The yeast nuclear exosome contains multiple 3′→5′ exoribonucleases, raising the question of why so many activities are present in the complex. All components are required during the 3′ processing of the 5.8S rRNA, together with the putative RNA helicase Dob1p/Mtr4p. During this processing three distinct steps can be resolved, and hand‐over between different exonucleases appears to occur at least twice. 3′ processing of snoRNAs (small nucleolar RNAs) that are excised from polycistronic precursors or from mRNA introns is also a multi‐step process that involves the exosome, with final trimming specifically dependent on the Rrp6p component. The spliceosomal U4 snRNA (small nuclear RNA) is synthesized from a 3′ extended precursor that is cleaved by Rnt1p at sites 135 and 169 nt downstream of the mature 3′ end. This cleavage is followed by 3′→5′ processing of the pre‐snRNA involving the exosome complex and Dob1p. The exosome, together with Rnt1p, also participates in the 3′ processing of the U1 and U5 snRNAs. We conclude that the exosome is involved in the processing of many RNA substrates and that different components can have distinct functions.
Eukaryotic cells contain a large number of stable RNA species, nearly all of which are synthesized by post‐transcriptional processing from larger precursors. This has long been known for the highly abundant cytoplasmic RNAs, tRNAs and rRNAs, but more recently it has become clear that this is also the case for the small nuclear RNAs (snRNAs), which participate in pre‐mRNA splicing, and the small nucleolar RNAs (snoRNAs), which participate in rRNA processing and modification.
Analyses of 3′ processing of the 5.8S rRNA in Saccharomyces cerevisiae led to the identification of the exosome complex of 3′→5′ exonucleases (Mitchell et al., 1996, 1997; Allmang et al., 1999). Originally reported to contain five different 3′→5′ exonucleases, it is now likely that the exosome contains at least 10 exonucleases (Allmang et al., 1999). These include six homologues of Escherichia coli RNase PH (Rrp41p, Rrp42p, Rrp43p, Rrp45p, Rrp46p and Mtr3p), a homologue of E.coli RNase R and RNase II (Rrp44p/Dis3p), and a homologue of E.coli RNase D (Rrp6p). Two other components, Rrp4p and Rrp40p, are homologous to each other, and Rrp4p has been shown to be a 3′→5′ exonuclease (Mitchell et al., 1997). The remaining component is Csl4p, which has not been reported to have nuclease activity but does contain a potential S1 RNA binding domain (S.Mian, personal communication), indicating that it is also likely to bind RNA directly. All components of the exosome are essential for viability (Mitchell et al., 1996, 1997; Noguchi et al., 1996; Baker et al., 1998; Allmang et al., 1999) with the exception of Rrp6p, the absence of which results in temperature‐sensitive (ts) lethality and impaired growth at all temperatures (Briggs et al., 1998). Normal processing of the 7S pre‐rRNA to the mature 5.8S rRNA requires all components of the exosome, but the phenotype of the rrp6‐Δ strain differs substantially from that of the other mutants, making it unclear whether these function in the same or parallel pathways. In addition to the components of the exosome, the yeast genome contains at least six other open reading frames that are predicted to encode 3′→5′ exonucleases, based on sequence comparisons with E.coli enzymes (Mian, 1997; Moser et al., 1997).
5′ processing of the 5.8S rRNA requires the activity of two homologous 5′→3′ exonucleases, Rat1p and Xrn1p, with the major role probably being played by Rat1p (Henry et al., 1994). The same exonucleases are required for the 5′ processing of several snoRNA species, many of which are either synthesized from polycistronic pre‐snoRNA transcripts, or are excised from the introns of pre‐mRNAs following intron lariat debranching (Ooi et al., 1998; Petfalski et al., 1998). All characterized yeast polycistronic snoRNAs are initially processed by endonuclease cleavage by Rnt1p (Chanfreau et al., 1998a,b; Qu et al., 1999), the yeast homologue of E.coli RNase III (Abou Elela et al., 1996), which separates the individual pre‐snoRNAs. Rnt1p also processes the pre‐rRNA in the 3′ external transcribed spacer (3′‐ETS) (Abou Elela et al., 1996; Kufel et al., 1999) and cleaves 3′ extended precursors to the U1, U2 and U5 snRNAs (Chanfreau et al., 1997; Abou Elela and Ares, 1998; Seipelt et al., 1999). In the absence of Rnt1p cleavage, polyadenylated forms of U1 and U2 are synthesized (Abou Elela and Ares, 1998; Seipelt et al., 1999). In rnt1‐Δ strains the processing of the 3′‐ETS and polycistronic pre‐snoRNAs is almost completely inhibited, with severe effects on rRNA and snoRNA synthesis. However, 3′ processing of the snRNAs continues, indicating the existence of alternative processing pathways or activities. The existence of alternative 3′ processing pathways has also been shown for yeast tRNAs (Yoo and Wolin, 1997) and multiple activities can carry out 3′ processing of many small RNAs in E.coli (see Li et al., 1998 and references therein).
Here, we investigate the roles of the different exosome components in the 3′ processing of the 5.8S rRNA and pre‐rRNA spacer degradation, and present data indicating that the exosome also participates in the 3′ processing of many snRNA and snoRNA species.
