus of six patients with maple syrup urine disease: multiple-exon skipping as a secondary effect of the mutations

We have identified two novel mutant alleles in the transacylase (E2) gene of the human branched-chain alpha-keto acid dehydrogenase (BCKAD) complex in 6 of 38 patients with maple syrup urine disease (MSUD). One mutation, a 2-bp (AT) deletion in exon 2 of the E2 gene, causes a frameshift downstream of residue (-26) in the mitochondrial targeting presequence. The second mutation, a G-to-T transversion in exon 6 of the E2 gene, produces a premature stop codon at Glu-163 (E163*). Transfection of constructs harboring the E163* mutation into an E2-deficient MSUD cell line produced a truncated E2 subunit. However, this mutant E2 chain is unable to assemble into a 24-mer cubic structure and is degraded in the cell. The 2-bp (AT) deletion and the E163* mutant alleles occur in either the homozygous or compound-heterozygous state in the 6 of 38 unrelated MSUD patients studied. Moreover, an array of precise single- and multiple-exon deletions were observed in many amplified E2 mutant cDNAs. The latter results appear to represent secondary effects on RNA processing that are associated with the MSUD mutations at the E2 locus.

The skipping of constitutive exons in vivo induced by nonsense mutations

Nonsense mutations create a premature signal for the termination of translation of messenger RNA. Such mutations have been observed to cause a severe reduction in the amount of mutant allele transcript or to generate a peptide truncated at the carboxyl end. Analysis of fibrillin transcript from a patient with Marfan syndrome revealed the skipping of a constitutive exon containing a nonsense mutation. Similar results were observed for two nonsense mutations in the gene encoding ornithine delta-aminotransferase from patients with gyrate atrophy. All genomic DNA sequences flanking these exons that are known to influence RNA splicing were unaltered, which suggests that nonsense mutations can alter splice site selection in vivo.

The mutational spectrum of single base-pair substitutions in mRNA splice junctions of human genes: causes and consequences

A total of 101 different examples of point mutations, which lie in the vicinity of mRNA splice junctions, and which have been held to be responsible for a human genetic disease by altering the accuracy of efficiency of mRNA splicing, have been collated. These data comprise 62 mutations at 5' splice sites, 26 at 3' splice sites and 13 that result in the creation of novel splice sites. It is estimated that up to 15% of all point mutations causing human genetic disease result in an mRNA splicing defect. Of the 5' splice site mutations, 60% involved the invariant GT dinucleotide; mutations were found to be non-randomly distributed with an excess over expectation at positions +1 and +2, and apparent deficiencies at positions -1 and -2. Of the 3' splice site mutations, 87% involved the invariant AG dinucleotide; an excess of mutations over expectation was noted at position -2. This non-randomness of mutation reflects the evolutionary conservation apparent in splice site consensus sequences drawn up previously from primate genes, and is most probably attributable to detection bias resulting from the differing phenotypic severity of specific lesions. The spectrum of point mutations was also drastically skewed: purines were significantly over-represented as substituting nucleotides, perhaps because of steric hindrance (e.g. in U1 snRNA binding at 5' splice sites). Furthermore, splice sites affected by point mutations resulting in human genetic disease were markedly different from the splice site consensus sequences. When similarity was quantified by a 'consensus value', both extremely low and extremely high values were notably absent from the wild-type sequences of the mutated splice sites. Splice sites of intermediate similarity to the consensus sequence may thus be more prone to the deleterious effects of mutation. Regarding the phenotypic effects of mutations on mRNA splicing, exon skipping occurred more frequently than cryptic splice site usage. Evidence is presented that indicates that, at least for 5' splice site mutations, cryptic splice site usage is favoured under conditions where (1) a number of such sites are present in the immediate vicinity and (2) these sites exhibit sufficient homology to the splice site consensus sequence for them to be able to compete successfully with the mutated splice site. The novel concept of a "potential for cryptic splice site usage" value was introduced in order to quantify these characteristics, and to predict the relative proportion of exon skipping vs cryptic splice site utilization consequent to the introduction of a mutation at a normal splice site.

