HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS		QUICK SEARCH:   [advanced] Author: Keyword(s): Year:  Vol:  Page:

J. Biol. Chem., Vol. 276, Issue 52, 48908-48914, December 28, 2001

This Article Abstract Full Text (PDF) All Versions of this Article: 276/52/48908    most recent M103967200v1 Submit a Letter to Editor Alert me when this article is cited Alert me when eLetters are posted Alert me if a correction is posted Citation Map Services Email this article to a friend Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Request Permissions Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Cowper, A. E. Articles by Screaton, G. R. Search for Related Content PubMed PubMed Citation Articles by Cowper, A. E. Articles by Screaton, G. R. Social Bookmarking                      What's this?

Serine-Arginine (SR) Protein-like Factors That Antagonize Authentic SR Proteins and Regulate Alternative Splicing^*

Alison E. Cowper, Javier F. Cáceres^§, Akila Mayeda^¶, and Gavin R. Screaton

From the  Human Immunology Unit, Weatherall Institute of Molecular Medicine, John Radcliffe Hospital, Headington, Oxford OX3 9DU, United Kingdom, ^§ Medical Research Council Human Genetics Unit, Western General Hospital Edinburgh, EH4 2XU Scotland, United Kingdom, and ^¶ Department of Biochemistry and Molecular Biology, University of Miami School of Medicine, Miami, Florida 33136-1019

Received for publication, May 3, 2001, and in revised form, September 27, 2001

	ABSTRACT

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS AND DISCUSSION REFERENCES

We have characterized two RNA-binding proteins, of apparent molecular masses of ~40 and 35 kDa, which possess a single N-terminal RNA-recognition motif (RRM) followed by a C-terminal domain rich in serine-arginine dipeptides. Their primary structures resemble the single-RRM serine-arginine (SR) protein, SC35; however their functional effects are quite distinctive. The 40-kDa protein cannot complement SR protein-deficient HeLa cell S100 extract and showed a dominant negative effect in vitro against the authentic SR proteins, SF2/ASF and SC35. Interestingly, the 40- and 35-kDa proteins antagonize SR proteins and activate the most distal alternative 5' splice site of adenovirus E1A pre-mRNA in vivo, an activity that is similar to that characterized previously for the heterogeneous nuclear ribonucleoprotein particles A/B group of proteins. A series of recombinant chimeric proteins consisting of domains from these proteins and SC35 in various combinations showed that the RRM, but not the C-terminal domain rich in serine-arginine dipeptides, has a dominant role in this activity. Because of the similarity to SR proteins we have named these proteins SRrp40 and SRrp35, respectively, for SR-repressor proteins of ~40 and ~35 kDa. Both factors show tissue- and cell type-specific patterns of expression. We propose that these two proteins are SR protein-like alternative splicing regulators that antagonize authentic SR proteins in the modulation of alternative 5' splice site choice.

	INTRODUCTION

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS AND DISCUSSION REFERENCES

Pre-mRNA splicing is a fundamental biological process involved in the expression of most genes in higher eukaryotes. Splicing takes place in a large protein·RNA complex known as the spliceosome, which includes the small nuclear ribonucleoprotein particles (snRNPs)¹ U1, U2, U4/U6, and U5, together with a large number of non-snRNP protein factors. The snRNPs and non-snRNP factors are involved in forming sophisticated protein-protein and protein-RNA contacts within the spliceosome for the catalytic activity (reviewed in Refs. 1-3). Many non-snRNP factors have been described, but the best characterized of these are the serine/arginine-rich protein (SR protein) family (reviewed in Refs. 4-6). SR proteins contain either one or two N-terminal RNA-recognition motifs and a C-terminal RS domain of varying length rich in arginine (R)/serine (S) dipeptides. The prototype for this family, SF2/ASF, was initially purified and cloned by two groups using in vitro splicing to look for factors involved in constitutive splicing and the regulation of alternative splicing, (7-10). Since then a number of other members of the SR protein family have been cloned by protein purification, yeast two-hybrid screening, or degenerate PCR approaches. To date 10 SR proteins have been characterized in humans, SRp20, 9G8, SF2/ASF (SRp30a), SC35 (SRp30b), SRp30c, SRp40, SRp46, p54, SRp55, and SRp75 (9-17). A class of related RS domain-containing proteins has recently been described; however, their characters and functions are poorly understood (reviewed in Refs. 4 and 18).

SR proteins have been shown to play a role in constitutive and alternative splicing reactions using both in vitro and in vivo systems. The precise mechanisms whereby the SR proteins mediate these effects have been the subject of much research and appear highly complex and multifactorial (reviewed in Ref. 18). In this respect, SR proteins have been demonstrated to bind exonic and intronic enhancer sequences to form the early E or commitment complex. SR proteins facilitate the recruitment of U1-snRNP to the 5' splice site (19) and of U2AF^65/35 to the 3' site through RS domain-RS domain interactions (20-22). In addition to these effects, SR proteins have also been shown to act at later stages of splicing by recruiting the U4/U6·U5 tri-snRNP complex and may also participate in the second step of splicing (23-25).

