Pseudouridylation of yeast U2 snRNA is catalyzed by either an RNA-guided or RNA-independent mechanism
2005; Springer Nature; Volume: 24; Issue: 13 Linguagem: Inglês
10.1038/sj.emboj.7600718
ISSN1460-2075
AutoresMA Xiao-ju, Chunxing Yang, Andrei Alexandrov, Elizabeth J. Grayhack, Isabelle Behm‐Ansmant, Yi‐Tao Yu,
Tópico(s)RNA Research and Splicing
ResumoArticle16 June 2005free access Pseudouridylation of yeast U2 snRNA is catalyzed by either an RNA-guided or RNA-independent mechanism Xiaoju Ma Xiaoju Ma Department of Biochemistry and Biophysics, University of Rochester Medical Center, Rochester, NY, USA Search for more papers by this author Chunxing Yang Chunxing Yang Department of Biochemistry and Biophysics, University of Rochester Medical Center, Rochester, NY, USA Search for more papers by this author Andrei Alexandrov Andrei Alexandrov Department of Biochemistry and Biophysics, University of Rochester Medical Center, Rochester, NY, USA Search for more papers by this author Elizabeth J Grayhack Elizabeth J Grayhack Department of Biochemistry and Biophysics, University of Rochester Medical Center, Rochester, NY, USA Search for more papers by this author Isabelle Behm-Ansmant Isabelle Behm-Ansmant Laboratoire de Maturation des ARN et Enzymologie Moleculaire, UMR 7567 CNRS-UHP Nancy I, Faculte des Sciences, Vandoeuvre-les-Nancy, France Search for more papers by this author Yi-Tao Yu Corresponding Author Yi-Tao Yu Department of Biochemistry and Biophysics, University of Rochester Medical Center, Rochester, NY, USA Search for more papers by this author Xiaoju Ma Xiaoju Ma Department of Biochemistry and Biophysics, University of Rochester Medical Center, Rochester, NY, USA Search for more papers by this author Chunxing Yang Chunxing Yang Department of Biochemistry and Biophysics, University of Rochester Medical Center, Rochester, NY, USA Search for more papers by this author Andrei Alexandrov Andrei Alexandrov Department of Biochemistry and Biophysics, University of Rochester Medical Center, Rochester, NY, USA Search for more papers by this author Elizabeth J Grayhack Elizabeth J Grayhack Department of Biochemistry and Biophysics, University of Rochester Medical Center, Rochester, NY, USA Search for more papers by this author Isabelle Behm-Ansmant Isabelle Behm-Ansmant Laboratoire de Maturation des ARN et Enzymologie Moleculaire, UMR 7567 CNRS-UHP Nancy I, Faculte des Sciences, Vandoeuvre-les-Nancy, France Search for more papers by this author Yi-Tao Yu Corresponding Author Yi-Tao Yu Department of Biochemistry and Biophysics, University of Rochester Medical Center, Rochester, NY, USA Search for more papers by this author Author Information Xiaoju Ma1,‡, Chunxing Yang1,‡, Andrei Alexandrov1, Elizabeth J Grayhack1, Isabelle Behm-Ansmant2 and Yi-Tao Yu 1 1Department of Biochemistry and Biophysics, University of Rochester Medical Center, Rochester, NY, USA 2Laboratoire de Maturation des ARN et Enzymologie Moleculaire, UMR 7567 CNRS-UHP Nancy I, Faculte des Sciences, Vandoeuvre-les-Nancy, France ‡These authors contributed equally to this work *Corresponding author. Department of Biochemistry and Biophysics, University of Rochester Medical Center, 601 Elmwood Avenue, Rochester, NY 14642, USA. Tel.: +1 585 275 1271; Fax: +1 585 275 6007; E-mail: [email protected] The EMBO Journal (2005)24:2403-2413https://doi.org/10.1038/sj.emboj.7600718 PDFDownload PDF of article text and main figures. ToolsAdd to favoritesDownload CitationsTrack CitationsPermissions ShareFacebookTwitterLinked InMendeleyWechatReddit Figures & Info Yeast U2 small nuclear RNA (snRNA) contains three pseudouridines (Ψ35, Ψ42, and Ψ44). Pus7p and Pus1p catalyze the formation of Ψ35 and Ψ44, respectively, but the mechanism of Ψ42 formation