Nuclear fate of yeast snoRNA is determined by co-transcriptional Rnt1 cleavage

Grzechnik, Pawel; Szczepaniak, Sylwia A.; Dhir, Somdutta; Pastucha, Anna; Parslow, Hannah; Matuszek, Zaneta; Mischo, Hannah E.; Kufel, Joanna; Proudfoot, Nicholas J.

doi:10.1038/s41467-018-04094-y

Download PDF

Article
Open access
Published: 03 May 2018

Nuclear fate of yeast snoRNA is determined by co-transcriptional Rnt1 cleavage

Pawel Grzechnik^1,2^na1,
Sylwia A. Szczepaniak^3,4^na1,
Somdutta Dhir¹,
Anna Pastucha³,
Hannah Parslow²,
Zaneta Matuszek³,
Hannah E. Mischo¹,
Joanna Kufel³ &
…
Nicholas J. Proudfoot¹

Nature Communications volume 9, Article number: 1783 (2018) Cite this article

4844 Accesses
21 Citations
6 Altmetric
Metrics details

Subjects

Abstract

Small nucleolar RNA (snoRNA) are conserved and essential non-coding RNA that are transcribed by RNA Polymerase II (Pol II). Two snoRNA classes, formerly distinguished by their structure and ribonucleoprotein composition, act as guide RNA to target RNA such as ribosomal RNA, and thereby introduce specific modifications. We have studied the 5ʹend processing of individually transcribed snoRNA in S. cerevisiae to define their role in snoRNA biogenesis and functionality. Here we show that pre-snoRNA processing by the endonuclease Rnt1 occurs co-transcriptionally with removal of the m⁷G cap facilitating the formation of box C/D snoRNA. Failure of this process causes aberrant 3ʹend processing and mislocalization of snoRNA to the cytoplasm. Consequently, Rnt1-dependent 5ʹend processing of box C/D snoRNA is critical for snoRNA-dependent methylation of ribosomal RNA. Our results reveal that the 5ʹend processing of box C/D snoRNA defines their distinct pathway of maturation.

Improving prime editing with an endogenous small RNA-binding protein

Article Open access 03 April 2024

Mechanism of single-stranded DNA annealing by RAD52–RPA complex

Article Open access 24 April 2024

DNA double-strand break–capturing nuclear envelope tubules drive DNA repair

Article 17 April 2024

Introduction

Small nucleolar RNA (snoRNA) are classified based on conserved RNA sequences as either box C/D or H/ACA snoRNA¹. These sequences form specific secondary structures which are associated with class-specific proteins. Small ribonucleoprotein complexes (snoRNP) are essential for many cellular processes including RNA processing, mRNA splicing, chromatin maintenance and RNA editing^2,3. However, their predominant function is in ribosome biogenesis where they guide ribosomal RNA (rRNA) 2ʹ-O-methylation (box C/D snoRNA) and pseudouridinylation (box H/ACA snoRNA)².

Genomic organization of snoRNA genes differs between organisms. In higher eukaryotes they exist as polycistronic transcription units (TUs) or are located within the introns of protein-coding genes. Few are expressed as single independent genes^4,5,6,7. In Saccharomyces cerevisiae (S.cerevisiae) they are predominantly independent TUs with only a minority as polycistrons or within pre-mRNA introns⁸. All yeast snoRNA, except Pol III-dependent SNR52, are Pol II transcribed with their transcription termination mediated by the combined action of Nrd1-Nab3-Sen1 (NNS) and cleavage and polyadenylation complexes^9,10,11,12. This is followed by 3ʹend processing with rounds of oligoadenylation and exonucleolytic digestion performed by TRAMP4/5, the core nuclear exosome and associated exonuclease Rrp6^13,14.

Like 3ʹend processing, snoRNA 5ʹend maturation in yeast is distinct from mRNA processing since mature snoRNA lack the mRNA-specific m⁷G cap (Fig. 1a). Most box C/D pre-snoRNA are synthesized with a 5ʹcapped extension forming a stem-loop with the AGNN loop consensus sequence. The endonuclease Rnt1 recognizes and cleaves this stem-loop at specific Rnt1 cleavage sites (RCS)^15,16. The remaining, uncapped 5ʹextensions are further processed by Rat1 and Xrn1 exonucleases. In some instances 5ʹend maturation may occur independently of Rnt1 cleavage, by exonucleolytic trimming¹⁶. All snoRNA clusters are box C/D class and excised from their TU by Rnt1, thereby removing the m⁷G cap structure from the first snoRNA^15,17,18,19. Only four box C/D snoRNA are transcribed without a 5ʹend extension like most box H/ACA snoRNA. Here the 5ʹend cap structure remains on the mature snoRNA, but is converted into trimethylated m^2,7,7G (TMG) cap by Tgs1 to distinguish it from mRNA caps (Fig. 1a)^15,20.

In S. cerevisiae physical interaction between the Cap Binding Complex (CBC) and NNS has linked 5ʹcap structure with 3ʹend processing²¹. Similarly, in human cells, CBC collaborates with the Nrd1 orthologue ARS2 in transcriptional termination and 3ʹend processing of small nuclear RNA (snRNA)^22,23. This suggests that in both eukaryotes 5ʹ and 3ʹend processing may be functionally interdependent for capped small non-coding RNA. We have set out to test if 5ʹend processing of Rnt1-dependent (RD) snoRNA (mainly box C/D) is also coupled to its 3ʹend processing. This was an attractive hypothesis as Rnt1 has been also reported to interact with the NNS complex²⁴. We show that removal of the mRNA-like 5ʹend cap structure is required for RD box C/D snoRNP maturation and function. Furthermore, Rnt1 co-transcriptionally cleaves the 5ʹend of snoRNA precursors removing the 5ʹcapped extension. This facilitates further steps in box C/D snoRNA maturation. Notably, lack of snoRNA 5ʹend processing affects multiple downstream events, including 3ʹsnoRNA trimming and nuclear localization, and ultimately inhibits rRNA methylation. This connection between maturation and functionality may explain why intron-encoded snoRNA, lacking independent transcription start sites, have evolved in higher eukaryotes.

Results

Co-transcriptional 5ʹend processing of snoRNA precursors

The mature 5ʹends of both box H/ACA and C/D snoRNA are generated either by Rnt1-dependent cleavage followed by exonucleolytic trimming or by m⁷G cap trimethylation (Fig. 1a)^15,16. To systematically catalogue snoRNA 5ʹend processing, we employed RNA-seq analysis to detect 5ʹextended pre-snoRNA in the rnt1Δ strain. We identified three additional box H/ACA (snR81, 83, 85) and one box C/D snoRNA (snR87) with 5ʹextensions that form Rnt1 cleavage structures (Supplementary Fig. 1a). Combining this and previously published data^{15,16,17,18,19}, we show that each snoRNA class generally employs a different 5ʹend processing pathway (Fig. 1b and Supplementary Table 1). The 5ʹends of box C/D snoRNA are generated by Rnt1 dependent (RD) cleavage with only four snoRNA out of 24 independent and five polycistronic genes retaining caps. In contrast, 21 box H/ACA snoRNA have capped 5ʹends with only six cleaved by Rnt1 at their 5ʹends. These results indicate that cap and associated proteins play different roles in the biogenesis of each snoRNA class.

In human cells, CBC interacts with transcription termination factors and is known to be involved in 3ʹend processing of small nuclear RNA (snRNA)^22,23. Similarly, both yeast CBC and Rnt1 interact with the NNS termination complex^21,24. We therefore established the profiles of CBC and Rnt1 across all snoRNA genes in S. cerevisiae using ChIP-seq analysis. Yeast strains were generated with endogenous myc-tagged RNT1 and CBC20 genes. Since Cbp20 and Cbp80 form a heterodimer²⁵, the presence of Cbp20 is indicative of the whole CBC complex. Notably, CBC is recruited across the entire TU of all box C/D and H/ACA snoRNA (Fig. 1c,d and Supplementary Fig. 1b), as with protein-coding genes²⁶. Instead, Rnt1 is more enriched towards the 3’ends of RD snoRNA TUs (Fig. 1c,d and Supplementary Fig. 1b), downstream of Nrd1 binding sites (NBS)²⁷. Differential Cbp20 and Rnt1 co-transcriptional binding profiles are especially evident for longer polycistronic snoRNA TUs (Fig. 1d and Supplementary Fig. 1b). Thus, CBC peaked in the proximal parts of polycistrons while Rnt1 accumulated over transcription termination regions, even though RCS are proximal to transcription start sites (TSS). 3ʹend-specific Rnt1 recruitment may rely on interaction with snoRNA associated proteins and NNS^24,28,29 since it is also detectible on Rnt1-independent box C/D and H/ACA snoRNA as well as over NNS-dependent protein coding genes (Fig. 1e and Supplementary Fig. 1c,d). Rnt1 was also present over many protein-coding genes (see database), suggesting broader functions for Rnt1^30,31. In contrast, CBC did not show a recruitment bias towards gene 3ʹends and NNS binding sites (Fig. 1c,d and Supplementary Fig. 1b). While both CBC and Rnt1 interact with NNS, RNase treatment reduced their ChIP signals over Rnt1-dependent SNR47 (Fig. 1f). Possibly CBC and Rnt1 interact with NNS containing RNA, which is displaced from Pol II CTD prior to transcriptional termination.

