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

J. Biol. Chem., Vol. 275, Issue 48, 38059-38066, December 1, 2000

This Article Abstract Full Text (PDF) All Versions of this Article: 275/48/38059    most recent M001483200v1 Alert me when this article is cited 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 Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Guth, S. Articles by Valcárcel, a. J. Search for Related Content PubMed PubMed Citation Articles by Guth, S. Articles by Valcárcel, a. J. Social Bookmarking                      What's this?

Kinetic Role for Mammalian SF1/BBP in Spliceosome Assembly and Function after Polypyrimidine Tract Recognition by U2AF^*

Sabine Guth and and Juan Valcárcel

From the Gene Expression Programme, European Molecular Biology Laboratory, Meyerhofstrasse 1, D-69117 Heidelberg, Germany

Received for publication, February 22, 2000, and in revised form, August 9, 2000

	ABSTRACT

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS DISCUSSION REFERENCES

Two sequences important for pre-mRNA splicing precede the 3' end of introns in higher eukaryotes, the branch point (BP) and the polypyrimidine (Py) tract. Initial recognition of these signals involves cooperative binding of the splicing factor SF1/mammalian branch point binding protein (mBBP) to the BP and of U2AF⁶⁵ to the Py tract. Both factors are required for recruitment of the U2 small nuclear ribonucleoprotein particle (U2 snRNP) to the BP in reactions reconstituted from purified components. In contrast, extensive depletion of ST1/BBP in Saccharomyces cerevisiae does not compromise spliceosome assembly or splicing significantly. As BP sequences are less conserved in mammals, these discrepancies could reflect more stringent requirements for SF1/BBP in this system. We report here that extensive depletion of SF1/mBBP from nuclear extracts of HeLa cells results in only modest reduction of their activity in spliceosome assembly and splicing. Some of these effects reflect differences in the kinetics of U2 snRNP binding. Although U2AF⁶⁵ binding was reduced in the depleted extracts, the defects caused by SF1/mBBP depletion could not be fully restored by an increase in occupancy of the Py tract by exogenously added U2AF⁶⁵, arguing for a role of SF1/mBBP in U2 snRNP recruitment distinct from promoting U2AF⁶⁵ binding.

	INTRODUCTION

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS DISCUSSION REFERENCES

The expression of eukaryotic genes requires the accurate removal of intervening sequences (introns) and the concomitant fusion of the flanking exons via RNA splicing. The correct recognition of 5' and 3' splice sites occurs in the spliceosome, a large and dynamic macromolecular complex of small nuclear ribonucleoprotein particles (snRNPs)¹ and non-snRNP proteins that assembles in a stepwise manner on the pre-mRNA (1, 2). Five snRNPs play a role in removal of canonical GU/AG introns, each composed of a different U snRNA, a set of polypeptides common to most spliceosomal snRNPs and a set of proteins specific for each snRNP (reviewed in Refs. 3 and 4).

The first ATP-dependent step in spliceosome assembly is the stable association of U2 snRNP with the 3' part of the intron (reviewed in Refs. 5 and 6). This part of the pre-mRNA contains three sequence elements important for the splicing process as follows: the branch point region (BP), the polypyrimidine tract (Py tract), and the conserved dinucleotide AG at the 3' splice site. The BP is highly conserved in yeast (UACUAAC) and is more degenerate in higher eukaryotes (consensus YNCURAY, Y is pyrimidine, R is purine, N is any nucleotide). The BP establishes base pairing interactions with a specific sequence of U2 snRNA, bulging out the nucleotide, usually adenosine, that forms a 2'-5' phosphodiester bond with the 5' end of the intron (7-10). The Py tract, particularly important for splicing in higher eukaryotes, is a pyrimidine-rich sequence located immediately downstream of the BP and upstream of the AG dinucleotide (1).

Biochemical fractionation of mammalian nuclear extracts indicated that stable binding of purified U2 snRNP to the pre-mRNA requires four activities as follows: SF3a, SF3b, U2AF, and SF1 (11). SF3a and SF3b were subsequently found to be integral components of the 17 S U2 snRNP that dissociate during purification (12). U2AF is composed of two subunits of 65 and 35 kDa (13). U2AF⁶⁵ binds to the Py tract (14), whereas U2AF³⁵ recognizes the 3' splice site AG and helps to stabilize the U2AF⁶⁵/Py tract interaction (15-17). SF1, also called mammalian branch point binding protein (mBBP), was purified as a 75-kDa polypeptide and found to recognize specifically the BP (18, 19).

The mechanisms by which U2AF and SF1/mBBP facilitate U2 snRNP binding are not well understood. The amino-terminal arginine-serine (RS) rich domain of U2AF⁶⁵ contacts the BP, and it has been proposed that its positively charged residues can stabilize the limited base pairing interactions that can be established between U2 snRNA and the BP (20-22). An interaction between U2AF⁶⁵ and SAP 155, a protein subunit of SF3b, can provide an additional mechanism for U2 recruitment (23). U2AF⁶⁵ also interacts with UAP56, a DEAH box helicase found to be important for U2 snRNP binding (24).

SF1/mBBP and U2AF⁶⁵ interact with each other, and this interaction can facilitate cooperative recognition of the BP and Py tract that are adjacent to each other in the pre-mRNA (25, 26). It is currently unclear whether SF1/mBBP plays other roles in promoting U2 snRNP assembly.