remains unclear. Using a U2 substrate containing a single 32P radiolabel at position 42, we screened a GST-ORF library for pseudouridylase activity. Surprisingly, we found a Ψ42-specific pseudouridylase activity that coincided with Nhp2p, a protein component of a Box H/ACA sno/scaRNP (small nucleolar/Cajal body-specific ribonucleoprotein). When isolated by tandem affinity purification (TAP), the other protein components of the H/ACA sno/scaRNP also copurified with the pseudouridylase activity. Micrococcal nuclease-treated TAP preparations were devoid of pseudouridylase activity; however, activity was restored upon addition of RNAs from TAP preparations. Pseudouridylation reconstitution using RNAs from a Box H/ACA RNA library identified snR81, a snoRNA known to guide rRNA pseudouridylation, as the Ψ42-specific guide RNA. Using the snR81-deletion strain, Nhp2p- or Cbf5p-conditional depletion strain, and a cbf5 mutation strain, we further demonstrated that the pseudouridylase activity is dependent on snR81 snoRNP in vivo. Our data indicate that snRNA pseudouridylation can be catalyzed by both RNA-dependent and RNA-independent mechanisms. Introduction All five spliceosomal small nuclear RNAs (snRNAs) (U1, U2, U4, U5, and U6) contain a large number of post-transcriptional modifications, including a 5′ cap and internal 2′-O-methylation and pseudouridylation sites (Massenet et al, 1998). Remarkably, the internally modified nucleotides are often clustered in regions important for pre-mRNA splicing, suggesting their functional role in this process (Yu et al, 2005). Among all snRNAs, vertebrate U2 is the most extensively modified (Massenet et al, 1998). In fact, more than 10% of its total ∼189 nucleotides are either 2′-O-methylated or pseudouridylated. Most noticeably, all six uridines in the branch site recognition region and its adjacent 3′ region are converted to pseudouridines following transcription. Three of these pseudouridines (Ψ34, Ψ41, and Ψ43) have counterparts in Saccharomyces cerevisiae U2 snRNA (Ψ35, Ψ42, and Ψ44, as numbered in yeast U2) (Figure 1). Figure 1.Sequences of the branch site recognition region of vertebrate U2 and S. cerevisiae U2. The three pseudouridines numbered 34, 41, and 43 in vertebrate U2 are equivalent to the three pseudouridines numbered 35, 42, and 44 in yeast U2 (three arrows). Cm and Um in vertebrate U2 indicate 2′-O-methylated residues. The branch site recognition sequences are boxed. Download figure Download PowerPoint The functional importance of U2 modifications in pre-mRNA splicing and small nuclear ribonucleoprotein (snRNP) biogenesis was recently confirmed experimentally in vertebrates (Yu et al, 1998; Donmez et al, 2004; Zhao and Yu, 2004) and in S. cerevisiae (Yang et al, 2005). Consistently, the NMR study from the Greenbaum group (Newby and Greenbaum, 2002) also showed that, when paired with the pre-mRNA branch site, a pseudouridine in the U2 branch site recognition region (equivalent to Ψ35 of yeast U2) is favored over uridine for maintaining the bulge of the branch point nucleotide adenosine for nucleophilic attack in the first step of splicing reaction. The importance of the pseudouridine in splicing is further supported by the recent work of Valadkhan and Manley (2003), who showed that pseudouridylation at the same position greatly enhances the production of X-RNA, a product generated by a splicing-related reaction in a cell- and protein-free system (Valadkhan and Manley, 2003). Experimental and bioinformatic data accumulated thus far suggest that vertebrate U2 snRNA and the other vertebrate spliceosomal snRNAs are modified by an RNA-guided mechanism (Yu et al, 2005). Specifically, pseudouridylation is catalyzed by a class of RNA–protein