Structure Probing of U4 snRNA: Investigations into the Mechanism of U4/U6 di-snRNP Formation by Tara Ann Wong B.Sc, University of Northern British Columbia, 2007 THESIS SUBMITTED IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF MASTER OF SCIENCE IN MATHEMATICAL, COMPUTER, AND PHYSICAL SCIENCES THE UNIVERSITY OF NORTHERN BRITISH COLUMBIA August 2010 ©Tara Ann Wong, 2010 1*1 Library and Archives Canada Bibliotheque et Archives Canada Published Heritage Branch Direction du Patrimoine de I'edition 395 Wellington Street OttawaONK1A0N4 Canada 395, rue Wellington OttawaONK1A0N4 Canada Your file Votre reference ISBN: 978-0-494-75144-2 Our file Notre reference ISBN: 978-0-494-75144-2 NOTICE: AVIS: The author has granted a nonexclusive license allowing Library and Archives Canada to reproduce, publish, archive, preserve, conserve, communicate to the public by telecommunication or on the Internet, loan, distribute and sell theses worldwide, for commercial or noncommercial purposes, in microform, paper, electronic and/or any other formats. L'auteur a accorde une licence non exclusive permettant a la Bibliotheque et Archives Canada de reproduire, publier, archiver, sauvegarder, conserver, transmettre au public par telecommunication ou par I'lnternet, preter, distribuer et vendre des theses partout dans le monde, a des fins commerciales ou autres, sur support microforme, papier, electronique et/ou autres formats. The author retains copyright ownership and moral rights in this thesis. Neither the thesis nor substantial extracts from it may be printed or otherwise reproduced without the author's permission. L'auteur conserve la propriete du droit d'auteur et des droits moraux qui protege cette these. Ni la these ni des extraits substantiels de celle-ci ne doivent etre imprimes ou autrement reproduits sans son autorisation. In compliance with the Canadian Privacy Act some supporting forms may have been removed from this thesis. Conformement a la loi canadienne sur la protection de la vie privee, quelques formulaires secondaires ont ete enleves de cette these. While these forms may be included in the document page count, their removal does not represent any loss of content from the thesis. Bien que ces formulaires aient inclus dans la pagination, il n'y aura aucun contenu manquant. 1+1 Canada ABSTRACT Precursor messenger RNA splicing is catalyzed by a dynamic complex termed the spliceosome. An essential step in spliceosome assembly occurs when U4 snRNA base pairs with U6 snRNA to form the U4/U6 di-snRNP. Although failure to form the di-snRNP is lethal, the mechanism by which di-snRNP formation occurs is still unknown, in part due to lack of information about the structure of free U4 snRNP prior to binding U6. To investigate the contribution of U4 in di-snRNP formation, I carried out a series of structure probing experiments to determine the secondary structure of a Brr2 released U4 snRNP. The structural model establishes the presence of four stem loops in yeast U4 snRNA, including a novel short stem loop at the extreme 5' end of the molecule. To determine which nucleotides of U4 are required for base pair formation, I carried out a modification/interference experiment. Modification of the 5' stem loop uridines (U5, U6, and U8) interfered with disnRNP formation, while modification of uridines within the central and 3' regions of U4 snRNA did not inhibit di-snRNP formation. Based on these results, I propose that intermolecular base-pairing between U4 loop nucleotides (U6 - Al 1) of the novel stem loop and U6 snRNA nucleotides (U70 - A75) may initiate di-snRNA formation. The U6 specific protein Prp24 would catalyze the subsequent annealing and stabilization of the U4/U6 intermolecular helices I and II. u TABLE OF CONTENTS Abstract ii Table of Contents iii List of Tables v List of Figures vi Acknowledgement viii Chapter One - Introduction 1 1.1 Pre-mRNA Splicing 1.2 U6 snRNP 1.3 U4 snRNP 1.4U4/U6di-snRNP 1.5 Kissing Loop Model of U4/U6 di-snRNP Formation 1.6 Overall Research Objective Chapter Two - Generation of Free U4 snRNP via the Brr2 Release Preparation 1 4 7 9 11 13 15 2.1 Materials and Methods 16 2.2 Results and Discussion 19 Chapter Three - Structure Probing of Brr2 Released U4 snRNP 23 3.1 Materials and Methods 23 3.2 Results and Discussion 26 Chapter Four - Genetic Analysis of U4 at the C12 Position 36 4.1 Materials and Methods 36 4.2 Results and Discussion 40 Chapter Five - Modification/Interference Analysis with CMCT 47 5.1 Materials and Methods 48 5.2 Results and Discussion 49 Chapter Six - Future Directions and Concluding Remarks 6.1 Future Directions 6.2 Concluding Remarks 59 59 62in Works Cited ix IV LIST OF TABLES Table 1. Doubling time and standard deviation determined from growth curves of wild-type and mutant SNR14 yeast strains grown at 37°C, 30°C, andl6°C. 44 Table 2. F-test results for each of the mutant strains compared to wild-type. 44 Table 3. Un-paired t-test results for determination of doubling time differences between mutant and wild-type strains. 45 Table 4. Interference Index used to determine if a modified nucleotide inhibits di-snRNP formation. 55 Table 5. Interference Index calculated from two independent experiments for U5, U6,U8, U19, andU25. 56 LIST OF FIGURES Figure 1. The excision of an intron from pre-mRNA by means of two consecutive transesterification reactions. 2 Figure 2. Spliceosome assembly and catalysis occurs in an ordered and sequential manner. 3 Figure 3. Secondary structure models of yeast U6 snRNA in free U6 snRNP. 6 Figure 4. Chemical modification and enzymatic cleavage data of Saccharomyces cerevisiae U4 snRNA phenol/chloroform extracted from U4/U6.U5 triple-snRNP. 9 Figure 5. Proposed secondary structure for yeast U4/U6 di-snRNA. 10 Figure 6. The short kissing stem loop can be formed at the 5' end of U4 snRNA from the organisms Saccharomyces cerevisiae, Homo sapiens, Caenorhabditis elegans, Viciafaba, and Drosophila melanogaster. 12 Figure 7. The kissing loop interaction may initiate formation of U4/U6 13 intermolecular base-pairing. Figure 8. The Brr2 release preparation generates free U4 snRNP. 20 Figure 9. Brr2 release with 2mM ATP gives the best enrichment for free U4 snRNP. Figure 10. Chemical structure probing of Brr2 released U4 snRNPs with DMS and CMCT supports the novel 5' stem loop. 22 28 Figure 11. Enzymatic probing of Brr2 released U4 snRNPs with RNase Vi and RNase A corroborates the novel 5' stem loop. 30 Figure 12. Yeast strains with HIS3 marked plasmids containing wild-type or 41 mutant SNR14 grow on -his and 5-FOA, but not on -ura selective media. Figure 13. Wild-type and mutant growth phenotypes assayed by dot dilution. 42 Figure 14. Northern blots of total RNA prepared from wild-type and mutant 46 SNR14 strains indicate a lack of U4 accumulation. Figure 15. Purification of 10His-taggedPrp24. 50 Figure 16. lOHis-tagged Prp24 anneals Brr2 released U4 and U6 snRNPs. 51 vi Figure 17. Addition of CMCT impairs di-snRNP formation. 51 Figure 18. Modification/Interference analysis with CMCT reveals that modification of U5, U6, and U8 interferes with di-snRNP formation. 54 vn ACKNOWLEDGEMENT I would like to thank my parents, Karen and Hui Wong, my brother, Travis, and my fiance, Trevor Lipinski, for their ongoing support and encouragement throughout my academic endeavors. I am very grateful to Stephen Rader for allowing me the opportunity to do graduate research in his lab. I would like to acknowledge Dr. Cecilia Alstrom-Rapaport, Dr. Daniel Erasmus, and Dr. Chow Lee for their mentorship as committee members. I would also like to thank all past and present members of the Rader lab for their help and support. In addition, I would like to thank Dr. Chow Lee and members of the Lee lab for the kind sharing of equipment. I gratefully acknowledge the financial support from the Michael Smith Foundation for Health Research, the Natural Sciences and Engineering Research Council of Canada, and the University of Northern British Columbia. VI11 Chapter One - Introduction Precursor messenger RNA splicing (pre-mRNA splicing) is the process by which non-coding introns are removed from transcribed pre-mRNA and coding exons are ligated to form mature mRNA, a template for protein synthesis. It is estimated that over 90% of human genes contain introns, and on average a gene is thought to contain 8.8 exons and 7.8 introns (Jurica 2008, Sakharkar et al. 2004). Thus in humans and other eukaryotes pre-mRNA splicing plays an important role in gene regulation and proteome diversification, and it is therefore important to understand how splicing is accomplished. Of the five small nuclear ribonucleoprotein particles (snRNPs) that help to facilitate the removal of introns, U4 is the least well characterized. Due to the difficulty in purifying free U4 snRNP, the RNA secondary structure as well as the protein composition of this snRNP has not been determined biochemically. It is known that U4 snRNA must first base pair to U6 snRNA, then subsequently dissociate for splicing to occur, but it is unknown how these interactions are initiated and regulated. A major goal of this thesis was to determine the secondary structure of the U4 snRNA in the free snRNP form as an effort towards understanding how U4 can interact with and regulate U6. 