Journal of Cell Science, Supplement 19,101 -105 (1995) 101 Printed in Great Britain © The Company of Biologists Limited 1995

The role of PRP8 protein in nuclear pre-mRNA splicing in yeast

Jean D. Beggs1, Stefan Teigelkamp1’* and Andrew J. Newman2 institute of Cell and Molecular Biology, University of Edinburgh, King’s Buildings, Mayfield Road, Edinburgh EH9 3JR, UK 2MRC Laboratory for Molecular Biology, Hills Road, Cambridge CB2 2QH, UK

'Present address: Institut für Molekularbiologie und Tumorforschung, Philipps-Universität Marburg, Emil-Mannkopff-Strasse 2, 35037 Marburg, Germany

SUMMARY

The removal of from precursor messenger RNAs has been used to map the binding sites, and shows extensive occurs in a large complex, the , that contains interaction between PRP8 protein and the 5' prior to many proteins and five small nuclear RNAs (snRNAs). The the first step of splicing and with the 3' splice site region snRNAs interact with the -containing substrate RNA subsequently. It is proposed that PRP8 protein may and with each other to form a dynamic network of RNA stabilize fragile interactions between the U5 snRNA and interactions that define the intron and promote splicing. exon sequences at the splice sites, to anchor and align them There is evidence that protein splicing factors play in the catalytic centre of the spliceosome. important roles in regulating RNA interactions in the spliceosome. PRP8 is a highly conserved protein that is associated in particles with the U5 snRNA and directly Key words: protein-RNA interaction, photo-crosslinking, RNA binds the substrate RNA in . UV crosslinking splicing, snRNP protein

INTRODUCTION pre-mRNAs into a conformation suitable for catalysis, and may also have catalytic roles in the splicing reactions. As charac­ Nuclear precursor messenger RNA (pre-mRNA) splicing is the terised in metazoans each snRNP, with the exception of U6, is removal of introns and joining of exon sequences to form composed of a single small nuclear RNA (snRNA) with a tri- mRNA. The excision of each intron involves two sequential methylguanosine cap, a set of eight common or ‘core’ proteins, transestérification reactions that occur within a large, dynamic and a variable number of snRNA-specific proteins (reviewed complex termed the spliceosome. Spliceosome assembly by Lührmann et al., 1990; Will et al., 1993). Unlike the others, requires ATP and multiple trans-acting factors that interact U6 snRNA is transcribed by RNA polymerase III, has a y- with one another and with conserved d.s-elements in the pre- monomethyl guanosine cap structure and does not directly bind mRNA. The mechanism of the two-step splicing reaction is the core proteins that are common to the other since highly conserved from yeast to mammals (reviewed by Green, it lacks the appropriate structural motif, the Sm-site. However, 1991; Rymond and Rosbash, 1992; Moore et al., 1993; Sharp, at least in Saccharomyces cerevisiae, U6 RNA associates with 1994), as are at least some of the splicing factors (reviewed by proteins that are structurally related to the core proteins Guthrie and Patterson, 1988; Hodges and Beggs, 1994; Hodges (Cooper et al., 1995; Séraphin, 1995). et al., 1995). First, the scissile phosphate at the 5' end of the In vitro, the snRNPs and other protein splicing factors intron (5' splice site) is attacked by the 2' OH of an adenosine assemble on the substrate pre-mRNA in precisely defined con­ (the branchpoint) residue in the 3' region of the intron. As the secutive steps to form the spliceosome, within which a network 3'-5' phosphodiester bond at the 5' splice site is cleaved there of RNA interactions develops (summarised below and in Fig. is concomitant formation of a 2'-5' phosphodiester bond 1 ; for more details and references see reviews by Madhani and between the phosphate at the 5' end of the intron and the Guthrie, 1994; Newman, 1994; Nilsen, 1994). The U l snRNP attacking adenosine to form a branched structure, and is the first to associate with the pre-mRNA at the 5' splice site, producing the reaction intermediates: the linear 5' exon and the and subsequently the U2 snRNP assembles at the branchpoint lariat intron-3' exon. An important question that will be sequence of the intron. There is substantial evidence from both addressed here is - how is the cleaved-off 5' exon retained in biochemical and genetic suppression studies that highly the catalytic centre of the spliceosome? In the second transes­ conserved sequences in the Ul and U2 snRNAs interact térification reaction the phosphate at the 3' splice site is through Watson-Crick basepairing with conserved sequences attacked by the 3' OH of the 5' exon, resulting in joining of the at the 5' splice site and branchpoint, respectively, in the two , and excision of the intron in lariat form. substrate RNA. The U4 and U6 snRNAs contain extensive The major subunits of the spliceosome are five small nuclear sequence complementarity with each other and are predomi­ ribonucleoprotein particles (snRNPs); U l, U2, U4, U5 and U6. nantly found base paired within a U4/U6 snRNP complex. The These snRNPs play critical roles in defining introns and folding U4/U6 snRNP interacts with the U5 snRNP to form a 102 J. D. Beggs, S. Teigelkamp and A. J. Newman