The pathway of 5.8S 3′ processing
Three different 3′ extended forms of the 5.8S rRNA can be detected in wild‐type strains of yeast. The 7S pre‐rRNAs are 3′ extended by ∼134 nt to site C2 in ITS2 (Veldman et al., 1981), while the 6S pre‐rRNAs represent 5.8S species with short, probably heterogeneous, 3′ extensions of ∼8 nt (Mitchell et al., 1996) (see Figure 1 for a schematic showing the pre‐rRNA and processing sites). In addition, the 5.8S + 30 species (Briggs et al., 1998) can be detected at low levels in the wild type (Figure 1D). The 5′ end of the 5.8S rRNA is heterogeneous, with two major forms that differ by 8 nt, designated 5.8SL and 5.8SS (Henry et al., 1994). Since 5′ processing of the 5.8S rRNA precedes 3′ processing, the 7S pre‐rRNA, 6S pre‐rRNA and 5.8S + 30 species all show long and short forms, e.g. 5.8S + 30L and 5.8S + 30S (Figure 1A and D).
Processing of the 5.8S rRNA was compared in strains carrying conditional mutations for the 10 essential components of the exosome, using the ts‐lethal mtr3‐1 allele and GAL‐regulated constructs allowing depletion of Rrp4p, Rrp40p, Rrp41p, Rrp42p, Rrp43p, Rrp44p, Rrp45p, Rrp46p or Csl4p (Figure 1A). As previously reported (Mitchell et al., 1997; Allmang et al., 1999), similar 3′ extended intermediates were observed in each case, forming a ladder up to the position of the 7S pre‐rRNAs. The GAL::rrp41 strain underexpresses Rrp41p in permissive, RSG medium, and therefore shows some accumulation of the extended species in the 0 h sample. Strong accumulation of the 6S pre‐rRNA was seen in the strains depleted of Rrp40p or Rrp45p, while 6S was reduced in the Rrp41p‐, Rrp44p‐ or Rrp46p‐depleted and mtr3‐1 strains and little altered in strains depleted of Rrp4p or Csl4p. Moreover, in strains depleted of Rrp41p, Rrp42p or Rrp43p the position of the 6S pre‐rRNA appeared to be displaced up the gel, corresponding to an increase in size of ∼3 nt. We conclude that different components of the exosome do not play identical roles in processing of the 6S pre‐rRNA.
In rrp6‐Δ strains, a distinctly different pattern of processing was observed (Figure 1A and D) (Briggs et al., 1998) with accumulation of high levels of the 5.8S + 30 pre‐rRNAs. To determine whether Rrp6p and the other exosome components act on the same pre‐rRNA processing pathway or function in independent parallel pathways, double‐mutant strains were constructed carrying the rrp6‐Δ allele and either the GAL::rrp41 (Mitchell et al., 1997) or the GAL::rrp45 allele. Depletion of either Rrp41p or Rrp45p from a strain lacking Rrp6p led to the progressive loss of the 5.8S + 30 processing intermediate, clearly showing that Rrp41p and Rrp45p act epistatically to Rrp6p in the 5.8S processing pathway (Figure 1D). Metabolic labelling of an rrp6‐Δ strain also indicated that Rrp6p participates in the major 5.8S rRNA processing pathway (Briggs et al., 1998).
Mutations in the putative ATP‐dependent RNA helicase Dob1p (Mtr4p) also interfere with 3′ processing of the 5.8S rRNA (de la Cruz et al., 1998). A GAL::dob1 strain genetically depleted of Dob1p accumulated both the 5.8S + 30 species and larger intermediates that are seen in other exosome mutants (Figure 1D). The 6S pre‐rRNAs accumulated in the GAL::dob1 strain 2 and 6 h after transfer to glucose medium but were reduced after 24 h. This probably occurred because processing of the pre‐rRNA is strongly inhibited prior to synthesis of 6S, as shown by the high accumulation of the 7S and 5.8S + 30 pre‐rRNAs.
Strains depleted of exosome components or Dob1p, or carrying the mtr3‐1 or rrp6‐Δ mutations, accumulated the excised 5′ ETS region and degradation intermediates (Figure 1B and F) (de la Cruz et al., 1998; Allmang et al., 1999). The levels of the degradation intermediates were quite variable among the exosome mutants. This indicates that, while degradation of the 5′ ETS involves the entire exosome complex, different components do not have identical functions during this activity.
Many yeast snoRNAs are synthesized by post‐transcriptional processing, either from the excised introns of pre‐mRNAs or from polycistronic transcripts that include multiple snoRNAs. In higher eukaryotes, both 5′ and 3′ processing of pre‐snoRNAs involves exonuclease activities (Caffarelli et al., 1994, 1996; Cecconi et al., 1995; Cavaillé and Bachellerie, 1996; Kiss et al., 1996). 5′ processing of several yeast pre‐snoRNAs was shown to require the 5′→3′ exonucleases Rat1p and Xrn1p, with the major role being performed by Rat1p (Larimer et al., 1992; Kenna et al., 1993; Petfalski et al., 1998; Villa et al., 1998). Northern analysis of RNA extracted from the rrp6‐Δ strain showed many species with a discrete shift in gel mobility that would correspond to an increase in length of 3 nt (Figure 2A). This was observed for the intronic snoRNAs U18, U24 and snR39, as well as U14 and snR41, which are encoded in dicistronic and polycistronic transcripts, respectively (Figure 2A). In contrast, the gel mobilities of snoRNAs that are transcribed from their own promoter and terminator, snR10 (Figure 2A) and U3 (data not shown), were unaffected by deletion of RRP6. The dicistronic snoRNA, snR190, which is cotranscribed with U14, was also not affected (Figure 2A).