1.1 Pre-mRNA Splicing Pre-mRNA splicing is catalyzed by a large and dynamic complex known as the spliceosome, and occurs via two consecutive transesterification reactions (Figure 1). In the first, the 2' hydroxyl group of the bulged branchpoint adenosine attacks the 5' phosphoryl group of the 5' exon/intron junction (5' splice site). The result is the simultaneous formation of a free 5' exon and a lariat intron/exon 2 intermediate. The second step involves an attack by the 3' hydroxyl group of the free 5' exon on the 5' phosphoryl group of the 3' exon/intron 1 junction (3' splice site), resulting in ligation of the two exons and removal of the intron as a lariat (Ruskin et al. 1984, Padgett et al. 1984, Domdey et al. 1984). The spliceosome is composed of five snRNPs and additional protein factors. Each snRNP consists of a uridine rich small nuclear RNA (snRNA) designated U l , U2, U4, U5, or U6, and a set of specific and tightly associated proteins. Assembly of a fully functional spliceosome and excision of an intron involves both the formation of snRNP particles, and the association and rearrangement of snRNPs and other protein components with respect to the pre-mRNA. First Transesterification o-p-oGU 0- A AG o|o Exon 2 A AGogo Exon 2 I o- Second Transesterification A -OH Lariat Intron Exon 1 Exon 2 Mature mRNA Figure 1. The excision of an intron from pre-mRNA by means of two consecutive transesterification reactions. The 5' and 3' consensus nucleotides are indicated along with the branchpoint adenosine. Spliceosome assembly is initiated when Ul snRNP recognizes and base pairs to the 5' splice site of the pre-mRNA (Ruby and Abelson 1988; Figure 2). Concomitantly, U2 snRNP recognizes and binds to the branch point sequence, resulting in the branchpoint adenosine bulge. The U6 and U4 snRNPs base pair to form the U4/U6 di-snRNP, which subsequently interacts with U5 to form the U4/U6.U5 triple snRNP. The U4/U6.U5 triple snRNP is then integrated into a precatalytic complex where numerous RNA: RNA, RNA: protein, and protein: protein rearrangements occur to form the catalytic complex (Cheng and 2 Abelson 1987). During these structural rearrangements the Ul snRNP dissociates from the 5' splice site while the U4 snRNP is unwound from the U6 snRNP, allowing U6 and U5 to base pair with the 5' splice site, and U6 to base pair with U2 snRNA (Brow 2002). Presently, it is unclear whether Ul and U4 snRNPs remain loosely associated with the activated spliceosome or whether they dissociate entirely from this complex, but splicing has been shown to occur in the absence of these snRNPs (Cheng and Abelson 1987, Yean and Lin 1991). The resultant active spliceosome consisting of U2, U5, and U6 snRNPs facilitates the transesterification reactions. From the current model of spliceosome assembly it can be seen that the snRNPs are structurally dynamic, and as the spliceosome is thought to be assembled de novo onto each intron that is removed, recycling of the snRNPs is critical. Figure 2. Spliceosome assembly and catalysis occurs in an ordered and sequential manner. 3 1.2 U6 snRNP U6, the most highly conserved snRNA, is essential and is believed to play a central role in the catalysis of splicing (Brow and Guthrie 1988, Valadkhan and Manley 2001, Yean et al. 2000). In splicing extract, U6 is thought to be largely present in free snRNP form (-80%), where it is associated with the proteins Lsm2-Lsm8, and Prp24 (Li and Brow 1993, Jandrositz and Guthrie 1995, Achsel et al. 1999, Shannon and Guthrie 1991). Before incorporation into the active spliceosome, U6 must first base pair with U4 to form the U4/U6 di-snRNP. Di-snRNP formation is catalyzed by the U6 snRNP protein Prp24, and results in conformational changes in both U6 and U4. The conversion of free U6 snRNP into a disnRNP is necessary for spliceosome assembly and subsequent catalysis, but the mechanism by which these conformational changes occur is still unknown. To understand how di-snRNP initiation can occur between free U6 and free U4 snRNPs, the secondary structures of the free snRNPs must first be determined. Although a number of structural models for both the yeast and mammalian U6 snRNA in the free snRNP form have been proposed (Fortner et al. 1994, Vidavcr et al. 1999, Ryan et al. 2002, Karaduman et al. 2006, Dunn and Rader 2010; Figure 3), and the RNA secondary structure has been characterized via chemical modification several times as both free RNA (Mougin et al. 2002), and in the free U6 snRNP both in vitro (Jandrositz and Guthrie 1995, Karaduman et al. 2006) and in vivo (Fortner et al. 1994), there is still debate over the structure of free U6 snRNP. Until recently, published models of free U6 snRNP mostly agreed on the presence of 5' and 3' intramolecular stem loops (ISL), with discrepancies in the central portion of the models (Figure 3). Support for the 3' ISL comes from dimethylsulfate (DMS) probing in 4 vivo (Former et al. 1994). DMS can diffuse into living cells and methylate adenines and cytosines of single-stranded RNA. Two of the possible three nucleotides in the 3' loop were strongly modified by DMS, providing support for a free loop structure in U6 encompassing nucleotides G71 - A75. However, modification patterns obtained from this experiment cannot be unequivocally assigned to the free U6 snRNP, as U6 undergoes multiple interactions including formation of the U4/U6 di-snRNP, and the U6/U2 catalytic core where the 3' ISL is thought to be a critical structural feature. U6 snRNP purified by TAP-tagged Prp24 and glycerol gradient centrifugation was analyzed by chemical modification and found to have a more accessible structure compared to in vitro transcribed U6 snRNA, especially in the bulge and 3' stem loop regions (Karaduman et al. 2006). The results showed that Prp24 and the Lsm proteins induce a conformational change in the U6 snRNP that allows the 3' stem loop and positions A79 and U80 to become available for interaction with U4 snRNA. This highlights the importance of protein binding inducing a conformational change in an snRNA. Also, this is in agreement with a number of biochemical studies suggesting that the Lsm proteins on the 3' end of U6 snRNA play a key role during U4/U6 biogenesis (Achsel et al. 1999, Vcrdone et al. 2004), and that Prp24 is necessary for di-snRNP formation (Raghunathan and Guthrie 1998). Hydroxyl radical probing of the U6 snRNP suggested that the binding site of Prp24 occurs over the range of nucleotides 4-60 encompassing the 5' stem loop, but not the 3' stem loop, and UV-crosslinking indicated that Prp24 is bound to nucleotides 28, 29, 38, and 55 (Karaduman et al. 2006). Similarly, structure probing of glycerol gradient purified U6 snRNP indicated that Prp24 was likely bound to nucleotides A40-C43 (Jandrositz and 5 Guthrie 1995). However, it is unclear how Prp24 helps to facilitate the conformational rearrangements necessary for di-snRNP formation. In spite of past models for the free U6 snRNP suggesting the presence of a 3' ISL, the secondary structure of the snRNA was not conserved among species, which lead to the proposal of a novel U6 secondary structure which is conserved among all known species (Dunn and Rader 2010; Figure 3). In the Dunn/Rader model the 3' ISL has been dissolved, and instead forms two intramolecular helices denoted Stem/loop A and B. Further experiments are needed to resolve the structure of the free U6 snRNP, however all models propose the U6 nucleotides C72, A73, U74, and A75 (S. cerevisiae numbering) to be unpaired in the free U6 as either an apical loop or a bulge, and consequently available for initial base pairing with the U4 snRNP. A Stem/loop A C " U G A U —A C —G C —G R u A G c ^- » U U U A" C G—CC U A A C Stem/loop E G A A A .A G.- - c. HA u u uu, ur G • '/ A A C C U c u-- A U-- A G-- c 5'Stem/loop A- ~ u u-„ A Central Stem A- - u A • G G-- c A A c-- G c-- G . U G c-- G u-- A G. U U • G G• U u-- A UAUUUCGUUUU3' .' G-- C A U U U Ounn-Rader2010 A G U A G A G A C A *Au C —G U-A U-A uCAG C AG-C A C G U U U u uA A C A A A G - C u 3'Stem/loop A C c A U —A U —A U —A G —C 5'Stem/loop A — U A G A —U A A & —C C G C —G G• U U A C —G G U U »G G U U—A 5' G—CAUUU UAUUUCGUUUU3' Karaduman-Luhrmann 2006 Figure 3. Secondary structure models of yeast U6 snRNA in free U6 snRNP. A) New model of free U6 snRNA secondary structure proposed by Dunn and Rader, 2010. B) Main model of free U6 snRNA secondary structure accepted as of 2009. 