Fig. 1. RNA interactions in the spliceosome. This is a highly schematic representation of interactions between the substrate RNA and snRNAs in the spliceosome. Boxes represent exon sequences, the intron is represented by a line between the boxes, A in the intron indicates the branchpoint, and the bold arrows indicate the formation of helix I and helix II between U2 and U6 snRNAs after destabilization of the U4/U6 interaction. For simplicity, loop I of U5 snRNA is discontinuous.

U4/U6.U5 triple snRNP which then associates with the U1-U2- changes. For example, in higher eukaryotes, certain members pre-mRNA complex to form the spliceosome. Of the spliceo- of the SR and hnRNP protein families have RNA annealing somal snRNAs, U6 is the most highly conserved in primary activities and function as constitutive or alternative splicing sequence, and it has been proposed that essential conserved factors, influencing splice site usage by modifying snRNP motifs in U6 snRNA might be involved in the catalysis of interactions with the pre-mRNA (reviewed by Dreyfuss et al., splicing. Prior to the first transestérification reaction (step 1 of 1993; Lamm and Lamond, 1993; Burd and Dreyfuss, 1994; splicing) the U4/U6 basepairing appears to be destabilized, Horowitz and Krainer, 1994; Norton, 1994). which led to the suggestion that U4 sequesters U6 in an In yeast, there are five protein splicing factors, PRP2, PRP5, inactive form until the spliceosomal function of U6 is required. PRP16, PRP22 and PRP28, that are members of the DEAD/H- Following this conformational change, U6 snRNA anneals box protein family of putative RNA and it has been with U2 snRNA to form two helices, one of them (helix I) proposed that they might influence RNA-RNA interactions in immediately upstream of the branchpoint-binding domain of splicing (Wassarman and Steitz, 1991). PRP2 protein is U2. A conserved sequence, ACAGAG, in U6 immediately required for the first transesterification reaction and interacts adjacent to the helix I region interacts with a conserved intron only transiently with spliceosomes at that time. A dominant sequence at the 5' splice site, bringing the branchpoint negative mutant form of PRP2 protein has been isolated that adenosine into close proximity with the scissile phosphate for blocks the first step of splicing and remains associated with the first transestérification reaction. The U5 snRNA primary stalled spliceosomes, directly bound to the substrate pre- sequence is not phylogenetically conserved except for a single­ mRNA (Plumpton et al.. 1994; Teigelkamp et al., 1994). This stranded loop (loop I) consisting of an invariant 9 nucleotide protein contains a mutation in the conserved SAT motif that is pyrimidine-rich sequence. Genetic suppression studies and proposed to be important for the RNA unwinding activity of photo-crosslinking experiments have shown that this DEAD/H proteins. Thus, the pre-mRNA may be a substrate for conserved loop interacts with the last three nucleotides of the the putative RNA unwinding activity of PRP2 protein. PRP16 5' exon prior to and following its cleavage from the remainder was identified through suppression of a branchpoint mutation of the pre-mRNA (step 1), and with the first two nucleotides in an intron-containing reporter gene (Couto et al., 1987). It of the 3' exon prior to the second step, thus maintaining contact has been proposed that PRP16 influences the accuracy of with the free upstream exon after the first transestérification branchpoint recognition by regulating the use of a discard reaction and possibly aligning it with the downstream exon for pathway for aberrant lariat intermediates (Burgess and Guthrie, the second