Mutagenesis Suggests Several Roles of Snu114p in Pre-mRNA Splicing
2003; Elsevier BV; Volume: 278; Issue: 30 Linguagem: Inglês
10.1074/jbc.m303043200
ISSN1083-351X
AutoresCornelia Bartels, Henning Urlaub, Reinhard Lührmann, Patrizia Fabrizio,
Tópico(s)RNA and protein synthesis mechanisms
ResumoSnu114p, a yeast U5 small nuclear ribonucleoprotein (snRNP) homologous to the ribosomal GTPase EF-2, was recently found to play a part in the dissociation of U4 small nuclear RNA (snRNA) from U6 snRNA. Here, we show that purified Snu114p binds GTP specifically. To test the possibility that binding and hydrolysis of GTP by Snu114p are required to stimulate the unwinding of U4 from U6, we produced several mutations of Snu114p. Residues whose mutations led to lethal phenotypes were all clustered in the P loop and in the guanine-ring binding sequence (NKXD) of the G domain, which in elongation factor-G is required for the binding and hydrolysis of GTP. An arginine residue in domain II, which in EF-G forms a salt bridge with a residue of the G domain, when mutated in Snu114p (R487E), led to a temperature-sensitive phenotype. The substitution D271N in the NKXD sequence is predicted to bind XTP instead of GTP. Spliceosomes containing this mutant, isolated by affinity chromatography after heat treatment, retained U4 snRNA paired with the U6 snRNA. U4 snRNA was released efficiently only when these arrested spliceosomes were reactivated by lowering the temperature in the presence of a mixture of ATP and XTP. Because non-hydrolyzable XTP analogues did not consent the release of U4, we conclude that the release requires hydrolysis of XTP. This suggests that Snu114p needs GTP to influence, directly or indirectly, the unwinding of U4 from U6. An additional role for Snu114p is also demonstrated: after growth of the D271N and R487E strains at high temperatures, we observed decreased levels of the U5 and the U4/U6·U5 snRNPs. This indicates that, before splicing, Snu114p plays a part in the assembly of both particles. Snu114p, a yeast U5 small nuclear ribonucleoprotein (snRNP) homologous to the ribosomal GTPase EF-2, was recently found to play a part in the dissociation of U4 small nuclear RNA (snRNA) from U6 snRNA. Here, we show that purified Snu114p binds GTP specifically. To test the possibility that binding and hydrolysis of GTP by Snu114p are required to stimulate the unwinding of U4 from U6, we produced several mutations of Snu114p. Residues whose mutations led to lethal phenotypes were all clustered in the P loop and in the guanine-ring binding sequence (NKXD) of the G domain, which in elongation factor-G is required for the binding and hydrolysis of GTP. An arginine residue in domain II, which in EF-G forms a salt bridge with a residue of the G domain, when mutated in Snu114p (R487E), led to a temperature-sensitive phenotype. The substitution D271N in the NKXD sequence is predicted to bind XTP instead of GTP. Spliceosomes containing this mutant, isolated by affinity chromatography after heat treatment, retained U4 snRNA paired with the U6 snRNA. U4 snRNA was released efficiently only when these arrested spliceosomes were reactivated by lowering the temperature in the presence of a mixture of ATP and XTP. Because non-hydrolyzable XTP analogues did not consent the release of U4, we conclude that the release requires hydrolysis of XTP. This suggests that Snu114p needs GTP to influence, directly or indirectly, the unwinding of U4 from U6. An additional role for Snu114p is also demonstrated: after growth of the D271N and R487E strains at high temperatures, we observed decreased levels of the U5 and the U4/U6·U5 snRNPs. This indicates that, before splicing, Snu114p plays a part in the assembly of both