6 1.3 U4 snRNP Contrary to the free U6 snRNP, in splicing extract the predominant species of U4 is found complexed to U6. Although U4 is essential, little is known about the structure and composition of this particle due to the minimal amounts of free U4, and the difficulty in isolating and purifying free U4 snRNP. It is unknown if any snRNP specific proteins associate with free U4, and if so, whether or not these proteins stabilize a particular conformation of the RNA. Currently it is believed that the U4 snRNP may function as a negative regulator of U6 by binding to and sequestering catalytically important regions of U6 snRNA (Brow and Guthrie 1988). It is also possible that U4 helps to activate U6 by positioning U6 in the pre-catalytic spliceosome in a manner that helps to facilitate the conversion to an active spliceosome (Dunn and Rader 2010). Although, no snRNP specific proteins have been identified in the free U4 snRNP, it is possible that the U4/U6 specific proteins may be brought to the di-snRNP complex by association with the free U4 snRNP. In humans, proteins 15.5K, 61K, 20K, 60K, and 90K have been shown to interact with the U4/U6 di-snRNP in a hierarchical manner (Nottrott et al. 2002). Except for the 20K protein, which does not have a homologuc in yeast, the proteins Snul3 (15.5K), Prp4 (60K), Prp3 (90K), and Prp31 (6IK) have also been shown to interact with the yeast U4/U6 particle. Analysis of the interactions of these proteins in humans has led to the observation that 15.5K binding to the U4 5' kink turn stem loop is necessary for the subsequent binding of 6IK and the triple protein complex of 20/60/90K (Nottrott et al. 1999, Nottrott et al. 2002, Schultz et al. 2006). In addition, binding of 20/60/90K appears to occur through interactions of the 90K protein with stem II of the U4/U6 duplex (Nottrott et al. 2002). In humans all of the U4/U6 di-snRNP specific proteins 7 are destabilized or released in the rearrangement to an activated spliceosomal complex (Makarov et al. 2002). These proteins may be released bound to U4 snRNA, and given the putative binding site locations in the di-snRNP, 15.5K (Snul3) and 61K (Prp31) appear to be the strongest candidates for binding the U4 snRNA in the free snRNP. It is known that the U4 snRNA interacts with the Sm proteins B/B' (alternatively spliced protein products that differ by only 11 residues at the C-termini), Dl, D2, D3, E, F, and G. The Sm proteins are common to U l , U2, U4 and U5 snRNAs, and these proteins interact with a uridine rich Sm binding site, which in U4 is located at the 3' end of the snRNA (Reviewed in Nagai et al. 2001). Anti-Sm antibodies have been shown to immunoprecipitate U4/U6 and U4/U6.U5 particles, but the Sm proteins do not bind directly to the U6 snRNA in the free U6 snRNP (Luhrmann et al. 1990, Seraphin 1995). Therefore they are not brought to the U4/U6 particle by free U6, and may be associated with the free U4 snRNP, or associate after U4/U6 formation. The structure of phenol/chloroform extracted U4 snRNA from yeast U4/U6.U5 trisnRNP has been analyzed by chemical and enzymatic probing, and fit onto a previously proposed secondary structure by the Branlant lab (Mougin et al. 2002, Krol and Branlant 1981, Myslinski and Branlant 1991; Figure 4). The data shows support for a 5' (kink-turn) stem loop structure encompassing nucleotides 13-60, a central stem loop (61-82), and a 3' stem loop structure (91-142). Human and other eukaryotic U4 snRNAs can also adopt this secondary structure confirmation, and many eukaryotes encompass an additional stem loop 3' of the Sm binding site that is not present in yeast. Notably, the structure of the first 10 nucleotides has not been analyzed by structure probing techniques, and this region has been proposed to be single stranded, despite the high degree of sequence conservation (Myslinski 8 and Branlant 1991). In addition, the structure of U4 snRNA has yet to be analyzed in the snRNP form, and it is conceivable that association with proteins might stabilize an alternate conformation of the RNA. A. 0 A c-e A U4 RNA || M> Mf6"< /* AAu , f m * J 'GpppAUCCUOWjstAC *R!yG glycerol, xylene cyanol, bromophenol blue) for Northern blot analysis. 5'End Labeling ofDNA Oligonucleotides Two and a half microliters of 10X T4 Polynucleotide Kinase Buffer (NEB) and 14.8uL of ddF^O was added to 7 picomoles of DNA oligonucleotide. Five microliters of [yP] ATP (PerkinElmer) was added to this along with 20 units of T4 Polynucleotide Kinase (NEB) to a total volume of 25uL and incubated at 37°C for one hour. The reaction was 17 diluted to 50LIL with ddHbO and unincorporated [y- P] ATP was removed with a G25 or G50 microspin column (GE Healthcare) according to the manufacturer's instructions. Northern Blot Samples were run on a pre-chilled 4.5% non-denaturing gel at 400V for 40 minutes at 4°C. The gel was then transferred to Amersham Hybond-N+ nylon transfer membrane (GE Healthcare) for 15 minutes at 450mA (32mA per cm2) using a semi-dry electroblotter (Owl Panther Hep-3). The RNA was crosslinked to the membrane in a UV Stratalinker 1800 (Stratagene) with 120,000 J of ultraviolet radiation. The membrane was then blocked with lOmL of Rapid-Hyb Buffer (GE Healthcare) for at least 30 minutes at room temperature, followed by incubation with labeled DNA oligonucleotide (14b) for at least 1 hour at room temperature. The membrane was washed 3 times for 3 minutes with lOmL of Wash Buffer (6X SSC, 0.2% SDS), and then exposed to a phosphor screen (PerkinElmer) overnight. The autoradiogram was visualized and quantified with Cyclone © phosphorimager and OptiQuant © Software (Packard Instruments). Solution Hybridization (Brow Gel) Following RNA extraction samples were resuspended in 9uL of ddP^O and luL of 10X Brow hybridization buffer (50mM Tris-HCl pH 7.5, 150mM NaCl, ImM EDTA), incubated with 100,000 counts of labeled oligonucleotide 14b, and heated to 42°C for 8 minutes. Two microliters of 4X native gel loading buffer was added to the sample and loaded onto a pre-chilled 6% non-denaturing gel, run at 300V for 45 minutes at 4°C, and exposed and quantified (as above). Labeled Oligonucleotide Usedfor Northern Blots and Solution Hybridization Gels 14b: 5' - AGGTATTCCAAAAATTCCCTAC - 3' (binds U4 nucleotides 158 - 137) 18 2.2 Results and Discussion In the Brr2 release, U4/U6.U5 triple-snRNPs and possibly higher orders of snRNP complexes (that may include Ul and U2 snRNPs) bind to the resin through an interaction with the polyoma tagged Brr2 and the monoclonal polyoma antibodies linked to Protein G Sepharose resin. Binding of these snRNP complexes to the resin is characterized by decreased amounts of U4/U6 di-snRNP in the flowthrough (material that has not bound to the resin) fraction. After washing the resin to disrupt proteins and RNA that bind nonspecifically, ATP is used to promote dissociation of the di-snRNP via Brr2. Elution fractions should ideally contain only free U4 and free U6, and any remaining di-snRNP should be retained on the resin. If the Brr2 release is optimized to generate maximal amounts of free U4 and U6 snRNPs, only a small amount of di-snRNP should remain associated with the resin. Extracted RNA from splicing extract, flowthrough, elution fractions and the remaining resin can be visualized by means of Northern blot, or solution hybridization. For Northern blots, the extracted RNA is run on a non-denaturing gel, transferred to a membrane, and the membrane is probed with labeled oligonucleotides complementary to U4 or U6. In solution hybridization, the labeled oligonucleotide is added to a fraction prior to separation on a non-denaturing gel. The advantage of a Northern blot is that the membrane can be stripped of labeled oligonucleotide, and re-probed with different oligonucleotides complimentary to another species, allowing for easy determination of different molecules in the same fraction. Solution hybridization is a less time consuming technique, and I have found solution hybridization to be a more sensitive technique for visualizing small amounts of RNA. 19 A) B) fmol IVT U4 5 SE FT El E2 RES IVT U4 10 40 U4/U6 U4 %U4 V*. WJp 93 Figure 8. The Brr2 release preparation generates free U4 snRNP. A) Schematic diagram of the Brr2 release preparation. Ul and U2 snRNPs that may or may not associate with the U4/U6.U5 triple-snRNP are indicated with a dotted outline. B) Solution hybridization gel probed for U4 with labeled oligonucleotide 14b, with 1.5% of the total loaded in each lane. (SE - splicing extract, FT - flowthrough, El and E2 - elutions, RES - resin). The Brr2 release preparation consistently enriched for free U4 snRNP in the elution fractions compared to splicing extract (Figure 8). A small amount of U4/U6 di-snRNP was found to be present in the elutions, likely from leaching off of the resin. At maximum the U4/U6 di-snRNP accounted for 16% of the U4 snRNP species present in an elution, but it 20 was frequently found to be between 2% and 10%. Any elution that was found to contain more than 16% U4/U6 di-snRNP was not used for subsequent structure probing. A number of Brr2 release conditions were tested to attain elution fractions with a high concentration of free U4 snRNP. While addition of GTP is thought to stimulate the ATPasc activity of Brr2 via U5 associated GTP binding protein Snul 14 (Small et al. 2006), the addition of GTP to a final concentration of 2mM along with ATP did not reproducibly enhance the amount of free U4 snRNP generated (Figure 9A). Moreover, the amount of U4/U6 di-snRNP present in the elution fractions was found to be higher than the ATP only elution fractions. Alternatively, increasing the ATP concentration in the elution mix to lOmM resulted in a significant amount of U4/U6 leaching off of the column (Figure 9B). In addition, a band that migrates faster than U4/U6, but slower than U4 is also enhanced in the lOmM ATP elution. While the identity of this band has not been confirmed, it is also present when probing for U6, indicating that this may be a di-snRNP with a truncated U4 or U6 snRNA. Varying other assay conditions including tRNA concentration, wash volume and length of washes, and incubation time with the resin did not improve U4 snRNP release compared to the conditions described above (Section 2.1; data not shown). I found that for this particular assay, 2mM of ATP in the elution mix gives the best enrichment of U4 snRNP, with the smallest amount of U4/U6 di-snRNP species. 