Primer extension revealed that the position of the 5′ end was unaffected for each of these snoRNAs (Figure 2B) indicating that the altered gel mobility represents a failure in the 3′ trimming of the snoRNA. For U24 the presence of a 3′ extension was confirmed by RNase protection (data not shown).
Strains individually depleted for each of the other exosome components or Dob1p, or carrying the rrp4‐1, mtr3‐1 or dob1‐1 mutations at non‐permissive temperature, were analysed for processing of U14, U18, U24 and snR190 (shown for GAL::rrp41 and GAL::rrp45 in Figure 2C and dob1‐1 in Figure 2A). No clear alteration in the length of the mature snoRNAs was observed, showing that depletion or mutation of other individual components of the exosome or Dob1p does not inhibit snoRNA 3′ end trimming. Double‐mutant strains lacking Rrp6p and depleted of either Rrp41p or Rrp45p were also analysed (shown for GAL::rrp41/rrp6‐Δ in Figure 3I). The length of the ‘almost‐mature’ snoRNAs in these strains was the same as in strains lacking only Rrp6p. We conclude that 3′ trimming of the snoRNAs requires specifically the Rrp6p component of the exosome complex. In addition, longer extended forms were observed for U14, U18 and U24, but not for snR190 (Figure 3 and data not shown).
Yeast U18 and U24 are intron encoded (Maxwell and Fournier, 1995; Qu et al., 1995; Kiss‐László et al., 1996) and are synthesized predominantly from the debranched intron lariats (Ooi et al., 1998; Petfalski et al., 1998). As previously reported (Petfalski et al., 1998), the species corresponding to the introns that are 3′ unprocessed but 5′ processed to the end of the mature snoRNAs [U18‐3′ (253 nt) and U24‐3′ (192 nt)] are detected in the wild‐type strain (Figure 3IIB and IIC, lanes 1, 3, 5 and 7). Both U24‐3′ and U18‐3′ can be detected on Northern blots with probes that hybridize specifically with 3′ extended species (Figure 3II, lanes 7 and 8) and both are lost in strains carrying mutations in the 5′→3′ exonucleases Rat1p and Xrn1p (Petfalski et al., 1998). In strains lacking Rrp6p, the level of U24‐3′ was increased and ladders of intermediates appeared both below and above this species (Figure 3IC, lane 2). To test the possibility that the species observed above U24‐3′ correspond to polyadenylated forms, RNA from the wild‐type and the GAL::rrp41/rrp6‐Δ strain 2 h after transfer to glucose medium was deadenylated in vitro with oligo(dT) and RNase H. On deadenylation the heterogeneous species observed above the U24‐3′ band were lost and the U24‐3′ signal was increased (Figure 3IIC, compare lanes 4 and 6). Strains depleted of Rrp41p (Figure 3IC, lane 3) did not clearly accumulate U24‐3′, whereas accumulation of U24‐3′ and a ladder of smaller intermediates were detected on depletion of Rrp45p and in a rrp4‐1 strain (data not shown), demonstrating that processing of U24‐3′ is not specific for Rrp6p. Extended forms of U24‐3′ were not observed in these strains, and the formation of polyadenylated species may be specific for rrp6‐Δ. It is notable that RRP6 was originally identified as a mutation that suppressed a defect in polyadenylation (Briggs et al., 1998).
For U18, heterogeneous 3′ extended species were observed in the rrp6‐Δ strain, and these were longer and more discrete in the GAL::rrp41/rrp6‐Δ double mutant (Figure 3IB). These were estimated to be in the range 250–300 nt. Following deadenylation, the U18‐3′ RNA is clearly seen to be strongly accumulated in the GAL::rrp41/rrp6‐Δ strain, with a ladder of intermediates extending up to this position (Figure 3IIB, lanes 4 and 8). The same species are detected in the non‐deadenylated sample (Figure 3IIB, lane 6), but are less clear.
snR190 and U14 are cotranscribed, separated by Rnt1p cleavage and then 5′ processed by Rat1p (Zagorski et al., 1988; Chanfreau et al., 1998b; Petfalski et al., 1998). Extended forms of U14 were observed in rrp6‐Δ strains but were both more abundant and longer in GAL::rrp41/rrp6‐Δ strains that were also depleted of Rrp41p (Figure 3A). On deadenylation these species formed a ladder up to the U14‐3′ RNA, which was estimated to be ∼525 nt in length. It seems probable that U14‐3′ extends to the transcription termination site, which has not yet been located. Extended snoRNAs were not observed in strains depleted of Dob1p (data not shown).
To determine whether U18‐3′ and U24‐3′ species are processing intermediates or dead‐end products, a transcription inhibition experiment was performed (Figure 4). RNA was recovered from a rbp1‐1 strain, carrying a ts mutation in RNA polymerase II, at time points after shift to the non‐permissive temperature. Following transcription inhibition, U18‐3′ and U24‐3′ were progressively lost, indicating that they are normal processing intermediates.
We conclude that the exosome participates in the 3′ processing of snoRNAs, processing the primary transcript of the dicistronic snR190‐U14 snoRNAs and the debranched intron lariats containing the U18 and U24 snoRNAs. As seen for 5.8S synthesis, snoRNA processing is at least biphasic. Initial processing is partially inhibited, but not blocked, by different mutations in the exosome, whereas trimming of the final 3 nt specifically requires Rrp6p. The additive effect of the rrp41 and rrp6 mutations on U18 and U14 synthesis is, however, in contrast to the epistatic interactions observed for 5.8S processing. In the absence of Rrp6p some polyadenylation of processing intermediates also occurs.