According to the initial characterization of SF2 (26), authentic members of the SR protein family have been identified by their ability individually to complement splicing in an S100 extract deficient in SR proteins. Another early observation was that SR proteins tended to favor the utilization of proximal alternative 5' splice sites (7, 8). However, it is now well documented that individual SR proteins can show unique specificities in both constitutive and alternative splicing in vitro and in vivo (14, 27-30). This observation is consistent with the fact that a single organism possesses multiple SR proteins and that sequences of individual proteins including the RS domains are highly conserved between different species. In addition, knockout or mutations of individual proteins have been shown to have deleterious effects. One such example is the deletion of SF2/ASF by homologous recombination in the chicken B cell line DT40. This led to cell cycle arrest and death and could be complemented by human SF2/ASF (31). Interestingly complementation with a number of mutant SR proteins, which show full activity in vitro, failed to rescue the knockout DT40 cell lines. Other SR protein knockout experiments have also shown deleterious phenotypes in mouse, fruit fly (Drosophila melanogaster), and nematode (Caenorhabditis elegans), although in the latter case there is functional redundancy of SR proteins other than the SF2/ASF orthologue (32-35).

In this paper we report the characterization of two RNA-binding proteins of apparent molecular masses of ~40 and ~35 kDa, whose primary structures are significantly homologous to SC35. We demonstrate that these two proteins have properties that are quite distinctive from other well characterized authentic SR proteins.

	EXPERIMENTAL PROCEDURES

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS AND DISCUSSION REFERENCES

DNA Cloning and Northern Blotting Analyses-- The expressed sequencing tag clones containing SRrp40 (accession number A167835) and SRrp35 (accession number AW270156) were obtained from the IMAGE consortium and fully sequenced. For expression of recombinant protein the DNA sequences encoding amino acids 3-263 of SRrp40 were subcloned into the BamHI site of baculovirus transfer vector pAcG2T (PharMingen) in frame with, and C-terminal to, GST. Recombinant baculovirus was produced by co-transfecting Sf9 cells with the SRrp40 transfer construct and BacPAK6 viral DNA (CLONTECH). Recombinant virus was plaque-purified, and a high titer virus stock was used to infect 1 liter of cells in serum-free media. Following lysis, SRrp40 was purified over glutathione-Sepharose beads. GST-tagged SF2/ASF and GST-tagged SC35 recombinant proteins expressed in baculovirus were prepared as described (30).

For mammalian expression sequences encoding SRrp40 amino acids 3-262 and SRrp35 amino acids 3-261 were subcloned into the pCGNHCF_FL (g10) plasmid (36) to form an N-terminal fusion with the 11-amino acid epitope tag (MASMTGGQQMG) from T7 phage gene 10 protein. Multiple human tissue Northern blots (CLONTECH) were probed with ³²P-labeled random hexamer DNA encompassing the SRrp40 coding region. The chimeric protein constructs were made by fusing DNA sequences encoding RRM and the RS domains of the indicated proteins. Amino acid sequences used in these fusion proteins were SRrp40 RRM amino acids 3-119, SC35 RRM amino acids 1-115, SRrp40 RS amino acids 119-262, SRrp35 RS amino acids 114-261, and SC35 RS amino acids 116-221. These sequences were used to construct four chimeric proteins; SC35RRM/SRrp40RS, SRrp40RRM/SC35RS, SRrp40RRM/SRrp35RS, and SC35RRM/SRrp35RS amino acids 108-119 have been deleted in the construct SRrp40YRD. To ascertain molecular masses of the new proteins, lysates of HeLa cells transfected with epitope-tagged proteins by Western blot were probed with anti-T7 tag monoclonal antibody (Novagen) and revealed by chemiluminescense (ECL kit; Amersham Biosciences, Inc.).

Transfection and in Vivo Splicing Analyses-- HeLa cells were transfected using Lipofectin (Invitrogen). Briefly, cells were grown to 50% confluency in 6-well plates and then incubated with 3 µg of SR protein-encoding plasmid and 1 µg of adenovirus E1A reporter plasmid. The medium was changed after 6 h, and following 48 h of incubation total RNA was isolated using Tri-Reagent (Sigma) and treated with RQ DNase I (Promega). The corresponding cDNA was prepared by reverse transcription-PCR (25-30 cycles) with oligo(dT) primer, Taq polymerase, and the adenovirus E1A primer pair (forward, ctcttgagtgccagcgagtagag; reverse, gtcttgcaggctccggttctgg). Amplified products were resolved by non-denaturing polyacrylamide gel electrophoresis, transferred to nylon membranes, and probed with a ³²P-5'end-labeled internal E1A primer (forward, gttttctcctccgagccgctccga) followed by autoradiography. Autoradiograms were scanned by densitometry, and the relative amounts of 13, 12, and 9 S products were expressed as a ratio to each other.

