Utp25p, a nucleolar Saccharomyces cerevisiae protein, interacts with U3 snoRNP subunits and affects processing of the 35S pre-rRNA Mauricio B. Goldfeder and Carla C. Oliveira Department of Biochemistry, University of Sa˜o Paulo, SP, Brazil Introduction Ribosome biogenesis is a complex and energy-consum- ing process in eukaryotic cells that demands tight regulation between rRNA transcription and process- ing, r-protein translation and rRNA ⁄ r-protein assem- bly. Three of the Saccharomyces cerevisiae rRNAs are transcribed by RNA polymerase I as a polycistronic 35S precursor that undergoes endo- and exonucleolytic cleavage reactions and nucleotide modifications, before originating the mature rRNAs 18S, 5.8S and 25S which will be assembled into the small and large ribo- somal subunits, respectively. At least 200 factors are predicted to be involved in pre-rRNA processing in yeast, and a large number of them are small nucleolar ribonucleoproteins (snoRNPs) [1,2]. Most snoRNPs are classified as members of two major families, box C ⁄ D (that guide 2¢-O-ribose-methylation at specific Keywords nucleolus; pre-40S; ribosome synthesis; rRNA processing; Saccharomyces cerevisiae Correspondence C. C. Oliveira, Department of Biochemistry, Chemistry Institute, University of Sa˜o Paulo, Av. Prof. Lineu Prestes, 748, Sa˜o Paulo, SP, Brazil CEP 05508-900 Fax: +55 11 3815 5579 Tel: +55 11 3091 3810; Ext. 208 E-mail: ccoliv@iq.usp.br (Received 19 November 2009, revised 31 March 2010, accepted 28 April 2010) doi:10.1111/j.1742-4658.2010.07701.x In eukaryotes, pre-rRNA processing depends on a large number of nonribo- somal trans-acting factors that form intriguingly organized complexes. Two intermediate complexes, pre-40S and pre-60S, are formed at the early stages of 35S pre-rRNA processing and give rise to the mature ribosome subunits. Each of these complexes contains specific pre-rRNAs, some ribosomal proteins and processing factors. The novel yeast protein Utp25p has previously been identified in the nucleolus, an indication that this protein could be involved in ribosome biogenesis. Here we show that Utp25p interacts with the SSU processome proteins Sas10p and Mpp10p, and affects 18S rRNA maturation. Depletion of Utp25p leads to accumulation of the pre-rRNA 35S and the aberrant rRNA 23S, and to a severe reduction in 40S ribosomal subunit levels. Our results indicate that Utp25p is a novel SSU processome subunit involved in pre-40S maturation. Structured digital abstract l MINT-7889901: SAS10 (uniprotkb:Q12136) physically interacts (MI:0915) with Utp25p (uni- protkb: P40498)bypull down (MI:0096) l MINT-7889915: NIP7 (uniprotkb:Q08962) physically interacts (MI:0915) with RRP43 (uni- protkb: P25359)bytwo hybrid (MI:0018) l MINT-7889852: Utp25p (uniprotkb:P40498) physically interacts (MI:0915) with MPP10 (uniprotkb: P47083)bytwo hybrid (MI:0018) l MINT-7890065: NOP1 (uniprotkb:P15646) and Utp25p (uniprotkb:P40498) colocalize ( MI:0403)byfluorescence microscopy (MI:0416) l MINT-7889865: Utp25p (uniprotkb:P40498) physically interacts (MI:0915) with SAS10 (uni- protkb: Q12136)bytwo hybrid (MI:0018) Abbreviations GFP, green fluorescent protein; GST, glutathione S-transferase; snoRNP, small nucleolar ribonucleoprotein; SSU, small subunit; UTP, U three-protein complex; YP, yeast extract–peptone medium. 2838 FEBS Journal 277 (2010) 2838–2852 ª 2010 The Authors Journal compilation ª 2010 FEBS positions in nascent rRNAs) and box H ⁄ ACA (that guide pseudouridylation of specific nucleotides in rRNAs). Some snoRNPs, however, are involved in endonucleolytic cleavage reactions of the pre-rRNA, among them the endonuclease MRP (responsible for the cleavage at site A 3 in ITS1) [3], the box C ⁄ D snoRNPs U3 and U14, and the box H ⁄ ACA snoRNPs snR10 and snR30, involved in the cleavage reactions at sites A 0 ,A 1 and A 2 [4–7]. All box C ⁄ D snoRNAs are bound by four core pro- teins, Nop1p, Nop58p, Nop56p and Snu13p [8]. In addition to the core proteins, the U3 snoRNP is asso- ciated with other proteins specific for this snoRNP. The first proteins to be identified in the U3 snoRNP complex were Sof1p, Mpp10p, Lcp5p, Imp3p, Imp4p, Dhr1p and Rrp9p [9–14]. Later experiments have shown that U3 is associated with at least 28 proteins, forming a large multisubunit complex also known as the small subunit (SSU) processome [15]. The mecha- nism of U3 snoRNP complex assembly in the 90S par- ticle is unknown. However, recent evidence suggests that stable subcomplexes bind the nascent 35S pre-rRNA sequentially [16]. Interestingly, electron microscopy analyses have shown that early preriboso- mal particles undergo time-dependent changes in size and shape upon binding to the primary pre-rRNA pre- cursor, suggesting that their components are sequen- tially assembled [17]. Accordingly, recent studies have revealed the presence of discrete 90S particle subcom- plexes that have been named U three-protein com- plexes (UTP) UTP-A ⁄ t-UTP, UTP-B and UTP-C [18,19]. t-UTP complex binds very early during tran- scription of the pre-rRNA, followed by the UTP-B complex, U3 snoRNP and the Mpp10p complex, and later by Rrp5p and the UTP-C complex [16]. It is pre- dicted, however, that the SSU processome interacts with other proteins in order for the cleavage of the pre-rRNA to occur. The S. cerevisiae protein coded by the