particles. Splicing of mRNA precursors (pre-mRNA) proceeds by two consecutive transesterification steps that are catalyzed by the spliceosome. The spliceosome is formed by the ordered interaction of the U1, U2, U5, and U4/U6 snRNPs 1The abbreviations used are: snRNP, small nuclear ribonucleoprotein; XTP, xanthosine triphosphate; XMP-PNP, xanthosine-5′-[(β,γ)-imido]triphosphate; Ni-NTA, nickel-nitrilotriacetic acid; GTPγS, guanosine 5′-O-(3-thiotriphosphate); ATPγS, adenosine 5′-O-(3-thiotriphosphate); EF-G, elongation factor G.1The abbreviations used are: snRNP, small nuclear ribonucleoprotein; XTP, xanthosine triphosphate; XMP-PNP, xanthosine-5′-[(β,γ)-imido]triphosphate; Ni-NTA, nickel-nitrilotriacetic acid; GTPγS, guanosine 5′-O-(3-thiotriphosphate); ATPγS, adenosine 5′-O-(3-thiotriphosphate); EF-G, elongation factor G. (in which the U4 and U6 RNAs are base paired) and of several splicing factors, with the pre-mRNA (1Burge C.B. Tuschl T. Sharp P.A. Gesteland R.F. Cech T.R. Atkins J.F. The RNA World. Cold Spring Harbor Laboratory Press, Cold Spring Harbor, NY1999: 525-560Google Scholar). The spliceosome is assembled by the initial interaction of the U1 snRNP with the 5′ splice site, thereafter the U2 snRNP recognizes and binds to the branch site, forming the pre-spliceosome. Spliceosome assembly is completed by the subsequent association of the U4/U6 and U5 snRNPs in the form of a U4/U6·U5 tri-snRNP complex.Activation of the assembled spliceosome into a catalytically functional machine requires extensive reorganization of the components. Thus, during spliceosome activation, base-pairing between U4 and U6 is disrupted and a new base-pairing between U2 and U6 occurs (reviewed in Ref. 2Nilsen T.W. Simons R.W. Grunberg-Manago M. RNA Structure and Function. Cold Spring Harbor Laboratory Press, Cold Spring Harbor, NY1998: 279-307Google Scholar). Concomitantly with these events, base-pairing of U1 with the 5′ splice site is exchanged for base-pairing between U6 and the 5′ splice site (3Kandels-Lewis S. Séraphin B. Science. 1993; 262: 2035-2039Crossref PubMed Scopus (220) Google Scholar, 4Lesser C.F. Guthrie C. Science. 1993; 262: 1982-1988Crossref PubMed Scopus (239) Google Scholar). After these rearrangements, U1 and U4 snRNPs are released from the spliceosome prior to catalysis. The conserved loop I of U5 RNA makes contact with exonic sequences at the 5′ and 3′ splice sites, while the splicing reaction proceeds (5Sontheimer E.J. Steitz J.A. Science. 1993; 262: 1989-1996Crossref PubMed Scopus (293) Google Scholar).The unwinding of the U4/U6 snRNA duplex is an important step toward the activation of the spliceosome and depends on the functioning of a large number of protein factors. In addition to the putative U4/U6 helicase Brr2p, five splicing factors in yeast have already been implicated in the release of U4 snRNA during spliceosome activation. These are the U4/U6 snRNP protein Prp4p (6Ayadi L. Miller M. Banroques J. RNA (N. Y.). 1997; 3: 197-209PubMed Google Scholar), the non-snRNP protein Prp19p (7Tarn W.Y. Lee K.R. Cheng S.C. Proc. Natl. Acad. Sci. U. S. A. 1993; 90: 10821-10825Crossref PubMed Scopus (82) Google Scholar), the tri-snRNP protein Prp38p, and the U5 snRNP proteins Prp8p and Snu114p (8Xie J. Beickman K. Otte E. Rymond B.C. EMBO J. 1998; 17: 2938-2946Crossref PubMed Scopus (47) Google Scholar, 9Kuhn A.N. Li Z. Brow D.A. Mol. Cell. 1999; 3: 65-75Abstract Full Text Full Text PDF PubMed Scopus (88) Google Scholar, 10Fabrizio P. Laggerbauer B. Lauber J. Lane W.S. Lührmann R. EMBO J. 1997; 16: 4092-4106Crossref PubMed Scopus (180) Google Scholar, 11Bartels C. Klatt C. Lührmann R. Fabrizio P. EMBO Rep. 2002; 3: 875-880Crossref PubMed Scopus (62) Google Scholar).Previously