We note that CBC signals decreased over the 3ʹends of RD snoRNA genes and especially the much longer polycistronic TUs (Fig. 1d) but not for Rnt1-independent snoRNA, such as 606 nt long box H/ACA SNR30 and 194 nt long box C/D SNR4 (Supplementary Fig. 1e). Since Tgs1 is localised in the nucleolus²⁰ and cap trimethylation most likely occurs post-transcriptionally, the decrease in Cbp20 signal over RD snoRNA gene 3ʹends suggests that Rnt1 cleaves pre-snoRNA co-transcriptionally, thereby removing cap associated CBC. To further investigate Rnt1 co-transcriptional cleavage we analysed available NET-seq data³². Although NET-seq was primarily developed to map nascent RNA 3ʹends in the Pol II active centre, it also detects RNA 3ʹends generated by co-transcriptional cleavage within protein complexes associated with Pol II^32,33. Notably, NET-seq displayed pronounced signals corresponding to known or predicted Rnt1 cleavage sites for both box C/D and H/ACA snoRNA (Fig. 1g and Supplementary Fig. 1f). NET-seq peaks co-localized with 27 out of 29 RCS present in snoRNA 5ʹextensions (24 independent and 5 polycistronic snoRNA genes) (Fig. 1h and Supplementary Table 2). Meta-analysis of snoRNA 5ʹends placed these RCS within 12–18 nucleotides of AGNN hairpin loops (Fig. 1i). No NET-seq peaks were detected upstream of snoRNA transcribed without 5ʹend extensions (Supplementary Fig. 1g). We predict that Rnt1 recruited to RD snoRNA 3ʹends remove capped 5ʹends from pre-snoRNA containing specific RCS. This implies cross-talk between both ends of the transcription unit.

Influence of cap and associated CBC on snoRNA expression levels

To investigate the requirement of cap and associated CBC for snoRNA maturation, a temperature-sensitive ceg1-63 mutant was employed to inactivate the essential guanyl transferase, Ceg1^34,35. Promoters of endogenous snR13 (C/D) and snR3 (H/ACA), were replaced by the inducible GAL1 promoter in WT and ceg1-63 strains. Note that both snoRNA lack RCS, to avoid unprotected 5ʹend degradation. Galactose induction of box C/D SNR13 in WT at 37 °C accumulated mature (M) and a decapped, 5ʹtruncated (Mt)³⁶ snR13 as well as previously described 3ʹextended processing intermediates¹⁴: shorter oligoadenylated substrates (Me) for the Rrp6 exonuclease and longer polyadenylated precursors (Pa) processed by both Rrp6 and the exosome (Fig. 2a)¹⁴. The basal activity of the GAL1 promoter yielded Mt under all conditions. Its function is unknown³⁶. GAL1 induced SNR13 in ceg1-63 cells at non-permissive temperature (37 °C) revealed a similar accumulation of mature snR13 transcripts over time (Fig. 2a,d), but reduced levels of 3ʹextended pre-snR13 precursors (Me)¹⁴. This indicates an altered Rrp6-dependent 3ʹend processing phenotype in this strain. Notably, RRP6 deletion restored the accumulation of Me snR13 precursors in ceg1-63 (Supplementary Fig. 2a). Also, precursors were shortened in rrp6Δ ceg1-63, suggesting that uncapped precursors are more sensitive to 3ʹ–5ʹexonucleolytic digestion.

In contrast to snR13, GAL1 induced box H/ACA snR3 synthesis was severely restricted by CEG1 mutation at 37 °C (Fig. 2b,d). Compared to WT, total snR3 levels expressed from the GAL1 promoter were substantially lower in ceg1-63. SnR3 precursors were mapped by RNase H treatment followed by Northern blot analysis, confirming that the Me species does not originate from an alternative TSS (Supplementary Fig. 2b). Thus, box H/ACA snR3 behaves like a mRNA, such as GAL1, requiring a cap for transcript stability (Fig. 2c,d)³⁵. We also confirmed that at permissive temperature in the ceg1-63 mutant, galactose-induced snR13 was unaffected, while snR3 levels were slightly decreased over 1 h (Supplementary Fig. 2c).

Since SNR13 and SNR3 lack RCS, we also tested RD box C/D SNR65 and SNR68 as well as RD box H/ACA SNR43 and SNR46 transcribed from GAL1 promoters in WT and ceg1-63 cells. GAL1 promoter insertion excluded RCS to prevent possible co-transcriptional degradation from the unprotected 5ʹend. The accumulation (after 60 min induction) of mature snR65 was low but comparable between WT and ceg1-63 cells (1.5× and 1.78× fold respectively) (Fig. 2e), Moreover, long pre-snR65 were absent and processing intermediates were detected in ceg1-63. Transcription of SNR68 from GAL1 gave very weak accumulation of long 3ʹextended pre-snR68 (Supplementary Fig. 2d), unsuitable for further analysis. In contrast, GAL1 promoted box H/ACA SNR43 and SNR46 levels were strongly activated by galactose but this was inhibited by CEG1 mutation. Accumulation of all snR43 and snR46 species in ceg1-63 was respectively 5.5× and 4.8× fold lower than in WT (Fig. 2e and Supplementary Fig. 2d). These results indicate that 5’extensions and m⁷G cap play differential roles in RD box C/D and H/ACA snoRNA biosynthesis.

We next tested if yeast CBC is required for snoRNA 3ʹend processing. RNA-seq analysis detected no unprocessed box C/D or H/ACA snoRNA in the cbp20Δ cbp80Δ mutant (Fig. 2f). Increased snoRNA levels observed in the double mutant (Supplementary Fig. 2e) indicate that CBC influences snoRNA abundance by the regulation of exonucleolytic nuclear degradation³⁷. Consistently, ChIP analysis employing anti-Pol II antibody showed unchanged levels of Pol II between WT and cbp80Δ cells (Supplementary Fig. 2f).

Overall, we conclude that the synthesis of box H/ACA snoRNA requires m⁷G cap. In contrast, this structure appears dispensable for the expression of box C/D snoRNA, but may still affect some steps in their 3ʹend maturation. The reduction of snR13 and snR65 3ʹextended precursors in ceg1-63 suggests that cap removal from box C/D pre-snoRNA accelerates 3ʹprocessing.

Rnt1 cleavage controls 3ʹends of RD box C/D snoRNA

Our data above suggest that RD box C/D snoRNA have their 5ʹcaps removed co-transcriptionally by Rnt1. Furthermore, cap presence is not required for the expression of box C/D snoR13 and snR65 though it may affect 3ʹend maturation. We tested by RNA-seq analysis if 5ʹend processing regulates 3ʹprocessing differently between WT and rnt1Δ strain where pre-snoRNA 5ʹends remain uncleaved. Notably, boxC/D snoRNA transcribed with 5ʹextensions as well as the last snoRNA from polycistronic TUs have short unprocessed extensions at their 3ʹends in rnt1Δ cells (Fig. 3a, Supplementary Fig. 3a and Supplementary Table 3). In contrast, 3ʹends of box C/D snoRNA transcribed without 5ʹextension and all box H/ACA snoRNA (including snoRNA transcribed with 5ʹextensions) were unaffected by RNT1 deletion (Fig. 3b and Supplementary Fig. 3a). This RNA-seq analysis was confirmed by Northern blot analysis (Supplementary Fig. 3b). To compare 3ʹends in WT and rnt1Δ, we generated homogenous 5ʹends by prior RNase H digestion with oligonucleotides complementary to mature snoRNA 5ʹends. These data also show that in the rnt1Δ strain, although 5ʹextended pre-snoRNA are dominant, some 5ʹprocessed snoRNA are still generated (Supplementary Fig. 3b) by an alternative pathway (Fig. 1a).

The proportion of transcripts unprocessed at their 3ʹend in rnt1Δ was measured by circular RNA RT-PCR (CR-RT-PCR) technique for box C/D snR68 and snR65 to simultaneously sequence the 5ʹ and 3ʹends of the same RNA molecule. This required snoRNA cap removal by RNase H digestion. PCR primers directed against the snoRNA mature sequence allowed detection of both pre- and mature snoRNA (Fig. 3c). CR-RT-PCR on RNA isolated from rnt1Δ showed that most 5ʹextended box C/D snR68 and snR65 possessed short, oligonucleotide extensions with an oligo(A) tail. In contrast, snoRNA with mature 5ʹends (presumably processed by the Rnt1-independent alternative pathway) have normally processed 3'ends. This analysis reveals that removal of RD snoRNA 5ʹextensions is critical for their 3ʹend trimming. Also, these RNA are oligoadenylated, implicating TRAMP4/5-activated Rrp6- and exosome-dependent digestion³⁸. However, accumulation of oligo(A) tails in the rnt1Δ strain suggests that pre-snoRNA, which have not been processed at their 5ʹends, are stalled in 3ʹ–5ʹexonucleolytic processing.

