Surveillance of nuclear-restricted pre-ribosomes within a subnucleolar region of Saccharomyces cerevisiae
2006; Springer Nature; Volume: 25; Issue: 7 Linguagem: Inglês
10.1038/sj.emboj.7601035
ISSN1460-2075
AutoresChristophe Dez, Jonathan Houseley, David Tollervey,
Tópico(s)RNA and protein synthesis mechanisms
ResumoArticle16 March 2006free access Surveillance of nuclear-restricted pre-ribosomes within a subnucleolar region of Saccharomyces cerevisiae Christophe Dez Christophe Dez Search for more papers by this author Jonathan Houseley Jonathan Houseley Search for more papers by this author David Tollervey Corresponding Author David Tollervey Wellcome Trust Centre for Cell Biology, University of Edinburgh, Edinburgh, UK Search for more papers by this author Christophe Dez Christophe Dez Search for more papers by this author Jonathan Houseley Jonathan Houseley Search for more papers by this author David Tollervey Corresponding Author David Tollervey Wellcome Trust Centre for Cell Biology, University of Edinburgh, Edinburgh, UK Search for more papers by this author Author Information Christophe Dez, Jonathan Houseley and David Tollervey 1 1Wellcome Trust Centre for Cell Biology, University of Edinburgh, Edinburgh, UK *Corresponding author. Institute of Cell and Molecular Biology, University of Edinburgh, University of Edinburgh, Swann Building, King's Buildings, Mayfield Road, Edinburgh EH9 3JR, UK. Tel.: +44 131 650 7092; Fax: +44 131 650 7040; E-mail: [email protected] The EMBO Journal (2006)25:1534-1546https://doi.org/10.1038/sj.emboj.7601035 Correction(s) for this article Surveillance of nuclear-restricted pre-ribosomes within a subnucleolar region of Saccharomyces cerevisiae07 June 2006 PDFDownload PDF of article text and main figures. ToolsAdd to favoritesDownload CitationsTrack CitationsPermissions ShareFacebookTwitterLinked InMendeleyWechatReddit Figures & Info We previously hypothesized that HEAT-repeat (Huntington, elongation A subunit, TOR) ribosome synthesis factors function in ribosome export. We report that the HEAT-repeat protein Sda1p is a component of late 60S pre-ribosomes and is required for nuclear export of both ribosomal subunits. In strains carrying the ts-lethal sda1–2 mutation, pre-60S particles were rapidly degraded following transfer to 37°C. Polyadenylated forms of the 27S pre-rRNA and the 25S rRNA were detected, suggesting the involvement of the Trf4p/Air/Mtr4p polyadenylation complex (TRAMP). The absence of Trf4p suppressed polyadenylation and stabilized the pre-rRNA and rRNA. The absence of the nuclear exosome component Rrp6p also conferred RNA stabilization, with some hyperadenylation. We conclude that the nuclear-restricted pre-ribosomes are polyadenylated by TRAMP and degraded by the exosome. In sda1–2 strains at 37°C, pre-40S and pre-60S ribosomes initially accumulated in the nucleoplasm, but then strongly concentrated in a subnucleolar focus, together with exosome and TRAMP components. Localization of pre-ribosomes to this focus was lost in sda1–2 strains lacking Trf4p or Rrp6p. We designate this nucleolar focus the No-body and propose that it represents a site of pre-ribosome surveillance. Introduction Ribosome synthesis largely occurs within the nucleolus but late maturation steps take place in the nucleoplasm, following release of the pre-ribosomes from the nucleolus, and in the cytoplasm, following nucleocytoplasmic export; reviewed in Venema and Tollervey (1999), de la Cruz et al (2003), Fromont-Racine et al (2003) and Dez and Tollervey (2004). During maturation, a complex processing pathway converts the 35S pre-rRNA into the mature 18S, 5.8S and 25S rRNAs (see Figure 1A and Supplementary Figure S1), whereas the 5S rRNA is independently