21 (A) (B) ATP/GTP IVT SE FT El U4 ATP E2 RES FT El E2 RES IT (mM) IVT El U4 10 2 E2 El E2 '/i U4/U6 U4/U6 U4 •#* !*#» I * J ** Mi U4 Ml ui Ml % U4/U6 % U4/U6 15 15 1 1 Hi 4 it* » 75 75 14 12 4 Figure 9. Brr2 release with 2mM ATP gives the best enrichment for free U4 snRNP. A) Northern blot of a Brr2 release with both 2mM GTP and 2mM ATP, or with 2mM of ATP only. B) Northern blot of a Brr2 release with lOmM or 2mM of ATP. (SE - splicing extract, FT - flowthrough, El and E2 - elutions, RES - resin). 22 Chapter Three - Structure Probing of Brr2 Released U4 snRNP To gain insight into the mechanism of di-snRNP formation, the U4 snRNA secondary structure in the free snRNP must first be determined. Single stranded regions of RNA can be differentiated from double stranded regions with the use of chemical modifiers that modify nucleotide bases at Watson-Crick base pair positions (Moazed et al. 1986). l-cyclohexyl-3(2-morpholinoethyl) carbodiimide metho-p-toluene sulfonate (CMCT) can be used to alkylate N3 of uridine, and dimethylsulfate (DMS) can be used to methylate Nl of adenine, and N3 of cytosine. The modifiers cannot react with bases that are hydrogen bonded at these positions; therefore double stranded RNA is protected from chemical modifiers. To verify the results of chemical modification, ribonucleases that cleave RNA in double or single stranded regions can be used. Double strand specific ribonuclease Vi has been shown to cleave the RNA backbone of nucleotides hydrogen-bonded in Watson-Crick or wobble base-pairing, including G-U wobble base pairs (Lockard and Kumar 1981), and ribonuclease A has been shown to cleave phosphate bonds 3' of single-stranded pyrimidines (Reviewed in Raines 1998). This chapter summarizes the structure probing experiments with DMS, CMCT, RNase V,, and RNase A in a model of free U4 snRNA. 3.1 Materials and Methods Chemical Modification The chemical modifiers used were dimethylsulfate (DMS; Sigma) and 1-cyclohexyl3-(2-morpholinocthyl) carbodiimide metho-p-toluene sulfonate (CMCT; Sigma). Fifty microliters totaling approximately 7.5fmol of Brr2 released U4 snRNP was used for structure probing experiments. Modification with DMS was carried out in a total volume of 200uL 23 with HNM buffer (80mM K-HEPES pH 7.9, 20mM MgCl2, 0.3M NaCl) in the presence of lOug of yeast tRNA. The reaction was allowed to proceed in the presence of 2 to 100(iL of 10.6M DMS (neat) for 50 minutes on ice, and stopped with 50uL of DMS stop buffer (1M Tris-HCl pH 7.5, 0.1M EDTA, 1M B-mercaptoethanol) and 650|uL of 100% ethanol. The RNA was pelleted and washed with 70% ethanol, resuspended in 500uL of resuspension buffer (0.3M NaOAc, 0.5% SDS, 6mM EDTA), and extracted before primer extension analysis. Modification with CMCT was carried out in a total volume of 200uL with BKM buffer (50mM borate/KOH pH 8.0, 50mM KC1, lOmM MgCl2) in the presence of 5ug of yeast or E. coli tRNA at pH 8.0. The amount of CMCT added to the reaction ranged from 20uL to 50uL of 0.4M CMCT, and the reaction was carried out for 5 minutes at room temperature. The reaction was stopped with 800uL of 100% ethanol, and the RNA was precipitated and washed with 70% ethanol, then resuspended in 500uL of resuspension buffer for RNA extraction. Ribonuclease Reactions The enzymes used were Ribonuclease Vi (RNase Vi; Ambion) and Ribonuclease A (RNase A; Sigma). Fifty microliters totaling approximately 7.5fmol of Brr2 released U4 snRNP in the presence of 5|ig of tRNA was used for enzymatic probing experiments. Both RNase A and RNase Vj reactions were carried out in a total volume of 200|LIL TNM buffer (30mM Tris-HCl pH 7.5, 300mM NaCl, 20mM MgCl2). RNase Vi reactions were carried out in the presence of 1 JJ.L of O.lU/uL RNase Vi at room temperature for 3 to 15 minutes. RNase A reactions were carried out in the presence of 3.4uL of 5 x 10 4g/L RNase A (> 70 Kunitz units/mg protein) at room temperature for 3 to 15 minutes. The reactions were stopped by the addition of 150uL of 0.3M NaOAc, and 650uL of 100% ethanol. The RNA 24 was precipitated, washed with 70% ethanol, resuspended in 500uL of 0.3M NaOAc and phenol extracted before primer extension analysis. Primer Extension Phenol extracted RNA samples were resuspended in 8uL of water, and brought to a total volume of 12uE with luL of lOmM dNTPs, luL of lOmg/mL tRNA, and 2uL of labeled primer. Samples were incubated at 65°C for 6 minutes and subsequently cooled on ice for 2 minutes. Eight microliters of RT mix (0.5uL of AffmityScript Reverse Transcriptase (Stratagene), 0.5uL of SUPERase-In, 2uL of 10X AffinityScript buffer, luL of 0.1M DTT, and 4uL of ddH 2 0) was added to the sample, and the sample was incubated at 37°C - 52°C for at least 1 hour (dependent on the Tm of the labeled primer). Reactions were stopped with 150uL of TE (lOmM Tris-HCl pH 7.5, ImM EDTA), 20uL of 3M NaOAc, 0.5uL of (20mg/mL) glycogen, and 500uL of 100% ethanol. Samples were precipitated and washed with 70% ethanol, resuspended in 5uL of 2X formamide loading buffer (deionized formamide, 50mM EDTA, 2X TBE, xylene cyanol, bromophenol blue), and heated to 95°C for three minutes before they were run on an 8%/7M denaturing sequencing gel. Dideoxy sequencing lanes contained 25 fmol of IVT U4 (T7U4 with two extra G nucleotides at the 5' end), and were treated identical to sample reactions except that the RT mix contained 1 uL of 20mM ddNTP per lane. Oligonucleotides Used for Primer Extension of U4 snRNA 675: 5' - AAAGGTATTCCAAAAATTC - 3' (U4 nucleotides 160-142) 14b: 5' - AGGTATTCCAAAAATTCCCTAC - 3' (U4 nucleotides 158 - 137) SSU4: 5' - ACCATGAGGAGACGGTCTGG - 3' (U4 nucleotides 100 - 8 1 ) CM6: 5' - TCAACCAGCAAAAACACAATCTCG - 3' (U4 nucleotides 66 - 43) 25 CM2: 5 ' - C T C G G A C G A A T C C T C A C - 3 ' (U4 nucleotides 46 - 30) 3.2 Results and Discussion Chemical modification and enzymatic cleavage were optimized such that on average only one modification or cleavage should be present per snRNA molecule. This was accomplished by titration of the probes to a suitable amount, and with the addition of tRNA as another substrate for modification or cleavage. DMS structure probing was performed on ice because a low temperature was thought to minimize conformational changes. However structure probing of the Brr2 released U4 with CMCT did not result in any modifications when the reaction was performed on ice. In addition, the ribonucleases are not optimal below 4°C (Ambion; Sigma), thus the structure probing experiments with CMCT, RNase Vi, and RNase A were carried out at room temperature. Structure probing of U4 with DMS, CMCT, RNase Vi, and RNase A shows strong support for the three stem loops (3' stem loop, central stem loop, and the 5' kink-turn stem loop) that were previously proposed for yeast U4 snRNA (Myslinski and Branlant 1991, Mougin ct al. 2002). In addition, the modification pattern indicates the presence of a short stem loop at the extreme 5' end of U4 snRNA (Figure 10). This stem loop, which I will refer to as the 5' kissing stem loop of U4, is thought to consist of four base pairs encompassing nucleotides G15 - C12 and U2 - U5, with 6 nucleotides forming an apical loop. Stem nucleotides C4 and C12 are protected from modification, C3 is very weakly modified, and U5 is weakly modified. In addition, a fifth base pair may transiently form between the very weakly modified Al 1 and U6. Loop nucleotides CIO and U8 are strongly modified, demonstrating the availability of these nucleotides for an initial interaction with U6. This is 26 further supported by RNase A cleavage 3' of loop nucleotides TJ6, U8, and CIO, indicating that this region is indeed single stranded (Figure 11). Enzymatic cleavage with RNase Vi supports the presence of the 5' kissing stem loop with moderate cleavages on the 5' side of the stem, and weak cleavages on the 3' side. The 5' kissing stem is a short stem structure, but the minimum substrate size for RNase Vi is thought to be between 4 and 6 nucleotides (Lowman and Draper 1985), so cleavage in this region is credible. RNase Vi is thought to recognize a substrate that is in an approximately helical conformation, and it does not require that the nucleotide bases be hydrogen bonded in a helix (Lowman and Draper 1985). Therefore stabilization of a helical arrangement near this stem, like the potential base pair A l l - U6, might further contribute to the cleavage site for RNase Vi. Notably the observed Vi cuts in the 3' and central regions of U4 (3' of positions A122, Ul 12, Ul 11, U109, UlOO, U97, U63, and U49 ) were also found in the free phenol/chloroform