complexes called Box H/ACA sno/scaRNPs (small nucleolar/Cajal body-specific RNPs), each containing four core proteins and a single RNA that base-pairs with the target RNA, thereby directing pseudouridylation at specific sites (Ganot et al, 1997; Ni et al, 1997). Similarly, 2′-O-methylation is catalyzed by another class of RNA–protein complexes, named Box C/D sno/scaRNPs, in which a single RNA component serves as a guide for site-specific modification (Cavaille et al, 1996; Kiss-Laszlo et al, 1996). By contrast, current experimental data indicate that modification of S. cerevisiae U2 snRNA is catalyzed by an RNA-independent mechanism (Massenet et al, 1999; Ma et al, 2003). It has been reported that pseudouridylation at positions 35 and 44 is catalyzed by Pus7p (Ma et al, 2003) and Pus1p (Massenet et al, 1999), respectively. Thus, two of the three pseudouridines in yeast U2 are introduced into the branch site recognition region via a protein-only mechanism. By analogy, it has thus far been assumed that the other pseudouridine (Ψ42) in yeast U2 is also produced via an RNA-independent mechanism, although this has yet to be demonstrated. Consistent with this idea, extensive yeast database searches have failed to reveal any guide RNAs specific for spliceosomal snRNA modifications, although a near-complete set of sno/scaRNA guides have been identified for the modification of yeast ribosomal RNAs (Balakin et al, 1996; Lowe and Eddy, 1999; Samarsky and Fournier, 1999; Bachellerie et al, 2002; Edvardsson et al, 2003; Schattner et al, 2004). To address the mechanism by which pseudouridylation is achieved at position 42 in yeast U2, we used U42-radiolabeled U2 snRNA in an in vitro system to screen a yeast GST-ORF fusion protein library and a Box H/ACA RNA library for the candidate pseudouridylase. We show that snR81 snoRNP, a Box H/ACA snoRNP particle, is necessary for pseudouridylation at position 42, indicating that the formation of Ψ42 in yeast U2 is catalyzed by an RNA-dependent mechanism that is completely different from that for the other two pseudouridines, Ψ35 and Ψ44 (Massenet et al, 1999; Ma et al, 2003). Thus, pseudouridylation within yeast U2 is catalyzed by two different mechanisms—an RNA-independent mechanism for Ψ35 and Ψ44 and an RNA-guided mechanism for Ψ42. Results A pseudouridylation activity responsible for the formation of Ψ42 coincides with the GST-Nhp2p fusion protein To identify the enzymatic activity responsible for the formation of Ψ42 in yeast U2 snRNA, we used a previously established in vitro system in which U2 snRNA, radiolabeled only at the 5′ side of U42, was used to screen a yeast GST-ORF fusion protein library for the candidate pseudouridylase (Ma et al, 2003). Briefly, a yeast GST-ORF genomic library, consisting of 6144 strains, each expressing a unique GST-ORF fusion protein under the control of the PCUP1 promoter, was grown in 64 pools of 96 strains each (Phizicky et al, 2002). After induction, the expressed GST-ORF fusion proteins in every pool were purified via glutathione agarose chromatography (Phizicky et al, 2002) and incubated under the pseudouridylation conditions with the U2 substrate containing 32P-labeled U42 (see Materials and methods). U2 was then recovered, digested to completion with nuclease P1, and analyzed by thin-layer chromatography (TLC). As shown in Figure 2A, only pool 44 was tested positive for the pseudouridylation activity. The 96 strains in pool 44 were then arranged individually into 12 columns (columns 1–12) and eight rows (rows A–H), and the combined strains in each column or row (i.e., 'subpools') were further assayed for pseudouridylation. Two of the subpools, column 2 and row A, tested positive for pseudouridylation, pinpointing strain 