transcribed. Figure 1.Sda1p is a component of late nuclear 60S pre-ribosomes. (A) The 35S pre-rRNA contains the sequences of the mature 18S, 5.8S and 25S rRNAs, which are separated by internal transcribed spacers 1 and 2 (ITS1 and ITS2) and flanked by the 5′ and 3′ external transcribed spacers (5′ETS and 3′ETS). Location of the different oligonucleotides used in this study is shown. (B, C) Northern analysis of rRNAs co-precipitated with Sda1-TAP or from extracts of cells lacking a tagged protein. Immunoprecipitation was performed on cell extracts using IgG-Sepharose. RNAs were extracted from the pellet after precipitation (lanes IP) or from an amount of cell extract corresponding to 1/20 (B, 1.2% denaturing agarose gel) or 1/10 (C, 8% polyacrylamide/urea gel) of the total amount used for the precipitations (lanes T). Following separation, RNAs were transferred to a nylon membrane and hybridized with the antisense oligonucleotides indicated to the right of each panel. Download figure Download PowerPoint The pre-ribosomal subunits are exported through the nuclear pore complex (NPC) to the cytoplasm prior to final maturation. In fast-growing strains of yeast, each NPC exports around 25–30 ribosomal subunits per minute, implying a very efficient transport mechanism. Ribosomal subunit export requires several pre-ribosomal proteins, as well as the exportin Crm1p/Xpo1p (Ho et al, 2000; Gadal et al, 2001; Moy and Silver, 2002; Thomas and Kutay, 2003; Trotta et al, 2003; Oeffinger et al, 2004), suggesting that multiple factors are required to allow effective transport of the ribosomal subunits. Free diffusion of molecules through the lumen of the NPC is restricted by the hydrophobic, FG-repeat regions of nucleoporins (Ribbeck and Gorlich, 2002; Gorlich et al, 2003). The translocation of large molecules through the NPC is strongly inhibited unless they can interact with the FG-repeats. However, ribosome subunits are large, highly structured and have many exposed, hydrophilic rRNA regions, and are very unlikely to interact favorably with the NPC. The ability to pass through the lumen of the NPC can be conferred by association with members of the importin-β/karyopherin-β family of transport factors, which consist of repeated, predominantly α-helical motifs termed HEAT-repeats (Huntington, elongation A subunit, TOR) that are related to Armadillo-like repeats (Andrade et al, 2001). We previously identified a family of ribosome synthesis factors that are made up of HEAT-repeats, and predicted that they would function in ribosomal subunit export (Dlakic and Tollervey, 2004; Oeffinger et al, 2004). Specifically we suggested that these factors might form extended structures that help cover the hydrophilic RNA surfaces of the ribosomal subunits, allowing their passage through the hydrophobic mesh formed by the FG-repeats (Dlakic and Tollervey, 2004; Oeffinger et al, 2004). Consistent with this model, four of the HEAT-repeat ribosome synthesis factors (Noc2p, Noc3p, Noc4p and Rrp12p) were shown to be required for subunit export, while Noc1p was implicated in intranuclear movement of the pre-ribosomes (Milkereit et al, 2001, 2003; Oeffinger et al, 2004). Moreover, Rrp12p was shown to interact directly with FG-repeat nucleoporins, Ran and RNA (Oeffinger et al, 2004). Yeast strains with defects in ribosome synthesis generally fail to accumulate high levels of the pre-rRNAs, indicating that an active surveillance system exists but little is known about the mechanisms involved; reviewed in Venema and Tollervey (1999). A key player in RNA degradation is the exosome complex of 3′–5′ exonucleases (Mitchell et al, 1997). In the nucleus, the exosome is implicated in elimination of defective pre-mRNAs (Bousquet-Antonelli