Since a number of snRNAs have been shown to undergo 3′ processing we analysed snRNA species for alterations in exosome mutants.
Analysis of the U4 snRNA showed the existence of low levels of two longer forms (U4‐3′i and U4‐3′ii) in the wild type (Figure 5, lanes 1 and 12). Both species were detected with either the internal U4 probe (oligo 243) (Figure 5B) or a probe complementary to the sequence across the 3′ end of the mature U4 (oligo 246), which hybridizes only to 3′ extended species (Figure 5A). U4‐3′i is a set of heterogeneous species carrying short 3′ extensions similar in length to those detected in the 6S pre‐rRNA. Oligo 246 detects only the longer forms of U4‐3′i. The U4‐3′ii species is ∼140 nt larger than mature U4 and is presumably a normal processing intermediate since it is detected in wild‐type cells. In strains depleted of the exosome components Rrp41p (Figure 5, lanes 2 and 3) or Rrp45p (Figure 5, lanes 7 and 8), or lacking Rrp6p (Figure 5, lane 15), the levels of U4‐3′i and U4‐3′ii were increased and a ladder of intermediates was observed that extended from the size of U4‐3′i to that of U4‐3′ii. The GAL::rrp41 strain shows some accumulation of the extended species in the 0 h sample due to underexpression of Rrp41p (Figure 5, lane 2). The same species were accumulated in a strain depleted of Dob1p (data not shown). The accumulation of these intermediates was not stronger in the rrp6‐Δ strain that was also depleted of Rrp41p or Rrp45p than in the rrp6‐Δ single‐mutant strain (data not shown). In strains lacking Rnt1p, U4‐3′ii RNA was absent whilst the levels of the U4‐3′i species were increased. These observations suggested that the U4‐3′ii is generated by Rnt1p cleavage and acts as an entry site for the exosome complex.
Strains lacking Rnt1p also accumulated longer 3′ extended forms of U4 (Figure 5, lanes 13 and 14; 4‐fold more RNA was loaded in lane 14 to allow visualization of the longer 3′ extended species). These were more abundant in rnt1‐Δ strains that were also depleted of Rrp41p (Figure 5, lanes 4–6), Rrp45p (Figure 5, lanes 9–11) or Rrp6p (Figure 5, lane 16). Particularly in the rnt1‐Δ/rrp6‐Δ strain (Figure 5B, lane 16) these were seen to extend to a distinct species (U4‐3′iii) of ∼590 nt. It seems likely that these represent intermediates in the 3′ processing of U4 from U4‐3′iii, which most probably extends to the transcription termination site. Little effect on the 3′ extended forms of U4 was seen on treatment with RNase H and oligo(dT) (data not shown).
Rnt1p cleaves on both sides of extended, imperfect stems with closing AGNN tetraloops (Chanfreau et al., 1998a). Inspection of the U4 3′ flanking region identified good consensus Rnt1p cleavage sites (Figure 6D). To determine whether this represented a genuine substrate for Rnt1p, cleavage was assayed in vitro. The region of the U4 3′ flanking sequence from positions +49 to +235, which includes the putative cleavage sites, was transcribed in vitro (see Materials and methods). The sites cleaved in vitro by recombinant His6–Rnt1p were identified by primer extension with oligo 249 (Figure 6A). Two major stops, corresponding to sites of cleavage between nucleotides +135/+136 (site I in Figure 6D) and +169/+170 (site II in Figure 6D) with respect to the 3′ end of the mature U4, were detected following incubation with His6–Rnt1p (Figure 6A, lane 5); these sites were not detected in the no‐enzyme controls (Figure 6A, lane 6). The sites lie on both sides of the predicted stem–loop structure and are in good agreement with the consensus for other Rnt1p cleavage sites (Figure 6D). Additional minor stops were detected one nucleotide 5′, corresponding to positions +134/+135 and +168/+169 (smaller arrows in Figure 6D). To confirm that these stops represented sites of endonuclease cleavage, internally labelled RNA transcripts (Figure 6B) were also assayed using either recombinant His6–Rnt1p or extracts prepared from RNT1+ and rnt1‐Δ strains of yeast (see Materials and methods). Incubation with either recombinant Rnt1p (Figure 6B, lanes 3–6) or the yeast extract containing Rnt1p (Figure 6B, lane 7), resulted in the appearance of discrete cleavage products that were not observed in the input RNA (Figure 6B, lane 2) or with the yeast extract lacking Rnt1p (Figure 6B, lane 8). The estimated sizes of these products are in good agreement with the predicted sizes of the products of cleavage at both sites I and II (Figure 6B; the predicted locations and sizes of the cleavage products are indicated below the gel).
To identify the position of in vivo processing, we performed primer extension using an oligo that hybridizes to the U4 flanking sequence 3′ to the Rnt1p cleavage site (oligo 249). In the wild‐type strain, a primer extension stop was detected between nucleotides +135 and +136, precisely matching in vitro cleavage site I (Figure 6C, lane 1). This stop was absent from the rnt1‐Δ strain (Figure 6C, lane 2). In vivo cleavage could not be detected at site II, but the primer gave a high background in this region (data not shown). Following cleavage of the pre‐rRNA in the 3′‐ETS by Rnt1p, the excised 3′ fragment is degraded 5′→3′ by Rat1p (Kufel et al., 1999). The primer extension stop observed in the rat1‐1 strain at position +135/+136 was stronger than in the wild type, supporting the conclusion that this is a site of in vivo endonuclease cleavage (Figure 6C, lane 3).