Exon sequences distant from the splice junction are required for efficient self-splicing of the Tetrahymena IVS

The presence of a natural rRNA secondary structure element immediately preceding the 5' splice site of the Tetrahymena IVS can inhibit self-splicing by competing with base pairing between the 5' exon and the guide sequence of the IVS (P1). Formation of this alternative hairpin is preferred in short precursor RNAs, and results in loss of G-addition to the 5' splice site. Pre-rRNAs which contain longer exons of ribosomal sequence, however, splice rapidly. As many as 146 nucleotides of the 5' exon and 86 nucleotides of the 3' exon are required for efficient self-splicing of Tetrahymena precursors. The presence of nucleotides distant from the 5' splice site apparently alters the equilibrium between the alternative hairpins, and promotes formation of active precursors. This effect is dependent on the specific sequences of the ribosomal pre-RNA, since point mutations within this region reduce the rate of splicing as much as 50-fold. This system provides an opportunity to study the way in which long-range interactions can influence splice site selection in a highly structured RNA.

Mutations which alter splicing in the human hypoxanthine-guanine phosphoribosyltransferase gene

A large proportion of mutations at the human hprt locus result in aberrant splicing of the hprt mRNA. We have been able to relate the mutation to the splicing abnormality in 30 of these mutants. Mutations at the splice acceptor sites of introns 4, 6 and 7 result in splicing out of the whole of the downstream exons, whereas in introns 1, 7 or 8 a cryptic site in the downstream exon can be used. Mutations in the donor site of introns 1 and 5 result in the utilisation of cryptic sites further downstream, whereas in the other introns, the upstream exons are spliced out. Our most unexpected findings were mutations in the middle of exons 3 and 8 which resulted in splicing out of these exons in part of the mRNA populations. Our results have enabled us to assess current models of mRNA splicing. They emphasize the importance of the polypyrimidine tract in splice acceptor sites, they support the role of the exon as the unit of assembly for splicing, and they are consistent with a model proposing a stem-loop structure for exon 8 in the hprt mRNA.

Alternative splicing of beta-tropomyosin pre-mRNA: cis-acting elements and cellular factors that block the use of a skeletal muscle exon in nonmuscle cells

The rat beta-tropomyosin (beta-TM) gene encodes both skeletal muscle beta-TM and fibroblast TM-1 by an alternative RNA-splicing mechanism. This gene contains 11 exons. Exons 1-5, 8, and 9 are common to all mRNAs expressed from the gene. Exons 6 and 11 are used in fibroblasts as well as smooth muscle cells, whereas exons 7 and 10 are used in skeletal muscle cells. In this study we have carried out an extensive mutational analysis to identify cis-acting elements that block the use of the skeletal muscle-specific exon 7 in nonmuscle cells. These studies localize the critical elements for regulated alternative splicing to sequences within exon 7 and the adjacent upstream intron. In addition, mutations that inactivate the 5'- or 3'-splice sites of exon 6 do not result in the use of the skeletal muscle-specific exon 7 in nonmuscle cells, suggesting that splice-site selection in vivo is not regulated by a simple cis-acting competition mechanism but, rather, by a mechanism that inhibits the use of exon 7 in certain cellular environments. In support of this hypothesis we have identified sequence-specific RNA-binding proteins in HeLa cell nuclear extracts using native gel electrophoresis and binding competition assays. Mutations in the pre-mRNA that result in the use of the skeletal muscle exon in vivo also disrupt the binding of these proteins to the RNA in vitro. We propose that the binding of these proteins to the pre-mRNA is involved in regulated alternative splicing and that this interaction is required for blocking the use of the skeletal muscle exon in nonmuscle cells.

Identification of a specific exon sequence that is a major determinant in the selection between a natural and a cryptic 5' splice site

The first intron of the early region 3 from adenovirus type 2 contains a cryptic 5' splice site, Dcr1, 74 nucleotides downstream from the natural site D1. The cryptic site can be activated when the natural site is inactivated by mutagenesis. To investigate the basis for selection between a natural and a cryptic 5' splice site, we searched for cis-acting elements responsible for the exclusive selection of the natural site. We show that both the relative intrinsic strength of the sites and the sequence context affect the selection. A 120-nucleotide segment located at the 3' end of exon 1 enhances splicing at the proximal site D1; in its absence the two sites are used according to their strength. Thus, three cis-acting elements are involved in the silencing of the cryptic site: the sequence of D1, the sequence of Dcr1, and an upstream exonic sequence. We show that the exonic element folds, in solution, into a 113-nucleotide-long stem-loop structure. We propose that this potential stem-loop structure which is located 6 nucleotides upstream of the exon 1-intron junction is responsible for the preferential use of the natural 5' splice site.