we showed that the G domain-containing protein Snu114p and its human orthologue protein U5–116K are close homologues of the ribosomal elongation factor EF-2 (10Fabrizio P. Laggerbauer B. Lauber J. Lane W.S. Lührmann R. EMBO J. 1997; 16: 4092-4106Crossref PubMed Scopus (180) Google Scholar). Consistently with this, the U5–116K protein cross-links specifically to GTP in purified U5 snRNP particles, and this cross-linking is stimulated by poly(U) (10Fabrizio P. Laggerbauer B. Lauber J. Lane W.S. Lührmann R. EMBO J. 1997; 16: 4092-4106Crossref PubMed Scopus (180) Google Scholar). Remarkably, as predicted by its homology with EF-2, we found that Snu114p is involved in rearranging the spliceosomal RNP structure. Indeed, a temperature-sensitive deletion mutant of Snu114p (snu114ΔN), led to the accumulation of arrested, precatalytic spliceosomes in which U4 is still base-paired with the U6 snRNA. This suggested that Snu114p is involved in the transition to an active form of the spliceosome by influencing factors that are directly required for the unwinding of the U4/U6 duplex (11Bartels C. Klatt C. Lührmann R. Fabrizio P. EMBO Rep. 2002; 3: 875-880Crossref PubMed Scopus (62) Google Scholar).As described above, Snu114p is involved in this rearrangement step together with several additional factors, and there is increasing evidence for a link between Snu114p and some of these factors, for example, between Prp8p and the putative U4/U6 helicase Brr2p. The most direct evidence for this link is the strong RNA-independent protein-protein interaction that has been detected in the human system between the proteins 220K/Prp8p, 200K/Brr2p, and 116K/Snu114p after dissociation of these proteins as a stable complex from the U5 snRNP in high salt buffer (12Achsel T. Ahrens K. Brahms H. Teigelkamp S. Lührmann R. Mol. Cell. Biol. 1998; 18: 6756-6766Crossref PubMed Scopus (120) Google Scholar). In yeast, several studies have shown that Snu114p is in contact with Prp8p and that Prp8p interacts with Brr2p (13Dix I. Russell C.S. O'Keefe R.T. Newman A.J. Beggs J.D. RNA (N. Y.). 1998; 4: 1239-1250Crossref PubMed Scopus (62) Google Scholar, 14van Nues R.W. Beggs J.D. Genetics. 2001; 157: 1451-1467Crossref PubMed Google Scholar). Therefore, we had hypothesized (11Bartels C. Klatt C. Lührmann R. Fabrizio P. EMBO Rep. 2002; 3: 875-880Crossref PubMed Scopus (62) Google Scholar) that Snu114p is a crucial factor, acting in an earlier step to trigger the function of Prp8p/220K and/or of the Brr2p/200K RNA unwindase, in this intricate network of factors involved in the U4/U6 unwinding during spliceosome activation.To define the details of the reaction that Snu114p catalyzes in concert with other factors, and primarily to test the possibility that binding and hydrolysis of GTP by Snu114p are required for spliceosome activation, we replaced or deleted several amino acids in evolutionarily conserved domains of Snu114p. We found that residues that led to lethal phenotypes are all clustered in the G domain, more precisely, in the P loop and in the guanine-ring-binding sequence (NKXD), which in EF-G are required for GTP binding and hydrolysis (15Aevarsson A. Brazhnikov E. Garber M. Zheltonosova J. Chirgadze Y. Al-Karadaghi S. Svensson L.A. Liljas A. EMBO J. 1994; 13: 3669-3677Crossref PubMed Scopus (332) Google Scholar, 16Czworkowski J. Wang J. Steitz T.A. Moore P.B. EMBO J. 1994; 13: 3661-3668Crossref PubMed Scopus (358) Google Scholar). We also found that an arginine in domain II, which in EF-G forms a salt bridge with a residue of the G domain, when mutated in Snu114p (R487E), led to a weakly viable, temperature-sensitive phenotype even at 25 °C. Surprisingly, all of the point mutations made in the other domains did not lead to any significant effect, except for a residue in domain IV.We also replaced aspartic acid 271 with asparagine, in the NKLD sequence of the G domain of Snu114p, to give the mutant snu114-D271N. An analogous substitution in several GTPases changes their specificity of binding and hydrolysis from GTP to xanthosine triphosphate (XTP) (17Bishop A. Buzko O. Heyeck-Dumas S. Jung I. Kraybill B. Liu Y. Shah K. Ulrich S. Witucki L. Yang F. Zhang C. Shokat K.M. Annu. Rev. Biophys. Biomol. Struct. 