SF1/mBBP contains at least five distinct structural domains. The amino-terminal region contacts U2AF⁶⁵ (27), an essential interaction for SF1/mBBP function that is disrupted by phosphorylation by the serine/threonine kinase PKG-1 in response to cGMP (28). This region of the protein is followed by an hnRNP K homology domain (KH domain) and a zinc knuckle, two motifs implicated in RNA binding (reviewed in Ref. 29). The KH domain is sufficient for specific recognition of the pre-mRNA BP (27, 30), and the zinc knuckle seems to confer additional RNA binding affinity, probably through interaction with the negatively charged phosphate backbone (30). The RNA binding domains are followed by a proline-rich region and a carboxyl-terminal domain. Alternatively spliced variants of human SF1/mBBP are expressed in a cell type-specific manner that differ in the length of the proline-rich region and have distinct carboxyl termini (31, 32). The carboxyl terminus and the proline-rich region are not only dispensable for SF1/mBBP activity in spliceosome assembly in vitro but also for viability in yeast (27) and might be required for other functions of SF1/mBBP in vivo.

A Saccharomyces cerevisiae SF1/mBBP homolog was genetically identified by its synthetic lethality with MUD2, the gene encoding the functional homolog of U2AF⁶⁵ in yeast (25). Biochemical experiments showed that ySF1/BBP also contacts the 3rd RNA Recognition Motif of Mud2p and binds the BP (19, 25). Direct interaction of ySF1/BBP with prp40, a component of U1 snRNP, suggests that ySF1/BBP can form a bridge between the 5' splice site and the 3' splice site region in one of the first detectable complexes able to commit the pre-mRNA to undergo splicing (25, 33). Human SF1/mBBP has also been found to be a component of such complexes in mammals (19, 25). SF1/mBBP is not present, however, in pre-spliceosome complexes, suggesting that BP recognition by the protein is no longer needed and/or is incompatible with the stable association of U2 snRNP with this sequence (19, 33). Recent results suggest that ySF1/BBP plays a kinetic role in the progression through the two commitment complexes observable in yeast extracts (33). Depletion of ySF1/BBP resulted in a reduction in the accumulation of commitment complex 2 (CC2), in which the 5' splice site is recognized by U1 snRNP and the BP by ySF1/BBP and MUD2. Surprisingly, however, spliceosome formation was not affected by ySF1/BBP depletion. These results suggest that under conditions of reduced amounts of ySF1/BBP, formation of CC2 became rate-limiting and the CC2 complexes formed were immediately chased into spliceosomes (33).

In this report we have developed an immunodepletion protocol allowing removal of SF1/mBBP from HeLa nuclear extracts, and we used this method to answer two questions. First, is the activity of SF1/mBBP essential for U2 snRNP binding in mammals but more dispensable in yeast, as suggested by the differences between biochemical reconstitution experiments in HeLa extracts and depletion experiments in S. cerevisiae? Second, is the only role of SF1/mBBP in mammalian spliceosome assembly to assist U2AF⁶⁵ binding to the Py tract? Consistent with the results in yeast and in contrast with the requirement of SF1/mBBP in reconstituted reactions with purified components, extensive depletion of SF1 had relatively small effects in the accumulation of complexes containing U2 snRNP. Although substitution of the weak BP of an IgM pre-mRNA by the yeast BP consensus conferred a kinetic advantage in spliceosome assembly, this advantage was no longer observed in SF1-depleted nuclear extracts. These effects correlated with a reduced cross-link of U2AF⁶⁵ to the RNA and could be reversed by supplementing the depleted extract with recombinant SF1/mBBP. Interestingly, addition of U2AF⁶⁵ also increased the cross-linking signal but was unable to restore splicing complex formation. These observations imply additional functions for SF1/mBBP in spliceosome assembly distinct from facilitating U2AF⁶⁵ binding.

	EXPERIMENTAL PROCEDURES

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS DISCUSSION REFERENCES

Plasmids-- IgM-yBp was obtained by replacing the IgM branch point sequence with the yeast consensus branch point (TACTAAC) via polymerase chain reaction-based site-directed mutagenesis of plasmid pµM (34) as described elsewhere (35). Mutant clones were confirmed by sequencing. IgM mutPy was described in Ref. 36. AdML was described in Ref. 37.

Anti-SF1 Antisera-- Antibodies were raised in rabbits against an SF1/mBBP amino-terminal peptide (MATGANATPLDFPS) coupled to keyhole limpet hemocyanin (38) or against recombinant SF1C4 (SF1 amino acid residues 1-320, as described in Ref. 27).

Affinity Purification of Anti-SF1 Antibodies-- Anti-SF1 peptide antibodies were affinity-purified from the -SF1 peptide antiserum. 100 µg of recombinant SF1C4 were blotted onto polyvinylidene difluoride membrane (Immobilon-P, Millipore). The membrane was blocked with PBST (phosphate-buffered saline, 0.1% Tween 20) + 5% non-fat dry milk, incubated with 500 µl of -SF1 peptide serum for 4-5 h at room temperature, and then washed with PBST 3 times for 10 min. Bound -SF1 antibodies were eluted with 500 µl of 100 mM triethylamine, pH 11.5, at room temperature for 15 min. The solution was immediately neutralized by adding 1 ml of 1 M Tris·HCl, pH 7. Purified antibodies were stored at 80 °C.

Preparation of HeLa Nuclear Extract-- HeLa nuclear extract was prepared as described by Dignam et al. (39).