Identification of a specific exon sequence that is a major determinant in the selection between a natural and a cryptic 5' splice site

The first intron of the early region 3 from adenovirus type 2 contains a cryptic 5' splice site, Dcr1, 74 nucleotides downstream from the natural site D1. The cryptic site can be activated when the natural site is inactivated by mutagenesis. To investigate the basis for selection between a natural and a cryptic 5' splice site, we searched for cis-acting elements responsible for the exclusive selection of the natural site. We show that both the relative intrinsic strength of the sites and the sequence context affect the selection. A 120-nucleotide segment located at the 3' end of exon 1 enhances splicing at the proximal site D1; in its absence the two sites are used according to their strength. Thus, three cis-acting elements are involved in the silencing of the cryptic site: the sequence of D1, the sequence of Dcr1, and an upstream exonic sequence. We show that the exonic element folds, in solution, into a 113-nucleotide-long stem-loop structure. We propose that this potential stem-loop structure which is located 6 nucleotides upstream of the exon 1-intron junction is responsible for the preferential use of the natural 5' splice site.

Does steric interference between splice sites block the splicing of a short c-src neuron-specific exon in non-neuronal cells?

The neuron-specific splicing of the mouse c-src N1 exon was analyzed. Model src genes, transiently expressed in HeLa and LA-N-5 neuroblastoma cells, were assayed for the insertion of the 18-nucleotide neuron-specific N1 exon into their product mRNA. The normal clone fails to use this exon in HeLa cells but inserts the exon into 50% of the mature mRNA in LA-N-5 cells. When the exon and flanking intron sequences are placed between two adenovirus exons, the N1 exon is still only inserted in the neural cells. Thus, the neural specificity is a property of the exon itself and its immediate flanking sequences. Simply extending the length of the N1 exon to 109 nucleotides allows its efficient use in HeLa cells, implying that the exon is normally skipped because it is too short to allow spliceosomes to assemble at both ends simultaneously. This model predicts that exclusion of the exon should be sensitive to proteins or mutations that alter the relative strength of the flanking splice sites. Mutations that change these splice sites support this hypothesis.

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.

Molecular analysis of ultraviolet-induced mutations in a xeroderma pigmentosum cell line

We have isolated and characterized 47 ultraviolet light-induced hprt mutants from a simian virus 40-transformed excision-repair-deficient xeroderma pigmentosum cell line (complementation group A). Twenty-one independent mutations were found, of which the majority were point mutations. Eleven of these were identified as base changes, nine of which could be attributed to ultraviolet damage on the transcribed DNA strand. Both transitions and transversions were found among the single base changes. A large proportion of the mutations (13/21) resulted in aberrant splicing of the hprt gene, suggesting that the target size for mutations resulting in aberrant splicing must be quite large. A small number of spontaneous mutations were identified, most of which were large deletions. Our data provide a spectrum for the intrinsic mutations resulting from ultraviolet damage in human cells in the absence of repair.

Influences of separation and adjacent sequences on the use of alternative 5' splice sites

Single nucleotide changes to the sequence between two alternative 5' splice sites, separated by 25 nucleotides in a beta-globin gene derivative, caused substantial shifts in pre-mRNA splicing preferences, both in vivo and in vitro. An activating sequence for splicing was located. Models for the recognition by U1 small nuclear ribonucleoproteins (snRNPs) of competing 5' splice sites were tested by altering the distance separating the two sites. Use of the upstream splice site declined sharply when it was separated from the downstream (natural) site by distances of 40 nucleotides or more. This effect was reversed in vivo, but not in vitro, by altering the upstream sequence to that of a consensus 5' splice site sequence. Dilution of an extract used for splicing in vitro shifted preferences when the sites were close towards the downstream site. We conclude that the mechanism of selection depends on the distance apart of the potential splice sites and that with close sites steric interference between factors bound to both sites may impede splicing and affect splicing preferences.