2A (column 2 and row A) as the source of the activity (Figure 2B). Indeed, the GST-ORF fusion protein purified from strain 2A was sufficient to catalyze the conversion of U42 to Ψ42 (Figure 2C). Surprisingly, strain 2A was found to express a fusion protein containing Nhp2p, which is one of the four core protein components of Box H/ACA sno/scaRNPs (Henras et al, 1998; Watkins et al, 1998; Dragon et al, 2000; Pogacic et al, 2000; Watanabe and Gray, 2000; Rozhdestvensky et al, 2003; Wang and Meier, 2004). This result suggests that Nhp2p alone, or a Box H/ACA sno/scaRNP that copurifies with Nhp2p, is responsible for the pseudouridylation activity. Figure 2.Ψ42-specific pseudouridylase activity coincides with Nhp2p. Yeast U2 snRNA containing a single 32P label at the 5′ side of U42 was used to screen a GST-ORF library (Phizicky et al, 2002) for pseudouridylase activity. After the pseudouridylation reaction, U2 snRNA was digested with nuclease P1 and analyzed by TLC. (A) Of all 64 pools of GST-ORF fusion proteins, pool #44 was positive. (B) The 96 strains in pool #44 were arranged into 12 columns and eight rows, with each column and row representing a 'subpool'. The column #2 subpool and row A subpool were positive, implicating strain 2A as the source of the pseudouridylase activity. (C) Individual strains were further screened to confirm that strain 2A, expressing GST-Nhp2p, was indeed positive. The migration of labeled 32PU and 32PΨ is indicated. Control reactions (Con.) contained either yeast cell extract (+) or water (−). Download figure Download PowerPoint The Nhp2p-associated pseudouridylase activity specifically catalyzes Ψ42 formation Having identified a pseudouridylase activity, we then tested whether it was specific for the formation of Ψ42. For this purpose, we prepared three synthetic U2 substrates, each containing a single 32P label at the 5′ side of U42 (as described above), U35, or U44 (all of these sites are naturally occurring pseudouridylation sites). The three U2 snRNAs were incubated separately with the Nhp2p-associated pseudouridylase prepared from strain 2A. As shown in Figure 3, the Nhp2p-associated pseudouridylase activity converted U42 to Ψ42, and the extent of conversion correlated directly with the amount of pseudouridylase added to the reaction (lanes 2–4). In contrast, the Nhp2p-associated activity failed to catalyze the conversion of uridine into pseudouridine at position 35 or 44 (lanes 6–8 and 10–12). As controls, yeast cell extracts were able to pseudouridylate U2 snRNA at all three positions (lanes 5 and 9, and see Figure 2, control lanes). Taken together, these results indicate that the Nhp2p-associated pseudouridylase activity is specific for position 42. Figure 3.Nhp2p-associated pseudouridylase activity is specific for the formation of Ψ42 in yeast U2 snRNA. Yeast U2 snRNA containing a 32P label at the 5′ side of either U42 (lanes 1–4), U35 (lanes 5–8), or U44 (lanes 9–12) was used in pseudouridylation assays (see Materials and methods, and Figure 2 legend). The final concentration of the GST-Nhp2p preparation (purified from strain 2A) in the reactions was ∼15 ng/ml (lanes 2, 6, and 10), ∼75 ng/ml (lanes 3, 7, and 11), or ∼375 ng/ml (lanes 4, 8, and 12). As controls, the wild-type yeast cell extract (lanes 5 and 9) or water (lane 1) was used in place of the GST-Nhp2p preparation. Download figure Download PowerPoint Tandem affinity purification, targeting either Nhp2p, Cbf5p, or Gar1p, copurifies the pseudouridylase activity for Ψ42 formation Since the pseudouridylase activity coincided with GST-Nhp2p, it is possible that Nhp2p is a modifying enzyme. To test this possibility, GST-Nhp2p or His-Nhp2p was expressed in Escherichia coli and purified by affinity