extracted U4 RNA by Mougin et al. 2002. The fact that I also observe these cuts indicate a similarity in structure of the 3' and central regions of U4 regardless of the presence of proteins, and serve to validate the Vi cleavages I have observed at the 5' end. 27 U G U DMS(umol) G C A CMCT (umol) 20 iisft CIO CIO All L12 « m mm • -» «* L19 A16 A17 • A20** U G C A DMS (umol) 53 106 - 29A 33A w G40 2SC CMCT(umol) 16 8 36A U19 U25 59C U60 68A 67A 73A 72A lu LM8 U41 U49 86C 87t 90C 91C 95C U64 + U73 102 A 103 A UR9 U91 U94 U97 U100 121A 122A 124A 12SC 128C 129A 130A U63 L69 U 0 U7I UI00 m» " * liiffit 'fit** UJ7 U 1 U54 Lb7 U5fi U60 S(,A M •*> 97r >2196A IP7 ini 22 2T 8 8 Ullf U117 U118 Ul 19 LT4 B) 5' Kissing Loop (U6- A l l ) 150 90 3'-UUUCCAUAAGGUUUUUAAG—VC 160 ,. W 20 G—VC, 0-GS3U)UO-(A;U(AISA)G . UA- ' _ G_0 0»G 50G—© M—0U 0-[u! Steml G—C A- @ ^-JQ /TJ\ 5'(kink-turn) Stem Loop U • G 3' Stem Loop U • Giio [AH0 Figure 10. Chemical structure probing of Brr2 released U4 snRNPs with DMS and CMCT supports the novel 5' stem loop. A) Chemical modification of accessible nucleotide bases in the free U4 snRNP with DMS and CMCT. Primer extension reactions were carried out with radiolabeled CM2, 14b, or 675 oligonucleotides. The position where the reverse transcriptase was inhibited by a modification is shown on the right. Lane U, G, C, and A are dideoxy sequencing lanes made with the corresponding oligonucleotide. The amount of modifier used is given above the lane, where (-) indicates the omission of the modifier. DMS experiments were carried out on ice for 50 minutes, and CMCT experiments were performed at room temperature for five minutes. B) Secondary structure model of the free U4 snRNP with the modification pattern mapped on it. Boxes indicate protected nucleotides, semi circles indicate very weak modifications, circles indicate weak modifications, grey circles indicate medium modifications, and dark grey circles indicate strong modifications. Nucleotides at the 3' end of the molecule are where oligonucleotides 14b and 675 bind. Nucleotides lacking boxes or circles are nucleotides for which there is no information, or positions of strong reverse transcriptase stops. The model is based on probing experiments that were repeated two to more than five times, depending on the difficulty in reproducing modifications, and the region of U4. The 5' region was probed at least three times. 29 A) u G c A + RNase VI Time (min) w• n 3 s mm 4 * — *L -C3 -C4 -U5 . + 10 ^SWT"- ™t". G C A + 15 •^Bd^»- C3- U6- m <•» *» • + 5 • s <«•» CIO + <•» • • < • ) U8- -All -C12 C10- «•»•"•• ««•• <•*> U19- » 4* CIO «* *w A20 < • *** &* * -C21 -022 •m * : 4» A20 • » * «fc -«&> U RNase VI + Time (min) 3 + 3 G C RNase A Time (min) A 3 u G ^ wp* ^^B^ ^gsw A f 3 -«» t |, u iw m i , j ^ j . C «^ ««!l» 3 .^(W .«, *^*: * * * -im ^P0 G40 5 P -U38 sis W - US! IJ63 - -*&#t 4- -*^ LT60 ^_^ A80 •I (^jjj^ -C86 -C87 U97 UIOO U100- U l 11 U l 12 ^™Si >* " " CI 06 iAft A M I 9 B) 0 5' Kissing Loop (U6-A11) 000 150 90 o o 3'- UUUCCAUAAGGUUUUUAAG —;[CJR^ 160 G_p^l)TG+ 140 G elution buffer (column wash buffer with 500mM imidazole) over 10 column volumes at 0.5mL/min, followed by 100% elution buffer for an additional 10 column volumes. The peak fractions (total 2mL) were dialyzed against buffer TNB (500mM NaCl, 20mM Tris-HCl pH 8, 5mM B-mercaptoethanol, 30% glycerol) overnight. The pooled fractions were run alongside bovine serum albumin (BSA) on a 10% SDS polyacrylamide gel, and the concentration of Prp24 was determined by Bradford assay. The purified protein was stored at -80°C. 5.2 Results and Discussion Due to the nature of the Brr2 preparation, Prp24 is not associated with the Brr2 released U6 snRNP particle. Purified recombinant Prp24 must be added to catalyze disnRNP formation of Brr2 released snRNPs. lOHis-tagged Prp24 was partially purified from crude E. coli lysate with a lmL HisTrap column (Figure 15, lanes 2 - 5). Clearly some contaminants were still present after purification; however, the partially purified protein had the ability to anneal free U4 and U6 snRNPs to form the U4/U6 di-snRNP (Figure 16), so 49 further purification was deemed unnecessary. In total, lmL of dialyzed protein at a concentration of 196iiM was obtained after column purification. Ladder PrePrp24 column 1:10 1:100 1:1000 BSA 20 2 ug Figure 15. Purification of lOHis-tagged Prp24. (Lanel - ladder, 2 - crude E. coli lysate, 3 1:10 dilution of Prp24, 4 - 1:100 dilution of Prp24, 5-1:1000 dilution of Prp24, 6 - 20 ug of bovine serum albumin (BSA), 7 - 2 |^g of BSA). Brr2 released U4 and U6 snRNPs do not anneal in the absence of Prp24 over a period of 45 minutes at room temperature (Figure 16, lanes 2 - 5 ; data not shown), proving that Prp24 is indeed necessary for in vitro di-snRNP formation. The maximum amount of disnRNP produced was 73% using 25pmol of Prp24 (Figure 16, lane 7), and did not increase with the addition of more Prp24, even up to 196pmol (Figure 16, lanes 6 - 8 ; data not shown). Therefore all subsequent experiments were carried out with 25pmol of Prp24. The addition of 150iiL of CMCT buffer decreased the amount of di-snRNP formed (Figure 16, lanes 7 and 9; Figure 17, lane 8; data not shown), and more noticeably the addition of CMCT itself considerably inhibited the formation of the di-snRNP (Figure 17). Titration of CMCT in structure probing experiments (Chapter 3) showed that the addition of 20uL of CMCT (8iimoles) was near the minimum amount of CMCT that showed distinct modifications. Although the addition of 20uL of CMCT showed a decrease in the amount of di-snRNP 50 formed, this decrease is not as extreme compared to the addition of 50 or 40uL of CMCT. As a result, 20uL of CMCT was chosen for subsequent modification/interference experiments. IVTU4 + - - . . . . . . CMCT buffer - - - + + . . . + 50 25 Prp24(pmol) 0 0 0 0 0 10 25 ~~ —«» m mmm^mwm U4/U6 U4 1 %U4/U6 2 13 3 13 4 5 13 6 13 7 62 8 73 9 63 64 Figure 16. lOHis-tagged Prp24 anneals Brr2 released U4 and U6 snRNPs. Approximately 7.5fmol of Brr2 released U4 snRNP was annealed to Brr2 released U6 snRNP with Prp24 for 15 minutes at room temperature. Reactions were allowed to proceed in the presence (Lanes 6 - 9) or absence (Lanes 2 - 5 ) of annealing factor Prp24, for 15 minutes at room temperature (Lanes 3, 5, 6 - 9), or stopped immediately with Brow stop buffer (Lanes 2 and 4). Annealing reactions were performed with an additional 150uL of CMCT buffer (Lanes 4, 5, 9) to determine how this buffer affects the annealing reaction. CMCT(nL) 50 40 U4/U6 30 20 10 5 1 0 5 32 6 38 7 39 8 20 MS* t r f l U4 %U4/U6 1 7 2 10 3 16 4 20 Figure 17. Addition of CMCT impairs di-snRNP formation. Approximately 7.5fmol of Brr2 released U4 snRNP was annealed to Brr2 released U6 snRNP with 25pmol of Prp24 for 15 minutes at room temperature in the presence of 150uL CMCT buffer, and 0 to 50uT of 0.4M CMCT. The amount of U4 used for modification/interference was increased from 7.5fmol to 50fmol in hopes of increasing the amount of U4/U6 formed. The increase in starting 51 material consistently resulted in 50% U4/U6 di-snRNP formation with no CMCT added, and approximately 35% di-snRNP formation with 8fxmoles of CMCT added (data not shown). The increase in U4 snRNP did not change the modification pattern observed, as all of the uridine residues that have the potential to be modified continue to be modified by this protocol except for U91, which was observed to be a weak modification during structure probing (Chapter 3). To facilitate analysis and discussion of the degree of interference observed in these experiments, I developed an interference index whereby the strength of interference at a particular position can be reported as a numerical value. The interference index was calculated by first determining the intensity of the band in each lane by subtracting a background band and normalizing to the full length band for that lane. The interference index at each position indicated in Figure 18B was calculated by dividing the value of the U4 modified band subtracted by the U4 control band by the U4/U6 modified band subtracted by the U4/U6 control band. In principle, an interference index greater than 1 would indicate some interference in di-snRNP formation, and an interference index equal to 1 would indicate no interference in di-snRNP formation. However, some uridines in the central and 3' region of U4 had a calculated interference index as low as 0.49. While it is possible that some modifications alter the U4 structure in such a way as to allow it to be incorporated more readily into the di-snRNP compared to wild-type U4, it is unlikely that modification that should inhibit a base-pair interaction accomplishes this. It is more likely that +/- .5 indicates the amount of error associated with the calculation of the interference index (data not shown). Therefore, if the interference index was 1 +/- 0.5 the nucleotide was said to not 52 inhibit di-snRNP formation, and if the interference index was greater than 1.5 the nucleotide was said to inhibit di-snRNP formation. Most of the modified uridines are readily incorporated into the U4/U6 di-snRNP (interference index = 1 +/- 0.5), except for uridines 5, 6 and 8 (Figure 18; Table 4; Table 5). The interference indices indicate that these modified uridines are underrepresented in the disnRNP compared to the free U4 snRNP. Thus, modification at U5, U6, or U8 interferes with the ability of these residues to base pair with their corresponding partners in U6, and consequently inhibits the formation of the di-snRNP. The results support the kissing loop model, in which the loop region of the extreme 5' stem loop is critical for formation of the di-snRNP. Notably this is in agreement with literature regarding the necessity of stem II for di-snRNP formation in yeast, humans, and Xenopus. In humans, deletion of all nucleotides that form stem II, or half of these (nucleotides 1 - 8) resulted in complete abolishment of disnRNP formation in vitro (Wersig and Bindereif 1990), and in Xenopus deletion of stem II was found to inhibit U4/U6 di-snRNP assembly (Vankan et al. 1990, Vankan et al. 1992). Notably in Xenopus, substitution of nucleotides 2 - 6 or nucleotides 12 - 16 of U4 (analogous to the stem nucleotides of the short 5' stem loop) resulted in an intermediate phenotype, but substitution of nucleotides 7 - 1 1 (analogous to loop nucleotides) inhibited formation of the di-snRNP (Vankan ct al. 1990, Vankan et al. 1992). Therefore the loop nucleotides of the short stem loop are essential for di-snRNP formation. 