We conclude that Rnt1p cleaves the 3′ flanking sequence of the U4 snRNA. The predominant in vivo cleavage is at position +135/+136, generating the U4‐3′ii RNA. Cleavage may also occur at +169/+170, although no species corresponding to this cleavage was detected by Northern hybridization. The +135/+136 cleavage acts as a site of entry for the exosome complex. Since U4‐3′ii is detected in wild‐type cells it is presumably a normal processing intermediate. In the absence of Rnt1p, longer transcripts are detected; these are also substrates for the exosome since they accumulate at higher levels in double‐mutant strains, but can be efficiently processed to mature U4 by another pathway(s). It seems likely that the accumulation of the short 3′‐extended U4‐3′i species is not a direct product of inhibition of the exosome complex, but is associated with the activation of an alternative processing pathway, since these species are increased relative to the wild type in strains lacking either exosome components or Rnt1p.
Quantitation of the Northern data (Table I) revealed that the mature U4 accumulates above the wild‐type level in both the exosome and Rnt1p mutant strains relative to the cytoplasmic 7SL RNA homologue SCR1 (Table I and Figure 5D) or the nucleolar MRP RNA (data not shown). We conclude that a significant fraction of the U4 or pre‐U4 population is normally degraded by an exosome‐dependent pathway in wild‐type strains.
Two forms of the U5 snRNA, U5L and U5S, which differ at their 3′ ends, are observed in wild‐type yeast strains (Patterson and Guthrie, 1987). Species with short, heterogeneous 3′ extensions were observed for both U5L (U5L‐3′) and U5S (U5S‐3′). These species are detected with an internal U5 probe (oligo 244; Figure 7A and D), and also with probes across the 3′ end of U5L (oligo 247; Figure 7B) or across the 3′ end of U5S (oligo 248; Figure 7C), although only the longer forms are detected by oligos 247 and 248. Both U5L‐3′ and U5S‐3′ were detected at low levels in wild‐type strains (Figure 7, lanes 1 and 12). These, and a longer species, U5‐3′i, were accumulated in strains depleted of Rrp41p (Figure 7, lanes 2 and 3) or Rrp45p (Figure 7, lanes 7 and 8), or lacking Rrp6p (Figure 7, lane 15) and were mildly accumulated in strains depleted of Dob1p (data not shown). In rrp6‐Δ strains that were also depleted of Rrp41p, the accumulation of these intermediates was not clearly different from the rrp6‐Δ single‐mutant strains (data not shown). Cleavage sites for Rnt1p are present in the U5 3′ flanking region (Chanfreau et al., 1997). In the absence of Rnt1p, U5L‐3′ and U5‐3′i were absent and the level of U5L was strongly reduced (Figure 7D, lanes 13 and 14 and Table I). Based on their gel mobilities, the U5L‐3′ species extend up to a position close to the 5′ Rnt1p cleavage site [site λ in Figure 7 and Chanfreau et al. (1997)], while the larger U5‐3′i species extends to a position close to the 3′ Rnt1p cleavage site [site σ in Figure 7 and Chanfreau et al. (1997)]. The accumulation of these species in exosome mutants suggests that the Rnt1p cleavages normally act as entry sites for the exosome. In the absence of Rnt1p (Figure 7A, lanes 13 and 14), longer 3′ extended forms of U5 were detected in a ladder to a species designated U5‐3′ii of ∼690 nt (Figure 7A). These were strongly increased in rnt1‐Δ strains also depleted of Rrp41p, Rrp45p or Rrp6p (Figure 7A). It seems likely that U5‐3′ii extends to the transcription termination site and that the ladder represents intermediates in its 3′ processing to U5S. Little effect on the 3′ extended forms of U5 was seen on treatment with RNase H and oligo(dT) (data not shown).
The signal for U5S was strikingly increased relative to U5L, SCR1 RNA or MRP RNA (data not shown) in strains depleted of Rrp41p (Figure 7, lanes 2 and 3), Rrp45p (Figure 7, lanes 7 and 8) or Dob1p (data not shown), or lacking Rrp6p (Figure 7, lane 15) or Rnt1p (Figure 7D, lane 13). These results are quantified in Table I; the U5S:U5L ratio is changed >10‐fold on depletion of Rrp41p or Rrp45p and the total amount of U5L + U5S synthesized is substantially more than wild type. We conclude that when the normal processing pathway is active, a large fraction of the pre‐U5S or U5S population is degraded. U5S is unusual among snRNAs in not having a terminal stem–loop structure beyond the Sm binding site to stabilize the 3′ end, which may make it particularly liable to degradation.
