Sorting out the complexity of SR protein functions

A purine-rich intronic element enhances alternative splicing of thyroid hormone receptor mRNA

“…Purine-rich elements are required for SEa2 activity SEa2 is extremely purine rich and contains multiple copies of two motifs that have been described in other splicing enhancers+ First, three uninterrupted stretches of purine residues (underlined in Fig+ 1A), that include repeats of the trinucleotide GAR, are important for splicing+ When these GAR repeats were altered by eight purine-purine substitutions, a 50% decrease in splicing activity was observed+ This result is consistent with other studies showing that tandem repeats of GAR are essential for enhancer activity (Xu et al+, 1993;Tanaka et al+, 1994;Humphrey et al+, 1995)+ A second type of purine-rich element in SEa2 consists of several consecutive G residues (overlined in Fig+ 1A)+ Such G tracts have been found within introns close to intron/exon junctions (Nussinov, 1988;McCullough & Berget, 1997, 2000+ Consecutive G residues are found in certain sequence motifs present in other intronic enhancers including two, GGGGCUG and (A/U)GGG (Sirand-Pugnet et al+, 1995;Carlo et al+, 1996;Carstens et al+, 1998) that are also present in SEa2+ Because four of the five G tracts within SEa2 are disrupted in SE80G/U, which lacks enhancer activity, these elements may also be essential for SEa2 function+ Further deletion and substitution mutations demonstrate that sequences other than the GAR repeats are important for enhancer function+ However, the G/A substitutions disrupt the GAR repeats while extending, not disrupting, two overlapping runs of three consecutive G residues in SEa2+ Thus, the GAR repeats themselves may be needed for full enhancer function+…”

“…Inhibition of TRa2 splicing by excess wild-type competitor but not mutant transcripts suggests that SEa2 function requires sequence-specific binding of transacting factors in HeLa nuclear extract+ RNA-protein crosslinking reveals that 55-kDa, 34-kDa, and 22-kDa proteins bind to SEa2 (Fig+ 5A)+ Both the 22-kDa and 34-kDa proteins appear to be SR proteins (Fig+ 5B)+ There is a single well-characterized SR protein of 20 kDa (SRp20) and at least four SR proteins that migrate at about 34 kDa (SF2/ASF, SC35, 9G8, or SRp30c; reviewed in Graveley, 2000)+ SF2/ASF is at least one of the proteins interacting with SEa2, as indicated by UV crosslinking, immunoprecipitation, and northwestern blot analysis (Figs+ 5B and 6A)+ Binding of the 22-kDa protein is competed by both the wild-type and SE80G/U competitor RNA (Fig+ 5A)+ Although this result may indicate nonspecific binding, the protein also crosslinks much less efficiently to SE60 than to SE80 (Fig+ 9A), suggesting that the 22 kDa protein may be associated with the 59 end of SE80, which is absent in SE60 and unchanged in SE80G/U+ SR proteins have been shown to mediate the activity of other purine-rich exonic splicing enhancers (reviewed in Graveley, 2000)+ Consistent with this role in splicing, enhancer sequences have been shown to contain consensus sequences recognized by a number of SR proteins (Liu et al+, 1998;Tacke & Manley, 1999)+ SE80 contains an unusually high density of potential SF2/ASF interacting sites (not shown) as determined by sequence analysis with a score matrix generated by selection for SF2/ASF-dependent enhancer sequences in vitro (Liu et al+, 1998)+ SEa2 also includes two GGACAA sequences that were identified as SF2/ASF-dependent enhancers in b-globin (Schaal & Maniatis, 1999)+ This may explain the inhibition of b-globin splicing in the presence of SEa2 competitor RNA (Fig+ 4)+ In addition, both the SE80G/A and SE80G/U mutations alter potential SF2/ASF interactions+ Thus, it is likely that SF2/ASF plays a role in mediating enhancer activity of SEa2+…”