open reading frame YIL091C had not been previously characterized. However, analysis of essential yeast proteins had iden- tified in the YIL091C sequence a domain with low homology to RNA helicases. These studies have also shown that this protein is localized to the nucleolus [20]. Here we show that the protein named Utp25p is involved in pre-rRNA processing. Its depletion leads to accumulation of the pre-rRNA 35S and the aber- rant 23S, and subsequent decrease in the levels of pre- rRNA 20S and mature 18S rRNA. Consistent with its subcellular localization and involvement in 18S rRNA formation, Utp25p interacts with the SSU processome proteins Sas10p and Mpp10p. Utp25p also co-immu- noprecipitates U3 snoRNA, which strongly indicates that it is a novel SSU processome subunit. Results Previous global analyses of yeast protein localization have shown that the 83 kDa protein Utp25p, coded by the open reading frame YIL091C, localizes to the nucleolus [20]. In order to confirm the subcellular local- ization of Utp25p, the UTP25 gene was cloned into a plasmid, fused to green fluorescent protein (GFP), and cells were analyzed by fluorescence microscopy. The GFP–Utp25p signal is restricted to the nucleus and is concentrated in the nucleolus (Fig. 1). GFP, by con- trast, is present throughout the cell. RFP–Nop1p, used as a control, is restricted to the nucleolus (Fig. 1). The nucleolar localization of Utp25p suggested that this protein is involved in ribosome synthesis. GFP Hoechst GFP + RFPRFP-Nop1 GFP GFP- Utp25 GFP + RFP + Hoechst Fig. 1. Subcellular localization of GFP–Utp25p. Yeast strains expressing RFP–Nop1p and either GFP (upper) or a GFP–Utp25p N-terminal fusion (lower) were analyzed. Hoechst, indicates nuclei stained with the DNA dye Hoechst; GFP, indicates the localization of the green fluo- rescent protein; RFP, indicates the localization of the red fluorescent protein. GFP + RFP, merging of green and red signals. GFP + RFP + Hoescht, merging of all signals. M. B. Goldfeder and C. C. Oliveira Utp25p affects pre-rRNA processing FEBS Journal 277 (2010) 2838–2852 ª 2010 The Authors Journal compilation ª 2010 FEBS 2839 In order to characterize Utp25p function, we first obtained a conditional mutant strain. A heterozygous diploid strain (YIL091C ⁄ yil091c::KanMX4 – Euro- scarf) was transformed with a plasmid containing a copy of the UTP25 gene under control of the inducible promoter GAL1. After sporulation, a haploid deletion strain was obtained and its genotype was confirmed by PCR analysis of UTP 25 gene (data not shown). The conditional Dutp25 ⁄ GAL1::UTP25 strain was then analyzed for growth in glucose medium, compared with the otherwise isogenic parental strain, UTP25. Dutp25 ⁄ GAL1::UTP25 cells are not able to grow on glucose plates, showing that Utp25p is essential for growth (Fig. 2A). Dutp25 ⁄ GAL1::UTP25 cells were transformed with a second plasmid that harbors an extra copy of UTP25 under the control of a constitu- tive promoter, which rescues growth of the conditional mutant on glucose plates (Fig. 2A). As shown here, the fusion proteins GFP–Utp25p and Gal4AD–Utp25p (transcription activation domain of Gal4p) are functional. Growth of the conditional strain was also analyzed in liquid glucose medium and the results show that after 5 h in glucose, growth of Dutp25 ⁄ GAL1::UTP25 slows in comparison with the parental wild-type strain, but the difference in growth rate is more evident after 14 h in this medium (Fig. 2B). Based on the Utp25p nucleolar localization and its possible involvement in ribosome synthesis, we ana- lyzed the polysome profile of the conditional strain after depletion of Utp25p. When growing in galactose medium, Dutp25 ⁄ GAL1::UTP25 cells show a normal polysome profile on density gradients. However, after 20 h of growth in glucose, it is possible to see a severe reduction in the relative amounts of the 40S ribosomal subunit, as well as in 80S ribosomes and polysomes (Fig. 3A, lower). Accordingly, free 60S accumulate in the cells, resulting in a large peak that overlaps with 80S ribosomes (Fig. 3A, lower). Analysis of free ribo- some subunits in the presence of EDTA confirms a strong decrease in the relative amounts of 40S ribo- A B UTP25 AD-Utp25 GFP-Utp25 – Glucose GAL1::UTP2 5 UTP25 h, Glu0 5 10 15 20 Log(OD/OD 0 ) 5 4 3 2 1 0 Δ utp25/ GAL1::UTP25 Fig. 2. UTP25 is an essential S. cerevisiae gene. (A) Tenfold serial dilution of UTP25 and Dutp25 ⁄ GAL1::UTP25 strains growing on glu- cose-containing plates. Dutp25 ⁄ GAL1::UTP25 was transformed with plasmids containing an extra copy of the UTP25 gene under the control of a constitutive promoter, fused to Gal4AD or GFP. –, empty plasmid. (B) Growth curve of UTP25 and Dutp25 ⁄ GAL1::UTP25 strains in glucose medium. A B A 254 nm Polysomes 40S 60S 80S Polysomes 40S 60S 80S A 254 nm Galactose Polysomes 40S 60S 80S Glucose Polysomes 40S 60S 80S A 254 nm GAL1::UTP25 UTP25 GAL1::UTP25 Galactose 40S 60S 22.716.3 Glucose 40S 60S 22.3 3.4 Fig. 3. Analysis of the polysomal profile in strain Dutp25 ⁄ GAL1::UTP25. UTP25 and Dutp25 ⁄ GAL1::UTP25 strains were incu- bated either in galactose or in glucose medium for 20 h for the analysis of polysomal profile through sucrose gradient. (A) Upper, UTP25 strain. Lower, Dutp25 ⁄ GAL1::UTP25 strain shows very low levels of 40S ribosomal subunit, an