Immunoblot Analysis-- Proteins were separated by electrophoresis in a 10% SDS-polyacrylamide gel and transferred to nitrocellulose membrane (PROTRAN, Schleicher & Schuell). The membrane was first incubated with the -SF1 antiserum (1:3000 dilution) or affinity-purified -SF1 antibodies (1:100 dilution) followed by incubation with the secondary antibody (-rabbit IgG, horseradish peroxidase linked whole antibody from donkey, Amersham Pharmacia Biotech) at a 1:5000 dilution. Antigens were detected by enhanced chemiluminescence using the ECL Western blotting reagent (Amersham Pharmacia Biotech).

SF1 Immunodepletion-- -SF1 antibodies were covalently coupled to protein A-Sepharose beads (4Fast Flow, Amersham Pharmacia Biotech) using the dimethyl pimelimidate method (38).

SF1 protein was immunodepleted from HeLa nuclear extracts in two steps using different salt concentrations. Prior to immunodepletion the Sepharose-coupled antibodies were washed with buffer D (20 mM Hepes, pH 8, 0.2 mM EDTA, 1 mM dithiothreitol, 20% glycerol, 0.015% Nonidet P-40) + 0.1 M KCl. 600 µl of HeLa nuclear extract were incubated with 0.4 ml of sedimented beads at 4 °C for 1.5 h with agitation. The nuclear extract was removed and kept on ice. Bound SF1 was removed from the beads by incubating them with 1.5 ml of 0.1 M glycine, pH 3, twice for 5 min. The resin was washed 3 times with 1 ml of buffer D + 0.5 M KCl. Prior to adding the nuclear extract back to the beads for a second round of depletion, the salt concentration was raised to 0.5 M by adding 15 mg of KCl to the extract. After a second round of depletion the nuclear extract was dialyzed against buffer D + 100 mM KCl and kept at 80 °C.

In Vitro Transcription of Splicing Substrates-- Transcription templates for the full-length and 3'-half substrates were generated by polymerase chain reaction from the AdML, pµM, or pIgM-yBp plasmids including a T7 or SP6 promoter in the upstream primer.

For the full-length substrates, approximately 1 µg of template DNA was used in 25-µl transcription reactions containing 40 mM Tris·HCl, pH 7.9, 10 mM NaCl, 6 mM MgCl₂, 2 mM spermidine, 0.8 mM dithiothreitol, 0.4 mM ATP and CTP, 0.04 mM GTP, 0.08 mM UTP, 1.6 mM CAP analog (m7G(5')ppp(5')G, New England Biolabs), 5 µl of [-³²P]UTP (20 mCi/ml, 800 Ci/mmol, Amersham Pharmacia Biotech), and 40 units of SP6 or T7 RNA polymerase (Promega). For the 3'-half RNAs the CAP analog was omitted from the reaction mix, and the GTP concentration was increased to 0.4 mM. After a 2-h incubation at 37 °C the transcripts were gel-purified.

In Vitro Splicing Assays and Spliceosome Assembly Reactions-- Splicing reactions and splicing complementation assays were performed as described elsewhere (36). Spliced products were resolved on 13% denaturing polyacrylamide gels.

For time course experiments of spliceosome assembly the reaction mixture was incubated at 30 °C, and 5-µl samples were removed at different time points and pipetted into chilled tubes containing 0.5 µl of heparin (50 mg/ml; Sigma) to stop complex formation. Spliceosomal complexes were separated on native 4% acrylamide:bisacrylamide (80:1), 0.5% agarose gels in 50 mM Tris base, 50 mM glycine buffer.

Gels were exposed to film (Kodak X-Omat AR) and/or PhosphorImager screens (Fuji BAS-MP).

UV Cross-linking and Immunoprecipitation of U2AF⁶⁵-- The experiment was performed exactly as described elsewhere (36).

	RESULTS

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS DISCUSSION REFERENCES

Immunodepletion of SF1/mBBP from HeLa Nuclear Extracts-- To facilitate the functional analysis of SF1/mBBP, anti-SF1 antisera were obtained by inoculating rabbits with recombinant purified hSF1-C4 (amino acid residues 1-320 of human SF1/mBBP (27)) or with a peptide containing the amino-terminal 14 amino acids coupled to keyhole limpet hemocyanin. Western blot analyses revealed that both antisera and affinity-purified anti-peptide antibodies were able to recognize recombinant hSF1-C4 (Fig. 1A, lanes 3, 5, and 7) as well as two protein species with apparent molecular masses of 67 and 75 kDa in nuclear extracts from HeLa cells (Fig. 1A, lanes 4, 6, and 8), consistent with the size of the two reported HeLa SF1/mBBP isoforms (27, 31). No signal was detected with the pre-immune serum (Fig. 1A, lanes 1 and 2). These observations indicated that the antisera were able to specifically recognize SF1/mBBP. A protein species of approximately 150 kDa was recognized by the anti-SF1 peptide serum even after affinity purification of the antibodies (Fig. 1A, lanes 6 and 8). As this protein was not recognized by the anti-SF1-C4 antiserum (Fig. 1A, lane 4) nor by two additional antisera generated against hSF1-C4 or the amino-terminal SF1/mBBP peptide (data not shown), we suggest that it corresponds to a polypeptide unrelated to SF1/mBBP that cross-reacts with one out of the four anti-SF1 antisera used for these experiments.