et al, 2000; Hilleren et al, 2001; Torchet et al, 2002; Zenklusen et al, 2002; Milligan et al, 2005; Wyers et al, 2005), pre-tRNAs (Kadaba et al, 2004; Vanacova et al, 2005) and pre-rRNAs (Zanchin and Goldfarb, 1999; Allmang et al, 2000; Libri et al, 2002; Fang et al, 2004; LaCava et al, 2005). The activity of the nuclear exosome is stimulated by the TRAMP4 and TRAMP5 complexes (Trf/Air/Mtr4 polyadenylation complex) (Haracska et al, 2005; LaCava et al, 2005; Vanacova et al, 2005; Wyers et al, 2005; Houseley and Tollervey, 2006), which include a nuclear poly(A) polymerase, either Trf4p or Trf5p, respectively (Saitoh et al, 2002), the putative RNA helicase Mtr4p (Liang et al, 1996; de la Cruz et al, 1998) and one of two redundant zinc-knuckle proteins Air1p and Air2p (Inoue et al, 2000). Consistent with a widespread role for the TRAMP complexes in RNA surveillance, the polyadenylation of precursors to snoRNAs, snRNAs, tRNA and rRNA has been reported (Allmang et al, 1999; van Hoof et al, 2000; Fang et al, 2004; Kadaba et al, 2004; Kuai et al, 2004; Wyers et al, 2005). We report here the characterization of a surveillance and degradation pathway for ribosome subunits that fail to be exported due to a defect in the HEAT-repeat pre-ribosome component Sda1p. This process appears to occur preferentially within a subnucleolar structure, the No-body, and is mediated by the exosome following TRAMP-dependant polyadenylation of the rRNA substrates. Results Sda1p is a component of late 60S pre-ribosomes Previous proteomic analyses identified Sda1p as a putative component of 60S pre-ribosomes (Baßler et al, 2001; Gavin et al, 2002). To confirm this interaction, we analyzed RNA species that co-precipitated with epitope-tagged Sda1p, expressed under the control of its own promoter (Sda1-TAP; see Supplementary Table S1) (Rigaut et al, 1999). Following immunoprecipitation (see Materials and methods), bound RNAs were analyzed by Northern hybridization and compared with RNAs recovered in parallel from the nontagged control strain (Figure 1B and C). The 7S and 6S pre-rRNAs and the 25S, 5.8S and 5S rRNAs were efficiently co-precipitated with Sda1-TAP. These results are consistent with reports that Sda1p localizes to the nucleolus and nucleoplasm (Buscemi et al, 2000; Babbio et al, 2004), since 7S pre-rRNA is predominantly nucleolar, whereas the final maturation of 25S and 5.8S rRNA occurs in the nucleoplasm. This suggests that Sda1p associates with 60S pre-ribosomes during late nucleolar maturation steps and accompanies them into the nucleoplasm. Phosphorimager quantification indicated that approximately 5% of the total 25S rRNA population was bound with Sda1-TAP in the experiment shown in Figure 1B. This value would be consistent with previous reports that pre-60S particles undergo relatively slow conversion into mature 60S ribosomes after 25S rRNA synthesis (Tollervey et al, 1993). Together, these observations strongly indicate a role for Sda1p in late nuclear steps of 60S ribosome biogenesis. sda1–2 temperature-sensitive (ts) mutants exhibit a very fast ribosome export defect phenotype To assess the requirement for Sda1p in subunit export, we used the sda1–2 ts mutation (Zimmerman and Kellogg, 2001). The mutant and isogenic wild-type strains were transformed with CEN plasmids expressing GFP-tagged ribosomal proteins Rps2-GFP (40S subunit component), Rpl11b-GFP or Rpl25-GFP (60S subunit components) (Stage-Zimmermann et al, 2000; Grandi et al, 2002). Under permissive conditions (23°C), Rps2-GFP (Figure 2A), Rpl11b-GFP (Figure 2B) and Rpl25-GFP (data not shown) were predominantly found throughout the cell in both wild-type and sda1–2 strains. The sda1–2 cells growing at 23°C showed a mild growth defect (t2 