accumulation of free 60S sub- unit, and consequent low number of polysomes. (B) Analysis of ribosomal subunits through sucrose gradient in the presence of EDTA. Levels of 40S subunit are strongly decreased upon depletion of Utp25p. Numbers indicate area quantitation of subunits peaks. Utp25p affects pre-rRNA processing M. B. Goldfeder and C. C. Oliveira 2840 FEBS Journal 277 (2010) 2838–2852 ª 2010 The Authors Journal compilation ª 2010 FEBS somal subunits upon depletion of Utp25p (Fig. 3B). Indeed, estimation of the areas under free subunit peaks showed a change in the 60S :40S ratio from 1.4, under permissive conditions, to 6.5, after depletion of Utp25p (Figs 3B and S1). To investigate the possible association of Utp25p with ribosomal particles, the sedimentation profile of endogenous Utp25p on density gradients was analyzed. Total extracts prepared from wild-type strain UTP25 grown in glucose medium was loaded onto 5–47% sucrose gradients. Proteins isolated from the gradient fractions were analyzed by western blot using a poly- clonal antiserum raised against recombinant Utp25p. Total RNA isolated from the same fractions was ana- lyzed by northern blot to detect U3 snoRNA and the mature rRNAs 25S and 18S. The results show that endogenous Utp25p is concentrated in the fractions containing soluble proteins, co-fractionating with free snoRNA U3, but it is also present in higher molecular mass fractions (Fig. 4A). As a control, antiserum spe- cific for large ribosomal subunit protein Rpl5p was used, showing that it is concentrated in the fractions containing the 60S ribosomal subunits. Mature rRNAs 25S and 18S were used as controls for large and small subunit-containing fractions (Fig. 4A). U3 snoRNA shows a normal sedimentation profile in these sucrose gradients, being present in the soluble fractions but concentrated in fractions containing the 90S SSU processome (Fig. 4A and Fig. 4B, upper). The strong effect of Utp25p depletion on the 40S subunits levels, and its co-fractionation with free U3 snoRNA suggests that Utp25p is involved in 40S ribo- somal subunit maturation. To analyze whether Utp25p depletion might affect U3 snoRNP association with preribosomes, northern blot hybridization was performed to detect U3 snoRNA from sucrose gradient fractions. The results show that, after 20 h of growth in glucose, depletion of Utp25p leads to a distribution of U3 snoRNA in two different sets of fractions, those corresponding to soluble material and in larger complexes (Fig. 4B, lower). Interestingly, the 35S pre-rRNA distribution in these gradients is also shifted to larger complexes in the absence of Utp25p (Fig. 4B). To assess the possible involvement of Utp25p on pre-rRNA processing, the effect of its depletion on this pathway was analyzed by northern hybridization. The results show that upon depletion of Utp25p there is an accumulation of the pre-rRNA 35S and the aberrant 23S, and a decrease in pre-rRNA 20S and mature 18S rRNA (Fig. 5A). The large ribosomal subunit RNAs 25S, 5.8S and 5S are mostly unaffected by the deple- tion of Utp25p (Figs 5A and S2). The results shown here indicate the involvement of Utp25p in the early nucleolar reactions of pre-40S maturation. Pulse-chase RNA labeling experiments with [ 3 H]uracil were also performed with cells grown in glucose for 20 h. The results confirm the northern blot data and show that 25 S 40S 60S80S Polysomes 25 S 18 S 18 S U3 U3 GAL1::UTP25 UTP25 AB 35 S 35 S 40S 60S 80S Polysomes 25S 18S U3 Utp25p * * Rpl5p Fig. 4. Analysis of Utp25p and U3 snoRNA sedimentation profile on polysomal gradients. (A) Sedimentation of endogenous Utp25p was detected by western blot of fractions from the wild-type strain (UTP25) polysomal profile. Total RNA was analyzed using northern blotting to detect snoRNA U3. The sedimentation of mature rRNAs 25S and 18S were used as controls. Western blot with antiserum against Rpl5p was performed as a control. (B) The effect of Utp25p depletion on the sedimentation of U3 snoRNA was analyzed by northern hybridization. (Upper) Extract from UTP25 strain. (Lower) D utp25 ⁄ GAL1::UTP25 growing in glucose medium. Fractions corresponding to peaks of ribosome subunits are indicated. M. B. Goldfeder and C. C. Oliveira Utp25p affects pre-rRNA processing FEBS Journal 277 (2010) 2838–2852 ª 2010 The Authors Journal compilation ª 2010 FEBS 2841 the depletion of Utp25p slows the processing of 35S pre-rRNA, strongly inhibiting formation of mature 18S rRNA, although little affecting 25S rRNA forma- tion (Fig. 6). To gain insight into the effect of Utp25p depletion on early pre-rRNA cleavage reactions, primer exten- sion reactions were performed with total RNA extracted from either wild-type cells or Dutp25 ⁄ GAL1::UTP25 grown in glucose for 16 h. Reaction with a primer complementary to the 5¢ region of the mature 18S rRNA shows that the early cleavage reactions are strongly inhibited after the depletion of Utp25p, leading to an increased concentration of bands corresponding to the pre-35S rRNA 5¢-end (Fig. 7A). Although the accumulation of 35S and 23S rRNAs was detected by northern hybridization, little effect on cleavage at A 1 was observed by primer extension (Fig. 7A). This may be because of the high stability of mature 18S rRNAs formed prior to the depletion of Utp25p. Reaction with primer P 3 , which