Cell Culture, Heterokaryon Assays, and Immunofluorescence Microscopy Analyses-- The details of these assays have been described before (37). Briefly, HeLa cells were grown in Dulbecco's modified Eagle's medium supplemented with 10% fetal calf serum and transfected with 1 µmg of plasmid DNA per 60-mm dish of 60-75% confluent cells, in the presence of 20 µg of Lipofectin (Invitrogen) (38). For incubations of HeLa cells with transcriptional inhibitors, actinomycin D was used at 5 µg/ml, and cycloheximide was added at 20 µg/ml. Cells were fixed with 4% paraformaldehyde in phosphate-buffered saline for 15-30 min at room temperature, followed by incubation for 10 min in 0.2% Triton X-100. The fixed cells were incubated for 1 h at room temperature with 1:1000 anti-T7 monoclonal antibody (Novagen). The cells were washed with phosphate-buffered saline and incubated for 1 h at room temperature with 1:200 fluorescein-conjugated goat anti-mouse IgG (Cappel laboratories).

For transient transfections involving interspecies heterokaryons, because of the need for higher transfection efficiency, HeLa cells were transfected by electroporation using 10 µg of plasmid DNA per 60-mm dish of 90% confluent cells in the presence of 15 µg of carrier DNA and seeded on coverslips, followed by co-incubation with an excess number of untransfected mouse NIH 3T3 cells for 3 h in the presence of 50 µg/ml cycloheximide. The concentration of cycloheximide was then increased to 100 µg/ml, and the cells were incubated for an additional 30 min prior to fusion. Cell fusions were done with polyethylene glycol as described (39), and the heterokaryons were further incubated for 1 h in media containing 100 µg/ml cycloheximide prior to fixation. Immunofluorescence with the anti-T7 monoclonal antibody was performed as described above, except that 0.5 µg/ml 4,6-diamidino-2-phenylindole (DAPI) was included to reveal murine versus human nuclei. Samples were observed on a Axioskop microscope (Zeiss), and images were acquired with a Photometrics CH250 cooled CCD camera using Smartcapture extensions (Digital Scientific) within Spectrum software (IP Lab). The immunofluorescence figures show representative data. Each experiment was reproduced in multiple independent transfections, and the cells shown are representative of the large effects observed under each set of conditions.

In Vitro Splicing Assays-- Plasmids encoding b-globin (exons 1 and 2), HIV-tat (exons 2 and 3), and IgM (exons C3 and C4) mini-genes were used as templates for in vitro transcription as described previously (27). ³²P-Labeled pre-mRNA substrates were prepared by run-off in vitro transcription with SP6 RNA polymerase as described (40). HeLa cell nuclear and S100 extracts were prepared as described (41). In vitro splicing reactions were carried out in 25 µl with indicated volume of nuclear extract or S100 extract, indicated amounts of GST fusion proteins (GST-SRrp40, GST-SF2/ASF, and/or GST-SC35), and 20 fmol of ³²P-labeled pre-mRNA substrate, followed by incubation at 30 °C for 2 h as described (40). RNA products were analyzed by electrophoresis on a 5.5 (see Fig. 5) or 9% (see Fig. 6A) polyacrylamide/7 M urea gel followed by autoradiography with an intensifying screen at 70 °C.

Native Gel Analysis of Spliceosomal Complexes-- In vitro splicing reactions (25 µl) were carried out in 8 µl of S100 extract without or with GST-SRrp40 (~0.5 µM final concentration) as described above but for the shorter incubation time (30 min). Aliquots (4 µl) of each splicing reaction were loaded directly on 1.5% (w/v) low melting agarose gel as described (42) with running buffer, 50 mM Tris, 50 mM glycine (43). The horizontal slab gel (14 × 14 cm) was run at 150 V for 3.5 h at room temperature. Gel was dried on nitrocellulose membrane and analyzed by autoradiography with an intensifying screen at 70 °C.

	RESULTS AND DISCUSSION

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS AND DISCUSSION REFERENCES

Cloning and Sequence Analysis of SRrp40 and SRrp35-- Screening of expressed sequencing tag data bases using known SR protein sequences revealed two highly related coding sequences. SRrp40 (SR-repressor protein of ~40 kDa) was fully sequenced whereas the original expressed sequencing tag clone for SRrp35 (SR-repressor protein of ~35 kDa) was probably alternatively spliced and lacked an initiation codon. An alternative of this sequence has since been deposited in the data base that contains a complete N terminus and initiation codon. The sequences of SRrp40 (262 amino acids) and SRrp35 (261 amino acids) are shown in Fig. 1A, each protein possesses a single N-terminal RRM, and each protein has a predicted molecular mass of around 31 kDa. Overall the two proteins are 59% similar; however, sequence homology in the single RRM is very high (88% identity). An alignment of the single RRM with the canonical N-terminal RRM of 10 other SR proteins is shown in Fig. 1B. SRrp35 and SRrp40 both show the highest homology to SC35 (47/61 and 46/60% identity/similarity, respectively). SRrp40 is identical to TASR-2, which was cloned by yeast two-hybrid screen using TLS, a gene located on chromosome 16 that is the breakpoint for translocations to chromosomes 12 and 21, as a bait (44).