53 A) U G C A U4 j H u B y a w H | C M U4/U6 M C U G C A U4 C M U4/U6 M C -US -U6 jis cio ,$& ^fp w m urn *** ••-• ** * U60 A20 gfe 0(f m: • # ft. -U25 -U89 • — » . fl«w?< . . -U9I * w» ff 1 - * -U94 ^W^^ '^^^•r If f§ if ftp *W 1» p$ '-.'* «p -U97 -UiOO **" **»' H5 mm: -U104 -U105 B) 3-UUUCCAUAAGGUUUUUAAG—!;CY7]) 160 150 MffG r G—[CfSV 140 G — 0 U A-H G-0 U-0 G—0 G—£, p ^ ,E-a33®uJu]-(Au(MA)G . UA G-S G-0 vA^-060 0« G 0-G 50G—© S-G 0-0,, a 0-0 LA) Tu A-fl U• G C—G A—Oloo u' 70 cu u H-lul A G—FY (A) 2 — 1.5 then the nucleotide inhibits di-snRNP formation. Modification/interference experiments with CMCT were completed twice for all uridine residues indicated in Figure 18, with similar interference indices being obtained. Additional experiments showed that modified residues outside of the 5' kissing loop did not interfere with di-snRNP formation (interference indices not shown). Nucleotide U4 control U4 modified U4/U6 modified U4/U6 control Interference Index Inhibits di-snRNP formation U5 .163 .407 .211 .137 3.30 Yes U6 .490 .900 .363 .221 2.89 Yes U8 .330 .802 .354 .231 3.84 Yes U19 .679 1.04 .951 .372 .617 No U25 .702 1.45 1.47 .527 .792 No 55 Table 5. Interference Index calculated from two independent experiments for U5, U6, U8, U19, and U25. The Interference Index was determined as above (Table 4). Nucleotide Interference Index for Gel #1 Interference Index for Gel #2 Average Interference Index Inhibits di-snRNP formation U5 3.30 3.69 3.50 Yes U6 2.89 2.81 2.85 Yes U8 3.84 3.19 3.52 Yes U19 .617 .842 .730 No U25 .792 1.40 1.10 No All of the residues that compose Stem I of the U4/U6 di-snRNP are also found basepaired in the free U4 snRNP (Chapter 3), except for U57, which did not interfere with disnRNP formation. In humans deletion of all nucleotides that form stem I in U4 snRNA inhibited but did not abolish di-snRNP formation (~ 50% compared to wild-type), as deletion of stem II nucleotides did (Wersig and Bindereif 1990). Moreover, in vivo mutational analysis in yeast showed that stem II nucleotides ( 1 - 1 5 ) were the most sensitive to mutational change, but stem I nucleotides were very tolerant (Hu et al. 1995). It seems that base pairing in stem I may be optimal for di-snRNP formation, but it does not appear to be crucial like stem II. Interestingly modification of uridines 69, 70, 71, 74, and 75, that are predicted to be base paired to U6 in the putative stem III helix (Jakab et al. 1997) does not interfere with disnRNP formation. While it is unclear whether stem III formation is important during spliceosome assembly, the U4 uridine residues of this stem do not appear to be critical for U4/U6 di-snRNP formation. It is possible that stem III may function in U4/U6 di-snRNP dissociation by helping to position U6 correctly with the 5' splice site. 56 It is not surprising that uridines of the 5' kink-turn stem loop do not interfere with base pair formation given that none of these nucleotides are predicted to be base-paired to U6 in the di-snRNP, and also because deletion experiments in yeast and humans have found this region of U4 snRNA to be dispensable for di-snRNP formation. Deletion of human U4 snRNA 5' kink-turn stem loop reproducibly enhanced U4/U6 formation, but resulted in a block at subsequent spliceosome assembly stages. This is supported by yeast deletion experiments showing that deletion of U4 snRNA nucleotides 1 9 - 5 2 still had the ability to base pair to U6, but could not subsequently associate with the U5 snRNP (Bordonne et al. 1990). This suggests that the 5' kink turn stem loop is necessary for spliceosome assembly at stages following di-snRNP formation. No interference was detected from uridines in the 3' region of the molecule. Unfortunately a limitation of the assay is the inability to analyze nucleotides that are originally sequestered in the free snRNP, as many of the uridines in this region are. However, the few modified uridines 3' of the putative stem III interaction domain did not interfere with di-snRNP formation. The role of the 3' stem loop in di-snRNP formation appears to differ between species. The 3' portion of human U4 (nucleotides 91 - 145) including the Sm binding site has been found to be dispensable for both U4/U6 di-snRNP formation and subsequent spliceosome assembly and catalysis in vitro (Wersig and Bindereif 1990, Wersig and Bindereif 1992). Contrary to these results, a deletion mutant in yeast consisting of nucleotides 1 - 9 0 could not support di-snRNP formation, while U4 mutant 1 142 could, indicating a requirement for nucleotides 90-142 (Hayduk and Rader 2010). The discrepancy observed between human and yeast may reflect differences in the splicing machineries of the organisms. It is possible that one or more splicing factors in humans for 57 which there is no homologue in yeast fulfill the functional requirements that the 3' region of yeast U4 satisfies in di-snRNP formation. In conclusion, it is clear that stem II nucleotides are essential for di-snRNP interaction in vitro, and the loop nucleotides of the novel 5' stem loop appear to be the most significant. This supports a model of di-snRNP formation where loop nucleotides 6 - 11 of U4 snRNA and single stranded nucleotides 7 5 - 7 1 of U6 snRNA are the initial nucleation site of di-snRNP formation, and stem I nucleotides may contribute a smaller auxiliary function. However, it is not clear what steps of base-pair formation are monitored with the modification/interference experiment used here. While it is possible that the modifications inhibit the initial step of di-snRNP formation, they may also inhibit subsequent interactions with U6, or protein interactions that proceed through the Watson-Crick base pair positions of the RNA. Further experiments are needed to determine the temporal association of stem I, II, and III in the formation of the U4/U6 di-snRNP. 58 Chapter Six - Future Directions and Concluding Remarks 6.1 Future Directions The structure probing experiments presented here demonstrate the presence of a short stem loop at the 5' end of Brr2 released U4 snRNP in vitro. To complete chemical secondary structure probing of all positions in this molecule the chemical modifier kethoxal, which modifies guanine bases, can be used. Protection of the 5' kissing stem nucleotides G13 - G15 would strongly support the presence of this short stem, and preliminary tests suggest that they are indeed protected from modification. Modification/interference analysis of U4 with CMCT indicates that modification of 5' loop nucleotides U6 and U8, and closing stem nucleotide U5 inhibits di-snRNP formation. Modification of uridines throughout the remainder of the molecule does not interfere with disnRNP formation, suggesting that the 5' kissing loop is important in di-snRNP formation, possibly as a means of initial di-snRNP nucleation. Modification/interference assays would be complete with the use of DMS and kethoxal modifiers. The use of DMS would provide information for the 5' kissing loop nucleotides A7, CIO, and A l l , and modification/interference assays with kethoxal would provide information for the 5' kissing loop nucleotide G9. The use of these modifiers would confirm that the 3' and central regions of U4 do not inhibit di-snRNP formation, as indicated by CMCT modification/interference experiments. Although I have determined the secondary structure of a Brr2 released U4 snRNP, it remains unclear if this snRNP is structurally similar to a biogenesis U4 snRNP that has not been base-paired with U6, or if it is similar to U4 snRNP released from U6 in the spliccosome assembly pathway. The biogenesis U4 snRNP is thought to be a species that 59 has been transcribed and associates with the Sm and possibly U4 specific proteins, but has not base-paired to U6. The U4 snRNP species released from the spliceosome has been