Disruption of base pairing between the 5′ splice site and the 5′ end of U1 snRNA is required for spliceosome assembly

The C-terminal region of hPrp8 interacts with the conserved GU dinucleotide at the 5′ splice site

“…The close contact between the 59SS and Prp8 extends beyond the GU dinucleotide at the 59 end of the intron+ Introduction of relatively small acetamide groups (;3 Å) attached through the ribose backbone at positions flanking the 59SS junction (from positions Ϫ2 to ϩ3) inhibits spliceosome formation, suggesting a steric hindrance in the interaction of Prp8 with the derivatized 59SS RNA (Sha et al+, 1998)+ Furthermore, photoreactive azidophenacyl groups attached to the 59 exon (positions Ϫ3, Ϫ4) form crosslinks with hPrp8 (Sha et al+, 1998), consistent with earlier reports of the 59 exon: Prp8 crosslinks (Wyatt et al+, 1992;Teigelkamp et al+, 1995a)+ In addition to Prp8, U2 and U6 snRNAs have also been implicated in interactions with the GU dinucleotide at the 59SS (Sontheimer & Steitz, 1993;Kim & Abelson, 1996;Luukkonen & Séraphin, 1998a, 1998b)+ Finally, the 59SS intron nucleotides positions ϩ4 to ϩ7 were shown to interact with U6 snRNA (Kandels-Lewis & Séraphin, 1993;Konforti et al+, 1993;Lesser & Guthrie, 1993;Sontheimer & Steitz, 1993) and a number of other spliceosomal proteins (Sha et al+, 1998)+ The multiplicity of factors that contact the 59SS region may explain why the GU:hPrp8 crosslinking is affected by both the exon and intron sequences flanking the 59SS junction (Fig+ 2)+ All these different interactions involving the 59SS contribute to the overall recognition of the substrate, its joining with the spliceosome, and its proper positioning at the active site of the complex+ The 59SS RNA:hPrp8 crosslink maps to a small segment in the C-terminal region of the protein To gain more insight into the interaction between the 59SS RNA and hPrp8, we identified the region of the protein involved in contacting the GU dinucleotide+ In the first approach, we carried out immunoprecipitation experiments using antibodies raised against two overlapping C-terminal regions of hPrp8 (70R and KO5 Abs; Fig+ 4)+ From a mixture of proteolytic fragments containing the crosslink, peptides as small as ;8-10 kDa could be immunoprecipitated with these antibodies, indicating that the site of crosslink is located within the C-terminal portion of hPrp8+ Formally, the crosslink should be located near the region recognized by both antibodies, that is, positions 1876-1902+ Because of some cross-reactivity of the KO5 Ab, we only considered the results obtained with the highly specific 70R Ab, effectively defining the crosslink site to a 50-60-kDa region at the C-terminus of hPrp8+ However, the results of the second mapping approach using a battery of endoproteolytic reagents to analyze the 59SS RNA:hPrp8 crosslink were consistent with the immunoprecipitation experiments+ The crosslink peptides obtained from digestions with a number of proteolytic reagents constituted a single product, indicating that the site of crosslinking is highly homogeneous within the hPrp8 sequence, and allowing for the high-resolution mapping of the crosslink+ By comparing the...…”