We finally performed CR-RT-PCR on RNA from the WT strain using oligonucleotides located over the SNR68 coding sequence and 5ʹextension to focus the analysis on the 3ʹends of unstable 5ʹextended pre-snoRNA (Fig. 3d). All sequenced clones had 3ʹextension from 2 to ~100 nucleotides, giving a range of precursors that extend to the transcription termination sites. No pre-snR68 with fully processed 3ʹends were detected, indicating that snoRNA 5ʹend processing precedes 3ʹend processing. CR-RT-PCR with the same primer set but for the rnt1Δ strain detected only short (up to 10 nt) oligoadenylated 3ʹextensions (Supplementary Fig. 3c).

Since most box H/ACA snoRNA retain their cap structure, which is later trimethylated by Tgs1^15,20, we also tested if lack of Tgs1 affects 3ʹend formation of snoRNA. However, no 3ʹend processing defects were detected in the tgs1Δ mutant (Fig. 3e). TGS1 loss did not enhance box C/D-specific 3ʹend processing defects as observed in the rnt1Δ strain (Fig. 3f). Overall, a clear interplay between Rnt1-dependent 5ʹend processing and 3ʹend exosome-dependent processing for RD box C/D snoRNA is evident.

Cap retention impairs box C/D snoRNA processing

We wished to determine whether the 3ʹend processing defect in the rnt1Δ strain is caused by inactivity or absence of Rnt1. SnR68Δ strain was transformed with plasmids expressing either WT or a snR68 stem-loop mutant that is insensitive to Rnt1 processing³⁹. Northern blot analysis of RNase H cleaved snR68 showed that the stem-loop mutation results in the same 3ʹend trimming defect as observed in the rnt1Δ mutant (Fig. 4a and Supplementary Fig. 3b). snR68 processing in the rnt1-E320K catalytic mutant⁴⁰ generated aberrantly processed 3ʹends, like the rnt1Δ strain (Supplementary Fig. 4a). These results show that RD box C/D snoRNA 3ʹend processing requires removal of 5ʹend extensions. We next investigated if unprocessed stem-loops in the 5ʹextension somehow interfere with 3ʹend exonucleolytic processing. We therefore generated a strain with the GAL1 promoter and its 5ʹUTR integrated upstream of box C/D SNR13 (GAL1U::SNR13 strain) (Fig. 4b). GAL1 transcription formed a hybrid GAL1 UTR-snR13 RNA (snR13e). In a second strain, box H/ACA SNR3 was placed between the GAL1 UTR and SNR13 (GAL1U::SNR3::SNR13). Processing of these two artificially extended snoRNA genes was compared to the unmodified snR13 also expressed from a GAL1 promoter, lacking the UTR sequence (GAL1::SNR13) (Fig. 4b). Northern blot analysis revealed that the presence of GAL1 UTR at the 5ʹend of snR13e inhibited 3ʹend trimming, while snR13 synthesized without the 5ʹextension was normally processed. As before, we generated homogenous 5ʹends by specific RNase H digestion (Fig. 4b). Surprisingly, 5ʹextended hybrid box H/ACA snR3- C/D snR13e was also correctly 3ʹend trimmed. This indicates that the presence of box H/ACA snR3 enables correct processing of GAL1 UTR-snR3-snR13 dicistronic snoRNA by overriding the adverse impact of a capped 5ʹextension. Consistently, SNR3 transcribed with the GAL1 UTR did not display processing defects compared to WT (Fig. 4c and Supplementary Fig. 4c). It appears that the mere presence of a capped 5ʹextension affects Rrp6-dependent 3ʹprocessing of box C/D, but not H/ACA snoRNA. This is consistent with the rnt1Δ strain data where box H/ACA snoRNA 3ʹend processing is unaffected by the presence of 5ʹextensions (Fig. 3b).

We next tested if pre-snoRNA cap and CBC disturb the 3ʹend formation of box C/D snoRNA. Northern blot and RNA-seq analyses show that snR52, transcribed as an uncapped 5ʹextended precursor by Pol III, is normally processed at the 3ʹend in the rnt1Δ strain (Fig. 4d and Supplementary Fig. 4d,e). Also, snR52 transcribed from the Pol II-dependent GAL1 promoter was correctly trimmed at the 3ʹend when its transcription start was immediately followed by the snoRNA mature sequence. However, artificial extension of snR52 with the GAL1 UTR as in our modified snR13 constructs (Fig. 4b) resulted in accumulation of 3ʹextended RNA (Fig. 4d). Here, 3ʹunprocessed snR52 were longer and more heterogeneous than with other C/D snoRNA. Possibly Pol III-dependent SNR52 does not possess a typical NNS terminator, so that 3ʹextended snR52 species represent read-through transcripts.

To test if lack of m⁷G cap synthesis restores 3ʹend processing of snR13e, GAL1U::SNR13 transcription was tested in the ceg1-63 strain to impair m⁷G cap synthesis³⁵. Induced levels of snR13e in ceg1-63 decreased 3-fold compared to WT (Fig. 4e). Possibly exonucleolytic degradation over the GAL1 UTR compromised snoRNP formation, resulting in enhanced degradation of snR13e. Read quantitation over the mature and 3ʹuntrimmed snR13e show that in WT, 3ʹunprocessed snR13e was more abundant than the mature fraction, while in ceg1-63 this effect was reversed (Fig. 4e). This suggests that a retained cap structure causes the snR13e 3ʹend processing defect. Reduction in the 3ʹuntrimmed species was not observed during snR13e induction in the cbp80Δ strain, where CBC functions are compromised (Fig. 4f). Consistently, RNA-seq analysis of rnt1Δ/cbp80Δ did not reveal enhanced 3ʹend processing of RD box C/D snoRNA (Fig. 4g). Overall, we show that m⁷G cap, but not associated CBC, present on the box C/D snoRNA precursor interferes with the final step of 3ʹend processing.

5ʹend processing controls snoRNA cellular fate

SnoRNA 3ʹend processing is mediated by the combined exonucleolytic activities of the exosome core and Rrp6^13,14. Our results indicate that snoRNA capped 5ʹextensions block snoRNA 3ʹend processing. Possibly 5ʹunprocessed RD box C/D snoRNA are exported to the cytoplasm (mimicking mRNA), thereby preventing nuclear Rrp6-dependent trimming. To determine the cellular localization of mature and 5ʹextended box C/D and H/ACA snoRNA in rnt1Δ, we employed fluorescence in situ hybridization (FISH) analysis using fluorescent oligonucleotide probes complementary to mature or 5ʹextended snoRNA sequence. A probe targeting box C/D snR68 5ʹextension showed that pre-snR68 was distributed across the whole rnt1Δ cell (Fig. 5a). Hybridization with a probe against snR68 mature sequence also showed some cytoplasmic mislocalization of snR68. This indicates that 5ʹextended box C/D pre-snoRNA are exported from the nucleus in the rnt1Δ strain. The control FISH analysis in WT cells revealed that pre- and mature snR68 are localised in the nucleus (Fig. 5a). Note that the probe targeting mature snR68 showed no off-target specificity (Supplementary Fig. 5a). In contrast, precursor and mature RD box H/ACA snR43 and Rnt1-independent box C/D snR13 were exclusively nuclear in rnt1Δ cells (Fig. 5b,c). Moreover, box C/D-associated protein Nop1 was normally localized in rnt1Δ (Fig. 5d). Note that Rnt1-dependent snoRNA did not display strict nucleolar localization, reflecting the complex maturation of these snoRNA.

We also performed FISH analysis for snR13 in strains where SNR13 was expressed from the different GAL1 promoter variants (Fig. 5e and Fig. 4b). In the WT snR13 was localized as nuclear foci. Although snR13 transcribed from the GAL1 promoter was not restricted to the nucleolus, it was present exclusively in the nucleus. However, the 5’extended snR13e generated in the GAL1U::SNR13 strain was localized to both nucleus and cytoplasm. Consistent with previous results, the presence of box H/ACA snR3 upstream of snR13 supressed this effect, as snR3-snR13e hybrid was detected only in the nucleus. These reporter construct data reinforce and confirm FISH results for rnt1Δ showing that 5ʹend processing regulates nuclear localization of box C/D snoRNA.

Our cell imaging results demonstrate that 5ʹend processing of RD box C/D snoRNA precursors is critical for nuclear localization. In effect, Rnt1-dependent cleavage signals the cellular fate of snoRNA versus mRNA.