basepaired to U6, but is thought to be recycled in a way that it can undergo another association with U6, to be used for a future round of splicing. The literature suggests that Brr2 is the protein responsible for unwinding of the U4/U6 di-snRNP in the spliceosome, but it is unclear whether the released U4 snRNP undergoes compositional or conformational rearrangements in the recycling process prior to being re-annealed to U6 snRNP. Therefore, it is possible that the product of in vivo recycling is different from the product of the in vitro Brr2 release. An accumulated U4 snRNP, presumably a U4 snRNP species that has not base paired with U6, can be generated by specific mutations in Prp24 that inhibit di-snRNP formation. Structure probing of an accumulated U4 snRNP will provide insight into the possible conformations of U4 snRNA throughout the splicing cycle. Crystal structure determination and cross-linking experiments indicate that 15.5K and hPrp31 (61K) make direct contact with the 5' kink-turn stem loop of human U4 snRNA (Vidovic et al. 2000, Nottrott et al. 2002, Lui et al. 2007; Chapter 3). Notably, these experiments were only carried out with a small portion of the 5' region of U4, so it is possible that these proteins induce a slightly different conformation in the presence of full length RNA. Also, it will be interesting to see if the yeast proteins Snul3 and Prp31 bind to yeast U4 snRNA in a similar manner as 15.5K and 6IK are thought to bind to human U4 snRNA. A technique that may give some insight into the binding locations of these proteins in yeast is hydroxyl radical probing. If the U4 snRNA is bound by proteins (Snul3 or Prp31) then the RNA backbone will be protected from hydroxyl radical cleavage. However, if the RNA backbone is not bound by proteins it will be accessible to hydroxyl radicals. The 60 difference in hydroxyl radical cleavage pattern between proteinated and deproteinized U4 snRNP may indicate a protein binding site. Protection from hydroxyl radicals at and around the 5' kink turn stem loop of Brr2 released U4 would be a strong indication of the presence of Snul3 and Prp31. Alternatively, it may be possible to express recombinant Snul3 and Prp31 and incubate them with U4 snRNA both separately and in combination to determine the distinct footprint pattern obtained from each protein. Existing literature suggests that Snul3 and Prp31 may bind to the free U4 snRNP, but an accurate determination of the protein complement of the free U4 snRNP remains to be elucidated. Mass spectrometry of a purified U4 snRNP would return protein candidates, and can be verified with further biochemical experiments, including immunoprecipitation with tagged candidate proteins. First, a U4 snRNP of sufficient purity and quantity must be obtained for mass spectrometry analysis. The Brr2 released U4 snRNP contains large amounts of free U6 snRNP, and small amounts of U4/U6 di-snRNP (Chapter 2). A second purification of the Brr2 released U4 may result in a free U4 sample of sufficient purity for mass spectrometry. This may be accomplished with an additional affinity capture and release using biotinylated 2'-0-methyl oligonucleotides, and streptavidin agarose (Aukema and Rader, unpublished results). The biotinylated 2'-0-methyl oligonucleotide would be designed to bind to the free U4 snRNP, but not to U4 that is base-paired with U6. Free U4 that is bound by the biotinylated oligonucleotide can then be separated from the other snRNP species by interaction with streptavidin agarose. This procedure has been employed to yield purified U4 snRNP, although the amounts obtained may be insufficient for mass spectrometry (Aukema and Rader, unpublished results). Alternatively, it may be possible to purify the Brr2 released U4 snRNP, or an accumulated U4 snRNP, using glycerol gradients 61 that fractionate snRNPs based on composition and size. An accurate determination of the protein complement of free U4 snRNP is necessary for understanding the role of snRNP specific proteins in U4/U6 di-snRNP formation. There is still considerable debate about the structure of free U6 snRNP and whether free U6 contains a 3' intramolecular stem loop (ISL), a structural feature of U6 that is present in the catalytically active spliceosome. The structure of the Brr2 released U6 snRNP has yet to be determined, and it will be interesting to see if Brr2 released U6 is similar to a free U6 snRNP that does not contain a 3' ISL, or if it adopts a conformation comparable to U6 in the catalytic spliceosome. If Brr2 unwinds the di-snRNP in the spliceosome and allows U6 to fold into a conformation that facilitates assembly of the active spliceosome, free U6 from the Brr2 release may fold into a conformation that contains the 3' ISL. To address this question, the secondary structure of Brr2 released U6 snRNP should be probed with the chemical modifiers, and ribonucleases used here, and compared to structure probing experiments of free U6 snRNP (Former et al. 1994, Jandrositz and Guthrie 1995, Karaduman et al. 2006). 6.2 Concluding Remarks In the field of RNA splicing, there are no examples of RNA interactions whose genesis has been determined. The U4/U6 di-snRNP formation is a significant structural rearrangement in splicing that takes place outside of the active spliceosome, and it is therefore an important model for the regulation of spliceosomal RNA conformational changes. By the use of structure probing methods with chemical modifiers DMS, and CMCT, and with enzymatic probes RNase Vi and RNase A, I have provided evidence for the existence of a 5' kissing stem loop in Brr2 released U4 snRNP. Modification/interference experiments with CMCT show that modification of 5' kissing loop nucleotides U5, U6, and 62 U8 inhibits di-snRNP formation, yet modification of uridine bases in other regions of U4 do not inhibit di-snRNP formation. This suggests a model of di-snRNP initiation in which U4 kissing loop nucleotides 6 - 1 1 initiate contact with U6 single-stranded nucleotides 7 5 - 7 1 to provide the site for di-snRNP nucleation. The experiments presented in this thesis suggest a preliminary model for the initiation mechanism of di-snRNP formation that provides a foundation for future mechanistic analysis. 63 Works Cited Achsel T, Brahms H, Kastner B, Bachi A, Wilm M, Luhrmann, R. (1999) A doughnutshaped heteromer of human Sm-like proteins binds to the 3'-end of U6 snRNA, thereby facilitating U4/U6 duplex formation in vitro. EMBOJ. 18: 5789-5802. Anderson AA, Collins RA. (2001) Intramolecular secondary structure rearrangement by the kissing interaction of the Neurospora VS ribozyme. PNAS. 98: 7730-7735. Ansari A, Schwer B. (1995) SLU7 and a novel activity, SSF1, act during the PRP16dependent step of yeast pre-mRNA splicing. EMBOJ. 14: 4001-4009. Bell M, Schreiner S, Damianov A, Reddy R, Bindereif, A. (2002) pi 10, a novel human U6 snRNP protein and U4/U6 snRNP recycling factor. EMBOJ. 21: 2724-2735. Bordonne R, Banroques J, Abelson J, Guthrie C. (1990) Domains of yeast U4 spliceosomal RNA required for PRP4 protein binding, snRNP-snRNP interactions, and premRNA splicing in vivo. Genes Dev. 4: 1185-1196. Brody E, Abelson J. (1985) The "Spliceosome": Yeast pre-messenger RNA associates with a 40S complex in a splicing-dependent reaction. Science. 228: 963-967. Brow D. (2002) Allosteric cascade of spliceosome activation. Annu Rev Genet. 36: 333-360. Brow DA, Guthrie C. (1988) Spliceosomal RNA U6 is remarkably conserved from yeast to mammals. Nature. 334: 213-218. Brunei C, Marquet R, Romby P, Ehresmann C. (2002) RNA loop-loop interactions as dynamic functional motifs. Biochimie. 84: 925-944. Cheng SC, Abelson J. (1987) Spliceosome assembly in yeast. Genes Dev. 1: 1014-1027. Domdey H, Apostol B, Lin R-J, Newman A, Brody E, Abelson J. (1984) Lariat structures are in vivo intermediates in yeast pre-mRNA splicing. Cell. 39: 611-621. Dunn EA, Rader SD. (2010) The secondary structure of U6 snRNA: Implications for spliceosome assembly. Biochem Soc Trans. 38: 1099-1104. Ehresmann C, Baudin F, Mougel M, Romby P, Ebel J, Ehresmann B. (1987) Probing the structure of RNAs in solution. Nucleic Acids Res. 15: 9109-9128. Fortner DM, Troy RG, Brow DA. (1994) A stem/loop in U6 RNA defines a conformational switch required for pre-mRNA splicing. Genes Dev. 8: 221-233. Frazer L N, Nancollis V, O'Kecfe RT. (2008) The role of snul 14p during pre-mRNA splicing. Biochem Soc Trans. 36: 551-553. IX Hu J, Xu D, Schappert K, Xu Y, Friesen JD. (1995) Mutational analysis of Saccharomyces cerevisiae U4 small nuclear RNA identifies functionally important domains. Mol Cell Biol. 15: 1274-1285. Hu Qu L, Michot B, Bachellerie J-P. (1983) Improved methods for structure probing in large RNAs: a rapid 'heterologous' sequencing approach is coupled to the direct mapping of nuclease accessible sites. Application to the 5' terminal domain of eukaryotic 28S rRNA. Nucleic Acids Res. 11: 5903-5920. Inoue T, Cech TR. (1985) Secondary structure of the circular form of the Tetrahymena rRNA intervening sequence: A technique for RNA structure analysis using chemical probes and reverse transcriptase. Proc Natl Aca Sci USA. 82: 648-652. Jakab G, Mougin A, Kiss M, Pollak T, Antal M, Branlant C, Solymosy F. (1997) Chlamydomonas U2, U4, and U6 snRNAs. An evolutionary conserved putative third interaction between U4 and U6 snRNAs which has a counterpart in the U4atac-U6atac snRNA duplex. Biochimie. 