hybridizes down- stream of site D in ITS1, also shows the accumulation of pre-rRNAs, with increased extension bands corre- sponding to regions in the mature 18S rRNA upon depletion of Utp25p (Fig. 7B; asterisk). A primer extension reaction with an oligo complementary to a region downstream of A 2 shows that depletion of Utp25p causes a strong inhibition in the cleavage at this site (Fig. 7), and consequently the accumulation of extended products that correspond to regions within the mature 18S rRNA (Fig. 7C, asterisk). These results further indicate the involvement of Utp25p in the early steps of processing of pre-40S. To determine whether Utp25p might associate in vivo with pre-rRNAs, co-immunoprecipitation experiments were performed using a ProtA–Utp25p fusion. Total extract from cells expressing ProtA–Utp25p was sub- jected to affinity chromatography with IgG–Sepharose beads. Following co-immunoprecipitation, RNA was extracted from the different fractions and analyzed by northern hybridization, compared with RNAs recov- ered in parallel from the strain expressing only ProtA. The results show that ProtA–Utp25p co-precipitates B A 0 12 16 0 12 16 h, Glu P1 35S 23S P6 27S P7 25S P2 18S P3 20S P5 7S P4 5.8S 5S GAL1::UTP25UTP25 A 0 A 1 A 2 A 3 D B 1L A 3 →B 1S B 2 ←B 0 B 2 ←B 0 C 2 E←C 2 C 2 →C 1 C 2 E←C 2 C 2 →C 1 33S 32S 20S 27SA 2 18S 5.8S S 25S 5.8S L 25S 27SA 3 27SB S 27SB L 7S S 7S L A1 B2 D A 2 A3 B1L/B1S E C 2 C1 5´ETS A0 ITS1 3´ETSITS2 P 3 P 4 P 5 P 1 P 2 P 7 P 6 P 8 18S 5.8S 25S 35S or B0 Fig. 5. Northern blot analysis of pre-rRNA processing. (A) Total RNA (20 lg) extracted from cells incubated in glucose medium for different periods and hybridized against specific oligonucleotide probes. The relative positions of the probes on the 35S pre-rRNA are indicated in (B). Bands corresponding to the major intermediates and to the mature rRNAs are indicated on the right-hand side. The lower panel shows hybridization with a probe against the 5S rRNA, used as an internal control. (B) Structure of the 35S pre-rRNA and major intermediates of the rRNA processing pathway in S. cerevisiae. The positions of the probes used for northern hybridizations are indicated below the 35S pre- rRNA. Processing of 35S pre-rRNA starts with endonucleolytic cleavages at sites A 0 and A 1 in the 5¢-ETS, generating 32S pre-rRNA. Subse- quent cleavage at site A 2 , in ITS1, generates the 20S and 27SA 2 pre-rRNAs. The 20S pre-rRNA is then processed at site D to the mature 18S rRNA. The major processing pathway of the 27SA 2 pre-rRNA involves cleavage at site A 3 , producing 27SA 3 , which is digested quickly by exonucleases to generate the 27S B short (27SB s ) pre-rRNA. The subsequent processing step occurs at site B 2 , at the 3¢-end of the mature 25S rRNA. Processing at sites C 1 and C 2 separates the mature 25S rRNA from the 7S S pre-rRNA. This pre-rRNA is subsequently pro- cessed exonucleolytically to generate the mature 5.8S S rRNAs. A fraction of the 27SA 2 pre-rRNA is processed at the 5¢-end by a different mechanism and, following processing at the remaining sites, gives rise to the 5.8S long (5.8S L ) rRNA, which is 6-8 nucleotides longer than the 5.8S S rRNA at the 5¢-end. Utp25p affects pre-rRNA processing M. B. Goldfeder and C. C. Oliveira 2842 FEBS Journal 277 (2010) 2838–2852 ª 2010 The Authors Journal compilation ª 2010 FEBS the 35S pre-rRNA, the aberrant rRNAs 23S and 22S, and much less efficiently, the pre-rRNA 20S (Fig. 8). Mature 18S rRNA was not efficiently co-immunopre- cipitated with ProtA–Utp25p, further indicating that this protein is associated only with the early pre- rRNAs. This is in accordance with the hypothesis of the involvement of Utp25p in the early cleavages of the 35S pre-rRNA, ProtA–Utp25p co-immunoprecipi- tated U3 snoRNA (Fig. 8). Based on the above results, it seemed likely that Utp25p might interact with protein subunits of the SSU processome. To determine whether that interac- tion could occur, the two-hybrid assay was performed using Utp25p fused to the lexA DNA-binding domain and its interaction with Sas10p ⁄ Utp3p, Mpp10p, Imp3p and Imp4p was investigated [10,12,15]. Expres- sion of the reporter genes HIS3 and lacZ indicates a strong interaction of Utp25p with Sas10p ⁄ Utp3p, a weaker interaction with Mpp10p and no interaction with Imp3p or Imp4p (Fig. 9A, upper). The direct interaction between Utp25p and Sas10p was confirmed after expressing recombinant proteins in Escherichi- a coli and performing pull-down assays. The results show that glutathione S-transferase (GST)–Sas10p, immobilized in glutathione–Sepharose beads pulls 25 S 35 S 18 S 0 3 10 30 60 0 3 10 30 60 min 20 S 27 S 23 S UTP25 GAL1::UTP25 Fig. 6. Metabolic labeling of rRNA. Pulse-chase labeling with [ 3 H]uracil was performed after incubating Dutp25 ⁄ GAL1::UTP25 and control strain in glucose medium for 20 h. Total RNA (20 lg) was loaded onto agarose gel after [ 3 H]uracil labeling. The figure shows autoradiograph of RNA transferred to nylon membrane. Bands corresponding to major intermediates and mature rRNAs are indicated on the right-hand side. A 1 A 0 P 2 5’ GATC 0 16 0 16 h, Glu A * P 3 GATC 016016 h, Glu B GA TC 016016 h, Glu C * A 2 P 8 GAL1::UTP25 GAL1::UTP25 GAL1::UTP25 Fig. 7. Early cleavage reactions in 35S pre-rRNA were analyzed through primer extension