View larger version (43K): [in this window] [in a new window]   Fig. 1.   Primary structures of SRrp40 and SRrp35 proteins are homologous to the members of SR protein family. A, sequences of SRrp40 and SRrp35. Standard one-letter abbreviations are used for amino acids. B, alignment of RRM sequences of the SR protein family. The positions of the RNP2 and RNP1 submotifs are indicated with horizontal bars. Positions of identity or similarity are shaded black, and gaps are indicated by dots.

The main area of divergence between SRrp40 and SRrp35 is in the RS domains. Both domains are around 150 amino acids in length, which is longer than the RS domain of SC35 (110 amino acids). Despite this both proteins contain fewer SR/RS dipeptides than SC35, with 25, 17, and 27 for SRrp40, SRrp35, and SC35, respectively.

Northern blotting analysis of SRrp40 shows that it is widely yet differentially expressed across a variety of human tissues with the highest expression in skeletal muscle, testis, thymus, and spleen (Fig. 2A). The major isoform of SRrp40 migrates at about 3 kbp, whereas a higher 7.5-kbp species is seen in peripheral blood lymphocytes. SRrp35 seems more tightly controlled than SRrp40 and was only found in testis by Northern blot as a major 1.5-kbp species. Northern blotting of HeLa cell RNA showed expression of SRrp40 but not SRrp35 (data not shown). Tissue-specific or cell type-specific expression of these proteins is consistent with their potential role as the splicing regulator.

View larger version (52K): [in this window] [in a new window]   Fig. 2.   SRrp40 cDNA is differentially expressed across a variety of human tissues. A, Northern blotting analyses probed with SRrp40 and SRrp35 cDNAs in multiple human tissues. DNA size markers are indicated with their length (kb). -Actin cDNA was used as an internal control. PBL, peripheral blood lymphocytes. B, whole cell lysates of 293T cells transiently transfected with control (mock), SRrp40, SRrp35, SC35, and SRp40 cDNA, and expression of recombinant protein was analyzed by Western blots using an anti-T7 monoclonal antibody directed to the N-terminal epitope tag.

The coding regions of SRrp40 and SRrp35 were cloned into the expression vector in frame with a T7 gene 10 antibody epitope tag, so that all constructs encoded proteins with the T7 epitope tag at their N termini (see "Experimental Procedures"), allowing detection of the exogenous proteins with a monoclonal antibody that recognizes this epitope (45). We assayed the expression of these cDNAs by Western blotting analysis of whole-cell lysates in transiently transfected 293T cells (Fig. 2B). SR proteins are known to run with higher apparent molecular mass on SDS-PAGE because of phosphorylation of the RS domain (45). Surprisingly although SRrp40 and SRrp35 differ by only a single amino acid in length they run differently, and we predict this is partially related to the reduced RS/SR content and hence the level of serine phosphorylation of SRrp35.

SRrp40 Shuttles between the Nucleus and Cytoplasm-- We transiently overexpressed the T7 epitope-tagged SRrp40 in HeLa cells and observed the subcellular distribution and subnuclear localization of the tagged protein by indirect immunofluorescence microscopy. As expected SRrp40 is exclusively localized in the nucleoplasm and concentrated in the nuclear speckles in transfected cells (Fig. 3A). Nuclear speckles are thought to represent storage or assembly sites for splicing components and contain SR proteins and many other splicing factors (see Refs. 46-48; reviewed in Ref. 49). Accumulation in the speckles seems to be driven by the RS domain in single domain SR proteins such as SRp20 and SC35, as well as other RS domain-containing proteins such as the Suppressor-of-White-Apricot and Transformer genes from D. melanogaster (47, 50, 51). In the two RRM-containing SR proteins, such as SF2/ASF and SRp40, localization to the nuclear speckles is more complex being driven by the additive effects of the two RRMs and the RS domain, i.e. deletion of any one of these three domains reduces but does not abolish accumulation in the speckles. (47).

View larger version (41K): [in this window] [in a new window]   Fig. 3.   SRrp40 localizes to the nuclear speckles and can shuttle between nucleus and cytoplasm. A, subcellular localization is assayed by indirect immunofluorescence of HeLa cells transiently transfected with epitope-tagged SRrp40 or SC35 in the presence or absence of actinomycin D (Act D) and cycloheximide (CHX). B, nucleo-cytoplasmic shuttling ability was assayed by heterokaryons formed between transfected HeLa cells and murine 3T3 cells. Cells were stained for epitope-tagged SR proteins and counter-stained with DAPI to reveal murine nuclei (more punctate staining). The murine 3T3 cells are indicated with an arrowhead. mAb, monoclonal antibody.