“…Recognition of the 59SS by base pairing to the 59 end of U1 snRNA represents one of the initial steps of spliceosome assembly+ While this interaction controls the overall selection of the 59 splice site, it does not specify the actual cleavage site (reviewed in Moore et al+, 1993;Nilsen, 1994)+ Subsequently, the 59SS:U1 snRNA basepairing must be disrupted to allow the 59SS to interact with Prp8, among other components of the U4/U5/U6 snRNP (see Kandels-Lewis & Séraphin, 1993;Konforti et al+, 1993;Lesser & Guthrie, 1993;Sontheimer & Steitz, 1993)+ Both human and yeast Prp8 have been shown to crosslink to the 59SS region (Wyatt et al+, 1992;Teigelkamp et al+, 1995aTeigelkamp et al+, , 1995bChiara et al+, 1996;Reyes et al+, 1996)+ We have previously shown that a highly specific GU dinucleotide:hPrp8 crosslink at the 59SS can be detected within splicing complex B assembled in trans-splicing reactions, where the 59SS is provided as a short RNA oligonucleotide (Reyes et al+, 1996)+ The analogous 59SS:hPrp8 crosslink can also be detected within complex B formed in cis-splicing reactions, using a unimolecular pre-mRNA substrate (Fig+ 1)+ Also the kinetics of crosslinking indicate that in both trans-and cis-splicing, the 59SS:hPrp8 interaction takes place early in the reaction, after formation of complex B, but before the appearance of splicing intermediates and products+ The interaction of Prp8 with exon sequences at the 59SS is maintained beyond the first step of splicing, as evidenced by the persistence of crosslinks in both yeast and mammalian systems (Wyatt et al+, 1992;Teigelkamp et al+, 1995b)+ In contrast, the GU:hPrp8 interaction described here occurs within spliceosome complex B, but crosslink formation is not detected at later stages in the reaction (Fig+ 1)+ This crosslinking profile suggests that either the GU:hPrp8 interaction is disrupted at later stages of splicing, or that an alteration in the physical environment near the contact site may not permit UV crosslinking, even though the interaction itself may be maintained+…”

“…Removal of introns from eukaryotic pre-mRNA is catalyzed by a multicomponent complex termed the spliceosome+ A set of small nuclear ribonucleoprotein (snRNP) particles U1, U2, and U4/U5/U6 snRNPs, accompanied by a number of protein factors, bind to pre-mRNA in a stepwise fashion (reviewed in Moore et al+, 1993;Nilsen, 1994)+ Correct splicing of mRNA precursors depends on the precise recognition of sequence elements that direct assembly of the spliceosome and represent the substrates for the reaction+ One of them, the 59 splice site (59SS), defines the exon/intron junction and represents the substrate for the first step of splicing+ The initial base pairing between U1 snRNA and the 59SS formed in splicing complex E is subsequently disrupted and replaced by the interaction of the 59SS with U4/U5/U6 snRNP within splicing complex B+ A specific basepairing interaction between nucleotides in the conserved ACAGAG sequence of U6 snRNA and the intron bases at the 59SS was demonstrated both by genetic and biochemical analyses (Kandels-Lewis & Séraphin, 1993;Konforti et al+, 1993;Lesser & Guthrie, 1993;Sontheimer & Steitz, 1993)+ Furthermore, exon sequences at the 59SS are thought to interact with the conserved loop 1 in U5 snRNA (see Newman, 1997)+ In yeast, this interaction is dispensable for the first, but essential for the second catalytic step in vitro (O'Keefe et al+, 1996)+ Finally, the most highly conserved element in the 59SS sequence, the GU dinucleotide at the 59 end of the intron, is implicated in interactions with U6 and U2 snRNAs (Sontheimer & Steitz, 1993;Kim & Abelson, 1996;Luukkonen & Séraphin, 1998a, 1998b)+ The U5 snRNP-specific yeast Prp8 protein and its mammalian homolog hPrp8 (also termed p220) interact with the 59SS, branch site, polypyrimidine tract, and 39SS regions (e+g+, Wyatt et al+, 1992;MacMillan et al+, 1994;Teigelkamp et al+, 1995a;Chiara et al+, 1996;Reyes et al+, 1996;Umen & Guthrie, 1996)+ Thus, Prp8 is the only splicing factor known to interact with all the important pre-mRNA sequences+ Its role in the recognition of the 59SS is the subject of the present study+ A simplified in vitro system in HeLa cell nuclear extracts was used to study interactions between the 59SS and spliceosomal components+ A short RNA oligonucleotide comprising the 59SS consensus sequence (59SS RNA, A 5 G/GUAAGUAdTdC 3 , where / represents the exon/intron junction), in the presence of a longer RNA containing the branch site, polypyrimidine tract, and 39SS (39SS RNA), induces formation of splicing complex B and undergoes both steps of splicing in trans (Konforti & Konarska, 1995)+ Upon UV...…”