chromatography. The tagged Nhp2p proteins were subsequently assayed for pseudouridylation activity using the same procedure described above. However, no pseudouridylation activity was detected (data not shown), suggesting that Nhp2p alone does not have pseudouridylase activity. Another possibility is that GST-Nhp2p expressed in yeast copurified a Box H/ACA sno/scaRNP complex, and it is this complex that comprises the Ψ42-specific pseudouridylase activity. According to published data (Henras et al, 1998; Watkins et al, 1998; Dragon et al, 2000; Pogacic et al, 2000; Watanabe and Gray, 2000; Rozhdestvensky et al, 2003; Wang and Meier, 2004) that Box H/ACA sno/scaRNP contains four core proteins, Nhp2p, Cbf5p, Gar1p, and Nop10p, we used the tandem affinity purification (TAP) procedure (Rigaut et al, 1999) to insert a TAP tag at the 3′ end of each of these four genes in yeast chromosomes. Selection and sequencing revealed successful TAP tagging of all but the NOP10 gene, as no live cells were selected after transformation with a NOP10-specific DNA tag. After TAP purification, we assessed the pseudouridylase activity and protein composition. The pseudouridylation assay showed that each of the three TAP-tagged proteins copurified the pseudouridylase activity that effectively converted U42 to Ψ42 (Figure 4A, lanes 3–5). Importantly, when compared with yeast cell extract (lane 2), the TAP preparations were greatly enriched for the pseudouridylase activity (compare lanes 3–5 with lane 2, and see Discussion). As a control, TAP-tagged Mak5p, an unrelated protein, did not copurify the pseudouridylase activity (lane 6). Figure 4B presents an SDS–PAGE gel stained with Coomassie blue. Each of the three TAP-tagged proteins (Gar1p, Cbf5p, and Nhp2p) coselected the same set of four proteins (compare lanes 3–5; the band migrating below 42 kDa is Protease 3C, which was used during the TAP procedure). Based on previously published data and the sizes of the proteins (Wang and Meier, 2004), we inferred that they were the four core proteins, Cbf5p, Gar1p, Nhp2p, and Nop10p, of Box H/ACA sno/scaRNPs (Figure 4B). In fact, we confirmed the identity of each protein by protein sequencing (data not shown). Because the proteins used for selection contained a C-terminal tag, the tagged protein (Gar1p-TAP, Cbf5p-TAP, or Nhp2p-TAP) in each of the three lanes (lanes 3–5) migrated slower than its native form. As expected, TAP-tagged Mak5p, as a control, did not coselect any of the four core proteins (lane 6). Taken together, these results, coupled with our current knowledge of Box H/ACA sno/scaRNPs, suggest that each of the four core proteins contributes to the pseudouridylase activity. Figure 4.TAP-purified Box H/ACA sno/scaRNPs catalyze Ψ42 formation. (A) Yeast U2 snRNA containing a single 32P label at the 5′ side of U42 was used to assess the pseudouridylase activity of TAP preparations targeting Gar1p (lane 3), Cbf5p (lane 4), Nhp2p (lane 5), or Mak5p (as a control, lane 6). The final protein concentration of the TAP preparations in the pseudouridylation reaction (see Materials and methods, and previous figure legends) was ∼75 ng/ml. As controls, water (lane 1) or the wild-type yeast cell extract (lane 2) was used in place of the TAP preparations. (B) The protein compositions of TAP preparations were analyzed on an SDS–PAGE gel stained with Coomassie blue. Lanes 3, 4, and 5 contain samples of Gar1p-TAP, Cbf5p-TAP, and Nhp2p-TAP preparations, respectively. Lane 6 is a control containing the sample of Mak5p-TAP preparation. The positions of the Box H/ACA sno/scaRNP core proteins, Cbf5p, Gar1p, Nhp2p, and Nop10p (both TAP-tagged and native forms), are indicated on the left. The TAP-tagged proteins migrated more slowly than