transestérification reaction. Since exon sequences 1993). Based on genetic experiments, it has been suggested at the splice sites are highly variable, the predominance of that another DEAD box protein, PRP28, destabilizes the uridine residues in the U5 loop could be explained by their U4/U6 snRNA interaction prior to step 1 of splicing (Strauss capacity for promiscuous basepairing. and Guthrie, 1991). Thus, protein splicing factors have an essential impact on the formation, fidelity and stability of RNA-RNA interactions in early spliceosome formation. ROLES FOR PROTEINS IN MODIFYING RNA INTERACTIONS PRP8 PROTEIN Most of these RNA interactions involve rather short conserved motifs and are unlikely to be sufficiently stable by themselves Biochemical studies revealed that PRP8 of S. cerevisiae is a to build up and hold the complex and dynamic spliceosomal U5 snRNP-specific protein (Lossky et al., 1987; Whittaker et structure. The question arises what provides the necessary al., 1990), as is p220, the human homologue of PRP8 stability for such fragile, yet crucial interactions. There is (Anderson et al., 1989; Pinto and Steitz, 1989). Although PRP8 mounting evidence that proteins contribute to the stability and protein shows an extraordinarily high degree of conservation specificity of these interactions and regulate conformational among eukaryotes, it has no obvious homology to other PRP8 protein in pre-mRNA splicing 103 proteins (Hodges et al., 1995), suggesting a critical role in the 4-thiouridine - UpG splicing process. Since the high molecular mass (280 kDa in T7 in vitro transcription yeast) is also conserved among homologues (Jackson et al., with dinucleotide primer 1988; Anderson et al., 1989; Paterson et al., 1991; Kulesza et al., 1993), the large size might be essential for the function of UpG PRP8 protein. In vitro, PRP8 protein appears to play a role in 3' half substrate the formation of the U4/U6.U5 tri-snRNP complex and its assembly into spliceosomes, while in vivo depletion of PRP8 kinase protein results in degradation of U4, U5 and U6 snRNAs (Brown and Beggs, 1992). An allele of PRP8 has been isolated 3' half substrate as a suppressor of a prp28 mutation, and it was proposed that 5' half substrate PRP8 may counterbalance the putative U4/U6 destabilizing pU pG activity of PRP28 (Strauss and Guthrie, 1991). Following bridging DNA oligo incorporation of the U4/U6.U5 triple snRNP into spliceo­ somes, PRP8 protein as well as the human homologue p220 T4 DNA ligase interact directly with substrate RNA, as shown by UV- crosslinking using uniformly 32P-labelled pre-mRNA (Garcia- Bianco et al., 1990; Whittaker and Beggs, 1991). The PRP8 i ] p UpG - interaction with substrate RNA is established prior to the first t transesterification reaction, is maintained during both steps of 5 splice site splicing and continues with the excised intron after completion Fig. 2. The method of preparation of substrate RNA containing 4- of the splicing reaction (Teigelkamp et al., 1995a). RNase T1 thiouridine at a unique position. Boldface U represents 4-thiouridine, treatment of spliceosomes revealed that fragments of the and the neighbouring 5' 32P is indicated by an asterisk. In this substrate RNA from the 5' splice site and the branchpoint-3' example 4-thiouridine is incorporated as the last residue of the 5' splice site regions could be coimmunoprecipitated with PRP8- exon (5'SS-l). specific antibodies, indicating that these are potential sites of interaction of PRP8 protein with substrate RNA. Protection of the branchpoint-3' splice site region was detected only after complementary bridging oligodeoxynucleotide and joining step 1 of splicing (Teigelkamp et al., 