2.9 h for sda1–2 versus 1.7 h for wt) and around 5% showed detectable nuclear enrichment of Rps2-GFP, Rpl11b-GFP and Rpl25-GFP (data not shown). Cultures were shifted to 37°C by transfer to prewarmed flasks in a water bath. Following transfer, a very rapid inhibition of subunit export was observed. After 5 min at 37°C, over 80% of sda1–2 cells showed a strong nucleoplasmic accumulation of Rps2-GFP, Rpl11b-GFP and Rpl25-GFP (Figure 2A and B and data not shown). Sda1–2p is stable for at least 5 h at the nonpermissive temperature (Zimmerman and Kellogg, 2001), indicating that the protein is present but impaired in function. We conclude that functional Sda1p is required for efficient export of both pre-40S and pre-60S ribosomal subunits from the nucleoplasm to the cytoplasm. Figure 2.The sda1–2 mutant shows a rapid block in 40S and 60S subunit export. Analysis of 40S and 60S subunit export in sda1–2 strains expressing Rps2-GFP (A) or Rpl11b-GFP (B) reporter construct. Cells were pregrown at 23°C and then shifted to 37°C for 5 min. Cells were fixed, DAPI stained and viewed by fluorescence microscopy. Download figure Download PowerPoint Sda1p is required for 60S ribosomal subunit maturation and accumulation The fast and complete inhibition of export in sda1–2 strains allows us to relate the behavior of individual cells visualized in microscopy to RNA metabolism in the entire population, which was assessed by Northern hybridization (Figure 3A and B) and metabolic labeling (Figure 4). Figure 3.Northern analyses of pre-rRNA processing in sda1–2 mutant strains. Levels of pre-rRNAs and rRNAs were assessed by Northern blotting. Cells were pregrown at 23°C and transferred to 37°C. Samples were collected at 23°C (lanes 1, 6, 11 and 16) and after transfer to 37°C for 15 min (lanes 2, 7, 12 and 17), 30 min (lanes 3, 8, 13 and 18), 45 min (lanes 4, 9, 14 and 19) and 60 min (lanes 5, 10, 15 and 20). RNA was extracted and separated on denaturing 1.2% agarose gels (A) or on 8% polyacrylamide/urea gels (B) and transferred to a nylon membrane. Oligonucleotides used for Northern hybridization are indicated to the right of each panel (see Figure 1A for locations of oligos). (C) Quantification of data in panel A, standardized to the level of scR1 RNA. Download figure Download PowerPoint Figure 4.Pulse-chase analyses of pre-rRNA processing in sda1–2 mutant strains. Cultures were pregrown at 23°C and transferred to 37°C for 5 min. Cells were then pulse labeled with [3H]adenine for 2 min, followed by the addition of an excess of cold adenine. Samples were collected at the times indicated, and RNAs were extracted, separated on a denaturing 1.2% agarose gel (A) or a 8% polyacrylamide/urea gel (B) and transferred to a nylon membrane. Labeled RNAs were detected by fluorography. The exposure shown for the wild-type samples is four-fold shorter than the other panels. Download figure Download PowerPoint RNA was harvested from cells growing at 23°C, and at time points after transfer to 37°C. Northern hybridization (Figure 3A and B, lanes 1–10) revealed strong pre-rRNA accumulation defects in the sda1–2 strain 15–30 min after transfer to the restrictive temperature. Defects were most evident in the pathway leading to 25S rRNA synthesis, with substantial reductions in the levels of 35S and 27S pre-rRNAs as well as 25S mature rRNA (Figure 3A). Since an equal amount of total RNA was loaded in each lane, reduced levels of the very abundant rRNAs leads to overloading of the lanes when equal amounts of total RNA are loaded. Pre-rRNAs and rRNAs were therefore quantified relative to the scR1 RNA, the RNA component of the cytoplasmic signal recognition particle (SRP). The 35S and 27S pre-rRNA levels were reduced by 60% after 30 