reactions of total RNA extracted from cells grow- ing in media containing either galactose (0 h) or glucose (16 h). Relative positions of the primers used in the primer extension reactions are shown in Fig. 5B. (A) Primer extension with the primer P 2 allows the detection of the sites A 0 and A 1 . Processing inhibition in Dutp25 ⁄ GAL1::UTP25 strain allows the detection of the 5¢-end of 35S pre-rRNA. (B) Reaction with primer P 3 that hybridizes between sites D and A 2 shows the accumulation of pre-rRNA after depletion of Utp25p. (C) Primer extension reaction with primer P 8 shows that processing at site A 2 is inhibited upon depletion of Utp25p. Asterisks indicate longer extensions of the reactions due to inefficient processing. M. B. Goldfeder and C. C. Oliveira Utp25p affects pre-rRNA processing FEBS Journal 277 (2010) 2838–2852 ª 2010 The Authors Journal compilation ª 2010 FEBS 2843 down His–Utp25p, whereas the negative control GST does not (Fig. 9B). These results strongly suggest that Utp25p is part of the SSU processome, participating in the early stages of pre-rRNA maturation. In order to determine the portion of Sas10p that is responsible for the interaction with Utp25p, two Sas10p truncated mutants were fused to Gal4-AD and the interaction with Utp25p was investigated through the two-hybrid assay. The results show that the N-terminal portion of Sas10p is sufficient for interaction with Utp25p (Fig. 9A, lower). Many of the SSU processome protein subunits are conserved throughout evolution. In order to identify possible Utp25p orthologs in other organisms, a BLAST search was performed. Utp25p homologs are present in many organisms, including humans 0 10 20 30 40 50 60 70 80 90 100 ProtA ProtA ProtA-Utp25 TEFT FTWWBB U3 5S 25S 18S TE 35S 23S 22S/21 S 20S Bound/input ProtA A-Utp25 35S 23S 22S/ 21S 20S 18S 25S 5S U3 23S A 0 A 1 A 2 A 3 D 22S A 0 A 1 A 2 A 3 D 21S A 1 A 2 A 3 D 20S A 1 A 2 D A B C Fig. 8. RNA co-immunoprecipitation with ProtA–Utp25p. Total extracts from cells expressing either ProtA or ProtA–Utp25p were incubated with IgG–Sepharose beads. (A) After immunoprecipita- tion, RNA was extracted from fractions of total extract (TE), flow through (FT), wash (W) and bound material (B), separated by elec- trophoresis and subjected to northern hybridization with probes specific for the RNAs indicated on the right. The structures of the detected pre- and aberrant rRNAs are shown on the left. (B) Pro- teins isolated from the same fractions were subjected to western blot for detection of ProtA and ProtA–Utp25p. (C) Quantitation of the bands obtained from RNA co-ip by phosphorimaging. Ratio of bound ⁄ input is shown for all RNAs tested. BD-Utp25 AD-Sas10 AD-Mpp10 AD L40-61 – His X-Gal AD-Imp3 AD-Imp4 A B FT 1 FT 2 B GST + His-Utp25 GST-Sas10 + His-Utp25 TE 1 FT 1 BFT 2 TE 1 His-Utp25 GST-Sas10 GST AD-Sas10 (1–227) AD-Sas10 AD–Sas10 (226–610) AD BD-Utp25 Fig. 9. Utp25p interacts with SSU processome subunits. (A) Utp25p was fused to lexA DNA-binding domain (BD) and tested for interaction with Mpp10p, Sas10p, Imp3p and Imp4p, which were fused to Gal4p transcription activation domain (AD). Sas10p trun- cated mutants fused to Gal4p-AD are indicated by the amino acid positions relative to the full-length protein [Sas10(1–227) and Sas10(226–610)]. Protein interactions were analyzed using the two- hybrid system, testing for expression of the reporter genes HIS3 (left) and lacZ (right). BD–UTP25 + AD, negative control; strain L40-61, which harbors plasmids encoding BD–Nip7p and AD–Rrp43p, was used as a positive control. (B) Western blot for detection of proteins after pull-down assay. Total extract from cells expressing either GST or GST–Sas10p (TE 1 ) was incubated with a glutathione–Sepharose resin, the flow-through fraction was collected (FT 1 ) and after washing, total extract of cells expressing His–Utp25p (TE 2 , not shown) was loaded. The flow-through fraction was collected again (FT 2 ), resin was washed, and bound fraction obtained (B). His–Utp25p is only pulled-down by GST–Sas10p. His– Utp25p was detected with monoclonal anti-His IgG2a. GST–Sas10p and GST were detected with anti-GST serum. Utp25p affects pre-rRNA processing M. B. Goldfeder and C. C. Oliveira 2844 FEBS Journal 277 (2010) 2838–2852 ª 2010 The Authors Journal compilation ª 2010 FEBS (Fig. S3). Utp25p and hUtp25p (human Utp25p) show 37% sequence similarity and 28% sequence identity. Both proteins contain the domain of unknown func- tion DUF1253, which shows low similarity to DEAD box helicases [20]. To investigate whether hUtp25p and Utp25p might perform similar functions in the cell, the human gene (C1ORF7 ⁄ DEF) was cloned and expressed in strain Dutp25 ⁄ GAL1::UTP25 under the control of a constitutive promoter (MET25::GFP- hUTP25). Expression of the human protein could not rescue Dutp25 ⁄ GAL1::UTP25 growth under the restric- tive condition (Fig. 10A). To obtain higher levels of expression of hUtp25p in yeast cells, the gene was cloned under control of a stronger constitutive pro- moter, PGK1, without the GFP tag, but still could not rescue growth of the conditional strain in glucose med- ium (Figs 10A and S4), indicating that, although these proteins show sequence similarity in the C-terminal portion, divergences in the remaining sequence of the protein would render hUtp25p nonfunctional