2000; 29: 577-606Crossref PubMed Scopus (152) Google Scholar). Because this mutant does not lead to a lethal phenotype, we tried to address the interesting question of whether GTP hydrolysis (XTP in this case) is required to trigger the unwinding of U4 from U6. This mutant should allow the identification of the XTP-dependent function of mutated Snu114p and the characterization of this function in a crude extract containing a variety of ATPases, some of which are known to bind and hydrolyze GTP as well.As a prerequisite to these studies, we determined first that Snu114p, similar to its human orthologue, binds GTP in vitro in a specific manner, as analyzed by UV cross-linking. Thus, we show that Snu114p is a genuine GTP-binding protein. Next, we studied the D271N mutant in more detail, and found that this mutant leads to a slow growth phenotype that at high temperatures results in arrested splicing in vivo. Importantly, we show that heat treatment of the D271N extract leads also to a defect of splicing in vitro that is reversed by lowering the assay temperature. We establish that the in vitro thermal inactivation results in accumulation of arrested spliceosomes in which U4 is not released, because it is still base-paired with the U6 snRNA. These precatalytic complexes are isolated by affinity chromatography, depleted of residual nucleoside triphosphates (NTPs), and finally reactivated by adding back a mixture of ATP/XTP and by lowering the assay temperature. Using this system, we show that U4 is released efficiently from the spliceosome only when a mixture of ATP (required for the helicases) and hydrolyzable XTP (required by snu114-D271N) is added. Thus, these experiments suggest that snu114-D271N needs hydrolyzable XTP (and wild-type Snu114p needs GTP) to influence the unwinding of U4 from U6.We also found an additional role for Snu114p. Growth of the D271N and R487E mutant strains at high temperatures leads to extracts that have reduced levels of U5 and tri-snRNP particles, thus suggesting that Snu114p is, directly or indirectly, involved in maintaining proper amounts of U5 and tri-snRNP as well.EXPERIMENTAL PROCEDURESGlycerol Gradient Centrifugation—2.5 mg of yeast splicing extract was diluted to reduce the concentration of glycerol from 20% to 8% and loaded on a 10–30% (w/w) glycerol gradient containing 20 mm Hepes (pH 7.9), 200 mm KCl, 0.2 mm EDTA, 1.5 mm MgCl2, 0.5 mm dithiothreitol, and 0.5 mm phenylmethylsulfonyl fluoride. After centrifugation at 108,000 × g, for 18 h at 4 °C, the gradients were fractionated into 24 fractions of 500 μl. Each fraction was phenol-chloroform-extracted, and their RNA composition was analyzed by Northern blotting. The proteins were acetone-precipitated and assayed by Western blotting.Yeast Strains—Strains were YPF39, trp1-Δ1; his3-Δ; ura3–52; lys2–801; ade2–101; snu114Δ::HIS (TRP1 pRS424-GPD1/SNU114, 6his N-terminal, ARS, CEN); YPF37, trp1-Δ1; his3-Δ; ura3–52; lys2–801; ade2–101; snu114Δ::HIS (TRP1 pRS314/snu114D271N, 6his N-terminal, ARS, CEN); and YCK13, trp1-Δ1; his3-Δ; ura3–52; lys2–801; ade2–101; snu114Δ::HIS (TRP1 pRS314/snu114R487E, Flu-tag C-terminal, ARS, CEN).Mutagenesis—Most of the mutations in the SNU114 gene were introduced by the