their native forms. The label 3C indicates protease 3C, which was used during TAP purification. Lane M is a standard protein marker lane. Download figure Download PowerPoint Ψ42 formation is RNA dependent Given that a complete Box H/ACA sno/scaRNP contains an RNA guide as well as protein components, our results suggested that yeast U2 pseudouridylation at position 42 is catalyzed by an RNA-guided or RNA-dependent mechanism. To test this hypothesis experimentally, we treated the GAR1-TAP preparation with micrococcal nuclease in the presence of Ca2+. After inactivation of the nuclease by chelating Ca2+ with EGTA, we performed the pseudouridylation assay using the U2 substrate containing a single 32P label at the 5′ side of U42. As shown in Figure 5A, the mock-treated TAP preparation was highly active in converting U42 into Ψ42 (lane 2). However, upon micrococcal nuclease treatment, the TAP preparation completely lost its pseudouridylase activity (lane 3). Importantly, when supplemented with total Box H/ACA sno/scaRNA extracted from any of the three TAP preparations (GAR1-TAP, NHP2-TAP, or CBF5-TAP), the micrococcal nuclease-treated TAP preparation regained pseudouridylase activity—at least a substantial fraction of U42 was converted into Ψ42 (lane 4). As a control, a reconstitution reaction was carried out in parallel using a yeast 25S rRNA fragment containing a single 32P label at the 5′ side of U989, a naturally occurring pseudouridylation site known to be modified by a Box H/ACA snoRNP (Ofengand and Fournier, 1998). As expected, the mock-treated GAR1-TAP preparation effectively converted U989 into Ψ989 (Figure 5B, lane 2), whereas the micrococcal nuclease-treated GAR1-TAP preparation did not catalyze this pseudouridylation reaction (Figure 5B, lane 3). However, upon addition of total RNA extracted from the TAP preparations, the pseudouridylase activity of the micrococcal nuclease-treated GAR1-TAP preparation was regenerated (Figure 5B, lane 4). Taken together, these results demonstrate that a complete Box H/ACA sno/scaRNP, including both protein and RNA components, constitutes the pseudouridylase activity that catalyzes the formation of Ψ42. Figure 5.The formation of Ψ42 in yeast U2 requires an RNA component. (A) Pseudouridylation assay was performed using yeast U2 containing a single 32P label at the 5′ side of U42 (see Materials and methods, and previous figure legends). (B) Pseudouridylation assay was performed using a fragment of yeast 25S rRNA containing a single 32P label at the 5′ side of U989. In lane 2 (in panels A and B), the Gar1p-TAP preparation (∼75 ng/ml) was present in the reaction. In lane 3 (A, B), micrococcal nuclease (MN)-treated Gar1p-TAP preparation was used. The reaction in lane 4 (A, B) was identical to that in lane 3 except that it contained total RNA prepared from the Gar1p-TAP preparation (not treated with micrococcal nuclease). Lane 1 in each case shows a control in which no TAP preparation was added. Download figure Download PowerPoint SnR81 is the guide RNA responsible for the formation of Ψ42 To identify the guide RNA responsible for Ψ42 formation, we took advantage of our powerful reconstitution system discussed above (see Figure 5). Briefly, total Box H/ACA sno/scaRNA was isolated from the GAR1-TAP preparation, and subsequently fractionated on a denaturing polyacrylamide gel. The gel (entire lane) was then sliced, from top to bottom (based on size), into 11 fractions (Figure 6A), and RNAs were recovered from each fraction and added to the micrococcal nuclease-treated GAR1-TAP preparation to test their ability to reconstitute U2 pseudouridylation at position 42. As shown in Figure 6B, fraction E (ranging from ∼180 to ∼200 nucleotides) was able to reconstitute