and ⁄ or unstable in yeast. To analyze the possibility that the C-terminal DUF1253 domain of Utp25p might be sufficient for the protein function, truncation mutants were fused to GFP, cloned in a plasmid under the control of a con- stitutive promoter and transformed into Dutp25 ⁄ GA- L1::UTP25 strain. The results show that the DUF1253 domain does not complement growth of the condi- tional strain (Figs 10A and S4). The GFP-fused dele- tion mutants were also analyzed by western blot and the results show that all were expressed in the cell (Fig. 10B). Sequence analysis also predicted a possible phosphorylation site in Utp25p. Indeed, high-through- put analysis showed that Ser196 was phosphorylated [21]. To investigate whether this modification was important for function, a point mutation was intro- duced in Utp25(S196V) that would prevent phosphory- lation at this specific residue. The more conserved B 25 40 50 60 80 115 kDa α-GFP Coomassie A GFP GFP-Utp25 GFP-Utp25Δ243 GFP-Utp25Δ287 GFP-Utp25Δ411 1 721 aa300 DUF1253 Glu Δ utp25/GAL1::UTP25 GFP UTP25 GFP-Utp25 GFP-Utp25 (S196V) * GFP-hUtp25 GFP-Utp25 (S198A) * hUtp25 Fig. 10. Schematic representation of the different clones of Utp25p, full-length, truncated and the human ortholog, that were tested for complementation of growth of the conditional strain Dutp25 ⁄ GA- L1::UTP25 in glucose. (A) Tenfold serial dilution of UTP25 and Dutp25 ⁄ GAL1::UTP25 strains growing on glucose-containing plates. Dutp25 ⁄ GAL1::UTP25 was trans- formed with a plasmid containing an extra copy of the UTP25 gene, truncated mutants or hUtp25p under control of a constitutive promoter, fused to GFP. hUtp25 indicates PGK1::hUTP25 (without a GFP tag). (B) Analysis of GFP–Utp25p mutants and GFP–hUtp25p by western blot with anti-GFP serum, compared with wild-type GFP–Utp25p. Arrowheads indicate full-length proteins. Right, Coomassie Brilliant Blue-stained poly(vinylidene difluoride) membrane used in the immunoblot assay. M. B. Goldfeder and C. C. Oliveira Utp25p affects pre-rRNA processing FEBS Journal 277 (2010) 2838–2852 ª 2010 The Authors Journal compilation ª 2010 FEBS 2845 Ser198 was also mutated, originating Utp25(S198A). Interestingly, cells expressing Utp25(S196V) showed a growth rate similar to that of the wild-type strain, indi- cating that phosphorylation at Ser196 of Utp25p is not essential for function (Figs 10A and S4). Polysomal profile analysis of Dutp25 ⁄ GAL1::UTP25 ⁄ GFP- utp25(S196V) strain confirms that this mutant is fully functional (Fig. S4). Cells expressing Utp25(S198A), on the other hand, are not able to grow in glucose medium (Figs 10A and S4). Discussion Although various proteins have already been identified as components of the SSU processome [15,16,22], it is possible that many subunits remain to be isolated. Here, we report the characterization of Utp25p as a novel nucleolar protein required for efficient cleavage of 35S pre-rRNA at sites A 0 ,A 1 and A 2 . Depletion of Utp25p causes the accumulation of the pre-rRNA 35S and the aberrant 23S, a consequent decrease in the lev- els of mature 18S rRNA and strong depletion of 40S ribosomal subunit. In accordance with its nucleolar localization and effects on pre-rRNA processing, Utp25p interacts with the protein subunits of the SSU processome Sas10p and Mpp10p and co-immunopre- cipitates U3 snoRNA. High-throughput assays identified Utp25p in com- plexes with Mpp10p and Sas10p [23]. Mpp10p has been characterized as a nucleolar protein that interacts with the U3 snoRNP, depletion of which causes inhibi- tion of cleavages at sites A 0 ,A 1 and A 2 , leading to decreased levels of 18S rRNA [10]. Mpp10p is part of a ternary complex with Imp3p and Imp4p, and these proteins show interdependence for binding to U3 snoRNA [24]. Because Utp25p showed no interaction with Imp3p and Imp4p and was not isolated in the Mpp10p ternary complex, it is possible that its interac- tion with Mpp10p is transient or might occur in the context of the assembled SSU processome. Sas10- p ⁄ Utp3p is also part of the SSU processome, co-immu- noprecipitates U3 snoRNA and interacts with Mpp10p [15]. Depletion of Sas10p also causes a severe reduc- tion in 18S rRNA levels, without affecting 25S rRNA [15]. Interestingly, individual depletions of either U3 snoRNA or the U3 snoRNP protein subunits Nop1p, Nop58p, Mpp10p, Imp3p, Imp4p, Sof1p, Lcp5p, Utp23p, Utp24p and Enp1p all result in accumulation of the pre-rRNA 35S and the aberrant 23S, and decreased levels of the 20S pre-rRNA and the mature 18S rRNA, although to different degrees of severity [10–12,14,25–28]. These results indicate that the SSU processome must be fully assembled for the cleavage reactions at sites A 0 ,A 1 and A 2 to occur. The observa- tion that depletion of Utp25p leads to similar pheno- types and its interaction with U3 snoRNA, Mpp10p and Sas10p strongly indicate that this is a novel com- ponent of the SSU processome. The direct interaction between Utp25p and Sas10p was confirmed through protein pull-down assays, further indicating that Utp25p is a subunit of that complex. As shown here, in addition to interacting with SSU processome subunits, Utp25p co-immunoprecipitates the pre-rRNAs 35S, the aberrant rRNAs 23S and 22S,and much less efficiently the pre-rRNA 20S. Co-immunopre- cipitation of aberrant rRNAs with SSU