Next we assayed for nucleo-cytoplasmic shuttling. The transcriptional inhibition assay is based in the fact that many shuttling proteins accumulate in the cytoplasm when transcription is inhibited (52). SRrp40 appears to be a shuttling SR protein as cytoplasmic accumulation occurs upon actinomycin D treatment, whereas the control, SC35, does not shuttle (Fig. 3A 37). Cycloheximide is also used in these cultures to exclude new protein synthesis as a cause of the cytoplasmic accumulation. To confirm these results, we also performed a heterokaryon analysis (Fig. 3B). In this assay transiently transfected HeLa cells are fused to murine NIH 3T3 cells. The heterokaryons thus formed are stained for the transfected protein using the anti-g10 monoclonal antibody. Murine cells are then revealed using DAPI stain, which gives a more punctate staining on mouse compared with human nuclei. As shown in Fig. 3B, SRrp40 can be found in murine nuclei, implying that the protein has been exported from the HeLa nucleus to the cytoplasm and then imported back into the NIH 3T3 nucleus within a heterokaryon in the presence of cycloheximide. We conclude that SRrp40 is a nucleo-cytoplasmic shuttling protein.

Our previous studies have shown that a subset of SR proteins such as SF2/ASF, 9G8, and SRp20 shuttle between the nucleus and cytoplasm whereas others such as SC35 and SRp40 do not (37). Domain-swapped chimeric protein constructs have shown that this activity resides in the precise sequence of the RS domain and may be one of the evolutionary pressures causing the high degree of sequence conservation in individual SR proteins from a variety of species (37). The functional consequences of nucleo-cytoplasmic shuttling of SRrp40 and some SR proteins is not yet clear, but it is possible that these proteins may remain bound to the mRNA and play a role in mRNA export or provide some form of chaperone role in the cytoplasm.

SRrp40 and SRrp35 Activate Distal Alternative 5' Splice Sites in Vivo-- Modulation of alternative splice site selection can be tested in vivo by co-transfection of the expression vector encoding a protein of interest with an alternative spliced substrate reporter (38). We assayed the effects of SRrp40 and SRrp35 on the splicing of an adenovirus E1A reporter substrate that contains three alternative 5' splice sites. 48 h following transfection into HeLa cells, splicing of the E1A substrate was assayed by reverse transcription-PCR. Both SRrp40 and SRrp35 shift splicing of this substrate to the most distal 9 S 5' splice site (Fig. 4, lanes 4, 11, and 12), and SRrp40 additionally represses the proximal 13 and 12 S 5' splice sites, whereas SRrp35 affects these to a lesser degree. This is in contrast to the activities of SF2/ASF and SC35 (the closest homologue of SRrp40 and SRrp35), which shift splicing to the most proximal 13 S 5' splice site (lanes 2 and 3). This alternative splicing modulation effect could result from repression of the activity of SR proteins, such as SF2/ASF and SC35, which promote splicing to proximal alternative 5' splice sites.

View larger version (77K): [in this window] [in a new window]   Fig. 4.   SRrp40 and SRrp35 activate the distal alternative 5' splice site of adenovirus E1A pre-mRNA in vivo. In vivo splicing analyses were performed with HeLa cells transiently co-transfected with an adenovirus E1A reporter plasmid and the expression plasmid of schematically indicated proteins including chimeras created among SRrp40, SRrp35, and SC35. The E1A pre-mRNA contains three alternative 5' splice sites, 9, 12, and 13 S, from distal to proximal, shown schematically at the bottom. 10 and 11 S products are produced by aberrant splicing that do not involve simple competition between alternative 5' splice sites (66). Spliced products were analyzed by Southern blot of DNA products obtained from reverse transcription-PCR of the co-transfected HeLa cells. Quantitation was performed by densitometry of autoradiograms, and the relative 13, 12, and 9 S uses are expressed as a percentage of 13 + 12 + 9 S. Cont, control.

To test whether the shift to the 9 S 5' splice site was because of either the RRM or the RS domain, a series of domain-swap mutants between the RRM and RS domains of SC35, SRrp40, and SRrp35 were created. The results of these in vivo splicing assays conclusively show that the activity promoting the selection of distal 5' alternative splice sites resides in the RRM and not in the RS domain. All constructs containing the RRM of SRrp40 or SRrp35 promote splicing to the 9 S site whereas chimeras containing the RRM of SC35 splice to the 13 S site. SRrp40 contains a short linker rich in tyrosine (Y), arginine (R), and glutamic acid (D) residues between the RRM and RS domains. To examine the function of this YRD linker, we constructed a mutant, SRrp40YRD (Fig. 4, see lanes 7 and 8), in which this linker was deleted. In vivo splicing with the adenovirus E1A substrate showed no difference in activity between the deletion mutant and wild-type SRrp40 (lane 8). Glycine-rich linkers are found between the two RRMs of two-RRM SR proteins such as SF2/ASF and between the RRM and RS domains in single-RRM SR proteins such as SC35. Deletion of this domain from SF2/ASF had no effect on constitutive splicing in vitro or in vivo (31, 53).