1995a). with T4 DNA ligase (Moore and Sharp, 1992). For each test RNA a corresponding control RNA was generated using UpG instead of 4-thioUpG to prime synthesis of the 3' portion. As PHOTO-CROSSLINKING STUDIES USING 4- these RNAs containing only normal uridine were not stimu­ THIOURIDINE lated to crosslink by irradiation with long wavelength UV light, they acted as negative controls to confirm the site-specificity As PRP8 protein interacts with substrate RNA only in of any detected crosslink. Following UV-irradiation of splicing assembled spliceosomes, studies of PRP8 interactions with the reactions containing the test or control substrate RNAs, substrate RNA are complicated by the presence of other RNA- samples were treated with RNase Tl, and PRP8 protein was binding proteins in the spliceosome. To enhance the efficiency immunoprecipitated from the reaction and analysed by dena­ of crosslinking and to facilitate mapping of the PRP8 binding turing polyacrylamide gel electrophoresis and autoradiogra­ site(s), UV-crosslinking experiments were carried out using phy. If PRP8 protein became 32P-labelled this indicated substrate RNAs containing the photoactivatable uridine crosslinking to the 4-thiouridine residue at the test position in analogue, 4-thiouridine, incorporated at unique sites the substrate RNA. To investigate the timing of detected (Teigelkamp et al., 1995b). High efficiency crosslinking of 4- crosslinks with respect to the first step of the splicing reaction, thiouridine to contacting amino acids or nucleotides (usually results obtained with wild-type yeast splicing extracts were in non-Watson-Crick interactions) can be induced by relatively compared with crosslinks detected in heat-inactivated extracts low energy and brief duration UV-irradiation (Favre, 1990; produced from a temperature-sensitive prp2 -l strain. In such Sontheimer, 1994). In 4-thiouridine the keto oxygen at position extracts, which lack functional PRP2 protein, spliceosomes 4 on the pyrimidine ring is replaced by the fractionally larger assemble but do not carry out the first transesterification atom sulphur, a change that is unlikely to cause major pertur­ reaction unless supplemented with purified wild-type PRP2 bations. Thus crosslinking of RNA or proteins to pre-mRNA protein (Kim and Lin, 1993; Teigelkamp et al., 1994). containing 4-thiouridine is expected to reflect normal The results are summarised in Fig. 3. PRP8 protein became molecular interactions, and has been used to detect RNA-RNA crosslinked to 4-thiouridine at all three positions tested in the and RNA-protein interactions in mammalian spliceosomes 5' exon: the last and the penultimate position of the 5' exon (Wyatt et al., 1992; Sontheimer and Steitz, 1993). (5'SS-l and 5'SS-2), and 8 residues upstream of the 5' splice Splicing substrate RNAs containing 4-thiouridine were site (5'SS-8). In each case the crosslink was detected in prepared in several stages (Fig. 2). The 3' portion, starting at spliceosomes lacking functional PRP2 protein, demonstrating the test site was produced by T7 RNA polymerase transcrip­ that contact with the 5' exon was initiated prior to PRP2 action tion primed with the dinucleotide 4-thioUpG (Sigma), and then (that is prior to the first transesterification reaction). Compara­ 5' end-labelled with 32P using T4 polynucleotide kinase. The ble signals were obtained in control extracts supplemented with 5' portion of the substrate RNA was transcribed in vitro and functional PRP2 protein, indicating that the contact may be the two RNAs were directionally ligated by annealing to a maintained following PRP2 action. In contrast, PRP8 protein 104 J. D. Beggs, S. Teigelkamp and A. J. Newman