5ʹextended box C/D snoRNA precursors are non-functional

Efficient expression of box C/D snoRNA is required to achieve full methylation of rRNA. Consequently, if 5ʹextended box C/D snoRNA is mislocalized away from the nucleolus, this should impede rRNA synthesis. We therefore tested if disruption of snoRNA 5ʹend processing affects rRNA 2'-O-methylation in rnt1Δ strain. To determine the methylation levels of three clusters of methylated nucleotides in 25S rRNA (Fig. 6a), we employed a quantitative RT-PCR approach using low nucleotide concentrations for reverse transcription⁴¹ (RTLN-qP) (Fig. 6b). Low nucleotide concentration causes RT blockage at the sites of RNA methylation⁴¹. Thus, levels of cDNA synthesised at low dNTP concentration over a particular RNA position, compared to normal concentration, indirectly measure RNA methylation. In qPCR analyses, the number of PCR cycles (threshold cycle; Ct) required to detect DNA effectively quantitates cDNA levels. In WT, methylation of 25S rRNA delayed Ct from region 2 and 3 by ~2 cycles and for region 4 by almost 7 cycles. In contrast, in rnt1Δ Ct for RT-LN products were delayed by only 0.36 cycle for region 2 and 3 and by 1.7 cycle for region 4 (Fig. 6c). Overall, the amounts of cDNA synthesized over methylated nucleotide clusters in low dNTP conditions in rnt1Δ were approximately 4–16 times higher than in the WT, indicative of inefficient 25S rRNA methylation in the rnt1Δ strain. RTLN-qP analysis of 25S rRNA in rrp6Δ cells indicate that the lack of 3ʹend trimming itself has no effect on box C/D snoRNP enzymatic activity (Supplementary Fig. 6a).

We also tested site-specific 2'-O-methylation levels of 25S and 18S rRNA in the rnt1Δ strain using the 8–17 and 10–23 DNAzyme-dependent approaches⁴². DNAzymes are single stranded DNA which form a stem-loop structure around target RNA, cleaving RNA complementary to the junction of the loop arms, either downstream of guanine (8–17 DNAzyme) or between the purine and pyrimidine nucleotides (10–23 DNAzyme). Methylation of the nucleotide located downstream of the cleavage site strongly inhibits DNAzyme activity (Fig. 6d)^42,43. RNA isolated from rnt1Δ strain revealed that DNAzymes targeting snR56- and snR79-site specific methylation in 18S rRNA as well as snR68-site specific methylation in 25S rRNA (Supplementary Fig. 6b) partially cleaved rRNA, generating cleavage products cA and cB (Fig. 6e). RNA from the WT strain was resistant to DNAzyme-dependent cleavage, implying complete methylation of rRNA. To exclude a general defect in rRNA processing we analysed the snR13-specific methylation site of 25S rRNA in the rnt1Δ strain. No cleavage products were detected for RNA either in WT or rnt1Δ strains, indicating that Rnt1-independent snR13 is unaffected in the rnt1Δ strain (Fig. 6e). Overall, these data show that lack of 5ʹend processing affects RD box C/D snoRNA function. Note that rRNA in the rnt1Δ strain is still methylated to some extent as pre-snoRNA can be inefficiently processed at its 5ʹend by the alternative pathway. Consistently, endonucleolytic maturation of pre-rRNA mediated by U14 snoRNA (snR128)⁴⁴, which is processed from pre-snR190-snR128 dicistronic transcript¹⁷, is kinetically delayed in rnt1Δ⁴⁵. Moreover, our RNA-seq analysis shows that U14 levels are much higher when compared to other snoRNA in both WT and rnt1Δ strains (Supplementary Fig. 6c). We predict that U14 is processed by an alternative pathway in rnt1Δ cells, so it can separately, or as a 5ʹtruncated snR190-snR128 dicistron, process pre-rRNA.

We also tested if site-specific snR13-dependent methylation of 25S rRNA was affected in the strain expressing 5ʹextended snR13e (Fig. 6f). SnR13 guides methylation of two adjacent adenines at positions 2280 and 2281 in 25S rRNA⁸. Since the nucleotide following the second adenine is a pyrimidine (U), we analysed snR13-mediated methylation at this position using the 10–23 DNAzyme (Supplementary Fig. 6d)⁴². SnR13-dependent rRNA methylation in GAL1::SNR13 strains occurs at the same level as in WT when grown in galactose-containing medium (Supplementary Fig. 6e), so reflecting physiological conditions. The dynamics of snR13-depedendent methylation of 25S rRNA during transcriptional induction of normal snR13 and 5ʹextended snR13e was analysed by Northern blot of RNA samples collected at different time points, following incubation with the DNAzyme targeting snR13-dependent methylation site (Fig. 6f). All 25S rRNA species were detectible by RNA staining on the membrane, while hybridization with a radioactive probe complementary to the region downstream of the cleavage site, detected only full-length 25S and cB product. Prior to SNR13 induction, in both strains 25S rRNA was almost completely digested by the DNAzyme targeting the snR13-dependent site, indicating low levels of snR13-dependent rRNA methylation. During transcriptional induction of SNR13, 25S rRNA became gradually resistant to the DNAzyme activity, reflecting methylation of newly synthesized 25S rRNA. This process was clearly quantitatively slower in the strain expressing 5'extended snR13e (GAL1U::SNR13) (Fig. 6f). Analysis of snR13 and snR13e levels shows that snR13e did not accumulate as much as WT snR13 in the later time points (0.8× fold lower after 10 and 12 h of induction) (Supplementary Fig. 6f). However, 25S rRNA became equally resistant to DNAzyme cleavage after about 12 h, which suggests sufficient accumulation of functional snR13e. These data demonstrate that snR13e expression induces a specific rRNA methylation defect. We can infer that RD box C/D snoRNA requires 5ʹend processing to be fully functional.

Discussion

Pol II generates a wide range of ncRNA which display different cellular fates to mRNA. Mechanisms, which differentiate their RNA metabolism, are crucial to direct their correct maturation pathways. Distinguishing mRNA and ncRNA is often achieved by distinct transcription termination pathway leading either to RNA stabilization or degradation. We describe an additional mechanism for the snoRNA of S. cerevisiae which involves their co-transcriptional cleavage by Rnt1. While m⁷G cap is an essential element for mRNA, it is removed from Rnt1 dependent (RD) box C/D pre-snoRNA to facilitate their independent maturation and function.

NNS-dependent transcription termination releases Pol II from non-coding TUs and also recruits enzymatic activities required for their 3ʹend processing such as the TRAMP complex, which oligoadenylates RNA to stimulate degradation by the nuclear exosome^21,38. NNS also interacts with the Rnt1 endonuclease²⁴, implying a further role in Rnt1 recruitment. We show that Rnt1 is co-transcriptionally recruited over snoRNA 3ʹends together with NNS, even though it acts at the 5ʹends of these transcripts (Fig. 1c–e). The detection of RNA 3ʹends generated by Rnt1 over RCS while associated with transcribing Pol II (Fig. 1g) indicates that Rnt1 cleaves pre-snoRNA co-transcriptionally. Possibly, delayed recruitment of Rnt1 to snoRNA 3ʹends ensures that they remain unprocessed at their 5ʹends until snoRNP assembly occurs. NNS also co-purifies with the CBC²¹. However, CBC is recruited to snoRNA genes at an early stage of transcription (Fig. 1c–e) as occurs for protein-coding genes²⁶. CBC–NNS interactions may play post-transcriptional roles by recruiting the exosome to trigger nuclear RNA degradation³⁷. The pausing of Pol II over NNS-dependent terminators¹² may create a time window prior to RNA release from the DNA template. This allows snoRNP assembly and recruitment of factors required for subsequent maturation. Therefore, snoRNA terminators emerge as processing hubs that not only mediate transcription termination but also define this class of Pol II-dependent transcript as snoRNA.

Our data indicate that the removal of m⁷G cap is pivotal to inform the transcription machinery that the nascent RNA is not mRNA, so committing it to the RD box C/D snoRNA maturation pathway. Consequently, Rnt1-dependent cleavage at the 5ʹends of box C/D pre-snoRNA regulates 3ʹend processing and nuclear retention (Figs 3 and 5). In effect, the presence of cap (or associated proteins) emerges as a checkpoint in RD snoRNA synthesis. We hypothesise that m⁷G may compete with box C/D snoRNP proteins to signal the transcription machinery to direct the transcript to either mRNA or snoRNA maturation pathways (Fig. 7).