processome com- ponents has been reported previously [29,30]. Utp25p does not co-immunoprecipitate the mature 18S rRNA, however, which is in agreement with its involvement in the early cleavage of the 35S pre-rRNA. Analysis of endogenous Utp25p sedimentation on polysomal gradients shows that it is concentrated in the fractions corresponding to soluble material, frac- tions that also contain U3 snoRNA, which is consis- tent with Utp25p being part of U3 snoRNP complex. SSU processome subunits from different U3 snoRNP subcomplexes have also been reported to concentrate in the soluble fractions of polysomal gradients [16,31]. Combined, these results indicate that although Utp25p interacts with the SSU processome and is involved in pre-rRNA maturation, its interaction with the complex may be labile or transient. The question of whether Utp25p binds directly to the snoRNA U3 or associates via interaction with the proteins Sas10p and Mpp10p remains to be addressed. The fact that no known RNA-binding motifs can be distinguished in the Utp25p sequence, however, indicates that the latter is more likely. Analysis of the Utp25p sequence also shows that this protein contains the domain DUF1253, which occurs in several hypothetical eukaryotic proteins of unknown function and shows remote homology to DEAD box RNA helicases [20]. Attempts to gain insight into the role of the DUF1253 domain on Utp25p function, made by testing the complemen- tation of growth of the conditional strain Dutp25 ⁄ GAL::UTP25 with deletion mutants expressing only the DUF1253 domain, gave negative results. Interest- ingly, Utp25p shows some amino acid residues that are possible targets for phosphorylation. Indeed, Utp25p Ser196 has been previously shown to be phosphorylated [21]. Our data show that a point mutation in which Ser196 was replaced by a valine had no effect on Utp25p function. Interestingly, how- ever, substitution of Ser198 by alanine resulted in a nonfunctional protein. Utp25p affects pre-rRNA processing M. B. Goldfeder and C. C. Oliveira 2846 FEBS Journal 277 (2010) 2838–2852 ª 2010 The Authors Journal compilation ª 2010 FEBS During the final preparation of this article, a study was published on Utp25p [32]. In that work, a network- guided genetics approach was used to identify proteins involved in ribosome biogenesis, and Utp25p was char- acterized as a nucleolar protein associated with the 40S ribosomal subunit. Analysis of pre-rRNA processing also showed that Utp25p depletion causes an accumula- tion of 35S pre-rRNA. Those results are consistent with those shown here. Our data complement that study by showing the direct interaction of Utp25p with SSU pro- cessome subunits, and the analysis of the 35S pre-rRNA cleavage reactions that are affected by the depletion of Utp25p. Furthermore, we show that although a puta- tive human ortholog of Utp25p was identified, it does not complement the yeast protein function. Materials and methods DNA manipulation and plasmid construction The plasmids used in this study, described in Table 1, were constructed according to the cloning techniques described by Sambrook et al. [33] and sequenced by the Big Dye method (PerkinElmer, Waltham, MA, USA). Cloning strat- egies were as follows. UTP25 gene, encoded by the YIL091C open reading frame, was PCR amplified from S. cerevisiae genomic DNA using primers specific for UTP25: 5¢-CCCGGGTGGATCCATGAGTGACAGTTCT GTGAG-3¢ and 5¢-CTCGAGTTATTTAAATTCATAAAT TTCCTTTTGTGC-3¢. For the two-hybrid assays, the PCR product was digested with SmaI and XhoI and cloned into pBTM116 [34] and pGAD-C2 [35] digested with the same enzymes, generating pBTM–UTP25 and pGAD–UTP25 (which code for the fusions BD–Utp25p and AD–Utp25p respectively, where BD refers to the lexA DNA-binding domain and AD refers to the Gal4p transcription activation domain). MPP10 and SAS10 genes were PCR amplified, the products were digested with the enzymes PvuII and SmaI and cloned into pBTM116 and pGAD-C2 digested with SmaI. To obtain Sas10p truncation mutants, plasmid pGAD–SAS10 was cleaved with EcoRI, resulting in a frag- ment coding for Sas10p amino acid residues 1–226, which was cloned into pGADC2 generating pGAD–SAS10(1–226). The plasmid previously digested with EcoRI was religated, generating pGAD–SAS10(227–610). For the pull-down assays, BamHI–XhoI fragments of UTP25 and MPP10 genes were cloned into pET28a (Merck KGaA, Darmstadt, Ger- many) and pGEX-4T1 (GE Healthcare, Little Chalfont, UK), respectively. YCp111GAL–UTP25, which carries UTP25 under the control of GAL1 promoter, was obtained by inserting an EcoRV–SalI fragment into YCp111-GAL digested with NdeI (following T4 DNA polymerase treat- ment) and SalI. To determine the subcellular localization of yEGFP3–Utp25p by fluorescence microscopy, plasmid pUG34–UTP25 was constructed by inserting a BamHI–XhoI fragment into pUG34 (U. Gueldener & J. H. Hegemann, unpublished) digested with BamHI and SalI. Plasmid pUG36 (U. Gueldener & J. H. Hegemann, unpublished) was used to Table 1. List of plasmid vectors used. Plasmid Relevant characteristics Reference pBTM116 lexA DNA binding domain, TRP1,2lm34 pBTM–UTP25 lexA::UTP25, TRP1,2lm This study pGAD GAL4 activation domain, LEU2,2lm35 pGAD–MPP10 GAL4::SAS10, LEU2,2lm This study pGAD–SAS10 GAL4::MPP10, LEU2,2lm This study pGAD–IMP3 GAL4::IMP3, LEU2,2lm This