5'SS BP 3'SS U 1 I 5' exon I GUAUGU------UACUAAC ------cag | 3' exon -8 -2 -1 +4 +2 +1 +13

pre-PRP2: + + + - - -

post-PRP2: (+) (+) (+) + + W

Fig. 3. Summary of the results for crosslinking of PRP8 protein to CYH2 substrate RNAs containing 4-thiouridine (Teigelkamp et al., 1995). CYH2 RNA is represented by boxes (exons) and a line (intron), and the conserved sequences at both splice sites and at the branchpoint (BP). The positions investigated for crosslinking of PRP8 protein by incorporation of 4-thiouridine are indicated below. Discrimination of crosslinks detected before and after PRP2 action is indicated. + indicates a crosslink of PRP8 protein, - indicates no detectable crosslink, (+) refers to crosslinks that may be maintained after PRP2 action, and [+] represents a crosslink the timing of which was not examined.

did not crosslink to 4-thiouridine at the 4th position of the intron (5/SS+4). Because of the sequence constraints at the 5' splice site, this was the closest intron position to the 5' splice site that was testable by this method. PRP8 protein also crosslinked to 4-thiouridine at the first position of the downstream exon (3'SS+l) and two residues 3' to the branchpoint (BP+2), however, these contacts were detected only in extracts capable of undergoing the first trans­ estérification reaction (containing functional PRP2 protein). Crosslinking of PRP8 to 4-thiouridine next to a mutant 3' splice site (CAG>CUC) indicated that the primary sequence of the substrate RNA in this region did not direct the PRP8 interac­ tion. Also, as the 3' splice mutant RNA was unable to undergo the second transestérification reaction, contact with the 3' step 1 region was initiated after PRP2 function (required for the first V step) but prior to the second step of splicing. Further experi­ ments with substrate RNAs containing duplicate 3' splice sites or duplicated 3' exon sequences, suggested that PRP8 protein interacted with at least 13 residues of the 3' exon. Thus PRP8 protein has at least two distinct interactions with the substrate RNA that have different kinetics with respect to the progress of the splicing reaction. PRP8 protein interacts with at least eight exon residues at the 5' splice site during step 1 and with the 3' splice site region (defined as extending from the branchpoint to the 3' splice site and at least 13 nucleotides into the downstream exon) during step 2 of splicing, indicating its close proximity to the catalytic centre(s) of the spliceosome Fig. 4. A model showing PRP8 protein and the U5 snRNA during both steps of splicing. These contacts between PRP8 interacting with the pre-mRNA in spliceosomes before and after step protein and the substrate RNA resemble the interactions 1 of splicing. Exons are shown as boxes, the intron as a curved line. detected between the U5 snRNA and the substrate (Newman Stem-loop I of U5 snRNA is represented by a thick line, and the and Norman, 1991, 1992; Cortes et al., 1993; Sontheimer and interaction of loop I with exon residues adjacent to the splice sites is Steitz, 1993; A.J.N. unpublished). Considering that PRP8 shown by thin lines. PRP8 protein is drawn as an oval covering the protein is a specific component of U5 snRNPs, we propose a regions of interaction with the substrate RNA. model (Fig. 4) in which PRP8 cooperates with the U5 snRNA, stabilizing the extremely fragile interactions between U5 whether the proposed stabilizing properties of PRP8 protein snRNA and the non-conserved sequences at the ends of both also affect other RNA interactions in the spliceosome as previ­ exons, to anchor and align them in the active sites of the spliceo­ ously suggested (Strauss and Guthrie, 1991; Brown and Beggs, some. In HeLa spliceosomes a protein of approximately 220 1992). As such, PRP8 protein might be a pivotal component of kDa that is believed to be the human homologue of PRP8 has the spliceosome, responsible for the stability of multiple RNA been crosslinked to position-2 relative to the 5' splice site interactions that are subject to conformational change. (Wyatt et al., 1992; MacMillan et al., 1994) and to the branch­ point nucleotide with similar kinetics to those observed for J.D.B. is supported by a Royal Society Cephalosporin Fund Senior PRP8 (MacMillan et al., 1994). The strong conservation of the Research Fellowship. The work described was funded by a grant PRP8 protein sequence may therefore reflect the importance of from the Cancer Research Campaign. We thank Mary Plumpton for its function in spliceosomes. It will be interesting to determine Fig. 1. PRP8 protein in pre-mRNA splicing 105