A 3′ extended form of the U1 snRNA that is likely to extend to the Rnt1p cleavage site has been reported for wild‐type strains (Seipelt et al., 1999). This presumably corresponds to the species designated U1‐3′ii that we detect on Northern hybridization (Figure 8, lanes 3 and 14), which is absent in rnt1‐Δ strains (Figure 8, lane 15). The yeast U1 RNA is 568 nt in length (Kretzner et al., 1987; Siliciano et al., 1987), substantially larger than U4 and U5, and the precursors are therefore less well separated in Figure 8 than in Figures 5 and 7. In addition, we see shorter 3′ extended forms, U1‐3′i, which are strongly accumulated in strains depleted of Rrp41p (Figure 8, lanes 4 and 5) or Rrp45p (Figure 8, lanes 9 and 10), or lacking Rrp6p (Figure 8, lane 2). Unlike the longer U1‐3′ii species, the shorter U1‐3′i persists in rnt1‐Δ strains (Figure 8, lane 15). The rnt1‐Δ strain accumulates longer, heterogeneous 3′ extended species, U1‐3′iii, which are reported to be polyadenylated (Seipelt et al., 1999). As for U4 and U5, these long species are more strongly accumulated in rnt1‐Δ strains that also lack exosome components. In contrast to the results for U4 and U5, the mature U1 is not clearly accumulated above wild‐type levels in the exosome mutants.
We conclude that 3′ processing of U1, like U4 and U5, involves 3′ cleavage by Rnt1p and processing by the exosome, although mutations in individual components of the complex do not block processing.
The characterization of the exosome complex raised an obvious question: why are so many different exonucleases present in the complex? Possible explanations are that multiple enzymes might function in the processing of single RNA substrates or that different enzymes might be preferentially active on different substrates. We have presented initial evidence for both of these phenomena, as well as identifying a large number of additional substrates for the complex.
During the processing of the 7S pre‐rRNA to the 5.8S rRNA, the specificity of the exonuclease appears to change at least twice (see Figure 9A), a phenomenon that we refer to as exonuclease hand‐over. Moreover, the effects on the 6S pre‐rRNA vary between different mutants, showing that they do not have identical functions. The putative RNA helicase Dob1p/Mtr4p appears to be required for each of these processing steps. The 5.8S + 30 pre‐rRNAs extend to the 3′ side of a predicted stable stem–loop structure, which includes the two terminal nucleotides of the mature 5.8S rRNA (Yeh and Lee, 1990). It is, however, unclear whether processing is inhibited by the stem–loop structure itself, or by the consequent very close proximity to the 3′ region of the mature 5.8S rRNA, which is likely to be associated with ribosomal proteins. The simplest interpretation of the data would be that all of the essential exosome components, but not Rrp6p, are required for normal processing of the 7S pre‐rRNA to 5.8S + 30. At this point, Rrp6p may take over the major role and process the 5.8S + 30 species to 6S pre‐rRNA. In the absence of the exosome components some other activity is able to digest the 7S pre‐rRNAs partially, producing the observed ladder of intermediates. This alternative activity is not provided by Rrp6p, since these species are detected in the GAL::rrp41/rrp6‐Δ and GAL::rrp45/rrp6‐Δ strains. Moreover, both the remaining exosome components and any alternative activity can poorly process the 5.8S + 30 pre‐rRNAs since these accumulate to high levels in the rrp6‐Δ strain. It is notable that the 3′ end of this species is predicted to lie in a stem structure, which may inhibit its processing if released as a free RNA. Unlike the larger intermediates that are specifically detected in the exosome mutants, the 6S pre‐rRNAs are detected in the wild type. This may represent a site at which rapid processive processing is normally replaced by slower distributive trimming of the RNAs. The heterogeneity shown between different exosome mutant strains indicates that 6S pre‐rRNA is processed by the exosome, rather than being the product of an alternative pathway that is activated in the absence of exosome activity. The pattern of intermediates observed during degradation of the 5′ ETS region of the pre‐rRNA also varies amongst the different exosome mutants, suggesting that exonuclease hand‐over may be occurring during this processing.
Analysis of pre‐snRNA processing reveals a complex picture. Each of the RNA polymerase II‐transcribed snRNAs in yeast, U1, U2, U4 and U5, has a cleavage site for Rnt1p in the 3′ flanking region (Chanfreau et al., 1997; Abou Elela and Ares, 1998; Seipelt et al., 1999 and this work). For U1, U4 and U5 this may act as an entry site for the exosome complex, acting together with the Dob1p RNA helicase (see Figure 9C). However, in no case is synthesis of the snRNA blocked by the inhibition of cleavage by Rnt1p or by mutations in exosome components, indicating that alternative processing pathways exist.
In strains lacking Rnt1p the synthesis of long 3′ extended forms of U1 and U2 has been reported (Abou Elela and Ares, 1998; Seipelt et al., 1999) and we show here that this is also the case for U4 and U5. During pre‐rRNA processing, the 3′ end of the 35S pre‐rRNA is normally cleaved cotranscriptionally by Rnt1p (Abou Elela et al., 1996; Kufel et al., 1999). In the absence of Rnt1p the pre‐rRNA transcripts extend to a position close to the site of transcription termination. The extended U4 and U5 species form a ladder to a discrete size that we speculate represents the transcription termination site. These are not detected in the wild type, suggesting that pre‐U4 and pre‐U5 may also be cleaved cotranscriptionally. The extended species are substrates for the exosome since they accumulate at higher levels in rnt1‐Δ strains that are depleted for exosome components. For both the pre‐rRNA and U4, the excised 3′ fragments generated by Rnt1p cleavage are degraded by the 5′→3′ exonuclease Rat1p, which also processes the 5′ end of the 5.8Ss rRNA and many snoRNAs (see Figure 9B) (Amberg et al., 1992; Henry et al., 1994; Petfalski et al., 1998; Villa et al., 1998).