min at 37°C and by 80% after 1 h, while mature 25S rRNA was reduced by 25 and 45% (Figure 3C). Analysis of the same RNA preparations showed accumulation of the 7S pre-rRNA (five-fold after 30 min), accompanied by 1.5-fold underaccumulation of the mature 5.8S rRNAs (Figure 3B). The reduction in 25S levels was clearly more rapid than that of 5.8S rRNA. A similar observation was made for strains depleted of another HEAT-repeat protein Rrp12p (Oeffinger et al, 2004), suggesting that 25S rRNA is more rapidly degraded than 5.8S rRNA when export is inhibited. Relative to the scR1 RNA loading control, the 20S pre-rRNA accumulated 2.5-fold in sda1–2 ts mutants after 1 h at 37°C, whereas accumulation of the 18S rRNA was reduced by 25% (Figure 3A and C). The conversion of 20S pre-rRNA into mature 18S rRNA occurs in the cytoplasm, suggesting that this phenotype is a consequence of the 40S subunit export defect observed sda1–2 strains (Figure 2A). The signals for all snoRNAs tested were also increased in the sda1–2 strain (Figure 3B), presumably reflecting reduced levels of the rRNAs. Pre-rRNA processing in sda1–2 strains was also analyzed by pulse-chase labeling with [3H]adenine, commencing 5 min after temperature shift (Figure 4, lanes 1–12). Probably as a consequence of a rapid shift to 37°C, global pre-rRNA processing in both wild-type and sda1–2 strains was slower than is usually observed. Total incorporation of label was lower in each of the sda1–2 strains than the wild type (in Figure 4 the wild-type lanes are exposed four-fold less than the other panels), presumably due to the reduced growth rate prior to the temperature shift. Consistent with the Northern data, production of mature 18S rRNA was delayed in sda1–2 strain at 37°C relative to the wild-type strain (compare lanes 5 and 11), whereas 25S rRNA synthesis was strongly impaired. Notably, despite the substantial reduction in 25S accumulation, the appearance and loss of 27S pre-rRNA was relatively similar in both strains. This observation indicates that 27S pre-rRNA was synthesized in sda1–2 strains at 37°C, but the 27S pre-rRNA and/or the mature 25S were then degraded. The 7S pre-rRNA accumulation seen by Northern blot (Figure 3, lanes 6–10) and pulse-chase labeling (Figure 4B) shows that at least some of the 27SB pre-rRNA population continues to undergo endonuclease cleavage at site C2. We conclude that newly synthesized 25S rRNA and probably part of 27SB pre-rRNA population were degraded in the sda1–2 mutant at 37°C. Maturation of 20S pre-rRNA to 18S rRNA was slowed and 18S rRNA accumulation was reduced, consistent with slowed export. The loss of TRAMP or exosome activities stabilizes 27S pre-rRNA and 25S rRNA We speculated that the TRAMP and exosome complexes degrade the 27SB pre-rRNA and/or 25S rRNA in sda1–2 strains. We therefore created sda1–2 strains that also carry rrp6Δ, which eliminates a nuclear-specific exosome component, or trf4Δ, which eliminates the poly(A) polymerase component of the TRAMP4 complex. Northern analyses (Figure 3) showed that the absence of either Rrp6p or Trf4p partially rescued the (pre-)rRNA accumulation defect observed in sda1–2 single mutants. Accumulation of the 25S rRNA in sda1–2, trf4Δ and sda1–2, rrp6Δ strains was close to wild-type levels after the temperature shift (Figure 3). The 27S pre-rRNA pattern was more difficult to interpret, particularly in the sda1–2, rrp6Δ strain, which showed a precipitous decline in pre-rRNA levels 15 min after transfer to 37°C. The basis of this is not clear, but we have also seen dramatic, transient reductions in mRNA levels immediately following transfer of exosome mutants to 37°C (Milligan et al, 2005). However, 1 h after transfer to 37°C, 27S pre-rRNA levels were