View larger version (30K): [in this window] [in a new window]   Fig. 1.   Immunodepletion of SF1/mBBP. A, -SF1 antisera specifically recognize recombinant and endogenous SF1/mBBP. Western blot of HeLa nuclear extracts and recombinant SF1-C4. 5 ng of recombinant SF-C4 (rSF1) or 5 µl of HeLa nuclear extract (NE, 8 mg of total protein/ml) were fractionated on a 10% SDS-polyacrylamide gel and blotted onto nitrocellulose. The blot was probed with a 1:3000 dilution of pre-immune serum (lanes 1 and 2), polyclonal -SF1 serum (lanes 3 and 4), polyclonal -SF1 amino-terminal peptide serum (lanes 5 and 6), or a 1:100 dilution of affinity-purified -SF1 peptide antibodies (lanes 7 and 8). The positions of recombinant SF1-C4 and the two HeLa isoforms of SF1 are indicated on the right. M, size markers. B, immunodepletion of SF1 from HeLa nuclear extract. Nuclear extract (NE), serial dilutions of nuclear extract, SF1-immunodepleted nuclear extract (SF1NE), and mock-depleted nuclear extract (mockNE) were fractionated on a 10% SDS-polyacrylamide gel and blotted onto nitrocellulose. The membrane was probed with polyclonal -SF1 peptide serum. The position of HeLa SF1 isoforms are marked on the right. M, size markers.

Fig. 1B shows that HeLa nuclear extracts could be efficiently immunodepleted of SF1/mBBP using anti-SF1 anti-peptide antibodies covalently linked to protein A-Sepharose beads. At least 98% of SF1/mBBP were removed from the extract in a two-step depletion procedure (Fig. 1B, lane 6), whereas the levels of SF1/mBBP remained unaltered in the mock-depleted extract using the pre-immune serum (Fig. 1B, lane 7).

Effects of SF1 Immunodepletion on Splicing and Spliceosome Assembly-- The effects of SF1/mBBP depletion on splicing and spliceosome assembly of two model pre-mRNA substrates derived from an adenovirus gene (AdML (37)) and a mouse IgM gene (34) were tested.

The RNA substrates were incubated under splicing conditions in mock-depleted or SF1-depleted nuclear extracts in the presence or absence of recombinant purified SF1-C4 protein. This recombinant protein lacking the proline-rich region and carboxyl terminus had been previously shown to promote U2 snRNP binding in a reconstituted reaction with purified U2 snRNP and other auxiliary factors (27). After incubation, the reaction mixtures were either directly resolved on native gels to separate spliceosomal complexes or on denaturing gels after RNA isolation to resolve intermediates and products of the splicing reaction.

After 20 min of incubation at 30 °C, splicing complex formation on AdML remained largely unaltered in the SF1-depleted nuclear extract when compared with the mock-depleted extract (Fig. 2A, lanes 3 and 4). When the IgM substrate containing a predictable weaker branch point than that of AdML was used, splicing complex formation was reduced but not completely inhibited (Fig. 2A, lane 8). This reduction could be partially reversed by the addition of recombinant SF1-C4 (Fig. 2A, lane 9) but not with U2AF⁶⁵ or SR proteins (data not shown and see below), indicating that the effect observed is specifically related to SF1/mBBP function. The relatively weak effects of SF1/mBBP depletion were in contrast with the absolute requirement for SF1/mBBP in reconstitution reactions reported earlier (11).

View larger version (32K): [in this window] [in a new window]   Fig. 2.   Spliceosome assembly and complex A formation are only modestly reduced in SF1 immunodepleted nuclear extract. Radioactively labeled AdML and IgM full-length substrates (A) or 3'-half RNAs (B) were incubated in HeLa nuclear extracts (NE) in the presence or absence of ATP, in mock-depleted nuclear extract (mockNE), or SF1-immunodepleted nuclear extract (SF1NE) in the presence or absence of 5 ng/µl of recombinant SF1-C4 (27). The mixtures were loaded onto native polyacrylamide-agarose composite gels, allowing separation of ATP-independent hnRNP complexes (complex H), ATP-dependent pre-spliceosomes (complex A), and two conformations of the fully assembled spliceosome (complexes B and C) for the full-length substrate.

Similar results were obtained when the RNA splicing products and intermediates were analyzed on denaturing gels after 2 h of incubation at 30 °C (data not shown).

The effects of the SF1/mBBP depletion were slightly more pronounced when 3'-half RNAs of both substrates were used (Fig. 2B). These RNAs contain the 3' portion of the intron, including branch point and polypyrimidine tract, and the downstream exon and are capable of forming complexes with U2 snRNP (complex A3'). Reduction in complex A3' formation could be restored by addition of recombinant SF1-C4 to the depleted extract (Fig. 2B, lanes 3-5 and 7-9).

In summary, extensive depletion of SF1/mBBP results in relatively small decreases in U2 snRNP binding to two model pre-mRNA substrates. This is particularly striking compared with the strong reduction in U2 snRNP binding caused by a similar extent of depletion of U2AF⁶⁵ (e.g. Refs. 14, 21, 22, 24, and 36), despite the similar concentrations of SF1/mBBP and U2AF⁶⁵ in nuclear extracts (data not shown).