In the case of box C/D snoRNA where the TSS is adjacent to the mature snoRNA sequence (snR13, 4, 45 and 17), immediate association of snoRNP proteins may prevent conformational changes in the Pol II complex mediated by cap, which would otherwise classify the nascent transcript as mRNA. Therefore, these snoRNA remain capped at their 5ʹends¹⁵. In all cases the presence of m⁷G had no effect on 3ʹend processing. Consistently, snR13 artificially extended at the 5ʹend with the GAL1 UTR (snR13e) was unprocessed at the 3ʹend while snR13 transcribed from the GAL1 promoter lacking the UTR was normally processed (Fig. 4b). In contrast to most C/D snoRNA, 5ʹcaps are essential for box H/ACA snoRNA synthesis (Fig. 2b,e) and furthermore their 3ʹend processing is unaffected by cap retention (Fig. 3b). This reflects the genomic organization of box H/ACA snoRNA where Rnt1-dependent 5ʹend processing is rare and their 5ʹends are defined by transcription initiation with the cap structure remaining during subsequent processing. However, mature capped box H/ACA and Rnt1-indpendent box C/D snoRNA have an altered cap structure¹⁵ modified by Tgs1, which converts m⁷G cap into trimethylated cap (TMG)²⁰. It is plausible that Tgs1 modifies cap in order to disturb m⁷G-CBC interactions⁴⁶. Finally, our studies on yeast snoRNA synthesis indicate that CBC is not required for this process even though CBC was previously shown to enhance transcription termination and processing of human snRNA^22,23.

It is informative to consider the evolution of snoRNA gene arrangement and prevalence in eukaryotes. Notably, snoRNA numbers increase with organism complexity. S. cerevisiae has 77 while Drosophila melanogaster has 227 snoRNAs. Remarkably humans are estimated to possess 450–700 snoRNA genes ^47,48. Conversely, the fraction of independently transcribed snoRNA decreases with organism complexity, being replaced by either polycistronic TUs or intronic localization. In S. cerevisiae, only 10% of 77 snoRNA are located within the introns of protein coding genes^7,8, while in the human genome most snoRNA are intronic^7,47,48. Such an intronic localization of snoRNA may have evolved to facilitate co-expression with host ribosomal protein coding genes^47,49. However, host genes for snoRNA appear to correlate with expression levels rather than gene ontology. In general, snoRNA that modify abundant rRNA are located in the introns of highly transcribed protein-coding genes, so providing high expression levels⁵⁰. Why evolution has selected against independently transcribed snoRNA genes in higher eukaryotes and led to a loss of dedicated snoRNA promoters remains unknown. We show that the cap structure may be used to distinguish box C/D snoRNA from mRNA and so specify correct maturation pathways. Possibly, this may have contributed to the evolutionary pressure to remove independent transcription initiation sites for many snoRNA. Since m⁷G cap does not affect box H/ACA snoRNA maturation, apparently cap removal may be only one of several ways to shape genomic organization and processing pathways of snoRNA genes in higher organisms.

Methods

Yeast strains construction

Yeast strains used in this work are listed in Supplementary Table 4. The transformation procedure was as described⁵¹ using standard lithium acetate method. Strains were generated by a one-step PCR procedure⁵². To construct strains expressing snoRNA from an inducible GAL1 promoter, the region of the GAL1 promoter was amplified by PCR using the pFA6a-KanMx6-pGAL1 plasmid. The GAL1-snoRNA modules were then further transferred between strains by amplification of the GAL1-snoRNA cassette on the genomic DNA template followed by transformation into yeast.

Yeast growth conditions

Strains were grown at 23 °C or 30 °C in YPD medium (1% yeast extract, 2% Bacto-peptone, 2% glucose) to mid-exponential phase. Strains containing conditional temperature-sensitive alleles were pre-grown at 23 °C, up to mid-exponential phase and transferred to 37 °C. Transcription from GAL1 promoters was induced by addition of 2% galactose to yeast cultures pre-grown in minimal SC (0.67% yeast nitrogen base, supplemented with required amount of amino acids and nucleotide bases) containing 2% raffinose and 0.08% glucose. Each experiment was biologically replicated unless otherwise stated.

RNA methods

Total RNA from yeast cells was isolated using a hot phenol procedure⁵³. Northern hybridization was essentially as described⁵⁴. 8 μg of total RNA was separated on 6% denaturing polyacrylamide-urea gels, electro-transferred (TransBlot Biorad, 100 mA for 45 min) onto nylon membranes (GE Healthcare) and hybridized using PerfectHyb buffer (Sigma) with oligonucleotides labelled with ³²P at their 5ʹends. Overnight hybridization at 42 °C was followed by three washes with 6xSSPE. Hybridization signals were visualized and quantified using FLA5000 imaging system (Fuji). In most cases images were cropped on their edges in Photoshop. Raw, uncropped images from main figures are shown in Supplementary Fig. 7. Oligonucleotide probes used for hybridizations are listed in Supplementary Table 5. RNase H treatment was as described previously¹⁴, briefly 10 mg of RNA in 1 × RNase H buffer was incubated with 1 pmol of oligo for 10 min at 65 °C and cooled down to 30 °C. Next 10 U of RNaseH (NEB) was added and RNA was digested for 1 h. For RNA-seq, 5 μg of total RNA was rRNA-depleted using Ribo-Zero kit from Illumina. The so obtained rRNA-depleted fraction (100–500 ng) was used to prepare libraries employing Ion Total RNA-seq Kit v2 (Thermo Fisher) and subsequently sequenced using the Ion Proton system.

ChIP and ChIP-seq

Chromatin was precipitated as previously described¹⁴. 100 ml of culture (OD₆₀₀ = 0.4–0.8) was crosslinked with 1% (v/v) formaldehyde at room temperature for 20 min and quenched with 375 mM glycine for 5 min. Cells were resuspended in 1 ml of cold FA1-lysis buffer (50 mM HEPES-KOH pH 7.5, 150 mM NaCl, 1 mM EDTA, 1% Triton X-100, 0.1% sodium deoxycholate, protease inhibitors (Complete, Roche)) and disrupted with 300 µl of zirconia beads using a MagnaLyser (three times for 30 s at maximum speed with 5 min rest period on ice between runs). The lysate was diluted with 1 ml of FA1-lysis buffer and sonicated in a Bioruptor sonicator (Diagenode) for 15 min (15 s on, 15 s off) set at medium level. The lysate was clarified by 40 min spin at 16000 × g at 4 °C. 500 µl of the extract was diluted 5 times with FA-1 buffer and 1 ml was incubated overnight at 4 °C with 10 µl anti-Myc (ab9132, Abcam) or 1 µl anti-Rbp3 (1Y26, Neoclone) antibody. Next extract was incubated with a 1:1 mix of 60 µl of Dynabeads Protein G and A (Invitrogen) for 2 h at 4 °C. Beads were washed four times at room temperature with 1 ml of FA1-lysis buffer and once with 1 ml FA2-lysis buffer (50 mM HEPES-KOH pH 7.5, 500 mM NaCl, 1 mM EDTA, 1% Triton X-100, and 0.1% sodium deoxycholate), ChIP wash buffer (10 mM Tris-HCl pH 8.0, 250 mM LiCl, 1 mM EDTA, 0.5% Nonidet P-40, 0.5% sodium deoxycholate) and TE (10 mM Tris-HCl pH 8.0, 1 mM EDTA). Beads were resuspended in 100 µl of ChIP elution buffer (50 mM Tris-HCl pH 7.5, 10 mM EDTA, 1% SDS) and samples, including 20 µl of the input sample, were incubated with 40 µg of Proteinase K (Bioline) for 2 h at 50 °C and 6 h at 65 °C. For RNase treatment the diluted extract was incubated with or without RNase A (10 U, Qiagen) and T1 (500 U, Roche) at 37 °C for 1 h. Extracts were then incubated overnight at 4 °C, then for 2 h with 4 µl anti-Myc (ab9132, Abcam) or 1 µl anti-Rbp3 (1Y26, Neoclone) antibody. Next the extract was incubated with a 1:1 mix of 25 µl of Dynabeads Protein G and A (Invitrogen) for 2 h at 4 °C. Washes were performed as above. Beads resuspended in elution buffer with Proteinase K were incubated 2 h at 56 °C and 12 h at 65 °C. DNA was purified using the commercial clean-up kit (Qiagen). For ChIP-seq analysis at least four IPs were pooled and used for subsequent treatment. The libraries for ChIP-seq were prepared using NEBNext ChIP-seq Library Prep (NEB) and sequenced on Illumina HiSeq400 by the High Throughput Genomics Group, University of Oxford. ChIP followed by qPCR analyses are shown as an average of three independent biological replicates.

Bioinformatics methods

S. cerevisiae genome, scaCer3 (April 2011) was downloaded from UCSC (http://hgdownload-test.cse.ucsc.edu/goldenPath/sacCer3). Gene boundaries were obtained from the Saccharomyces Genome Database (SGD, http://www.yeastgenome.org/) for the same version.