QuikChange site-directed mutagenesis method (Stratagene), as described previously (10Fabrizio P. Laggerbauer B. Lauber J. Lane W.S. Lührmann R. EMBO J. 1997; 16: 4092-4106Crossref PubMed Scopus (180) Google Scholar). Two oligonucleotide primers, each complementary to opposite strands of SNU114 and containing the desired substitution, were designed. The PCR cycling parameter suggested by the QuikChange method was employed, using native Pfu DNA polymerase. The mutant allele was screened for the desired mutation by sequencing. A PCR-based strategy was used to delete 162 amino acids of domain IV, from position 700 to 862. Plasmids pRS314/SNU114, containing the desired substitution or deletion, were sequenced and separately transformed into strain YPF8. The plasmid-shuffling strategy was applied (18Boeke J.D. Trueheart J. Natsoulis G. Fink G.R. Methods Enzymol. 1987; 154: 164-175Crossref PubMed Scopus (1067) Google Scholar). After selection at 25 °C in medium lacking tryptophan, transformants were streaked once on medium lacking tryptophan and grown at 25 °C. Patches were streaked three times on 5-fluoroorotic acid plates to select for cells lacking of the URA3 plasmid. Cells that survived on 5-fluoroorotic acid plates were streaked on rich medium, and their growth phenotypes were analyzed by incubating cells and dilutions (5 × 104, 5 × 103, and 5 × 102 at 37, 30, 25, and 17 °C for 1–2 days or longer). A good temperature-sensitive strain was one that did not grow in -Ura plates at any of the temperatures and, in addition, did not grow in rich medium at 30 or 37 °C but grew at 25 °C.Antibody Production—The SNU114 BamHI SnabI fragment from plasmid pBSIISK(–)/SNU114, was cloned into vector pQE-30 (Qiagen) between BamHI and SmaI restriction sites. The resulting plasmid was transformed into Escherichia coli strain M15 (pREP4) and used to overexpress a histidine tag containing Snu114p fragment of 59.5 kDa bearing the N-terminal acidic domain, the G domain, and part of domain II. The protein was purified from crude cell lysates by Ni-NTA chelate chromatography, according to the instructions of the manufacturer (Qiagen), and used to immunize a rabbit. The polyclonal antiserum obtained was specific for Snu114p, as tested by Western blot analysis and ECL detection (Amersham Biosciences) of total yeast snRNPs and purified Snu114p. The protein could be clearly detected when the serum was diluted 1:5000.Purification of Snu114p—His-tagged Snu114p was purified from YPF39 extracts by Ni-NTA chelate chromatography according to the protocol supplied by Novagen. The binding buffer contained 20 mm Tris-HCl, 500 mm NaCl, and 5 mm imidazole (pH 8.0). In the wash buffer the imidazole concentration was increased to 10 mm. In a second washing step, 50 mm NaH2PO4, 300 mm NaCl, and 10 mm imidazole were used instead. Elution was performed with the latter buffer containing in addition 10% glycerol and 50, 100, 150, 200, or 250 mm imidazole. Most of Snu114p was eluted between 100 and 150 mm imidazole. To the eluted protein, glycerol and β-mercaptoethanol were added to give a final concentration of 20% and 20 mm, respectively. For further purification, the peak fractions of the nickel column eluate were loaded separately on a Superdex G-200 column (Amersham Biosciences). The running buffer contained 50 mm NaH2PO4 and 300 mm NaCl. Fractions of 100 μl were collected, and a 30-μl aliquot of each fraction was analyzed by SDS-PAGE.GTP Cross-linking Assays and Immunoprecipitation—15 pmol of purified Snu114p was incubated in 25 μl of 50 mm Tris-HCl (pH 8.0), 150 mm KCl, 1.3 mm dithiothreitol, 2.5 mm MgCl2, and 0.6 μg/μl poly(U) with 20 μCi (6.6 pmol) of [α-32P]GTP (Amersham Biosciences, specific activity, 3000 Ci/mmol) for 