These results with chimeric proteins are consistent with previous in vitro and in vivo splicing assays with various chimeric SR proteins, where the RRM confers their unique specificity in alternative splicing and constitutive splicing (27, 30, 54). The RRM is thought to confer sequence-specific RNA binding, which may be augmented by RS domain-RNA contacts that are likely less specific and regulated by serine phosphorylation (55). The major role of the RS domain appears to be the dynamic formation of a series of protein-protein interactions with other RS domain-containing proteins and to mediate subnuclear localization (47). This may serve to hold the spliceosome together but also allows the sequential recruitment of factors such as U1 snRNP, U2AF^65/35, and U4/U6·U5 tri-snRNP complex to the pre-mRNA (19, 20, 22-24). Indeed a heterologous protein created by fusing the bacteriophage MS2 RNA binding domain to a series of different RS domains was able to augment splicing of a pre-mRNA containing an inserted MS2 recognition sequence (56). The strength of this effect was related to the total number of RS/SR dipeptides and is thought to reflect the ability of the chimeric protein to recruit other RS-containing proteins, likely including the SR proteins themselves. Until recently it was thought that the presence of an RS domain was required for SR protein function in constitutive splicing, whereas RS-deletion mutants were still active in alternative splicing assays in the presence of basal amounts of wild-type SR protein in vitro. However, recently it was shown that SF2/ASF lacking an RS domain is able to complement an S100 extract deficient in SR protein activity in some circumstances (57). Particularly, the RS-deletion mutants are able to function in the splicing of pre-mRNAs with strong polypyrimidine tracts, whereas splicing of pre-mRNAs with weak polypyrimidine tracts cannot be restored. This may reflect the stabilization of U2AF^65/35 binding to the polypyrimidine tract and 3' splice site induced by RS domain-RS domain contacts (22). Although only 10 RS repeats are required for full SF2/ASF activity in vitro, it is possible that the unique sequences of the RS domains of different proteins are not purely degenerate but promote a degree of specificity in vivo. Indeed small deletions of the RS domain of SF2/ASF were not tolerated in the SF2-deficient cells (31).

SRrp40 Is a Repressor Protein of Splicing in Vitro-- We tested the effects of SRrp40 on constitutive splicing in vitro. One of the functional hallmarks of SR proteins is their ability to complement splicing in a splicing-deficient S100 extract. This complementation assay was originally used to purify the first canonical SR protein, SF2/ASF (26). To our surprise recombinant SRrp40 was unable to complement splicing of three substrates, -globin, HIV-tat (SF2/ASF-specific), and IgM (SC35-specific) pre-mRNAs (Fig. 5). In this assay, SF2/ASF was fully active with -globin pre-mRNA and tat pre-mRNA whereas SC35 was fully active in -globin and IgM pre-mRNA as reported previously (27). Interestingly, SRrp40 showed dominant negative effects upon SF2/ASF and SC35, because both the first and second steps of splicing are gradually repressed by addition of increasing amounts of SRrp40 (Fig. 6A, left panel). In the presence of splicing-competent nuclear extract, which contains a variety of SR proteins, splicing was more resistant to inhibition by SRrp40 but was again fully repressed at the higher concentrations (right panel). Native gel analysis of spliceosomal complex assembly shows that splicing is repressed at a very early step, because an initial ATP-dependent spliceosomal "A" complex is barely seen in the presence of SRrp40 (Fig. 6B).

View larger version (95K): [in this window] [in a new window]   Fig. 5.   SRrp40 does not complement splicing-deficient S100 extract in vitro and has a dominant negative effect against the SR proteins, SF2/ASF and SC35. In vitro splicing was performed with three representative substrates, b-globin (exons 1 and 2), tat (exons 2 and 3), and IgM (exons C3 and C4). The structures of these pre-mRNAs are shown schematically at the top. Indicated GST fusion proteins are added (+, ~0.4 µM) with SR protein-deficient S100 extract (8 µl). The positions of the spliced mRNAs are indicated by arrowheads. The asterisk indicates an aberrant cleavage product unrelated to splicing (26). The splicing products of these three substrates were characterized previously in detail (26, 67, 68). pBR322/HpaII DNA size markers are shown with lengths (nt, nucleotide).