Mapping RNA-seq sequencing reads: For Ion Torrent RNA-seq sequencing, single-end reads were mapped to sacCer3 genome using two-step alignments. First, the reads were aligned with TopHat⁵⁵. Second, the resulting unmapped reads from the first step were extracted and aligned with Bowtie2⁵⁶ with --very-sensitive-local and --local options. Uniquely mapped reads with no mismatches and mapping quality ≥30 from both steps were then combined using SAMtools merge⁵⁷. Number of reads mapped to each gene was normalized to its length and total number of genome-aligned reads (RPKM value, Reads Per Kilobase of exon model per Million mapped reads). The bigWig tracks from the resulting normalized samples were visualized in a custom UCSC Genome Browser track data hub, hosting the sacCer3 reference genome.

ChIP-seq data analysis: For Chip-seq, paired-end reads for each sample were mapped to the sacCer3 genome (UCSC, downloaded from http://hgdownload-test.cse.ucsc.edu/goldenPath/sacCer3) using the Bowtie2 alignment software. Uniquely mapped reads with a mapping quality ≥30 were retained for further analysis. Peaks were called using Model-based Analysis, MACS2⁵⁸ for ChIP-Seq with default options. Only those peaks with q-value below 0.05 were retained for further analysis. This resulted in 1238 peaks for Rnt1 and 3801 peaks for Cbp20.

Data visualization: Metagenes showing 3ʹend of Rnt1-dependent box C/D snoRNA and Rnt1-independent box C/D and H/ACA snoRNA was generated by plotting normalized read counts around annotated 3ʹend for sense strand relative to the direction of gene transcription. For box H/ACA snoRNA metagene snR30 and snR35 were discarded due to their very high signals relative to the other box H/ACA snoRNA. Next average reads from rnt1Δ were normalised to the average reads of the last 6 nucleotides from the coding sequence in WT by a factor 1.73. NET-seq data for wildtype yeast BY4741 was downloaded from GSM617027. Normalized read counts were calculated for sense and antisense strands, relative to the direction of gene transcription for a region of 25 bp upstream and downstream of annotated AGNN positions and plotted. SnoRNA used for metagene analysis are listed in Supplementary Table 3.

Circular RNA RT-PCR analysis

If required 5ʹcap was removed by RNAse H treatment in the presence of specific oligonucleotide. Next 10 μg of total RNA was circularized using 30 U of T4 RNA ligase (NEB) in 37 °C for 1 h in total volume of 30 μl. CR-RNA was purified by phenol followed chloroform extraction, precipitated and suspended in 10 μl H₂O. 1 μl of RNA was used for cDNA synthesis (AMV, Promega). Primers used for reverse transcription and PCR are listed in Supplementary Table 5 (primers for RT are marked as “RT”). PCR products were purified and then cloned into pGEM easy vector (Promega). Isolated clones were sequenced using T7 primers.

Fluorescent in situ hybridization

Strains were grown in YPD medium at 25 °C to log phase. Cells in 10 ml of medium were prefixed with 37% formaldehyde (final concentration 4%) for 15 min and harvested by spinning down. Cells were fixed in 5 ml of solution A (4% paraformaldehyde, 0,1 M KPO₄ (pH 6,5), and 5 mM MgCl₂) for 3 h. Next cells were washed twice with solution B (1,2 M sorbitol and 0,1 M KPO₄, pH 6,5), resuspended in 0,5 ml of solution B with 0,05% β-mercaptoethanol and freshly prepared Lyticase (Sigma Aldrich), digestion was performed at 37 °C for 20 min. Cells were washed tree times in ice cold solution B, and then resuspended in 0,3 ml of solution B and stored overnight at 4 °C. Spheroplasts were plated to the wells of white glass slides printed with Epoxy (Thermo Scientific™ Diagnostic Slides) that had been covered with a 0,1% poly-l-lysine-containing solution (Sigma Aldrich). Cells were washed with 70, 90 and 100% ethanol for 5 min respectively. Pre-hybridization was conducted in humid chamber for 2 h at 37 °C in buffer containing 10% dextran sulphate, 0,2% BSA, 2 × SSC), 125 μg Escherichia coli tRNA/ ml, 0.5 mg/ml single—strand DNA denatured (95 °C for 3 min.) and Ribolock 1 U/ul (Thermo Scientific). The same buffer was used for hybridization with probes at final concentration 100 pg/ml. 30 nt long probes (listed in Supplementary Table 5) were labelled at their 5ʹend with fluorescent dye Alexa fluor 488 and Alexa fluor 647 (Sigma Aldrich). Hybridization was performed in humid chamber overnight at 37 °C. After hybridization cells were washed three times for 10 min with 2 × SSC at 37 °C and three times with 1 × SSC at room temperature. Cells were briefly washed with 4 × SSC containing 1% Triton X-100 followed by two more washes with 4 × SSC, each wash lasting for 10 min. Nuclei were stained with 0.1 μg/ml DAPI in 1 × PBS. Slides were then mounted with VECTRASHIELD Mounting Medium (Vector laboratories) and stored at −20 °C. Images were acquired using Carl Zeiss Axio Imager Z2 confocal microscope with 63 × NA 1.4 oil objective and Zen software. 3D datasets were generated by multiple 200 nm z stacks covering the entire cell volume. 2D datasets were obtained in ImageJ by maximum projection function.

rRNA methylation analyses

Reverse Transcription at Low deoxyribonucleoside triphosphate concentrations followed by quantitative polymerase chain reaction (RTLN-qP) was performed as described previously⁴¹ with some modifications. Briefly 1 μg of total DNAse I treated RNA was incubated at 70 °C with 250 ng of random hexamers, chilled on ice and incubated at 25 °C for 5 min. Reverse transcription was performed using 20 U of AMV reverse transcriptase (Promega) in 1 mM (high concentration) or 5 μM (low concentration) of dNTPs at 42 °C for 1 h. This was followed by qPCR analysis using Sensimix master mix (Bioline). Oligonucleotides are listed in Supplementary Table 5.

Calculation of reverse transcription in low nucleotides concentration followed by qPCR values

In the RTLN-qP analysis we calculated how many cycles later the Ct is determined for the product of RT reaction performed in low dNTPs concentration (5 μM) as compared to the Ct for the product of RT in normal dNTPs concentration (1 mM). The Ct delay value (ΔCt) was determined for each cluster of methylated nucleotides in 25S rRNA (regions 2–4, R2–4) and for the region located in 25S 5ʹend, which is unmethylated (R1) (Fig. 6a). As cDNA for all regions was synthesized in the same reaction, we used ΔCt for region 1 to define RT efficiency for a particular reaction. To visualize how RNA methylation suppressed cDNA synthesis in low dNTPs concentration, ΔCt for the non-methylated region 1 was deducted from ΔCt obtained for regions 2–4.

DNAzyme-dependent assay

DNAzyme dependent analysis was performed as described⁴². For 10–23 DNAzyme treatment 5 μg of DNase I treated RNA was combined with 200 pmol of 10–23 DNazyme and 2.5 μl of incubation buffer (4 × concentrated: 24 mM Tris pH 8; 60 mM NaCl) in a final 10 μl volume. After heating at 95 °C for 3 min, the reaction was placed on ice for 5 min. Next 1 μl of Ribolock (Fermentas) was added and reaction was incubation at 25 °C for 10 min. Temperature was then raised to 37 °C and 4 μl of pre-warmed reaction buffer (4 × concentrated: 200 mM Tris pH 8; 600 mM NaCl) and 4 μl of 300 mM MgCl₂ was added. Finally, pre-warmed water was added to a final volume 20 μl. The reaction was continued for 1 hr at 37 °C. For reactions with 8–17 DNAzymes, 5 μg of DNAse I treated RNA and 400 pmol of DNAzyme (each in a volume of 4 μl) were heated separately at 95 °C for 2 min following by incubation at 25 °C for 10 min. 8 μl of reaction buffer (2 × concentrated: 200 mM KCl, 800 mM NaCl, 100 mM HEPES pH 7.0, 15 mM MgCl₂, and 15 mM MnCl₂) was then added to the DNAzyme. RNA and DNAzyme were mixed and incubated at 25 °C for 2 h. h. After incubation with either 10–23 or 8–17 DNazyme, RNA was extracted with phenol/chlorophorm, precipitated, separated on 1% agaose-formaldehyde gel and followed by Northern Blot analysis if necessary. Each experiment was replicated two times. Oligonucleotides are listed in Supplementary Table 5. Raw, uncropped images from main figures are shown in Supplementary Fig. 7.

Data availability

GEO accession number: GSE93240. All data are provided by the authors upon reasonable request.