study pGAD–IMP4 GAL4::IMP4, LEU2,2lm This study pET28a–UTP25 His 6 ::UTP 25, Kan R This study pGEX4T1–SAS10 GST::SAS10, Amp R This study YCplac33 ARS1, URA3, CEN4 51 YCp33GAL–UTP25 GAL1::UTP25, URA3, CEN4 This study pUG34 MET25::yEGFP3, CEN6, HIS3 U. Gueldener & J. H. Hegemann, unpublished pUG34–UTP25 MET25::yEGFP3-UTP25, HIS3, CEN6 This study pUG34–hUTP25 MET25::yEGFP3h-UTP25, HIS3, CEN6 This study pMET25–hUTP25 MET25:: UTP25, HIS3, CEN6 This study pPGK–hUTP25 PGK1:: UTP25, HIS3, CEN6 This study pUG36 MET25::yEGFP3, CEN6, URA3 U. Gueldener & J. H. Hegemann, unpublished pUG36–DsRed–NOP1 MET25::DsRED-NOP1, CEN6, URA3 This study YCp33GAL-A GAL1::PROTA, URA3, CEN4 52 YCp33GAL-A–UTP25 GAL1::PROTA-UTP25, URA3, CEN4 This study M. B. Goldfeder and C. C. Oliveira Utp25p affects pre-rRNA processing FEBS Journal 277 (2010) 2838–2852 ª 2010 The Authors Journal compilation ª 2010 FEBS 2847 [...]... Goldfeder and C C Oliveira Utp25p affects pre-rRNA processing Table 3 DNA oligonucleotides used for northern blot hybridization and primer extension analyses Oligo Sequence Ref P1 P2 P3 P4 P5 P6 P7 P8 anti -U3 5¢-GGTCTCTCTGCTGCCGGAAATG-3¢ 5¢-CATGGCTTAATCTTTGAGAC-3¢ 5¢-GCTCTCATGCTCTTGCCAAAAC-3¢ 5¢-CGTATCGCATTTCGCTGCGTTC-3¢ 5¢-CTCACTACCAAACAGAATGTTTGAGAAGG-3¢ 5¢-GTTCGCCTAGACGCTCTCTTC-3¢ 5¢-GCCGCTTCACTCGCCGTTACTAAGGC-3¢... sequence at the corresponding position To obtain Utp25(S19 8A) mutant, the Quickchange kit (Stratagene, La Jolla, CA, USA) was used The human ortolog hUTP25 gene (human C1ORF7, accession number BC022964) was PCR amplified from pCMV-SPORT6-C1ORF7 (Imagenes, Berlin, Germany) using primers: 5¢-GGATCCATGGGC AAACGCGGGAGCC-3¢ and 5¢-ATCGATGTCGACTCA TTTTTCTCCAGTAATGAAGAG-3¢ The gene was cloned in fusion with yEGFP3... fluorescence was observed as a nucleolar marker Localization of GFP was analyzed in the strain YMG-232 as a control Images were obtained on a Nikon TE300 inverted microscope equipped with a Roper CoolSnap HQ camera RNA analysis Exponentially growing cultures of yeast strains were shifted from galactose to glucose medium At various times, samples were collected and quickly frozen in a dry ice–ethanol bath Total... Osheim YN, Beyer AL & Baserga SJ (2006) The PINc domain protein Utp24, a putative nuclease, is required for the early cleavage steps in 18S rRNA maturation Proc Natl Acad Sci USA 103, 9464–9469 28 Chen W, Bucaria J, Band DA, Sutton A & Sternglanz R (2003) Enp1, a yeast protein associated with U3 and U14 snoRNAs, is required for pre-rRNA processing and 40S subunit synthesis Nucleic Acids Res 31, 690–699... Nop53p, an essential nucleolar protein that interacts with Nop17p and Nip7p, is required for pre-rRNA processing in Saccharomyces cerevisiae FEBS J 272, 4450–4463 FEBS Journal 277 (2010) 2838–2852 ª 2010 The Authors Journal compilation ª 2010 FEBS 2851 Utp25p affects pre-rRNA processing M B Goldfeder and C C Oliveira Supporting information The following supplementary material is available: Fig S1 Area quantitation... Baserga SJ (1999) Functional separation of pre-rRNA processing steps revealed by truncation of the U3 small nucleolar ribonucleoprotein component, Mpp10 Mol Cell Biol 19, 5441–5452 13 Venema J, Vos HR, Faber AW, van Venrooij WJ & ´ Raue HA (2000) Yeast Rrp9p is an evolutionarily conserved U3 snoRNP protein essential for early pre-rRNA processing cleavages and requires box C for its association RNA... centrifugation, and tested against purified His6–Utp25 and total yeast extract Yeast maintenance, transformation and sporulation Immunoblot analysis Yeast genetic techniques were conducted as described previously [37] Strains described in Table 2 were maintained in yeast extract–peptone (YP) medium or synthetic medium (YPD) with 2% (w ⁄ v) galactose or glucose as the carbon source, as indicated, and supplemented... Ignatchenko A, Li J, Pu S, Datta N, Tikuisis AP et al (2006) Global landscape of protein complexes in the yeast Saccharomyces cerevisiae Nature 440, 637– 643 Wehner KA, Gallagher JEG & Baserga SJ (2002) Components of an interdependent unit within the SSU processome regulate and mediate its activity Mol Cell Biol 22, 7258–7267 Tollervey D, Lehtonen H, Carmo-Fonseca M & Hurt EC (1991) The small nucleolar. .. Depletion of U14 small nuclear RNA (snR128) disrupts production of 18S rRNA in Saccharomyces cerevisiae Mol Cell Biol 13, 2469–2477 ´ ´ 8 Reichow SL, Hamma T, Ferre-D’Amare AR & Varani G (2007) The structure and function of small nucleolar ribonucleoproteins Nucleic Acids Res 35, 1452–1464 9 Colley A, Beggs JD, Tollervey D & Lafontaine DL (2000) Dhr1p, a putative DEAH-box RNA helicase, is associated with the. .. reverse transcriptase (Invitrogen, Carlsbad, CA, USA) and dNTPs (0.5 mm) for 30 min at 37 °C cDNA products were precipitated, resuspended in H2O, treated with Rnase A, denatured and analyzed on 6% denaturing polyacrylamide gels Gels were dried and analyzed in a Phosphorimager Oligonucleotides used in primer extension analyses are listed in Table 3 Polysome profile analysis For polysome profile analysis . Utp25p, a nucleolar Saccharomyces cerevisiae protein, interacts with U3 snoRNP subunits and affects processing of the 35S pre-rRNA Mauricio B an accumulation of the pre-rRNA 35S and the aberrant 23S, and a decrease in pre-rRNA 20S and mature 18S rRNA (Fig. 5A) . The large ribosomal subunit RNAs 25S, 5.8S and