reduced by only 10% in sda1–2, rrp6Δ and by 31% in sda1–2, trf4Δ, compared to a 80% reduction in the sda1–2 single mutant. The reduced 35S RNA levels seen in the sda1–2 strain was also largely suppressed in the strains also lacking Rrp6p or Trf4p. Sda1p is not predicted to be associated with 35S pre-rRNA, and we speculate that its reduction may be an indirect effect of activation of a pre-rRNA surveillance pathway, which is suppressed in the double mutants. Pulse-chase analyses of the sda1–2, trf4Δ and sda1–2, rrp6Δ strains revealed stabilization of the 27S pre-rRNA and 25S rRNA relative to the sda1–2 single mutant (Figure 4). In the sda1–2 single mutant, the 27S pre-rRNA signal dropped 15 min after the chase, whereas in the absence of Trf4p or Rrp6p it appeared much more stable. Accumulation of the mature 25S rRNA was also higher in the double mutants compared to sda1–2 single mutant. However, the trf4Δ and rrp6Δ mutations did not allow growth of sda1–2 strains at 37°C (data not shown), possibly because the 60S ribosomes remain trapped in the nucleus (see below). Consistent with the Northern data, the 7S pre-rRNA was much more stable than the 27S pre-rRNAs in the sda1–2 strains, and was not clearly further stabilized by the rrp6Δ or trf4Δ mutations. Also consistent with the Northern data, accumulation of the 18S rRNA was similar in sda1–2 strains with or without Rrp6p or Trf4p, indicating that it does not undergo substantial degradation by TRAMP/exosome activity. These results indicate that the degradation of the 27S pre-rRNA and 25S rRNA observed in sda1–2 strains requires the TRAMP complex and the nuclear exosome. The 27S pre-rRNA and 25S rRNA undergo TRAMP-dependent polyadenylation in sda1–2 strains The apparent involvement of the TRAMP complex in pre-ribosome degradation suggested the RNAs might undergo polyadenylation by the poly(A) polymerase activity of Trf4p prior to degradation by the exosome. To test this hypothesis, we analyzed the accumulation of polyadenylated RNA in sda1–2 mutant strains. RNA was extracted from strains growing at 23°C and at time points following transfer to 37°C, fractionated by oligo(dT) selection and analyzed by Northern hybridization (Figure 5). In the wild type (Figure 5, lanes 1–5), a low level of polyadenylated pre-rRNA was detected, as previously reported (Fang et al, 2004; Kuai et al, 2004). This did not comigrate with either the 27S pre-rRNA or 25S rRNA, and presumably reflects a low level of pre-rRNA degradation that normally occurs in wild-type cells. In the sda1–2 strain (Figure 5, lanes 6–10), polyadenylated forms of 27S and 25S RNA increased following transfer to 37°C, and were estimated to represented around 3% of the total pools. In contrast, the 7S pre-rRNA was not clearly polyadenylated in the sda1–2 strain (data not shown), consistent with its relative stability following pulse-labeling (Figure 4). The 20S pre-rRNA and 18S rRNA showed no increase in polyadenylated forms that were long enough to be retained on the column (∼A15) following temperature shift. This is consistent with the conclusion that the large ribosomal subunit precursors are preferentially degraded in sda1–2 strains. Figure 5.Polyadenylated rRNAs are accumulated in sda1–2 and rrp6Δ mutants but not in strains lacking Trf4p. The same RNA samples as used for Figure 4 were submitted to oligo(dT) selection. The resulting bound (poly(A)+ enriched) and unbound (poly(A)+ depleted) fractions were loaded on a denaturing 1.2% agarose gel and transferred to a nylon membrane. (A) Comparison of SDA1 and sda1–2 single mutants. (B) Comparison of SDA1 and sda1–2 in a trf4Δ background. (C) Comparison of SDA1 and sda1–2 in a rrp6Δ background. Oligonucleotides used for Northern