References

Matera, A. G., Terns, R. M. & Terns, M. P. Non-coding RNAs: lessons from the small nuclear and small nucleolar RNAs. Nat. Rev. Mol. Cell. Biol. 8, 209–220 (2007).
Article CAS PubMed Google Scholar
Dupuis-Sandoval, F., Poirier, M. & Scott, M. S. The emerging landscape of small nucleolar RNAs in cell biology. Wiley Interdiscip. Rev. RNA 6, 381–397 (2015).
Article CAS PubMed PubMed Central Google Scholar
Bratkovic, T. & Rogelj, B. The many faces of small nucleolar RNAs. Biochim. Biophys. Acta 438-43, 2014 (1839).
Google Scholar
Lestrade, L. & Weber, M. J. snoRNA-LBME-db, a comprehensive database of human H/ACA and C/D box snoRNAs. Nucl. Acids Res. 34, D158–D162 (2006).
Article CAS PubMed Google Scholar
Yang, J. H. et al. snoSeeker: an advanced computational package for screening of guide and orphan snoRNA genes in the human genome. Nucl. Acids Res. 34, 5112–5123 (2006).
Article CAS PubMed PubMed Central Google Scholar
Makarova, J. A. & Kramerov, D. A. SNOntology: myriads of novel snoRNAs or just a mirage? BMC Genom. 12, 543 (2011).
Article CAS Google Scholar
Yoshihama, M., Nakao, A. & Kenmochi, N. snOPY: a small nucleolar RNA orthological gene database. BMC Res. Notes 6, 426 (2013).
Article PubMed PubMed Central Google Scholar
Piekna-Przybylska, D., Decatur, W. A. & Fournier, M. J. New bioinformatic tools for analysis of nucleotide modifications in eukaryotic rRNA. RNA 13, 305–312 (2007).
Article CAS PubMed PubMed Central Google Scholar
Steinmetz, E. J., Conrad, N. K., Brow, D. A. & Corden, J. L. RNA-binding protein Nrd1 directs poly(A)-independent 3ʹ-end formation of RNA polymerase II transcripts. Nature 413, 327–331 (2001).
Article ADS CAS PubMed Google Scholar
Kim, M. et al. Distinct pathways for snoRNA and mRNA termination. Mol. Cell. 24, 723–734 (2006).
Article CAS PubMed Google Scholar
Schulz, D. et al. Transcriptome surveillance by selective termination of noncoding RNA synthesis. Cell 155, 1075–1087 (2013).
Article CAS PubMed Google Scholar
Grzechnik, P., Gdula, M. R. & Proudfoot, N. J. Pcf11 orchestrates transcription termination pathways in yeast. Genes Dev. 29, 849–861 (2015).
Article CAS PubMed PubMed Central Google Scholar
van Hoof, A., Lennertz, P. & Parker, R. Yeast exosome mutants accumulate 3ʹ-extended polyadenylated forms of U4 small nuclear RNA and small nucleolar RNAs. Mol. Cell. Biol. 20, 441–452 (2000).
Article PubMed PubMed Central Google Scholar
Grzechnik, P. & Kufel, J. Polyadenylation linked to transcription termination directs the processing of snoRNA precursors in yeast. Mol. Cell. 32, 247–258 (2008).
Article CAS PubMed PubMed Central Google Scholar
Chanfreau, G., Legrain, P. & Jacquier, A. Yeast RNase III as a key processing enzyme in small nucleolar RNAs metabolism. J. Mol. Biol. 284, 975–988 (1998).
Article CAS PubMed Google Scholar
Lee, C. Y., Lee, A. & Chanfreau, G. The roles of endonucleolytic cleavage and exonucleolytic digestion in the 5ʹ-end processing of S. cerevisiae box C/D snoRNAs. RNA 9, 1362–1370 (2003).
Article CAS PubMed PubMed Central Google Scholar
Chanfreau, G., Rotondo, G., Legrain, P. & Jacquier, A. Processing of a dicistronic small nucleolar RNA precursor by the RNA endonuclease Rnt1. Embo. J. 17, 3726–3737 (1998).
Article CAS PubMed PubMed Central Google Scholar
Qu, L. H. et al. Seven novel methylation guide small nucleolar RNAs are processed from a common polycistronic transcript by Rat1p and RNase III in yeast. Mol. Cell. Biol. 19, 1144–1158 (1999).
Article CAS PubMed PubMed Central Google Scholar
Ghazal, G., Ge, D., Gervais-Bird, J., Gagnon, J. & Abou Elela, S. Genome-wide prediction and analysis of yeast RNase III-dependent snoRNA processing signals. Mol. Cell. Biol. 25, 2981–2994 (2005).
Article CAS PubMed PubMed Central Google Scholar
Mouaikel, J., Verheggen, C., Bertrand, E., Tazi, J. & Bordonne, R. Hypermethylation of the cap structure of both yeast snRNAs and snoRNAs requires a conserved methyltransferase that is localized to the nucleolus. Mol. Cell. 9, 891–901 (2002).
Article CAS PubMed Google Scholar
Vasiljeva, L. & Buratowski, S. Nrd1 interacts with the nuclear exosome for 3ʹ processing of RNA polymerase II transcripts. Mol. Cell. 21, 239–248 (2006).
Article CAS PubMed Google Scholar
Hallais, M. et al. CBC-ARS2 stimulates 3ʹ-end maturation of multiple RNA families and favors cap-proximal processing. Nat. Struct. Mol. Biol. 20, 1358–1366 (2013).
Article CAS PubMed Google Scholar
Andersen, P. R. et al. The human cap-binding complex is functionally connected to the nuclear RNA exosome. Nat. Struct. Mol. Biol. 20, 1367–1376 (2013).
Article CAS PubMed PubMed Central Google Scholar
Finkel, J. S., Chinchilla, K., Ursic, D. & Culbertson, M. R. Sen1p performs two genetically separable functions in transcription and processing of U5 small nuclear RNA in Saccharomyces cerevisiae. Genetics 184, 107–118 (2010).
Article CAS PubMed PubMed Central Google Scholar
Izaurralde, E. et al. A nuclear cap binding protein complex involved in pre-mRNA splicing. Cell 78, 657–668 (1994).
Article CAS PubMed Google Scholar
Lidschreiber, M., Leike, K. & Cramer, P. Cap completion and C-terminal repeat domain kinase recruitment underlie the initiation–elongation transition of RNA polymerase II. Mol. Cell. Biol. 33, 3805–3816 (2013).
Article CAS PubMed PubMed Central Google Scholar
Creamer, T. J. et al. Transcriptome-wide binding sites for components of the Saccharomyces cerevisiae non-poly(A) termination pathway: Nrd1, Nab3, and Sen1. PLoS Genet. 7, e1002329 (2011).
Article CAS PubMed PubMed Central Google Scholar
Tremblay, A. et al. A physical interaction between Gar1p and Rnt1pi is required for the nuclear import of H/ACA small nucleolar RNA-associated proteins. Mol. Cell. Biol. 22, 4792–4802 (2002).
Article CAS PubMed PubMed Central Google Scholar
Giorgi, C., Fatica, A., Nagel, R. & Bozzoni, I. Release of U18 snoRNA from its host intron requires interaction of Nop1p with the Rnt1p endonuclease. EMBO J. 20, 6856–6865 (2001).
Article CAS PubMed PubMed Central Google Scholar
Rondon, A. G., Mischo, H. E., Kawauchi, J. & Proudfoot, N. J. Fail-safe transcriptional termination for protein-coding genes in S. cerevisiae. Mol. Cell. 36, 88–98 (2009).
Article CAS PubMed PubMed Central Google Scholar
Gagnon, J., Lavoie, M., Catala, M., Malenfant, F. & Elela, S. A. Transcriptome wide annotation of eukaryotic RNase III reactivity and degradation signals. PLoS Genet. 11, e1005000 (2015).
Article PubMed PubMed Central Google Scholar
Churchman, L. S. & Weissman, J. S. Nascent transcript sequencing visualizes transcription at nucleotide resolution. Nature 469, 368–373 (2011).
Article ADS CAS PubMed Google Scholar
Nojima, T. et al. Mammalian NET-seq reveals genome-wide nascent transcription coupled to RNA processing. Cell 161, 526–540 (2015).
Article CAS PubMed PubMed Central Google Scholar
Shibagaki, Y., Itoh, N., Yamada, H., Nagata, S. & Mizumoto, K. mRNA capping enzyme. Isolation and characterization of the gene encoding mRNA guanylytransferase subunit from Saccharomyces cerevisiae. J. Biol. Chem. 267, 9521–9528 (1992).
CAS PubMed Google Scholar
Jimeno-Gonzalez, S., Haaning, L. L., Malagon, F. & Jensen, T. H. The yeast 5ʹ–3ʹ exonuclease Rat1p functions during transcription elongation by RNA polymerase II. Mol. Cell. 37, 580–587 (2010).
Article CAS PubMed Google Scholar
Rasmussen, T. P. & Culbertson, M. R. The putative nucleic acid helicase Sen1p is required for formation and stability of termini and for maximal rates of synthesis and levels of accumulation of small nucleolar RNAs in Saccharomyces cerevisiae. Mol. Cell. Biol. 18, 6885–6896 (1998).
Article CAS PubMed PubMed Central Google Scholar
Das, B., Butler, J. S. & Sherman, F. Degradation of normal mRNA in the nucleus of Saccharomyces cerevisiae. Mol. Cell. Biol. 23, 5502–5515 (2003).
Article CAS PubMed PubMed Central Google Scholar
LaCava, J. et al. RNA degradation by the exosome is promoted by a nuclear polyadenylation complex. Cell 121, 713–724 (2005).
Article CAS PubMed Google Scholar
Chanfreau, G., Buckle, M. & Jacquier, A. Recognition of a conserved class of RNA tetraloops by Saccharomyces cerevisiae RNase III. Proc. Natl. Acad. Sci. USA 97, 3142–3147 (2000).
Article ADS CAS PubMed PubMed Central Google Scholar
Henras, A. K., Bertrand, E. & Chanfreau, G. A cotranscriptional model for 3ʹ-end processing of the Saccharomyces cerevisiae pre-ribosomal RNA precursor. RNA 10, 1572–1585 (2004).
Article CAS PubMed PubMed Central Google Scholar
Dong, Z. W. et al. RTL-P: a sensitive approach for detecting sites of 2ʹ-O-methylation in RNA molecules. Nucl. Acids Res. 40, e157 (2012).
Article CAS PubMed PubMed Central Google Scholar
Buchhaupt, M., Peifer, C. & Entian, K. D. Analysis of 2ʹ-O-methylated nucleosides and pseudouridines in ribosomal RNAs using DNAzymes. Anal. Biochem. 361, 102–108 (2007).
Article CAS PubMed Google Scholar
Silverman, S. K. In vitro selection, characterization, and application of deoxyribozymes that cleave RNA. Nucl. Acids Res. 33, 6151–6163 (2005).
Article CAS PubMed PubMed Central Google Scholar
Li, H. D., Zagorski, J. & Fournier, M. J. Depletion of U14 small nuclear RNA (snR128) disrupts production of 18S rRNA in Saccharomyces cerevisiae. Mol. Cell. Biol. 10, 1145–1152 (1990).
Article CAS PubMed PubMed Central Google Scholar
Kufel, J., Dichtl, B. & Tollervey, D. Yeast Rnt1p is required for cleavage of the pre-ribosomal RNA in the 3ʹ ETS but not the 5ʹ ETS. RNA 5, 909–917 (1999).
Article CAS PubMed PubMed Central Google Scholar
Schwer, B., Erdjument-Bromage, H. & Shuman, S. Composition of yeast snRNPs and snoRNPs in the absence of trimethylguanosine caps reveals nuclear cap binding protein as a gained U1 component implicated in the cold-sensitivity of tgs1Delta cells. Nucl. Acids Res. 39, 6715–6728 (2011).
Article CAS PubMed PubMed Central Google Scholar
Dieci, G., Preti, M. & Montanini, B. Eukaryotic snoRNAs: a paradigm for gene expression flexibility. Genomics 94, 83–88 (2009).
Article CAS PubMed Google Scholar
Jorjani, H. et al. An updated human snoRNAome. Nucl. Acids Res. 44, 5068–5082 (2016).
Article CAS PubMed PubMed Central Google Scholar
Maxwell, E. S. & Fournier, M. J. The small nucleolar RNAs. Annu. Rev. Biochem. 64, 897–934 (1995).
Article CAS PubMed Google Scholar
Hoeppner, M. P. & Poole, A. M. Comparative genomics of eukaryotic small nucleolar RNAs reveals deep evolutionary ancestry amidst ongoing intragenomic mobility. BMC Evol. Biol. 12, 183 (2012).
Article CAS PubMed PubMed Central Google Scholar
Gietz, D., St Jean, A., Woods, R. A. & Schiestl, R. H. Improved method for high efficiency transformation of intact yeast cells. Nucl. Acids Res. 20, 1425 (1992).
Article CAS PubMed PubMed Central Google Scholar
Longtine, M. S. et al. Additional modules for versatile and economical PCR-based gene deletion and modification in Saccharomyces cerevisiae. Yeast 14, 953–961 (1998).
Article CAS PubMed Google Scholar
Schmitt, M. E., Brown, T. A. & Trumpower, B. L. A rapid and simple method for preparation of RNA from Saccharomyces cerevisiae. Nucl. Acids Res. 18, 3091–3092 (1990).
Article CAS PubMed PubMed Central Google Scholar
Tollervey, D. & Mattaj, I. W. Fungal small nuclear ribonucleoproteins share properties with plant and vertebrate U-snRNPs. EMBO J. 6, 469–476 (1987).
CAS PubMed PubMed Central Google Scholar
Trapnell, C., Pachter, L. & Salzberg, S. L. TopHat: discovering splice junctions with RNA-Seq. Bioinformatics 25, 1105–1111 (2009).
Article CAS PubMed PubMed Central Google Scholar
Langmead, B. & Salzberg, S. L. Fast gapped-read alignment with Bowtie 2. Nat. Methods 9, 357–359 (2012).
Article CAS PubMed PubMed Central Google Scholar
Li, H. et al. The sequence alignment/map format and SAMtools. Bioinformatics 25, 2078–2079 (2009).
Article PubMed PubMed Central Google Scholar
Zhang, Y. et al. Model-based analysis of ChIP-Seq (MACS). Genome Biol. 9, R137 (2008).
Article PubMed PubMed Central Google Scholar