Time Course Experiments of Complex A Formation-- The modest effects of extensive SF1/mBBP depletion on U2 snRNP binding resemble the observations made in yeast by Rutz and Séraphin (33) and argue against a fundamental difference in SF1/BBP requirement between yeast and mammals. Rutz and Séraphin (33) observed that depletion of yBBP resulted in the absence of one of the commitment complexes observable in yeast (CC2) and proposed a role for ySF1 in promoting progression from CC1 to CC2. Although commitment complexes are not resolved in the native gel system commonly used to analyze spliceosome formation in mammalian extracts, we set out to compare the kinetics of spliceosome formation in SF1/mBBP-depleted versus mock-depleted extracts for two pre-mRNA substrates as follows: IgM and a mutant version of this substrate in which the BP (the branched nucleotide experimentally determined as position 22 from the 3' splice site; data not shown) was substituted by the yeast consensus BP region (IgM-yBP), which has been shown to also be the preferred BP in mammals (40). Time course experiments of spliceosome formation were performed as follows. Complex formation was stopped at different time points by addition of heparin, and the samples were analyzed on native gels that allow the separation of hnRNP complexes (complex H) from pre-spliceosomes (complex A) and two conformations of the fully assembled spliceosome (B/C). The gels were dried and exposed to PhosphorImager screens (Fig. 3A). Multiple preliminary experiments were performed to determine precisely the time, within 5-min intervals, at which complex A formation was maximal. Consistently, the maximum levels were achieved after 30 min for the wild-type IgM RNA in mock NE and SF1NE. The maximum level of complex A for the mutant IgM-yBP RNA was found after 20 min in the mock NE and 30 min in the SF1NE. Fig. 3A shows a representative example of a time course experiment. Fig. 3B compiles the results of four different experiments for each substrate under optimized conditions, the average normalized values and standard deviations plotted against time. To be able to compare directly the different values of PhosphorImager units from several experiments, the value obtained at 20 min in mock-depleted nuclear extracts was arbitrarily set to 1000 arbitrary scan units, and the rest of the values of the experiment were normalized accordingly. Equivalent results were obtained when the data were normalized to the 10- or 60-min time points.

View larger version (87K): [in this window] [in a new window]   Fig. 3.   Time course of complex A formation with IgM and IgM containing the yeast consensus branch point sequence (IgMyBp) in mock-depleted (mockNE) and SF1-immunodepleted nuclear extract (SF1NE). A, radioactively labeled IgM (left panel) and IgMyBp (right panel) full-length splicing substrates were incubated in mockNE or SF1NE. Samples were taken at the indicated time points and stopped by mixing them with heparin on ice. After all samples were collected, the mixtures were loaded onto native polyacrylamide-agarose composite gels to separate hnRNP complexes (complex H), pre-spliceosomes (complex A), and two conformations of the fully assembled spliceosome (complexes B and C). Gels were dried and exposed to PhosphorImager screens to allow quantification of complexes. B, quantification of complex A formation in time course experiments. Experiments as described in A were repeated several times to determine carefully the time of maximal levels of complex A. Four different experiments were then performed under optimized conditions. The results averaged and the standard deviation were calculated. Values obtained were plotted against time. The left panel shows the results for IgM wild-type RNA and the right panel for IgM-yBp, with the optimal binding site for SF1/mBBP (19). Black boxes represent values obtained in mock-depleted nuclear extracts, and open box values were obtained using SF1/mBBP-immunodepleted nuclear extracts.

The left part of Fig. 3B shows that reduced levels of complex A were formed on wild-type IgM in SF1/mBBP-depleted extracts compared with mock-depleted extracts. The kinetics of complex formation, however, were identical, with maximal levels of complex A formed after 30 min of incubation, which then decreased as pre-spliceosomes were converted into mature spliceosomes (complex B and C). This result indicates that depletion of SF1/mBBP to more than 98% does not affect the kinetics of spliceosome assembly of the IgM pre-mRNA.

A different result was obtained when the IgM-yBP pre-mRNA was used (Fig. 3B, right panel). In mock-depleted extract, maximal levels of complex A were reached 10 min earlier with this substrate than with the wild-type pre-mRNA. This observation suggested that the presence of a more consensus BP kinetically facilitated U2 snRNP binding. In contrast, this kinetic effect was not observed in SF1/mBBP-depleted extracts; maximal levels of complex A formation were detected at 30 min, as was also the case for the IgM substrate containing a wild-type branch point. Taken together these results indicate that a consensus BP, which is the optimal binding site for SF1/mBBP (19), allows faster recruitment of U2 snRNP and that this effect is influenced by the presence of SF1/mBBP.

SF1/mBBP Activity and U2AF⁶⁵ Binding-- Previous results demonstrated that purified recombinant SF1 and U2AF⁶⁵ bind cooperatively to pre-mRNAs in vitro (26). To determine whether (and to what extent) the activity of SF1/mBBP in complete extracts correlated with assisting U2AF⁶⁵ binding, UV cross-linking and immunoprecipitation experiments were performed.

Uniformly labeled IgM 3'-half RNA was incubated in nuclear extracts under splicing conditions, irradiated with UV light, digested with RNase A, and the reaction mix precipitated with antibodies specific for U2AF⁶⁵ as described previously (36). The products of immunoprecipitation were fractionated by SDS-polyacrylamide gel electrophoresis, exposed to PhosphorImager screens, and the U2AF⁶⁵ signals quantified. The results of Fig. 4A show that cross-linking of both endogenous or exogenously added purified recombinant GST-U2AF⁶⁵ could be detected and that U2AF⁶⁵ cross-linking was strongly reduced when the Py tract was mutated (U residues mutated to A residues). Together with previous evidence for the binding specificity of U2AF⁶⁵ to Py tracts (41), these data argue that specific cross-linking of U2AF⁶⁵ to the Py tract could be measured using this assay.