3′ maturation of snoRNAs that are excised from mRNA introns (U18 and U24) or from a dicistronic transcript (U14) also involves the exosome. U18 and U24 are predominately processed from the debranched intron lariat (Ooi et al., 1998; Petfalski et al., 1998; Villa et al., 1998). In the exosome mutants we see accumulation of the species in which the 5′ end of the snoRNA has been matured but the intron is 3′ unprocessed (U18‐3′ and U24‐3′), together with a ladder that probably represents intermediates in the 3′ processing of these to the mature snoRNAs. The U18‐3′ and U24‐3′ RNAs also undergo some polyadenylation in strains lacking Rrp6p. In Rnt1p mutants, the 3′ extended forms of U1 and U2 that are generated also become polyadenylated (Abou Elela and Ares, 1998; Seipelt et al., 1999), so this seems to be a general phenomenon in yeast. We conclude that 3′ processing of the debranched intron lariats containing U18 and U24 normally involves the exosome (see Figure 9B). Apparent intermediates in the 3′ processing of U14 are also observed, particularly in strains lacking both Rrp6p and Rrp41p. These may extend to the transcription termination site, but this has not yet been localized. Final trimming of each of these snoRNAs specifically requires Rrp6p. This activity apparently cannot be substituted by other exonucleases, since the entire snoRNA population is shifted in size by ∼3 nt. This trimming activity is not clearly inhibited by mutations in other single components of the exosome. It is notable that each of the enzymatic activities shown to be involved in pre‐snoRNA processing, the exosome, the 5′→3′ exonuclease Rat1p and the endonuclease Rnt1p, also participate in pre‐rRNA processing.
There are apparent similarities between 3′ processing of the 5.8S rRNA, snoRNAs and snRNAs. In each case there is a downstream cleavage, by endonuclease or splicing, which acts as a site of entry for exonucleases. Processing is at least biphasic with short 3′ extended forms accumulating, and each appears to involve the activities of more than one component of the exosome. In each case, other activities can partially substitute for the mutant exosome components. For the 7S pre‐rRNA, and particularly the 5.8S + 30 pre‐RNA, this is inefficient, and synthesis of mature 5.8S is strongly inhibited. The final trimming of snoRNAs is apparently completely dependent on the activity of Rrp6p, but the processing of further 3′ extended pre‐snoRNAs and pre‐snRNAs can be carried out by other activities with good efficiency, as shown by the wild‐type levels of the mature RNAs. Indeed, U4, and particularly U5S, are synthesized at substantially elevated levels in exosome mutants, indicating competition between synthesis of mature snRNA and degradation of the precursors in the wild type.
It is unclear whether residual processing in the exosome mutant strains is carried out by other components of the complex. It is notable that double‐mutant strains lacking both Rrp6p and Rrp41p show stronger phenotypes for some activities (e.g. 3′ processing of U14 and U18) than does either single mutant, indicating that the absence of one component does not necessarily inactivate the entire complex. Alternatively, the yeast genome contains several other predicted 3′→5′ exonucleases (Mian, 1997; Moser et al., 1997) that may be able to partially substitute for the exosome. The combination of mutations in these genes with mutations in exosome components will now be needed to analyse their interactions and substrates.
Materials and methods
Growth and handling of S.cerevisiae were by standard techniques. The transformation procedure was according to Gietz et al. (1992). Except where stated, strains were grown in liquid and solid minimal medium containing 0.67% yeast nitrogen base (Difco) and 2% glucose. GAL‐regulated strains were pre‐grown in RSG medium (2% raffinose, 2% sucrose, 2% galactose, 0.67% yeast nitrogen base) and harvested at intervals following a shift to 2% glucose.
Yeast strains used and constructed in this study are listed in Table II.
To construct the double mutants GAL10::rrp45/rrp6‐Δ (YCA30) and GAL10::rrp41/rrp6‐Δ (YCA31), the RRP6::TRP disrupted allele was amplified by PCR from genomic DNA extracted from rrp6‐Δ (YCA12). The PCR product was transformed into the corresponding conditional mutant strains P118 and YCA20, respectively. The amplification of RRP6::TRP was done with oligos: 5′RRP6, CAGTAATGAATATTAATGTTCATCTGAAGATAGACG; 3′RRP6, ATGGTGTGCATGGGGGAGCCATAACTCCATGACACA. Strains YDL401, P118 and YCA20 were used to construct the mutants rnt1‐Δ (YJK10), GAL10::rrp41/rnt1‐Δ (YJK11) and GAL10::rrp45/rnt1‐Δ (YJK12), using the same PCR strategy. Oligonucleotides 5′RNT1, 5′‐GAAGACATATCCGAAGTGACA and 3′RNT1, 5′‐GGATTTCTATACCCTCGAGGAG, complementary to sequences beyond the RNT1 gene, were used for the amplification with genomic DNA extracted from rnt1‐Δ strain generously provided by G.Chanfreau (Chanfreau et al., 1998b). Transformants were selected for Trp+ prototrophy and were screened by PCR. The phenotypes of respective constructs were confirmed by Northern hybridization. Strain rrp6‐Δ/rnt1‐Δ (YJK13) was constructed by crossing YJK10 and YCA12 strains. The double‐mutant strain was selected from dissected full tetrads by testing for the pre‐rRNA and snoRNA processing phenotypes by Northern hybridization. Wild‐type RNT1 and rnt1‐Δ sister strains (Chanfreau et al., 1998b) were used to prepare whole‐cell extract. Strain rat1‐1 was kindly provided by C.Cole (Amberg et al., 1992).