the pseudouridylase activity (lane 9), suggesting strongly that the guide RNA responsible for Ψ42 formation was present in fraction E and is about 180–200 nucleotides long. However, it appeared that fraction E contained a number of Box H/ACA RNAs, as a smeared signal ranging from ∼180 to 200 nucleotides in size was observed when the total RNA extracted from this fraction was 3′ radiolabeled with 32PCp (data not shown). Figure 6.snR81 snoRNA directs the formation of Ψ42 in yeast U2. (A) Total Box H/ACA RNA isolated from the Gar1p-TAP preparation was fractionated on an 8% polyacrylamide–8 M urea gel. The entire lane (lane 2) was sliced into 11 fractions (A–K). RNAs were eluted from each fraction for reconstitution analysis. Lane 1 is a size marker of MspI-digested pBR322 DNA. (B) Fractionated RNAs (A–K, lanes 5–15) were added to micrococcal nuclease-treated Gar1p-TAP preparation (lanes 3–15) to reconstitute U2 pseudouridylation at position 42. Lanes 3 and 4 are controls in which no RNA (lane 3) or unfractionated total Box H/ACA RNA (lane 4) was added to the reaction. Lane 1 is an unreacted singly radiolabeled U2 snRNA. In lane 2, U2 pseudouridylation was carried out using mock-treated Gar1p-TAP preparation. (C) Reconstitution of Ψ42 formation was carried out using micrococcal nuclease-treated Gar1p-TAP preparation (lanes 2–15) and individual RNAs (lanes 4–15) derived from a Box H/ACA RNA library (see Materials and methods). Lanes 2 and 3 are controls where the reaction was supplemented with no RNA (lane 3) or total RNA isolated from fraction E (see panel B) (lane 2). In lane 1, mock-treated Gar1p-TAP preparation was used. Sequencing data indicated that RNA 8 in lane 11 was snR81 snoRNA. (D) Primary sequence and secondary structure of snR81 snoRNA. As indicated, the 3′ pseudouridylation pocket guides 25S rRNA pseudouridylation at position 1051 (Schattner et al, 2004), whereas the 5′ pseudouridylation pocket directs U2 snRNA pseudouridylation at position 42 (this work). The conserved H and ACA boxes are shaded. Download figure Download PowerPoint To pinpoint which RNA in fraction E is responsible for the activity, a Box H/ACA RNA library was constructed. Briefly, RNAs recovered from fraction E were 3′ polyadenylated by poly(A) polymerase. The resulting RNAs were then used as templates for oligo(dT)-primed reverse transcription, generating complementary DNAs (cDNAs) that contained a 5′ poly(dT) tail. The cDNAs were further 3′ tailed with a stretch of guanosines using terminal deoxynucleotidyl transferase (TDT), and were subsequently amplified by PCR using an oligo(dC) and an oligo(dT) as primers. The amplified DNAs were cloned and transformed into an E. coli strain, resulting in a large number of colonies on a solid medium. Plasmids purified from single colonies were used as templates for in vitro RNA transcription. The transcribed RNAs were then tested for their ability to reconstitute U2 pseudouridylation at position 42 in the micrococcal nuclease-treated GAR1-TAP preparation. As shown in Figure 6C, one functional RNA (lane 11) was identified among 25 RNAs tested. Sequencing of its corresponding plasmid indicated that the RNA was snR81, a recently identified Box H/ACA snoRNA responsible for 25S rRNA pseudouridylation at position 1051 (Schattner et al, 2004). SnR81 has two hairpins, each containing a pseudouridylation pocket. According to Schattner et al (2004), the 3′ pocket is specific for Ψ1051 formation in 25S rRNA, whereas the 5′ pocket has no known target. Remarkably, sequence comparison indicated a perfect match between the guide sequence in the 5′ pocket and the sequence surrounding U42 of yeast U2 snRNA, positioning U42 perfectly at the pocket for pseudouridylation (see Figure 