hybridizations are indicated to the right of each panel. The PGK1 mRNA was used as a control for loading and polyadenylated RNA selection. Download figure Download PowerPoint In the strains lacking Trf4p, very little polyadenylated pre-rRNA or rRNA was detected (Figure 5, lanes 21–30). Overexposure of the Northern blots showed that a small amount of pre-rRNA and rRNA was still retained on the oligo(dT) column in the trf4Δ strains but the levels appeared to be independent of sda1–2 and temperature variation (data not shown). We concluded that polyadenylation of pre-rRNAs and rRNAs in sda1–2 mutant is largely the consequence of Trf4p activity. In the strain lacking only Rrp6p (Figure 5, lanes 41–45), polyadenylated forms were detected for all tested pre-rRNAs and rRNAs. Some hyperadenylation of the 27S pre-rRNA was apparent in the sda1–2, rrp6Δ strain (Figure 5, lanes 46–50) relative to the sda1–2 single mutant (Figure 5, lanes 6–10). From these results we concluded that exosome-mediated degradation of 27S pre-rRNA and 25S rRNA in sda1–2 mutants involves their prior polyadenylation by the TRAMP4 complex. Nuclear-restricted pre-ribosomes, TRAMP and the exosome concentrate in a nucleolar focus As shown above (Figure 2), nuclear accumulation of pre-ribosomes was seen 5 min after transfer to the nonpermissive temperature. However, the Northern analyses indicate that most degradation of the pre-ribosomes occurs after 15–30 min at 37°C. With the aim of determining the site of pre-ribosome degradation, the Rps2-GFP, Rpl11b-GFP and Rpl25-GFP reporters were analyzed in SDA1 and sda1–2 strains during 1 h following temperature shift. In wild-type strains, all three r-proteins showed a uniform distribution throughout the nucleus and the cytoplasm (Figures 2 and 6A, and Supplementary Figures S2D and S3A). In sda1–2 mutant strains, Rps2-GFP, Rpl11b-GFP and Rpl25-GFP each showed a striking pattern of localization. After 5 min at 37°C, the r-proteins accumulated in the nucleoplasm, with no clear nucleolar enrichment (Figures 2 and 6A and data not shown), consistent with a block in transport across the NPC. However, by 15 min after temperature shift, the r-proteins had become strongly enriched in a sharp focus (Figure 6A, Supplementary Figures S2D and S3A). Colocalization with the nucleolar marker Nop1p (Figure 6B) showed this focus to lie within the nucleolus. The focus persisted in the 30 min time point and was present in more than 80% of cells. The same localization was observed in living cells (data not shown). By 60 min after transfer to 37°C, the cell morphology had changed, with the formation of a very large cytoplasmic vacuole, while the nuclear GFP signal appeared weaker and more diffuse. This dynamic localization pattern was similar for both the 40S (Rps2p) and 60S (Rpl11bp and Rpl25p) ribosomal proteins (Figure 6A, Supplementary Figures S2D and S3A and data not shown) and was very reproducible. Figure 6.Rpl11p-GFP concentrates in a subnucleolar structure in sda1–2 strains 15 min after temperature shift. (A) Localization of Rpl11b-GFP in SDA1 wild-type and sda1–2 ts mutants. Aliquots were collected at different times following temperature shift and cells were fixed before they were viewed in the fluorescence microscope. (B) The Rpl11-GFP focus was visualized within the nucleolus using optical microscopy. The nucleolus was visualized using an anti-Nop1p antibody, which was subsequently recognized by an Alexafluor 555-conjugated secondary antibody. Download figure Download PowerPoint The cytoplasmic Rpl11b-GFP signal appeared unaltered in the sda1–2 strain at early time points after transfer to 37°C, but to fully confirm this, we mixed an equal quantity of SDA1 and sda1–2 cells (Supplementary Figure