RNA extraction, Northern hybridization and primer extension
RNA was extracted as described previously (Tollervey and Mattaj, 1987). Northern hybridization (Tollervey, 1987) and primer extension (Beltrame and Tollervey, 1992) were as described previously. Standard 6 or 8% acrylamide gels were used to analyse low molecular weight rRNA species and primer extension reactions.
For pre‐rRNA and rRNA analysis the following oligonucleotides were used: 017 5′‐GCGTTGTTCATCGATGC; 020 5′‐TGAGAAGGAAATGACGCT; 033 5′‐CGCTGCTCACCAATGG; 041 5′‐CTACTCGGTCAGGCTC.
The oligonucleotides used for Northern blot hybridization and primer extensions on other small RNAs were as follows: 031 (MRP) 5′‐AATAGAGGTACCAGGTCAAGAAGC; 201 (snR190) 5′‐CGTCATGGTCGAATCGG; 202 (U14) 5′‐TCACTCAGACATCCTAGG; 205 (U18) 5′‐GTCAGATACTGTGATAGTC; 206 (U18‐3′) 5′‐GCTCTGTGTGCTATCGTC; 210 (U14‐3′) 5′‐GTATACGATCACTCAGAC; 213 (U24‐3′) 5′‐AAACCATTCATCAGAG; 214 (U24) 5′‐TCAGAGATCTTGGTGATAAT; 218 (snR10, 2′‐O‐methyl RNA) 5′‐CUIUUAAAUUUICIUU; 236 (snR39) 5′‐GGTGATAAGTTACGACAGC; 238 (snR41) 5′‐GGGTTGTCGACATGTAGTTA; 242 (U1) 5′‐CACGCCTTCCGCGCCGT; 243 (U4) 5′‐CCGTGCATAAGGAT; 244 (U5) 5′‐AATATGGCAAGCCC; 245 (3′Ex‐U1) 5′‐TGTTCCATTTATTTCTGAAA; 246 (3′Ex‐U4) 5′‐AAAGAATGAATATCGGTAATG; 247 (3′Ex‐U5S) 5′‐GAGAAAAAGGGCAGAAAAG; 248 (3′Ex‐U5L) 5′‐TAGAAAAGATAAACGCCCT; 249 (U4DS) 5′‐GACACACAAGAAGGAGAACACTC; 250 (SCR1) 5′‐AAGGACCCAGAACTACCTTG.
RNase H treatment
Deadenylation was performed essentially as described (Decker and Parker, 1993). Samples of 30 μg of RNA were annealed with 750 ng oligo(dT) at 65°C for 1 h and digested with 6 U RNase H at 30°C for 1 h. The control samples were treated identically, except that the oligo(dT) was omitted.
In vitro processing reactions
Synthetic U4‐3′ RNA was obtained by in vitro transcription as described (Chanfreau et al., 1998b) using PCR product as template. PCR product was generated from genomic DNA using a forward primer carrying a T7 promoter (T7U4DS, 5′‐GCGAATTCTAATACGACTCACTATAGGAAGTAATATCAAAAAATAGG) and a reverse primer U4DS.
Whole‐cell extracts were prepared from wild‐type and rnt1Δ sister strains as described (Chanfreau et al., 1998b).
Recombinant His6–Rnt1p was prepared by cloning a PCR‐amplified RNT1 gene into pET16B (Novagen), using NdeI and BamHI restriction sites added into the primers (NdeI–RntI, 5′‐GGGAATTCCATATGGGCTCAAAAGTAGCAGG; Bam–RntI, 3′‐CGGGATCCTCAGCTTGTATCTGAGAATTTTCTTTTCTTATTC). Expression of His6–Rnt1p in E.coli strain BL21 was induced by addition of isopropyl‐β‐d‐galactopyranoside at OD0.5 (final concentration 0.5 mM). After 3 h of expression at 30°C, cells were harvested and pellets were kept at −80°C. Pellets were resuspended in 40 ml of Start buffer (20 mM sodium phosphate pH 7.0, 10 mM imidazole) and cells were further lysed by passing through a French press. Cell debris were pelleted and the supernatant was loaded into a Hi‐Trap Chelating column (Pharmacia) pre‐equilibrated with the Start buffer. The column was washed with 10 ml Start buffer and proteins were eluted with sodium phosphate buffer with increasing imidazole concentration (20, 40, 60, 100, 300 and 500 mM) or a linear gradient of imidazole (10–500 mM). Peak fractions were pooled and the protein was dialysed twice against the storage buffer (50% glycerol, 50 mM Tris–HCl pH 7.6, 200 mM KCl, 0.5 mM dithiothreitol, 0.5 mM EDTA pH 8.5). The protein was stored at a concentration of 2 mg/ml at −20°C and remained active for several months after storage.
In vitro processing of U4‐3′ RNA in cell extracts or with recombinant His6–Rnt1p, and mapping of the cleavage sites using primer extension, was performed as described (Chanfreau et al., 1998a,b). Prior to the reaction, gel‐purified RNA substrates (2 nM) were denatured for 2 min at 85°C in the Rnt1p buffer (50 mM Tris–HCl pH 7.6, 200 mM KCl, 0.1 mg/ml wheat‐germ tRNA, 5 mM MgCl2) and cooled to 23°C. The cleavage reaction was performed either at 23°C using from 50 to 200 fmol of recombinant His6–Rnt1 or by incubation in the whole‐cell extracts.
J.K. was the recipient of a long‐term EMBO fellowship. This work was supported by the Wellcome Trust.
- Copyright © 1999 European Molecular Biology Organization