A protein factor, ASF, controls cell-specific alternative splicing of SV40 early pre-mRNA in vitro

SV40 early pre-mRNA is alternatively spliced by utilization of two different 5' splice sites and a shared 3' splice site to produce large T and small t mRNAs. The ratio of small t to large T mRNAs produced in human embryonic kidney 293 cells is 10- to 20-fold greater than in other mammalian cells, suggesting the existence of a 293 cell-specific factor that modulates alternative splicing. Here we show that nuclear extracts from 293 cells give rise to significantly more small t splicing than do extracts from HeLa cells. Using an in vitro complementation assay, we have characterized and extensively purified a factor from 293 extracts that brings about striking increases in small t splicing with concomitant decreases in large T splicing. The factor is heat sensitive and micrococcal nuclease resistant, suggesting that it is a protein lacking an accessible RNA component. Purification of the alternative splicing factor indicates that the activity is contained in one of several possibly related polypeptides of 30-35 kd.

Control of retroviral RNA splicing through maintenance of suboptimal processing signals

The full-length retroviral transcript serves as genomic RNA for progeny virions, as an mRNA for structural proteins and enzymes, and as a pre-mRNA substrate for splicing that yields subgenomic mRNAs that encode other essential proteins. Thus, RNA splicing to form subgenomic mRNAs must be incomplete or regulated in order to preserve some of the full-length transcripts. We have used the avian sarcoma virus system to delineate the viral functions that are required in the regulation of the splicing event that forms the envelope glycoprotein (env) subgenomic mRNA. We observed previously that a specific insertion mutation just 5' of the env splice acceptor site resulted in nearly complete splicing to form env mRNA and a concomitant replication defect which is presumably due to a deficit of the full-length transcript. Replication-competent pseudorevertants contained second-site mutations that restored splicing control, and these mapped either just upstream or downstream of the env splice acceptor site. In this report, we show that splicing control at this site does not require expression of any known viral replication protein(s), nor does it appear to require the viral splice donor site. From these results and analysis of additional splicing mutations obtained by in vivo selection, we conclude that splicing is controlled through the maintenance of suboptimal cis-acting signals in the viral RNA that alter the efficiency of recognition by the cellular splicing machinery.

Sequences upstream of AAUAAA influence poly(A) site selection in a complex transcription unit

The adenovirus major late transcription unit (MLTU) encodes five colinear mRNA families, L1 through L5, each distinguished by a unique poly(A) site. Site selection is regulated during the course of infection, predominating early at the L1 site and late at the L2 through L5 sites. Two general mechanisms can be invoked to explain predominant usage of the L1 site early in infection. MLTU site selection may proceed in a first-come, first-serve manner whereby the L1 site is used most frequently because it is closest to the promoter. Alternatively, specific sequences flanking the L1 site may control predominant L1 site usage in a position-independent manner. To distinguish between these mechanisms, we constructed deletions in the L1 flanking sequences and inserted the mutated sites into either simple transcription units or mini-MLTUs encoding two poly(A) sites. The pattern of site selection for each construct was then quantitated by S1 nuclease analysis after transfection into 293 cells. The results indicated that L1 sequences upstream of AAUAAA define a novel selector element that can cause predominant L1 site usage at either position of a tandem transcription unit. The element did not significantly affect the stability or nucleocytoplasmic transport of L1 transcripts and was not required for efficient 3'-end processing in simple transcription units. Predominant L1 site usage required physical linkage of the processing signals and was independent of the major late promoter.

Exon definition may facilitate splice site selection in RNAs with multiple exons

Interactions at the 3' end of the intron initiate spliceosome assembly and splice site selection in vertebrate pre-mRNAs. Multiple factors, including U1 small nuclear ribonucleoproteins (snRNPs), are involved in initial recognition at the 3' end of the intron. Experiments were designed to test the possibility that U1 snRNP interaction at the 3' end of the intron during early assembly functions to recognize and define the downstream exon and its resident 5' splice site. Splicing precursor RNAs constructed to have elongated second exons lacking 5' splice sites were deficient in spliceosome assembly and splicing activity in vitro. Similar substrates including a 5' splice site at the end of exon 2 assembled and spliced normally as long as the second exon was less than 300 nucleotides long. U2 snRNPs were required for protection of the 5' splice site terminating exon 2, suggesting direct communication during early assembly between factors binding the 3' and 5' splice sites bordering an exon. We suggest that exons are recognized and defined as units during early assembly by binding of factors to the 3' end of the intron, followed by a search for a downstream 5' splice site. In this view, only the presence of both a 3' and a 5' splice site in the correct orientation and within 300 nucleotides of one another will stable exon complexes be formed. Concerted recognition of exons may help explain the 300-nucleotide-length maximum of vertebrate internal exons, the mechanism whereby the splicing machinery ignores cryptic sites within introns, the mechanism whereby exon skipping is normally avoided, and the phenotypes of 5' splice site mutations that inhibit splicing of neighboring introns.