“…Alternative mRNA processing is an important mechanism for generating multiple mRNAs and proteins from a single gene (reviewed in Smith & Valcárcel, 2000)+ In mammals, transcripts from the c-erbAa gene (NR1A1; Nuclear Receptors Committee, 1999) code for two major thyroid hormone receptor (TR) isoforms, TRa1 (NR1A1a) and TRa2 (NR1A1b), which are functionally antagonistic (for a review, see Lazar, 1993)+ TRa1 is an authentic thyroid hormone receptor that binds thyroid hormone (T3) and mediates its activity in many cells and tissues+ TRa2 is a variant receptor that lacks a functional hormone-binding site and antagonizes the ability of TRa1 to activate gene transcription in the presence of T3 (Koenig et al+, 1989;Lazar et al+, 1989a)+ Although both isoforms are expressed in most mammalian cells, the levels of each vary widely in different tissues and at different stages of development+ Thus, regulation of TRa alternative processing is important for determining cellular levels of TRa1 and TRa2 mRNAs, which, in turn, are critical for modulating the response to T3+ TRa1 and TRa2 mRNAs are distinguished by their distinct 39 terminal exons+ To produce TRa1 mRNA, TRa pre-mRNA is cleaved and polyadenylated at the penultimate exon (exon 9; Lazar et al+, 1989b; Laudet et al+, 1991)+ Alternatively, TRa2 mRNA results from use of a 59 splice site within exon 9 and ligation to an exon (exon 10) directly downstream of the TRa1-specific poly(A) site+ Thus, TRa2 lacks most of the TRa1 terminal exon but includes an alternative 39 exon+ Alternatively processed exons often are characterized by suboptimal splice signals that are recognized inefficiently in the absence of auxiliary sequences or transacting factors (for a review, see Berget, 1995;Smith & Valcárcel, 2000)+ In the case of TRa2 mRNA splicing, a single nucleotide deviation from the 59 splice site consensus sequence is responsible for weak splicing activity+ Mutation of the nonconsensus ϩ5C residue in the natural TRa2 59 splice site to a consensus ϩ5G results in constitutive processing of TRa2 mRNA with complete exclusion of the TRa1 mRNA processing (Hastings et al+, 2000)+ Efficient processing of such suboptimal splice sites often requires cis-acting sequence elements, referred to as splicing enhancers, that facilitate exon recognition through interactions with the splicing machinery (for a review, see Hertel et al+, 1997)+ Many splicing enhancers identified to date are purine-rich sequences located within downstream exons+ These enhancers interact with members of the serine-arginine (SR)-rich family of splicing proteins and promote the use of weak 39 splice sites in the upstream intron (for a review, see Graveley, 2000)+ The ability of individual SR proteins to recognize specific enhancer sequences and influence splicing in a dose-dependent manner may be an important mechanism for regulating alternative processing+…”

A regulator of Dscam mutually exclusive splicing fidelity

“…Given the well-established antagonism between hnRNP proteins and SR proteins 21,22 , the simplest explanation for these data is that when hrp36 is bound to the exon 6 variants, it prevents SR proteins from activating their splicing, perhaps by preventing the SR proteins from binding to the exons. Alternatively, as SR proteins are known to have exonindependent functions 23 , it is also possible that B52 functions without binding to the exons.…”

High-affinity hnRNP A1 binding sites and duplex-forming inverted repeats have similar effects on 5??? splice site selection in support of a common looping out and repression mechanism

“…The fidelity of pre-messenger RNA splicing relies upon the appropriate selection of specific splicing partners amid a pool of sequences that resemble splice sites+ The process of splice site recognition and pairing, however, has to remain sufficiently flexible to accommodate the generation of alternatively spliced variants+ The control of splice site utilization can operate during the step-wise process of spliceosome assembly, and each step, in principle, can serve as a regulatory point+ Although recent progress has led to the identification of factors that either promote or inhibit the use of a splice site, their mechanisms of action remain poorly understood+ In one of the best-documented cases, the assembly of a soma-specific complex containing PSI, hrp48, and U1 snRNP prevents the binding of U1 snRNP to the authentic 59 splice site in the transposase premRNA of the P-element in Drosophila (Adams et al+, 1997)+ In contrast, the binding of TIA-1 to intron sequences facilitates 59 splice site recognition by U1 snRNP on some pre-mRNAs (Del Gatto-Konczak et al+, 2000;Forch et al+, 2000;Le Guiner et al+, 2001)+ In the case of the adenovirus L1 splicing unit, the binding of ASF/SF2 immediately upstream of the branch site flanking the IIIa exon sterically prevents U2 snRNP binding and hence represses splicing (Kanopka et al+, 1996)+ Likewise, hnRNP I/PTB has been found to antagonize the binding of U2AF to some 39 splice sites (Valcárcel & Gebauer, 1997)+ In other situations, exon enhancer elements bound by SR proteins can promote inclusion of an exon by stimulating the interaction of U2AF and/or U2 snRNP with the 39 splice site region (reviewed in Tacke & Manley, 1999;Graveley, 2000)+ Although some control elements do not affect the initial recognition of splicing signals (Gontarek et al+, 1993;Chou et al+, 2000), the exact mechanism by which many elements and regulatory factors affect splice site selection is currently unknown+…”