5 min at room temperature. The subsequent UV cross-linking reaction was performed essentially as described previously (19Laggerbauer B. Lauber J. Lührmann R. Nucleic Acids Res. 1996; 24: 868-875Crossref PubMed Scopus (31) Google Scholar), except that the cross-linking time was extended to 7 min with the sample at a distance of 4.5 cm from the UV lamp. The cross-linked proteins were then immunoprecipitated with the anti-Snu114p antibody or non-immune serum. For the immunoprecipitations, 5–10 μl of anti-Snu114p antisera were coupled to 20 μl of protein A-Sepharose in 500 μl of buffer NET2–150 (50 mm Tris-HCl, pH 7.5, 150 mm NaCl, 0.05% Nonidet P-40) for 1.5 h at 4 °C and subsequently washed with three times 1 ml of NET2–150. The cross-linked samples or splicing extracts (400 μg of total proteins, see Figs. 3C and 5A) were added together with 500 μl of NET2–150 buffer to the washed beads, and Snu114p was precipitated from the reactions for 2 h at 4 °C. The beads were subsequently washed four times with 1.5 ml of NET2–150 and precipitated proteins were extracted with phenol/chloroform/isoamyl alcohol (50:49:1, v/v). After acetone precipitation, the proteins were separated by SDS-PAGE. The gel was stained with Coomassie Blue, and cross-linked proteins were detected by autoradiography.Fig. 5Strains containing the D271N and R487E snu114p mutants lead to low levels of U5 and U4/U6·U5 tri-snRNP particles when grown at high temperatures. A, immunoprecipitation of U5 and tri-snRNP particles obtained from wild-type, D271N, and R487E strains grown at 30 or 25 °C, respectively. Extracts (400 μg) were used in immunoprecipitation experiments using anti-Snu114p antibodies (lanes 1–6), anti-Prp8p antibodies (lanes 8, 9, 11, 12, 14, and 15) or non-immune serum (lanes 7, 10, and 13). After immunoprecipitation the reactions were washed with different salt concentrations, and the RNA content of the precipitates was analyzed by Northern blotting. The bands were quantified by PhosphorImager analysis. B, wild-type, D271N, and R487E extracts (2.5 mg) obtained from the corresponding strains grown at 30 or 25 °C were sedimented on a 10–30% glycerol gradient. The fractions were phenol-chloroform-extracted, and their snRNA content was analyzed by Northern blotting using U4, U5, and U6 snRNA probes or U1 and U2 snRNA probes (see D). C, Western blot of the gradient fractions from wild-type and D271N extracts using anti-Snu114p antibodies. The Western blot of the gradient fractions obtained from the R487E mutant extract showed very low levels of protein, and it is not shown.View Large Image Figure ViewerDownload Hi-res image Download (PPT)Splicing Extracts, in Vitro and in Vivo Splicing, Temperature Inactivation, and Complementation Assays—Splicing extracts of YPF36 (11Bartels C. Klatt C. Lührmann R. Fabrizio P. EMBO Rep. 2002; 3: 875-880Crossref PubMed Scopus (62) Google Scholar), YPF37, YPF39, and YCK13 were prepared by grinding frozen cells under liquid nitrogen (20Umen J.G. Guthrie C. Genes Dev. 1995; 9: 855-868Crossref PubMed Scopus (170) Google Scholar). For extract preparation the cells were grown at 25 or 30 °C. In vitro splicing reactions were performed as described previously (21Lin R.J. Newman A.J. Cheng S.C. Abelson J. J. Biol. Chem. 1985; 260: 14780-14792Abstract Full Text PDF PubMed Google Scholar), generally using actin pre-mRNA transcribed in the presence of [α-32P]UTP. For in vitro thermal inactivation according to Ref. 8Xie J. Beickman K. Otte E. Rymond B.C. EMBO J. 1998; 17: 2938-2946Crossref PubMed Scopus (47) Google Scholar, the prepared extracts were incubated in 120 mm KPO4 at 37 °C for the times indicated prior to the splicing reaction. When time-dependent measurements were made, the