View larger version (42K): [in this window] [in a new window]   Fig. 4.   Effects of SF1/mBBP on complex A formation with 3' half IgM and cross-linking of U2AF65 to the polypyrimidine tract. A, specific cross-linking of U2AF65 to the Py tract. Endogenous U2AF65 and recombinant GST-U2AF65 added to nuclear extract were cross-linked to radioactively labeled wild-type IgM3' RNA (lanes 1 and 2) or IgM3' with a mutant branch point sequence (U residues substituted by A residues, lanes 3 and 4) and the cross-linked proteins were immunoprecipitated with -U2AF65 specific antibodies. Positions of U2AF65 and GST-U2AF65 are indicated on the right. B, complex A3' formation. Radioactively labeled IgM3' RNA was incubated in HeLa nuclear extracts (NE) in the presence or absence of ATP, in mock-depleted NE (mockNE) or SF1/mBBP-immunodepleted NE (SF1NE) in the presence or absence of 5 ng/µl recombinant SF1-C4 or 10 ng/µl GST-U2AF65. The mixtures were loaded onto native polyacrylamide-agarose composite gels to separate complex A (indicated on the left) from hnRNP complexes. The dried gels were exposed to PhosphorImager screens (upper panel) and the amount of complex A was quantified for each lane (lower panel). C, quantification of U2AF65 cross-linking. Radioactively labeled IgM3' RNA was incubated under the same conditions as in (A). After incubation bound proteins were UV cross-linked and immunoprecipitated with -U2AF65 specific antibodies and subsequently fractionated on a 10% SDS-polyacrylamide gel. The dried gel was exposed to a PhosphorImager screen (upper panel). Positions of U2AF65 and GST-U2AF65 are indicated on the right. M, molecular weight marker. The amount of protein cross-linked in each lane was quantified (lower panel).

Spliceosome assembly and U2AF⁶⁵ binding to IgM 3'-half RNA were then analyzed in mock-depleted and SF1/mBBP-depleted extracts complemented with recombinant SF1-C4 or U2AF⁶⁵. The results shown in Fig. 4B indicate that the decrease in complex A formation caused by depletion of SF1-mBBP could be restored by addition of recombinant SF1-C4 but only marginally restored by addition of recombinant U2AF⁶⁵. The amount of GST-U2AF65 added in the complementation assay was optimized in previous experiments and was able to restore splicing activity of U2AF-depleted extracts prepared by either immunodepletion or chromatographic depletion (17, 22, 24, 36, 47). Control experiments showed that the presence of the GST moiety did not interfere significantly with the activity of the protein as a splicing factor (data not shown).

Fig. 4C shows the analysis of U2AF⁶⁵ binding by UV cross-linking and immunoprecipitation of aliquots from the same reaction mixtures used in Fig. 4B. SF1/mBBP depletion resulted in 50% reduction of U2AF⁶⁵ cross-linking, and this was partially restored by addition of recombinant SF1/mBBP (compare lanes 2-4). These data correlate with the activity of SF1/mBBP in complex A formation and would be consistent with a model in which SF1/mBBP acts by promoting the binding of U2AF⁶⁵, which, in turn, assists U2 snRNP binding. The comparison of lanes 5 from Fig. 4, B and C, however, argues against this possibility; addition of recombinant U2AF⁶⁵ to the SF1/mBBP-depleted extract resulted in a significant increase of the occupancy of the Py tract by U2AF⁶⁵ (both endogenous and recombinant) without a significant increase in complex A formation. The results shown in Fig. 4, B and C, were consistently obtained in independent experiments. Taken together, the results presented in Fig. 4 suggest that at least part of the activity of SF1/mBBP in promoting U2 snRNP binding does not correlate with its ability to promote binding of U2AF⁶⁵ to the Py tract.

	DISCUSSION

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS DISCUSSION REFERENCES

SF1/mBBP was first identified as a biochemical activity necessary, together with U2AF, to allow binding of purified human U2 snRNP to a model pre-mRNA (11). SF1/mBBP was subsequently found to be essential for the viability of S. cerevisiae and Caenorhabditis elegans (25, 42). The essential nature of the SF1/mBBP gene could be due to the activity of its protein products in pre-mRNA splicing. Depletion experiments in yeast (33) and in mammalian extracts (Figs. 2 and 3), however, indicate that significant reduction of SF1/mBBP levels cause only modest changes in the efficiency of formation of splicing complexes and in the accumulation of splicing products. It is possible that the splicing defects are more pronounced for pre-mRNAs of some essential gene(s) than for the model pre-mRNAs used in these studies. Indeed, the results of Figs. 2 and 3 are consistent with the possibility that the effects of SF1/mBBP depletion can be quantitatively different for different pre-mRNA substrates. Splicing defects upon ySF1/BBP depletion were also more readily noticeable in S. cerevisiae when reporters with mutated splice sites were analyzed.² Alternatively, the essential function(s) of SF1/BBP may be unrelated to pre-mRNA splicing. In line with the latter possibility, one SF1/mBBP isoform, ZFM1, was independently identified as a transcriptional repressor (43).

The low levels of ySF1/BBP required for viability and splicing of model substrates are particularly striking considering that S. cerevisiae cells in which transcription of the ySF1/BBP gene has been efficiently turned off can live for several generations with only some delay in their doubling time (33). Rutz and Séraphin (33) have interpreted their observations as revealing a transient interaction of ySF1/BBP with splicing complexes that kinetically facilitates their assembly. A similar transient interaction was also observed for MUD2p, the yeast homolog of U2AF⁶⁵. As yeast SF1/BBP and MUD2p interact with each other through protein domains conserved in their mammalian counterparts (25, 27), both proteins may bind to and leave the pre-mRNA as a complex, at least in the yeast system.