REFERENCES M adhani, H. D. and Guthrie, C. (1994). Dynamic RNA-RNA interactions in the spliceosome. Annu. Rev. Genet. 28, 1-26. Anderson, G. J., Bach, M., Liihrm ann, R. and Beggs, J. D. (1989). Moore, M. J. and Sharp, P. A. (1992). Site specific modification of pre- Conservation between yeast and man of a protein associated with U5 small mRNA: the 2'-hydroxyl groups at the splice sites. Science 256, 992-997. nuclear ribonucleoprotein. Nature 342, 819-821. Moore, M. J., Query, C. C. and Sharp, P. A. (1993). Splicing of precursors to Brown, J. D. and Beggs, J. D. (1992). Roles of PRP8 protein in the assembly of messenger RNAs by the spliceosome. In The RNA World (ed. R. F. Gesteland splicing complexes. EMBOJ. 11,3721-3729. and J. F. Atkins), pp. 303-357. Cold Spring Harbor, NY: Cold Spring Harbor Burgess, S. M. and Guthrie, C. (1993). A mechanism to enhance mRNA Laboratory Press. splicing fidelity: the RNA-dependent ATPase Prpl6 governs usage of a Newman, A. J. and Norman, C. (1991). Mutations in yeast U5 snRNA alter , discard pathway for aberrant lariat intermediates. Cell 73, 1377-1391. the specificity of 5' splice-site cleavage. Cell 65 115-123. Burd, C. G. and Dreyfuss, G. (1994). Conserved structure and diversity of Newman, A. J. and Norman, C. (1992). U5 snRNA interacts with exon functions of RNA-binding proteins. Science 265,615-620. sequences at 5' and 3' splice sites. Cell 68, 743-754. Cooper, M., Johnston, L. H. and Beggs, J. D. (1995). Identification and Newman, A. J. (1994). Small nuclear RNAs and pre-mRNA splicing. Curr. characterisation of Usslp (formerly Sdb23p): a novel U6 snRNA-associated Opin. Cell Biol. 6, 360-367. protein with significant similarity to core proteins of small nuclear Nilsen, T. W. (1994). RNA-RNA interactions in the spliceosome: unraveling ribonuclear proteins. EMBOJ. 14, 2066-2075. the ties that bind. Cell 78, 1-4 Cortes, J. J., Sontheimer, E. J., Seiwert, S. D. and Steitz, J. A. (1993). Norton, P. A. (1994). Alternative pre-mRNA splicing: factors involved in Mutations in the conserved loop of human U5 snRNA generate use of novel splice site selection. J. Cell. Sci. 107, 1-7. Paterson, T., Beggs, J. D., Finnegan, D. J. and Liihrmann, R. (1991). cryptic 5' splice sites in vivo. EMBO J. 12, 5181-5189. Polypeptide components of Drosophila small nuclear ribonucleoprotein Couto, J. R., Tamm, J., Parker, R. and Guthrie, C. (1987). A trans acting particles. Nucl. Acids Res. 19, 5877-5882. suppressor restores splicing of a yeast intron with a branch point mutation. Pinto, A. L. and Steitz, J. A. (1989). The mammalian analogue of the yeast Genes Dev. 1,445-455. PRP8 splicing protein is present in the U4/5/6 small nuclear Dreyfuss, G., Matunis, M. J., Pinol-Roma, S. and Burd, C. D. (1993). ribonucleoprotein particle and the spliceosome. Proc. Nat. Acad. Sci. USA hnRNP proteins and the biogenesis of mRNA. Annu. Rev. Biochem. 62, 289- 86, 8742-8746. 321. Plumpton., M., McGarvey, M. and Beggs, J. D. (1994). A dominant negative Favre, A, (1990). 4-Thiouridine as an intrinsic photoaffinity probe of nucleic mutation in the conserved RNA motif ‘SAT’ causes splicing factor acid structure and interactions. In Bioorganic Photochemistry, vol. I, PRP2 to stall in spliceosomes. EMBO J. 13, 879-887. Photochemistry and the Nucleic Acids (ed. H. Morrison), pp. 379-425. New Rymond, B. C. and Rosbash, M. (1992). Yeast pre-mRNA splicing. In The York: John Wiley & Sons. Molecular and Cellular Biology o f the Yeast Saccharomyces, vol. II (ed. E. Garcia-Bianco, M. A., Anderson, G. J., Beggs, J. D. and Sharp, P. A. W. Jones, J. R. Pringle and J. R. Broach), pp. 143-192. Cold Spring Harbor, (1990). A mammalian protein of 220 kDa binds pre-mRNA in the NY: Cold Spring Harbor Laboratory Press. spliceosome: a potential homologue of the yeast PRP8 protein. Proc. Nat. Seraphin, B. (1995). Sm and Sm-like proteins belong to a large family: Acad. Sci. USA 87, 3082-3086. identification of proteins of the U6 as well as the U 1, U2, U4 and U5 snRNPs. Green, M. R. (1991). Biochemical mechanisms of constitutive and regulated EMBOJ. 