79:387-395. Jandrositz A, Guthrie C. (1995) Evidence for a Prp24 binding site in U6 snRNA and in a putative intermediate in the annealing of U6 and U4 snRNAs. EMBOJ. 14:820-832. Jurica MS. (2008) Detailed close-ups and the big picture of spliceosomes. Curr Opin Struct Biol. 18:315-320. Karaduman R, Fabrizio P, Hartmuth K, Urlaub H, Luhrmann R. (2006) RNA structure and RNA-protein interactions in purified yeast U6 snRNPs. J Mol Biol. 356: 1248-1262. Kim C-H, Tinoco I. (2000) A retroviral RNA kissing complex containing only two G-C base pairs. PNAS. 97: 9396-9401. Kim D-H, Rossi JJ. (1996) The first ATPase domain of the yeast 246-kDa protein is required for in vivo unwinding of the U4/U6 duplex. RNA. 5: 959-971. Krol A, Branlant C. (1981) Primary and secondary structures of chicken, rat and man nuclear U4 RNAs. Homologies with Ul and U5 RNAs. Nucleic Acids Res. 9: 2699-2716. Kuhn AN, Li Z, Brow DA. (1999) Splicing factor Prp8 governs U4/U6 RNA unwinding during activation of the spliceosome. Molecular Cell. 3: 65-75. Ladner JE, Jack A, Robcrtus JD, Brown RS, Rhodes D, Clark BFC, Klug A. (1975) Structure of yeast phenylalanine transfer RNA at 2.5A resolution. Proc Nat Acad Sci USA. 72: 44144418. Laggerbauer B, Achsel T, Luhrmann R. (1998) The human U5-200kD DEXH-box protein unwinds U4/U6 RNA duplices in vitro. Proc Natl Acad Sci USA. 95: 4188-4192. x Lauber J, Fabrizio P, Teigelkamp S, Lane WS, Hartmann E, Luhrmann, R. (1996) The HeLa 200 kDa U5 snRNP-specific protein and its homologue in Saccharomyces cerevisiae are members of the DEXH-box protein family of putative RNA helicases. EMBOJ. 15: 40014015. Lehnert V, Jaeger L, Michel F, Westhof E. (1996) New loop-loop tertiary interactions in selfsplicing introns of subgroup IC and ID: a complete 3D model of the Tetrahymena thermophila ribozyme. Chemistry & Biology. 3: 993-1009. Li PTX, Bustamante C, Tinoco, I. (2006) Unusual mechanical stability of a minimal RNA kissing complex. PNAS. 103: 15847-15852. Li Z, Brow DA. (1993) A rapid assay for quantitative detection of specific RNAs. Nucleic Acids Res. 21: 4645-4646. Lockard RE, Kumar A. (1981) Mapping tRNA structure in solution using double-strandspecific ribonuclease VI from cobra venom. Nucleic Acids Res. 9: 5125-5140. Lowman HB, Draper DE. (1986) On the recognition of helical RNA by cobra venom VI nuclease. J Biol Chem. 261: 5396-5403. Luhrmann R, Kastner B, Bach M. (1990) Structure of spliceosomal snRNPs and their role in pre-mRNA splicing. Biochim Biophys Acta. 1087: 265-292. Liu S, Li P, Dybkov O, Nottrott S, Hartmuth K, Luhrmann R, Carlomagno T, Wahl MC. (2007) Binding of the human Prp31 Nop domain to a composite RNA-protein platform in U4 snRNP. Science. 316: 115-120. Makarov EM, Makarova OV, Urlaub H, Gentzel M, Will CL, Wilm M, Lurhmann R. (2002) Small nuclear ribonucleoprotcins remodeling during catalytic activation of the spliceosome. Science. 298: 2205-2208. Merryman C, Noller HF. (1998) Footprinting and modification-interference analysis of binding sites on RNA. RNA.Protein Interactions A Practical Approach. Oxford University Press. Oxford. 237-253 Moazed D, Stern S, Noller HF. (1986) Rapid chemical probing of conformation in 16 S ribosomal RNA and 30 S ribosomal subunits using primer extension. JMol Biol. 187: 399416. Mougin A, Gottschalk A, Fabrizio P, Luhrmann R, Branlant C. (2002) Direct probing of RNA structure and RNA-protein interactions in purified HeLa cell's and yeast spliceosomal U4/U6.U5 tri-snRNP particles. J Mol Biol. 317: 631-649. xi Myslinski E, Branlant C. (1991) A phylogenetic study of U4 snRNA reveals the existence of an evolutionarily conserved secondary structure corresponding to "free" U4 snRNA. Biochimie. 73: 17-28. Nagai K, Muto Y, Pomeranz Krummel DA, Kambach C, Ignjatovic T, Walke S, Kuglstatter A. (2001) Structure and assembly of the spliceosomal snRNPs. Biochem Soc Trans. 29: 1526. Nottrott S, Hartmuth K, Fabrizio P, Urlaub H, Vidovic I, Ficner R, Luhrmann R. (1999) Functional interaction of a novel 15.5kD [U4/U6.U5] tri-snRNP protein with the 5' stemloop of U4 snRNA. EMBO J. 18: 6119-6133. Nottrott S, Urlaub H, Luhrmann R. (2002) Hierarchical, clustered protein interactions with U4/U6 snRNA: a biochemical role for U4/U6 proteins. EMBO J. 21: 5527-5538. Oruganti S, Zhang Y, Li H (2005) Structural comparison of yeast snoRNP and spliceosomal protein Snul3p with its homologs. Biochem Biophys Res Commun. 333: 550 - 554. Padgett RA, Konarska MM, Grabowski PJ, Hardy SF, Sharp PA. (1984) Lariat RNA's as intermediates and products in the splicing of messenger RNA precursors. Science. 225: 898903. Paillart JC, Marquet R, Skripkin E, Ehresmann C, Ehresmann B. (1996) Dimerization of retroviral genomic RNAs: Structural and functional implications. Biochimie. 78: 639-653. Rader SD, Guthrie C. (2002) A conserved Lsm-interaction motif in Prp24 required for efficient U4/U6 di-snRNP formation. RNA. 8: 1378-1392. Raghunathan PL, Guthrie C. (1998) A spliceosomal recycling factor that reanneals U4 and U6 small nuclear ribonucleoprotein particles. Science. 279: 857-860. Raghunathan PL, Guthrie C. (1998) RNA unwinding in U4/U6 snRNPs requires ATP hydrolysis and the DEIH-box splicing factor Brr2. Curr Biol. 8: 847-855. Raines RT. (1998) Ribonuclease A. Chem Rev. 98: 1045-1065. Rastogi T, Beattie TL, Olive JE, Collins RA. (1996) A long-range pseudoknot is required for activity of the Neurospora VS ribozyme. EMBO J. 15: 2820-2825. Rinke J, Appel B, Digwccd M, Luhrmann R. (1985) Localization of a base-paired interaction between small nuclear RNAs U4 and U6 in intact U4/U6 ribonucleoprotein particles by psoralen cross-linking. J Mol Biol. 185:721-731. Ruby SW, Abelson J. (1988) An early hierarchic role of Ul small nuclear ribonucleoprotein in spliceosomc assembly. Science. 242: 1028-1035. xn Ruskin B, Krainer AR, Maniatis T, Green MR. (1984) Excision of an intact intron as a novel lariat structure during pre-mRNA splicing in vitro. Cell. 38: 317-331. Ryan DE, Stevens SW, Abelson J. (2002) The 5' and 3' domains of yeast U6 snRNA: Lsm proteins facilitate binding of Prp24 protein to the U6 telestem region. RNA. 8: 1011-1033. Sakharkar MK, Chow VTK, Kangueane P. (2004) Distributions of exons and introns in the human genome. In Silico Biology. 4: 0032. Schultz A, Nottrott S, Hartmuth K, Luhrmann R. (2006) RNA structural requirements for the association of the spliceosomal hPrp31 protein with the U4 and U4atac small nuclear rionucleoproteins. J Biol Chem. 281: 28278-28286. Schultz A, Nottrott S, Watkins NJ, Luhrmann R. (2006) Protein-protein and protein-RNA contacts both contribute to the 15.5K-mediated assembly of the U4/U6 snRNP and the box C/D snoRNPs. Mol Cell Biol. 26: 5146-5154. Seraphin B. (1995) Sm and Sm-like proteins belong to a large family: identification of proteins of the U6 as well as the U l , U2, U4 and U5 snRNPs. EMBO J. 14: 2089-2098. Shannon K, Guthrie C. (1991) Suppressors of a U4 mutation define a novel U6 snRNP protein with RNA binding motifs. Genes Dev. 5: 773-785. Small EC, Leggett SR, Winans AA, Staley JP. (2006) The ET-G-like GTPase Snul 14p regulates spliceosome dynamics mediated by Brr2p, a DExD/H box ATPase. Mol. Cell. 23: 389-839. Valadkhan S, James M. (2001) Splicing-related catalysis by protein-free snRNAs. Nature. 413: 701-707. Vankan P, McGuigan C, Mattaj IW. (1990) Domains of U4 and U6 snRNAs required for snRNP assembly and splicing complementation in Xenopus oocytes. EMBO J. 9: 3397-3404. Vankan P, McGuigan C, Mattaj IW. (1992) Roles of U4 and U6 snRNAs in the assembly of splicing complexes. EMBO J. 11: 335-343. Verdone L, Galardi S, Page D, Beggs JD. (2004) Lsm proteins promote regeneration of premRNA splicing activity. Curr Biol. 14: 1487-1491. Vidaver RM, Fortner DM, Loos-Austin LS, Brow DA. (1999) Multiple functions of Saccharomyces cerevisiae splicing protein Prp24 in U6 RNA structural rearrangements. Genetics. 153: 1205-1218. Vidovic I, Nottrott S, Hartmuth K, Luhrmann R. (2000) Crystal structure of the spliceosomal 15.5kD protein bound to a U4 snRNA fragment. Molecular Cell 6: 13311342. xni Wersig C, Bindereif A. (1990) Conserved domains of human U4 snRNA required for snRNP and spliceosome assembly. Nucleic Acids Res. 18: 6223-6229. Wersig C, Bindereif A. (1992) Reconstitution of functional mammalian U4 small nuclear ribonucleoprotein: Sm protein binding is not essential for splicing in vitro. Mol Cell Biol. 12: 1460-1468. Wolff T, Bindereif A. (1993) Conformational changes of U6 RNA during the spliceosome cycle: an intramolecular helix is essential both for initiating the U4-U6 interaction and for the first step of splicing. Genes Dev. 7: 1377-1389. Yean S-L, Lin R-J. (1991) U4 small nuclear RNA dissociates from a yeast spliceosome and does not participate in the subsequent splicing reaction. Mol Cell Biol. 11: 5571-5577. Yean SL, Wuenschell G, Termini J, Lin RJ: (2000) Metal-ion coordination by U6 small nuclear RNA contributes to catalysis in the spliceosome. Nature. 408: 881-884. xiv