Depletion of mammalian SF1/mBBP and of U2AF⁶⁵, however, have dramatically different effects in spliceosome assembly. The relatively modest effects of SF1/BBP depletion are in contrast with the substantial loss of activity of extracts chromatographically depleted or immunodepleted of U2AF⁶⁵ (14, 21, 22, 24, 36, 47). This is despite the fact that the two proteins bind cooperatively to the same region of higher eukaryotic pre-mRNAs, act at a similar step in spliceosome formation, their levels in HeLa nuclear extracts are comparable, and the extent of the reduction of their concentration in the depleted extracts is also comparable (compare, for example, Fig. 3B in Ref. 36 with Fig. 1 in this work). The dissociation constants for SF1/BBP binding to the yeast consensus BP range between 0.5 (yeast) and 30 µM (human). The affinity of U2AF65 for typical polypyrimidine tracts is significantly higher, the correspondingly lower dissociation constants ranging between 0.1 and 0.01 µM. If the effect of the depletion would depend solely on the relative affinities of these factors for their cognate sites, one would expect that reduction in the levels of the factor that has higher binding affinity for its binding site would be less detrimental for activity. However, the opposite is observed. These observations may imply a more critical function for U2AF⁶⁵ than for SF1/mBBP in mammalian spliceosome assembly. Alternatively, they may reflect additional roles that U2AF⁶⁵ plays in spliceosome assembly and that are independent of SF1/mBBP. In any case, the results argue against the possibility that the recognition of the more degenerate mammalian BP requires a more substantial contribution of SF1/mBBP compared with that of the conserved branch point by ySF1/BBP in the yeast system.

The strong dependence on SF1/mBBP to reconstitute U2 snRNP binding from purified components (11) is in contrast with the modest effects that SF1/BBP depletion has in the same process in extracts (Figs. 2-4). It is possible that other factors present in the nuclear extract can mitigate the effects of SF1/mBBP depletion by either facilitating alternative pathways of assembly or by taking over SF1/mBBP function in the immunodepleted extracts.

A key and still unsolved question is the molecular mechanism(s) by which SF1/BBP promotes U2 snRNP binding and/or other events in spliceosome assembly. SF1/BBP was proposed to be involved in initial cross-intron bridging by interacting simultaneously with the branch point sequence and Prp40p, which is associated with U1snRNP bound to the 5' splice site (25). Both synthetic lethality in yeast and GST-pulldown experiments were compatible with this proposal (25). The proline-rich region of yeast SF1/BBP and the WW domain of PrP40p were suggested to mediate this interaction. Analogous interactions were proposed in the mammalian system, involving the WW domain of the formin-binding protein FBP21 that can interact with human SF1/BBP, the U1-specific polypeptide U1C, and the snRNP core components SmB/SmB' (44). Several results, however, have challenged the role of these interactions in the initial bridging between splice sites. First, domain mapping analysis showed that the proline-rich region of ySF1/BBP located in the carboxyl-terminal half of the protein was not required for binding to Prp40p (25). Second, FBP21 has been found in pre-spliceosome and spliceosome (A and B) complexes, probably associated to U2 snRNP, but not in commitment (E) complexes (44). Third, the proline-rich region was found to be dispensable for promoting mammalian U2 snRNP binding in reconstituted reactions from purified components (27) and in SF1/mBBP-depleted extracts (Fig. 2) and also for yeast growth (27).

Given that SF1/mBBP recognizes the nucleotide bases at the BP and that these bases also establish base pairing interactions with U2 snRNA, it seems unlikely that SF1/mBBP and U2 snRNA can be bound simultaneously at the BP sequence. In fact, previous data are consistent with SF1/mBBP being replaced from the BP upon U2 snRNP binding (33, 45, 46). SF1/mBBP could interact with a protein component of U2 snRNP or with U2 snRNA to bring the snRNP in close proximity to the BP, or to induce conformational rearrangements within the snRNP necessary for its recruitment, and then be displaced when extensive RNA-RNA and RNA-protein interactions stabilize snRNP binding to the BP. SF1/mBBP could act in this case as a "guide" molecule that identifies the correct sequence for U2 snRNP binding, a task particularly demanding in higher eukaryotes where the BP sequence is rather degenerate and several base pairing arrangements between U2 snRNA and the BP region are possible. This activity could be at the basis of the improved kinetics of U2 snRNP binding afforded by an efficient BP-SF1/BBP interaction observed in the experiments presented in Fig. 3.

Alternatively, SF1/mBBP could help (with the assistance of U2AF⁶⁵) to clear the BP region from factors associated with the pre-mRNA before its commitment to spliceosome assembly, like hnRNP proteins. In fact, Chiara et al. (46) have shown that hnRNPI/PTB can be cross-linked to the BP sequence of a model pre-mRNA in those complexes. These factors may have more relaxed binding specificity but represent, because of their associated domains, activities or factors, a significant block to U2 snRNP binding and therefore their substitution by SF1/BBP could result in effective clearance of the BP region for U2 snRNP binding, particularly in combination with the cooperative binding and recruiting activities of U2AF⁶⁵.

The interaction between SF1/BBP and U2AF⁶⁵ mutually facilitates their interaction with the BP and Py tract (19). One possible scenario is that SF1 simply assists U2AF⁶⁵ binding to the Py tract and that only U2AF⁶⁵ has a recruitment function for U2 snRNP. RNA binding experiments using purified proteins have shown that SF1/BBP increases the affinity of U2AF⁶⁵ 5-fold (19). This increase in affinity may be higher for pre-mRNAs containing weak Py tracts and more consensus BPs. The results of Fig. 4, however, argue that the function of SF1/BBP cannot be justified exclusively by its effects on U2AF⁶⁵ binding, and argue that additional mechanisms like those discussed above can indeed play a role in SF1/BBP function. It is also possible that the interaction of SF1/BBP with U2AF⁶⁵ enhances the recruitment activities of the latter, for example by triggering a conformational change that will allow more extensive contacts between U2AF⁶⁵ and U2 snRNP components like SAP 155 (23).