Control of retroviral RNA splicing through maintenance of suboptimal processing signals

The full-length retroviral transcript serves as genomic RNA for progeny virions, as an mRNA for structural proteins and enzymes, and as a pre-mRNA substrate for splicing that yields subgenomic mRNAs that encode other essential proteins. Thus, RNA splicing to form subgenomic mRNAs must be incomplete or regulated in order to preserve some of the full-length transcripts. We have used the avian sarcoma virus system to delineate the viral functions that are required in the regulation of the splicing event that forms the envelope glycoprotein (env) subgenomic mRNA. We observed previously that a specific insertion mutation just 5' of the env splice acceptor site resulted in nearly complete splicing to form env mRNA and a concomitant replication defect which is presumably due to a deficit of the full-length transcript. Replication-competent pseudorevertants contained second-site mutations that restored splicing control, and these mapped either just upstream or downstream of the env splice acceptor site. In this report, we show that splicing control at this site does not require expression of any known viral replication protein(s), nor does it appear to require the viral splice donor site. From these results and analysis of additional splicing mutations obtained by in vivo selection, we conclude that splicing is controlled through the maintenance of suboptimal cis-acting signals in the viral RNA that alter the efficiency of recognition by the cellular splicing machinery.

Identification of two distinct intron elements involved in alternative splicing of beta-tropomyosin pre-mRNA

The rat beta-tropomyosin gene encodes two isoforms, termed skeletal muscle beta-tropomyosin and fibroblast last tropomyosim 1 (TM-1), via an alternative RNA processing mechanism. The gene contains 11 exons. Exons 1-5 and exons 8 and 9 are common to all mRNAs expressed from the gene. Exons 6 and 11 are used in fibroblasts, as well as smooth muscle, whereas exons 7 and 10 are used only in skeletal muscle. In the present studies we focused on the mutually exclusive internal alternative splice choice involving exon 6 (fibroblast-type splice) and exon 7 (skeletal muscle-type splice). We have identified two distinct elements in the intron, upstream of exon 7, involved in splice site selection. The first element is comprised of a polypyrimidine tract located 89-143 nucleotides upstream of the 3' splice site, which specifies the location of the lariat branchpoints used, 144-153 nucleotides upstream of exon 7. The 3' splice site AG dinucleotide has no role in selection of these branchpoints. The second element is comprised of intron sequences located between the polypyrimidine tract and the 3' splice site of exon 7. It contains an important determinant in alternative splice site selection, because deletion of these sequences results in the use of the skeletal muscle-specific exon in nonmuscle cells. We propose that the use of lariat branchpoints located far upstream from a 3' splice site may be a general feature of some alternatively excised introns, reflecting the presence of regulatory sequences located between the lariat branch site and the 3' splice site. The data also indicate that alternative splicing of the rat beta-tropomyosin gene is regulated by a somewhat different mechanism from that described for rat alpha-tropomyosin gene and the transformer-2 gene of Drosophila melanogaster.

Regulation of sex-specific RNA splicing at the Drosophila doublesex gene: cis-acting mutations in exon sequences alter sex-specific RNA splicing patterns

Sex-specific alternative RNA splicing of the doublesex (dsx) pre-mRNA results in sex-specific polypeptides that regulate both male and female somatic sexual differentiation in Drosophila melanogaster. We have molecularly characterized a class of dsx mutations that act in cis to disrupt the regulation of dsx RNA processing, causing the dsx pre-mRNA to be spliced in the male-specific pattern regardless of the chromosomal sex of the fly. These dsx mutations are associated with rearrangements in the female-specific exon just 3' to the female-specific splice acceptor. The mutations do not affect the female-specific splice sites or intron that are identical to wild-type sequences. These results indicate that sequences in the female-specific exon are important for the regulation of sex-specific RNA splicing, perhaps by acting as sites of interaction with trans-acting regulators. Furthermore, the data suggest that female-specific regulation of dsx RNA processing occurs by promoting the usage of the female splice acceptor site, rather than by repressing the usage of the alternative male-specific splice acceptor.