View larger version (57K): [in this window] [in a new window]   Fig. 6.   Splicing repression of SRrp40 is concentration-dependent, and SRrp40 blocks initial ATP-dependent splicing-complex formation. A, SRrp40 titration assay of in vitro splicing with -globin pre-mRNA. Splicing reactions contained either nuclear extract (NE; 6 µl) or S100 extract (7 µl) complemented with ~0.4 µM recombinant GST-SF2/ASF plus increasing amounts of the GST-SRrp40; 0 (), ~0.04, ~0.08, ~0.2, ~0.4, and ~0.8 µM final protein concentration. The positions of pre-mRNA, lariat intermediate, and spliced mRNA are indicated schematically at the left. To separate the lariat intermediate from other products, a higher polyacrylamide percentage (9%) gel was employed. DNA markers are the same as in Fig. 5. B, native gel analysis of spliceosomal complexes formed on -globin pre-mRNA. The positions of spliceosome complexes, H, A, B, and C (1,2), were indicated with brackets.

Because SRrp40 is highly homologous to the authentic SR proteins it is possible that the dominant negative effects are because of a direct competition. This competition may be for overlapping pre-mRNA binding sites, such as SF2/ASF- and SC35-specific exonic splicing enhancers (58, 59). The fact that SRrp40 and SRrp35 antagonize SR protein to activate the most distal 5' splice site of the adenovirus pre-mRNA is reminiscent of the effects observed with hnRNP A1, characterized previously as an antagonist of SR proteins in alternative splicing (38, 60). Interestingly, it was reported recently that hnRNP A1 also functions as a silencing factor for splicing of several pre-mRNAs that include exonic splicing-silencer sequences (61, 62). Alternatively, SRrp40 may bind to an unknown splicing silencer, not related to an SR protein binding site, and cause silencer-dependent splicing repression.

SRrp86 is an atypical SR-related protein that contains a single N-terminal RRM and two C-terminal RS-rich regions (63). Like SRrp40, SRrp86, which shows little sequence homology with SRrp40 or SRrp35, was unable to complement S100 and was also shown to repress activation of splicing in S100 extract complemented by other SR proteins such as SC35 and SF2/ASF. SRrp86 was also able to antagonize the effect of SR proteins in alternative splicing assays in vitro. Interestingly when SRrp86 activity was assayed in vivo, there was strong activation of distal 5' splice sites, which is similar to our results with SRrp40 and SRrp35. A version of SRrp86 lacking the RS domain displayed no inhibition of splicing in vitro and lost the activation of the distal 5' splice site in the adenovirus E1A transcript in vivo. SRrp86 can bind a subset of SR proteins via its RS domain, and it was proposed that its effects on splicing were because of the modulation of activity of a subset of the authentic SR proteins.

In contrast to SF2/ASF and SC35, one SR protein, p54, which complements S100 extract, promotes the use of a 5' distal site (16). Another factor, p32, which was initially co-purified with SF2/ASF (9), can inhibit binding of SF2/ASF to certain substrates and again causes a shift to the utilization of distal alternative splice sites (64). D. melanogaster RSF1, which contains a C-terminal domain rich in G, R, and S residues, is also a general repressor of splicing in vitro (65). RSF1 can bind SR proteins and seems to act at the early stages of splicing to prevent U1 snRNP binding to the 5' splice site. Interestingly, overproduction of RSF1 in the fly rescues several developmental defects caused by overexpression of the splicing activator SR protein B52/SRp55.

In summary, we have studied two proteins that are structurally homologous to SC35 but which reveal distinctive functions in constitutive and alternative splicing and provide a new example of splicing factors that seem to functionally antagonize SR proteins in splicing regulation.

	FOOTNOTES

^* This work was supported in part by the Association for International Cancer Research and Biotechnology and Biological Science Research Council (to G. R. S.), Medical Research Council (to G. R. S. and J. F. C.), and by funds awarded from the Lucille P. Markey Trust and Institutional Research Grant from A.C.S. (to A. M.). A. M. is a research member of the Sylvester Comprehensive Cancer Center.The costs of publication of this article were defrayed in part by the payment of page charges. The article must therefore be hereby marked "advertisement" in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.

The nucleotide sequence(s) reported in this paper has been submitted to the GenBank^TM/EMBL Data Bank with accession number(s) AF449427 and AF449428.

To whom correspondence should be addressed. Tel.: 44-1865-222442; Fax: 44-1865-222470; E-mail: Screaton@molbiol.ox.ac.uk.

Published, JBC Papers in Press, October 29, 2001, DOI 10.1074/jbc.M103967200

	ABBREVIATIONS

The abbreviations used are: snRNP, small nuclear ribonuclear protein; GST, glutathione S-transferase; SR protein, serine-arginine protein; RRM, RNA-recognition motif; RS, arginine-serine; DAPI, 4,6-diamidino-2-phenylindole; HIV, human immunodeficiency virus.