“…Crosslinking of the 59SS RNA to hPrp8 within the spliceosome-like complexes was detected using the 59SS consensus sequence+ Mutations of the GU dinucleotide, as well as minor modifications of the U at position ϩ2, interfere with complex formation and crosslinking to hPrp8 (Reyes et al+, 1996)+ However, what elements of the functional 59SS RNA are required, and whether the GU dinucleotide alone is sufficient to promote the interaction of the 59SS RNA with hPrp8 has not been analyzed in detail+ To determine whether crosslinking of hPrp8 to the 59 end of the intron requires a full-length consensus 59SS signal, we tested several derivatives of the 59SS RNA in crosslinking experiments+ The selected 59SS RNA oligonucleotides contained either a shortened intron that exhibits a diminished potential for base pairing with U6 snRNA or a shortened 59 exon sequence (Fig+ 2)+ All 59SS RNAs were incubated under trans-splicing conditions in the presence of the 39SS RNA and allowed to form splicing complex B+ Aliquots of the reactions were UV irradiated and resolved in a native gel to determine the efficiency of complex B formation (Fig+ 2A) or in a 12% SDS gel The full-length 59SS RNA (A 5 G/GUAAGUAdTdC 3 ), which efficiently assembles splicing complexes, participates in both catalytic steps, and crosslinks to hPrp8, was used as a reference for all subsequent experiments (Fig+ 2A-C, lane 1)+ First, we analyzed the effect of shortening the intron portion of the 59SS RNA+ Intron positions ϩ4 to ϩ8 in the 59SS have been shown to interact with U6 snRNA during splicing (Kandels-Lewis & Séraphin, 1993;Konforti et al+, 1993;Lesser & Guthrie, 1993;Sontheimer & Steitz, 1993;Crispino & Sharp, 1995)+ Deletion of these nucleotides is expected to decrease the ability of such 59SS RNAs to act as splicing substrates+ In fact, as the potential base pairing of the 59SS RNA to U6 snRNA is reduced, so is the efficiency of trans-splicing (data not shown), complex B formation, and crosslinking to hPrp8 (Fig+ 2, lanes 2 and 3)+ Deletion analysis of the 59 exon nucleotides showed a similar effect+ Specifically, exon length reduction from 6 to 3 or 1 nt also resulted in a marked decrease in trans-splicing efficiency (data not shown and Konforti & Konarska, 1995), complex formation, and crosslinking to hPrp8 (Fig+ 2, lanes 1, 4, and 5, respectively)+ Interestingly, a substrate with no exon nucleotides (intron only), but containing an 11 nt intron, exhibited a reduced level of complex B formation and crosslinking to hPrp8, comparable to that of the substrate with a single exon nucleotide and a shorter (8 nt) intron (compare Fig+ 2, lanes 5 and 6)+ For comparison, we have also used a 59SS RNA substrate containing the Gϩ5 r U mutation, which is expected to interfere with base pairing to U6 snRNA+ This mutation also results in a reduced level of splicing (data not shown), complex B formation, and crosslinking to hPrp8, but the effect is less dramatic than that observed with the above mentioned deletions (compare Fig+ 2, lanes 5 and 6 to lane 7)+ Thus, the sole presence of a GU at the 59 end of the intron does not suffice for its efficient interaction with hPrp8, as defined by the ability to form the crosslink+ These results indicate that the interaction of hPrp8 with the GU dinucleotide has to occur in the context of a functional 59SS capable of establishing other important interactions with the spliceosome components+ Immunoprecipitation with 70R Ab ...…”

Proximity of the U12 snRNA with both the 5′ Splice Site and the Branch Point during Early Stages of Spliceosome Assembly

“…Additional rearrangements of RNA-RNA interactions (39, 50), and changes in protein composition (33) result in the final catalytically active spliceosome (complex C) in which the U2 and U6 snRNAs together with the pre-mRNA form the catalytic core (6,17,32). In the catalytic core, U5 snRNA and Prp8/220-kDa protein interact with exon sequences near the 5Јss (38,49,57,69), while U6 and U2 snRNAs base pair with the 5Јss and BPS, respectively, and with each other to achieve a structure that can carry out likely RNA-based catalysis (21,25,59,60,74). During the assembly, U1 and U4 snRNPs are released from the final spliceosome prior to catalysis.…”