Exon definition may facilitate splice site selection in RNAs with multiple exons

Interactions at the 3' end of the intron initiate spliceosome assembly and splice site selection in vertebrate pre-mRNAs. Multiple factors, including U1 small nuclear ribonucleoproteins (snRNPs), are involved in initial recognition at the 3' end of the intron. Experiments were designed to test the possibility that U1 snRNP interaction at the 3' end of the intron during early assembly functions to recognize and define the downstream exon and its resident 5' splice site. Splicing precursor RNAs constructed to have elongated second exons lacking 5' splice sites were deficient in spliceosome assembly and splicing activity in vitro. Similar substrates including a 5' splice site at the end of exon 2 assembled and spliced normally as long as the second exon was less than 300 nucleotides long. U2 snRNPs were required for protection of the 5' splice site terminating exon 2, suggesting direct communication during early assembly between factors binding the 3' and 5' splice sites bordering an exon. We suggest that exons are recognized and defined as units during early assembly by binding of factors to the 3' end of the intron, followed by a search for a downstream 5' splice site. In this view, only the presence of both a 3' and a 5' splice site in the correct orientation and within 300 nucleotides of one another will stable exon complexes be formed. Concerted recognition of exons may help explain the 300-nucleotide-length maximum of vertebrate internal exons, the mechanism whereby the splicing machinery ignores cryptic sites within introns, the mechanism whereby exon skipping is normally avoided, and the phenotypes of 5' splice site mutations that inhibit splicing of neighboring introns.

cis-acting elements and a trans-acting factor affecting alternative splicing of adenovirus L1 transcripts

Expression of the L1 region of adenovirus is temporally regulated by alternative splicing to yield two major RNAs encoding the 52- to 55-kilodalton (52-55K) and IIIa polypeptides. The distal acceptor site (IIIa) is utilized only during the late phase of infection, whereas the proximal site (52-55K) is used at both early and late times. Several parameters that might affect this alternative splicing were tested by using expression vectors carrying the L1 region or mutated versions of it. In the absence of a virus-encoded or -induced factor(s), only the 52-55K acceptor was used. Decreasing the distance between the donor and the IIIa acceptor had no effect. Removal of the 52-55K acceptor induced IIIa splicing slightly, implying competition between the two acceptors. Fusion of the IIIa exon to the 52-55K intron greatly enhanced splicing of the IIIa junction, suggesting that the IIIa exon does not contain sequences that inhibit splicing. Thus, the lack of splicing to the IIIa acceptor in the absence of a virus-encoded or -induced factor(s) is probably due to the absence of a favorable sequence and/or the presence of a negative element 5' of the IIIa splice junction, or both. The presence of several adenovirus gene products, including VA RNAs, the E2A DNA-binding protein, and the products of E1A and E1B genes, did not facilitate use of the IIIa acceptor. In contrast, the simian virus 40 early proteins, probably large T antigen, induced IIIa splicing. This result, together with those of earlier studies, suggest that T antigen plays a role in modulation of alternative RNA splicing.

Sequences upstream of AAUAAA influence poly(A) site selection in a complex transcription unit

The adenovirus major late transcription unit (MLTU) encodes five colinear mRNA families, L1 through L5, each distinguished by a unique poly(A) site. Site selection is regulated during the course of infection, predominating early at the L1 site and late at the L2 through L5 sites. Two general mechanisms can be invoked to explain predominant usage of the L1 site early in infection. MLTU site selection may proceed in a first-come, first-serve manner whereby the L1 site is used most frequently because it is closest to the promoter. Alternatively, specific sequences flanking the L1 site may control predominant L1 site usage in a position-independent manner. To distinguish between these mechanisms, we constructed deletions in the L1 flanking sequences and inserted the mutated sites into either simple transcription units or mini-MLTUs encoding two poly(A) sites. The pattern of site selection for each construct was then quantitated by S1 nuclease analysis after transfection into 293 cells. The results indicated that L1 sequences upstream of AAUAAA define a novel selector element that can cause predominant L1 site usage at either position of a tandem transcription unit. The element did not significantly affect the stability or nucleocytoplasmic transport of L1 transcripts and was not required for efficient 3'-end processing in simple transcription units. Predominant L1 site usage required physical linkage of the processing signals and was independent of the major late promoter.