	REFERENCES

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS AND DISCUSSION REFERENCES

CiteULike

Complore

Connotea

Del.icio.us

Digg

Reddit

Technorati

This article has been cited by other articles:

				J. E. Swartz, Y.-C. Bor, Y. Misawa, D. Rekosh, and M.-L. Hammarskjold The Shuttling SR Protein 9G8 Plays a Role in Translation of Unspliced mRNA Containing a Constitutive Transport Element J. Biol. Chem., July 6, 2007; 282(27): 19844 - 19853. [Abstract] [Full Text] [PDF]

				J. Y. Wu, A. Kar, D. Kuo, B. Yu, and N. Havlioglu SRp54 (SFRS11), a Regulator for tau Exon 10 Alternative Splicing Identified by an Expression Cloning Strategy. Mol. Cell. Biol., September 1, 2006; 26(18): 6739 - 6747. [Abstract] [Full Text] [PDF]

				C. Guillouf, I. Gallais, and F. Moreau-Gachelin Spi-1/PU.1 Oncoprotein Affects Splicing Decisions in a Promoter Binding-dependent Manner J. Biol. Chem., July 14, 2006; 281(28): 19145 - 19155. [Abstract] [Full Text] [PDF]

				A. J. Tyson-Capper, J. Bailey, A. R. Krainer, S. C. Robson, and G. N. Europe-Finner The Switch in Alternative Splicing of Cyclic AMP-response Element Modulator Protein CREM{tau}2{alpha} (Activator) to CREM{alpha} (Repressor) in Human Myometrial Cells Is Mediated by SRp40 J. Biol. Chem., October 14, 2005; 280(41): 34521 - 34529. [Abstract] [Full Text] [PDF]

				D. Cazalla, K. Newton, and J. F. Caceres A Novel SR-Related Protein Is Required for the Second Step of Pre-mRNA Splicing Mol. Cell. Biol., April 15, 2005; 25(8): 2969 - 2980. [Abstract] [Full Text] [PDF]

				K. J. Liu and R. M. Harland Inhibition of neurogenesis by SRp38, a neuroD-regulated RNA-binding protein Development, April 1, 2005; 132(7): 1511 - 1523. [Abstract] [Full Text] [PDF]

				P. Stoilov, R. Daoud, O. Nayler, and S. Stamm Human tra2-beta1 autoregulates its protein concentration by influencing alternative splicing of its pre-mRNA Hum. Mol. Genet., March 1, 2004; 13(5): 509 - 524. [Abstract] [Full Text] [PDF]

				E. Sakashita, S. Tatsumi, D. Werner, H. Endo, and A. Mayeda Human RNPS1 and Its Associated Factors: a Versatile Alternative Pre-mRNA Splicing Regulator In Vivo Mol. Cell. Biol., February 1, 2004; 24(3): 1174 - 1187. [Abstract] [Full Text] [PDF]

				J. Li, I. C. Hawkins, C. D. Harvey, J. L. Jennings, A. J. Link, and J. G. Patton Regulation of Alternative Splicing by SRrp86 and Its Interacting Proteins Mol. Cell. Biol., November 1, 2003; 23(21): 7437 - 7447. [Abstract] [Full Text] [PDF]

				S. R. BRUCE, R.W. C. DINGLE, and M. L. PETERSON B-cell and plasma-cell splicing differences: A potential role in regulated immunoglobulin RNA processing RNA, October 1, 2003; 9(10): 1264 - 1273. [Abstract] [Full Text] [PDF]

				Q. Xu, D. Y. M. Leung, and K. O. Kisich Serine-Arginine-rich Protein p30 Directs Alternative Splicing of Glucocorticoid Receptor Pre-mRNA to Glucocorticoid Receptor {beta} in Neutrophils J. Biol. Chem., July 11, 2003; 278(29): 27112 - 27118. [Abstract] [Full Text] [PDF]

				Md. T. Nasim, H. M. Chowdhury, and I. C. Eperon A double reporter assay for detecting changes in the ratio of spliced and unspliced mRNA in mammalian cells Nucleic Acids Res., October 15, 2002; 30(20): e109 - e109. [Abstract] [Full Text] [PDF]

				J. Li, D. C. Barnard, and J. G. Patton A Unique Glutamic Acid-Lysine (EK) Domain Acts as a Splicing Inhibitor J. Biol. Chem., October 11, 2002; 277(42): 39485 - 39492. [Abstract] [Full Text] [PDF]

				A. J. Pollard, A. R. Krainer, S. C. Robson, and G. N. Europe-Finner Alternative Splicing of the Adenylyl Cyclase Stimulatory G-protein Galpha s Is Regulated by SF2/ASF and Heterogeneous Nuclear Ribonucleoprotein A1 (hnRNPA1) and Involves the Use of an Unusual TG 3'-Splice Site J. Biol. Chem., May 3, 2002; 277(18): 15241 - 15251. [Abstract] [Full Text] [PDF]

This Article Abstract Full Text (PDF) All Versions of this Article: 276/52/48908    most recent M103967200v1 Submit a Letter to Editor Alert me when this article is cited Alert me when eLetters are posted Alert me if a correction is posted Citation Map Services Email this article to a friend Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Request Permissions Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Cowper, A. E. Articles by Screaton, G. R. Search for Related Content PubMed PubMed Citation Articles by Cowper, A. E. Articles by Screaton, G. R. Social Bookmarking                      What's this?