	FOOTNOTES

^* 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.

To whom correspondence should be addressed. Tel.: 49-6221-387-156; Fax: 49-6221-387-518; E-mail: juan.valcarcel@embl-heidelberg.de.

Published, JBC Papers in Press, August 22, 2000, DOI 10.1074/jbc.M001483200

² B. Rutz and B. Séraphin, personal communication.

	ABBREVIATIONS

The abbreviations used are: U2 snRNP, U2 small nuclear ribonucleoprotein particle; snRNA, small nuclear RNA; Py, polypyrimidine; GST, glutathione S-transferase; BP, branch point; mBBP, mammalian branch point binding protein; CAP, m7G(5')ppp(5'; hnRNP, heterogeneous nuclear RNP; CC2, commitment complex 2.

	REFERENCES

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS DISCUSSION REFERENCES

CiteULike

Complore

Connotea

Del.icio.us

Digg

Reddit

Technorati

This article has been cited by other articles:

				V. Sridharan and R. Singh A Conditional Role of U2AF in Splicing of Introns with Unconventional Polypyrimidine Tracts Mol. Cell. Biol., October 15, 2007; 27(20): 7334 - 7344. [Abstract] [Full Text] [PDF]

				M. Sanchez-Alvarez, A. C. Goldstrohm, M. A. Garcia-Blanco, and C. Sune Human Transcription Elongation Factor CA150 Localizes to Splicing Factor-Rich Nuclear Speckles and Assembles Transcription and Splicing Components into Complexes through Its Amino and Carboxyl Regions. Mol. Cell. Biol., July 1, 2006; 26(13): 4998 - 5014. [Abstract] [Full Text] [PDF]

				G. Kol, G. Lev-Maor, and G. Ast Human-mouse comparative analysis reveals that branch-site plasticity contributes to splicing regulation Hum. Mol. Genet., June 1, 2005; 14(11): 1559 - 1568. [Abstract] [Full Text] [PDF]

				G. Tanackovic and A. Kramer Human Splicing Factor SF3a, but Not SF1, Is Essential for Pre-mRNA Splicing In Vivo Mol. Biol. Cell, March 1, 2005; 16(3): 1366 - 1377. [Abstract] [Full Text] [PDF]

				H. BANERJEE, A. RAHN, B. GAWANDE, S. GUTH, J. VALCARCEL, and R. SINGH The conserved RNA recognition motif 3 of U2 snRNA auxiliary factor (U2AF65) is essential in vivo but dispensable for activity in vitro RNA, February 1, 2004; 10(2): 240 - 253. [Abstract] [Full Text] [PDF]

				S. G. Khan, A. Metin, E. Gozukara, H. Inui, T. Shahlavi, V. Muniz-Medina, C. C. Baker, T. Ueda, J. R. Aiken, T. D. Schneider, et al. Two essential splice lariat branchpoint sequences in one intron in a xeroderma pigmentosum DNA repair gene: mutations result in reduced XPC mRNA levels that correlate with cancer risk Hum. Mol. Genet., February 1, 2004; 13(3): 343 - 352. [Abstract] [Full Text] [PDF]

				B. K. Abu Dayyeh, T. K. Quan, M. Castro, and S. W. Ruby Probing Interactions between the U2 Small Nuclear Ribonucleoprotein and the DEAD-box Protein, Prp5 J. Biol. Chem., June 7, 2002; 277(23): 20221 - 20233. [Abstract] [Full Text] [PDF]

				M. Cattaruzza, K. Schafer, and M. Hecker Cytokine-induced Down-regulation of zfm1/Splicing Factor-1 Promotes Smooth Muscle Cell Proliferation J. Biol. Chem., February 15, 2002; 277(8): 6582 - 6589. [Abstract] [Full Text] [PDF]

				A. C. Goldstrohm, T. R. Albrecht, C. Sune, M. T. Bedford, and M. A. Garcia-Blanco The Transcription Elongation Factor CA150 Interacts with RNA Polymerase II and the Pre-mRNA Splicing Factor SF1 Mol. Cell. Biol., November 15, 2001; 21(22): 7617 - 7628. [Abstract] [Full Text] [PDF]

				S. Guth, T. O. Tange, E. Kellenberger, and J. Valcarcel Dual Function for U2AF35 in AG-Dependent Pre-mRNA Splicing Mol. Cell. Biol., November 15, 2001; 21(22): 7673 - 7681. [Abstract] [Full Text] [PDF]

				Z. Liu, I. Luyten, M. J. Bottomley, A. C. Messias, S. Houngninou-Molango, R. Sprangers, K. Zanier, A. Kramer, and M. Sattler Structural Basis for Recognition of the Intron Branch Site RNA by Splicing Factor 1 Science, November 2, 2001; 294(5544): 1098 - 1102. [Abstract] [Full Text] [PDF]

				H. Peled-Zehavi, J. A. Berglund, M. Rosbash, and A. D. Frankel Recognition of RNA Branch Point Sequences by the KH Domain of Splicing Factor 1 (Mammalian Branch Point Binding Protein) in a Splicing Factor Complex Mol. Cell. Biol., August 1, 2001; 21(15): 5232 - 5241. [Abstract] [Full Text] [PDF]

This Article Abstract Full Text (PDF) All Versions of this Article: 275/48/38059    most recent M001483200v1