14,2089-2098. pre-mRNA splicing. Annu. Rev. Cell Biol. 7, 559-599. Sharp, P. A. (1994). Split genes and RNA splicing: Nobel lecture. C e llll, 805- Guthrie, C. and Patterson, B. (1988). Spliceosomal snRNAs. Annu. Rev. 815. Genet. 22, 387-419. Sontheimer, E. J. and Steitz, J. A. (1993). The U5 and U6 small nuclear RNAs Hodges, P. E. and Beggs, J. D. (1994). U2 fulfils a commitment. Curr. Biol. 4, as active site components of the spliceosome. Science 262, 1989-1996. 264-267. Sontheimer, E. J. (1994). Site specific RNA crosslinking with 4-thiouridine. Hodges, P. E., Jackson, S. P., Brown, J. D. and Beggs, J. D. (1995). Mol. Biol. Rep. 20, 35-44. Extraordinary sequence conservation of the PRP8 splicing factor. Yeast 11, Strauss, E. J. and Guthrie, C. (1991). A cold-sensitive mRNA splicing mutant 337-342. is a member of the RNA helicase gene family. Genes Dev. 5, 629-641. Horowitz, D. S. and Krainer, A. R. (1994). Mechanisms for selecting 5' splice Teigelkamp, S., McGarvey, M., Plumpton, M. and Beggs, J. D. (1994). The sites in mammalian pre-mRNA splicing. Trends Genet. 10, 100-106. splicing factor PRP2, a putative RNA helicase, interacts directly with pre- Jackson, S. P., Lossky, M. and Beggs, J. D. (1988). Cloning of the RNA8 gene mRNA. EMBO J. 13, 888-897. of Saccharomyces cerevisiae, detection of the RNA8 protein, and Teigelkamp, S., W hittaker, E. and Beggs, J. D. (1995a). Interaction of the demonstration that it is essential for nuclear pre-mRNA splicing. Mol. Cell yeast splicing factor PRP8 with substrate RNA during both steps of splicing. Biol. 8, 1067-1075. Nucl. Acids Res. 23, 320-326. Kim, S.-H. and Lin, R.-J. (1993). Pre-mRNA splicing within an assembled Teigelkamp, S., Newman, A. J. and Beggs, J. D. (1995b). Extensive yeast spliceosome requires an RNA-dependent ATPase and ATP hydrolysis. interactions of PRP8 protein with the 5' and 3' splice sites during splicing Proc. Nat. Acad. Sci. USA 90, 888-892. suggest a role in stabilization of exon alignment by U5 snRNA. EMBO J. 14, Kulesza, H., Simpson, G. G., Waugh, R., Beggs, J. D. and Brown, J. W. S. 2602-2612. (1993). Detection of a plant protein analogous to the yeast spliceosomal W assarman, D. A. and Steitz, J. A. (1991). Alive with DEAD proteins. protein, PRP8. FEBS Lett. 318, 4-6. Nature 349, 463-464. Lamm, G. M. and Lamond, A. I. (1993). Non-snRNP protein splicing factors. Whittaker, E. and Beggs, J. D. (1991). The yeast PRP8 protein interacts Biochem. Biophys. Acta 1173, 247-265. directly with pre-mRNA. Nucl. Acids Res. 19, 5483-5489. Lossky, M., Anderson, G. J., Jackson, S. P. and Beggs, J. D. (1987). Whittaker, E., Lossky, M. and Beggs, J. D. (1990). Affinity purification of Identification of a yeast snRNP protein and detection of snRNP-snRNP spliceosomes reveals that the precursor RNA processing protein PRP8, a interactions. Cell 51, 1019-1026. protein in the U5 small nuclear ribonucleoprotein particle, is a component of Liihrmann, R., Kastner, B. and Bach, M. (1990). Structure of spliceosomal yeast spliceosomes. Proc. Nat. Acad. Sci. USA 87, 2216-2219. snRNPs and their role in pre-mRNA splicing. Biochim. Biophys. Acta 1087, Will, C. L., Behrens, S. E. and Liihrmann, R. (1993). Protein composition of 265-292. mammalin spliceosomal snRNPs. Mol. Biol. Rep. 18, 121-126. MacMillan, A. M., Query, C. C., Allerson, C. R., Chen, S., Verdine, G. L. Wyatt, J. R., Sontheimer, E. J. and Steitz, J. A. (1992). Site-specific and Sharp, P. (1994). Dynamic association of proteins with the pre-mRNA crosslinking of mammalian U5 snRNP to the 5' splice site before the first step branch region. Genes Dev. 8,3008-3020. of pre-mRNA splicing. Genes Dev. 6, 2542-2553.