samples taken at each time point were placed on ice until the splicing reaction was initiated. As a control, the extract was incubated in 120 mm KPO4 on ice for the whole period of heat treatment. The splicing reactions after thermal inactivation were performed under standard splicing conditions (60 mm KPO4, 3% polyethylene glycol 8000, 2 mm ATP, 2.5 mm MgCl2, and 2 mm spermidine). Complementation was achieved by mixing equal volumes of the two thermally inactivated extracts before the splicing reaction was carried out. In the experiment shown in the left panel of Fig. 3B, thermal inactivation was performed by directly incubating the splicing reaction at 32 °C instead of at 25 °C. In vivo accumulation of unspliced pre-U3A and pre-U3B RNA transcripts was measured by primer extension as described previously (10Fabrizio P. Laggerbauer B. Lauber J. Lane W.S. Lührmann R. EMBO J. 1997; 16: 4092-4106Crossref PubMed Scopus (180) Google Scholar).Affinity Purification of Spliceosomes—For affinity purification of spliceosomes, actin pre-mRNA was transcribed in the presence of biotin-16-UTP (Roche Applied Science) with a UTP/biotin-16-UTP ratio of 20:1 (22Grabowski P.J. Sharp P.A. Science. 1986; 233: 1294-1299Crossref PubMed Scopus (115) Google Scholar). Standard splicing reaction mixtures (100 μl) contained 40 μl of extract and 270 fmol of biotinylated actin pre-mRNA. After incubation for 25 min at 32 °C, ATP was depleted by adding 4 mm glucose and incubating the samples for an additional 10 min. Subsequently, samples were added, together with 500 μl of NET2-75, to 25 μl of washed streptavidin beads (Roche Applied Science) in siliconized tubes. For affinity purification, the samples were rotated for 90 min at 4 °C. The precipitates were washed extensively with NET2-75, and 25 μl of 1× splicing buffer was added together with the indicated concentrations of different nucleotides. Subsequently, the samples were incubated for 25 min at 25 °C, and the precipitates were washed another three times with NET2-75. The precipitates were treated with proteinase K and extracted with phenol/chloroform. The RNAs were ethanol-precipitated and separated on a 10% denaturing gel. The gels were then blotted onto a nylon membrane (Amersham Biosciences), which was UV-irradiated at 120 mJ/cm2. Northern analyses with uniformly radiolabeled DNA probes specific for snRNAs U1, U2, U4, U5, and U6 was performed.RESULTSBinding of GTP by Snu114p—Because Snu114p exhibits the same domain structure as the ribosomal GTPase EF-2, including the GTP-binding and -hydrolyzing G domain (10Fabrizio P. Laggerbauer B. Lauber J. Lane W.S. Lührmann R. EMBO J. 1997; 16: 4092-4106Crossref PubMed Scopus (180) Google Scholar), we asked whether GTP is required for Snu114p function in splicing. First, we investigated whether Snu114p binds GTP. It was previously shown that the human homologue of Snu114p, the U5–116K protein, binds GTP when it is a part of the U5 snRNP (10Fabrizio P. Laggerbauer B. Lauber J. Lane W.S. Lührmann R. EMBO J. 1997; 16: 4092-4106Crossref PubMed Scopus (180) Google Scholar), but until now it was unclear whether the yeast protein Snu114p binds GTP independently of the other proteins and RNA present in the U5 snRNP complex. To address this question, we overexpressed Snu114p in yeast and purified it from the extract in two steps, nickel-agarose chromatography and size-exclusion chromatography (Fig. 1A).Fig. 1Purified Snu114p binds GTP specifically. A, SDS-PAGE of purified Snu114p, which was overexpressed in yeast and purified via Ni-NTA chromatography (Ni2+, lane 1) and size-exclusion chromatography (gel filtration, lane 2). For visualization the gel was stained with Coomassie Blue. The position of Snu114p is indicated