6D). Depletion of either Nhp2p or Cbf5p, or deletion of snR81 precludes Ψ42 formation in vivo To confirm that the snR81 Box H/ACA sno/scaRNP is the natural pseudouridylase for Ψ42, we next analyzed U2 pseudouridylation in vivo under conditions in which one of the Box H/ACA sno/scaRNP core proteins was depleted. Here, we used a conditional depletion approach because all four core proteins of the sno/scaRNP complex are essential for cell growth. Specifically, we constructed two strains in which the transcription of either NHP2 or CBF5 was under the control of the PGAL1 promoter. When grown in galactose-containing medium, the two mutant strains grew as well as the wild-type strain. However, when the galactose medium was replaced by glucose-containing medium, the Gal promoter was shut-off, and Nhp2p or Cbf5p was gradually depleted in the mutant cells. After about four generations (∼10 h after the medium switch), the mutant cells grew significantly slower compared with the wild-type cells (Figure 7A). Both mutant and wild-type cells were collected at three different time points (6, 16, and 48 h) after the medium switch. Consistent with previous report (Lafontaine et al, 1998), depletion for 48 h resulted in almost complete loss of Box H/ACA sno/scaRNP proteins (data not shown). Total RNA was then extracted from the cells, and pseudouridylation of endogenous U2 snRNA was analyzed by CMC (N-cyclohexyl-N′-(2-morpholinoethyl)-carbodiimid metho-p-toluolsulfonate) modification followed by primer extension (Bakin and Ofengand, 1993). Figure 7.Depletion of either Nhp2p or Cbf5p specifically abolishes Ψ42 formation in yeast U2 snRNA in vivo. (A) Growth curves for three yeast strains after the switch from galactose medium to glucose medium (YPD). Closed circles represent the nhp2-depletion strain in which chromosomal nhp2 was replaced by a plasmid containing a wild-type NHP2 gene under the control of the Gal promoter. Closed triangles represent the strain in which chromosomal cbf5 was replaced by a plasmid containing a wild-type CBF5 gene under the control of the Gal promoter. Closed squares represent the wild-type strain. (B) The U2 pseudouridylation assay (CMC modification followed by primer extension) was performed using total RNA isolated either from the wild-type strain that had been incubated in YPD for 6 h (lanes 1 and 2), from the nhp2-depletion strain that had been incubated in YPD for 6 h (lanes 3 and 4), 16 h (lane 5), or 48 h (lane 6), or from the cbf5-depletion strain that had been incubated in YPD for 6 h (lanes 7 and 8), 16 h (lane 9), or 48 h (lane 10). In lanes 1, 3, and 7, CMC modification was omitted. The lanes labeled U, C, G, and A represent a yeast primer-extension sequencing ladder. The lane labeled H2O contained no ddNTP. The three primer-extension stops/pauses representing the three pseudouridines (Ψ35, Ψ42, and Ψ44) in U2 snRNA are indicated. (C) A pseudouridylation assay for yeast 25S rRNA was performed exactly as in panel B except that the primer-extension primer was complementary to yeast 25S rRNA (see Materials and methods). The naturally occurring pseudouridylation sites are indicated. Download figure Download PowerPoint The pseudouridylation results are consistent with the growth data. As shown in Figure 7B, when RNA was isolated from cells that had been incubated for only 6 h after the medium switch (at this time point, no growth defect was observed for the mutant strains; see Figure 7A), primer extension clearly detected all three U2 pseudouridine stops/pauses, regardless of which strain was used (compare lanes 2, 4, and 8). This was not the case, however, when we isolated RNA from nhp2 (lanes 5 and 6) or cbf5 (lanes 9 and 10) depletion strains that had gro
Referência(s)