S2C: note that the site of pre-ribosome accumulation is out of the focal plane in some of the sda1–2 cells in the image). Western blotting (Supplementary Figure S2A) showed no substantial alterations in the level of the Rpl11b protein in the sda1–2 strain following transfer to 37°C. We next wanted to confirm that the GFP-tagged r-protein reporters were faithfully reporting the site of pre-ribosome accumulation. Sucrose density gradient fractionation showed that nearly all Rps2-GFP and Rpl11b-GFP are associated with pre-ribosomes and ribosomes in sda1.2 strain following transfer to 37°C (Figure 7A and Supplementary Figure S3B), confirming that the focus does not reflect an accumulation of free r-proteins. Figure 7.Pre-ribosome components and polyadenylated RNAs concentrate in the subnucleolar substructure. (A) Glycerol density gradient analysis of Rpl11p-GFP. A total cellular extract was prepared from wild-type and sda1.2 strains expressing Rpl11p-GFP, collected 15 min following temperature shift. Samples were loaded on a 10–30% glycerol gradient and 18 fractions were collected. Proteins and RNAs were extracted from the fractions and subjected to Western blot and Northern blot analysis, respectively. T: sample from the total unfractionated extract. (B) FISH localization of pre-rRNAs using a Cy3-labeled ITS2 probe (Leger-Silvestre et al, 2004). (C) FISH localization of polyadenylated RNAs using a Cy5-labeled oligo(dT). Wild-type and sda1.2 cells expressing Rpl11p-GFP were collected 15 min after temperature shift. Polyadenylated RNAs were imaged using a Deltavision system. These images were subjected to decovolution, and a single optical section is presented in each panel. Download figure Download PowerPoint Localization of the pre-rRNA was directly visualized by fluorescent in situ hybridization (FISH) as previously described (Bertrand et al, 1998) using a Cy3-labeled probe specific for the pre-rRNA ITS2 region (Leger-Silvestre et al, 2004). This region is present in the early 35S to the 32S pre-rRNAs and is retained in the 27S and 7S pre-rRNAs (see Supplementary Figure S1). In the wild-type cells, the signal was concentrated in the nucleolus and is also detected in the nucleoplasm (Figure 7B). In sda1–2 strains, clear accumulation of the ITS2 signal was seen in the nucleolar focus, together with weaker nucleolar and nucleoplasmic signals. Northern data indicated that the 27S pre-rRNA and 25S rRNA undergo polyadenylation in the sda1–2 strain (Figure 5). We therefore assessed whether the foci were enriched for polyadenylated RNA by FISH analysis, using a Cy5-labeled oligo(dT) probe (Figure 7C). As expected, a strong poly(A) signal was detected throughout the cytoplasm, but a focus was also seen to colocalize with the Rpl11b-signal (images in Figure 7C each show a single optical section through the cells). No clear nuclear poly(A) signal was observed in the wild-type (Figure 7C) or in the sda1–2 strain at 23°C (data not shown). This observation is consistent with the accumulation of polyadenylated pre-RNA and/or rRNA in the nucleolar focus in sda1.2 cells. To assess the colocalization of the degradation machinery with the GFP-tagged r-protein reporters, C-terminal RFP-fusions were constructed for exosome (Rrp45p), TRAMP (Trf4p, Air2p and Mtr4p) or proteasome (Rpn10p and Pre5p) components. These were expressed from the endogenous promoters at the endogenous loci. Strains were harvested and analyzed during growth at 23°C and 15 min after transfer to 37°C (Figure 8). The core exosome component Rrp45-RFP and all three TRAMP components Trf4-RFP, Mtr4-RFP and Air2-RFP each showed a broad distribution throughout the nucleus but were enriched in the pre-ribosome focus in sda1–2 cells
Referência(s)