with an arrow. The purity of the protein preparation was checked by mass spectrometry, which confirmed that Snu114p is 90–95% pure. The asterisk indicates a frequently detected contamination (alcohol dehydrogenase-1-P). In addition, traces of glyceraldehyde-3-phosphate dehydrogenase, cyclin-dependent protein kinase 33, and alkaline phosphatase 8 were found. B, binding of GTP to purified Snu114p. Snu114p was incubated with radioactively labeled GTP (lanes 1–12) or ATP (lane 13) in the absence or presence of 10× (66 pmol), 25× (165 pmol), and 50× (330 pmol) excess of unlabeled nucleotides. After cross-linking Snu114p was immunoprecipitated with anti-Snu114p antibody (lanes 1–10, 12, and 13) or non-immune serum (lane 11). The precipitates were loaded on a 10% SDS-PAGE and radioactive Snu114p (cross-linked to GTP or ATP) was detected by autoradiography. The lines indicate degradation products of Snu114p.View Large Image Figure ViewerDownload Hi-res image Download (PPT)Fig. 1B shows that [α-32P]GTP cross-links to purified Snu114p after UV irradiation (lanes 1 and 2, 6–10, and 12). Radiolabeled Snu114p, cross-linked to GTP, can be specifically immunoprecipitated with anti-Snu114p antibodies (Fig. 1B, lanes 1–10 and 12) but not with the non-immune serum (lane 11). The observed cross-link is stimulated, as in the case of the human U5 snRNP (10Fabrizio P. Laggerbauer B. Lauber J. Lane W.S. Lührmann R. EMBO J. 1997; 16: 4092-4106Crossref PubMed Scopus (180) Google Scholar), in the presence of poly(U) (compare lane 2 with lane 1). To determine the specificity of the cross-link obtained with labeled GTP, we performed competition experiments by adding an excess of unlabeled NTPs. Fig. 1B (lanes 3–5) shows that the [α-32P]GTP cross-linked to Snu114p can be competitively displaced by a 10-, 25-, and 50-fold excess of unlabeled GTP, but not by unlabeled ATP (lanes 6–8) or CTP or UTP (lanes 9 and 10). This result suggests that purified Snu114p has very high affinity for GTP only.Some [α-32P]ATP also became cross-linked to Snu114p, although to a lesser extent, corresponding to about 5% or less of the cross-linked [α-32P]GTP (Fig. 1B, lane 13). This cross-link could reflect a low affinity of Snu114p for ATP. Preliminary results indicate that this cross-link to ATP is prevented by competition from all of the unlabeled NTPs used, suggesting that Snu114p has unspecific affinity for ATP (data not shown).Mutagenesis of the G Domain of Snu114p—The finding that Snu114p indeed binds GTP in a highly specific manner (see above) and that a point mutation in Snu114p G domain leads to a lethal phenotype (10Fabrizio P. Laggerbauer B. Lauber J. Lane W.S. Lührmann R. EMBO J. 1997; 16: 4092-4106Crossref PubMed Scopus (180) Google Scholar), raised the intriguing question whether Snu114p hydrolyzes GTP, in analogy to the ribosomal translocases, during its function in splicing.Up to now GTP has never been reported as an essential cofactor for splicing. This could, however, be due to the fact that GTP is already available in the extract, or to the complexity of the splicing process. The non-hydrolyzable analogue of GTP, GTPγS, is not suitable to clarify this question, because it can inhibit splicing by binding not only to Snu114p but also to any of the numerous RNA unwindases that are involved in the splicing process and that are known to bind and hydrolyze promiscuously several of the NTPs (23Wang Y. Wagner J.D. Guthrie C. Curr. Biol. 1998; 8: 441-451Abstract Full Text Full Text PDF PubMed Scopus (83) Google Scholar).To address the complex question of GTP requirement in splicing, we performed site-directed mutagenesis of Snu114p G domain. This mutagenesis revealed that the G domain is far more se
Referência(s)