Download references

Acknowledgements

We thank A. Hinnebusch for CBP20-myc and CBP80-myc, G. Chanfreau for rnt1E320K and T.H. Jensen for ceg1-63 strains. The JK group was supported by the Polish Ministry of Science and Higher Education (N N301 065740 to J.K.), Polish-Swiss Research Programme (PSPB-183/2010 to JK) National Science Centre (UMO-2011/01/N/NZ1/04344 to SS) and EU Social Fund (UDAPOKL.04.01.01-00-072/09-00). The JK group used CePT infrastructure financed by the EU Regional Development Fund (Innovative economy 2007-13, POIG.02.02.00-14-024/08-00). The PG group is supported by a Sir Henry Dale Fellowship jointly funded by the Wellcome Trust and the Royal Society (200473/Z/16/Z). The NJP laboratory was supported by a Wellcome Trust Programme Grant (091805/Z/10/Z) and European Research Council Advanced Grant (339270). We thank the High-Throughput Genomics Group at the Wellcome Trust Centre for Human Genetics (funded by Wellcome Trust grant reference 203141/Z/16/Z) for the generation of the Sequencing data.

Author information

These authors contributed equally: Pawel Grzechnik, Sylwia A. Szczepaniak.

Authors and Affiliations

Sir William Dunn School of Pathology, University of Oxford, South Parks Road, Oxford, OX1 3RE, UK
Pawel Grzechnik, Somdutta Dhir, Hannah E. Mischo & Nicholas J. Proudfoot
School of Biosciences, University of Birmingham, Edgbaston, Birmingham, B15 2TT, UK
Pawel Grzechnik & Hannah Parslow
Institute of Genetics and Biotechnology, Faculty of Biology, University of Warsaw, Pawinskiego 5a, 02-106, Warsaw, Poland
Sylwia A. Szczepaniak, Anna Pastucha, Zaneta Matuszek & Joanna Kufel
College of Inter-Faculty Individual Studies in Mathematics and Natural Sciences, University of Warsaw, 02-089, Warsaw, Poland
Sylwia A. Szczepaniak

Authors

Pawel Grzechnik
View author publications
You can also search for this author in PubMed Google Scholar
Sylwia A. Szczepaniak
View author publications
You can also search for this author in PubMed Google Scholar
Somdutta Dhir
View author publications
You can also search for this author in PubMed Google Scholar
Anna Pastucha
View author publications
You can also search for this author in PubMed Google Scholar
Hannah Parslow
View author publications
You can also search for this author in PubMed Google Scholar
Zaneta Matuszek
View author publications
You can also search for this author in PubMed Google Scholar
Hannah E. Mischo
View author publications
You can also search for this author in PubMed Google Scholar
Joanna Kufel
View author publications
You can also search for this author in PubMed Google Scholar
Nicholas J. Proudfoot
View author publications
You can also search for this author in PubMed Google Scholar

Contributions

P.G. conceived the project. P.G., S.A.S., J.K. and N.J.P. designed experiments. P.G. and S.A.S. performed the majority of experiments. S.D. carried out the bioinformatics analyses, A.P. performed the FISH analysis, Z.M. assisted with strain construction and Northern Blot analyses while HP assisted with rRNA methylation analyses. H.E.M. advised on data analysis and experimental design. P.G., J.K. and N.J.P. wrote the manuscript.

Corresponding authors

Correspondence to Pawel Grzechnik, Joanna Kufel or Nicholas J. Proudfoot.

Ethics declarations

Competing interests

The authors declare no competing interests.

Additional information

Publisher's note: Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Electronic supplementary material

Supplementary Information

Peer Review File

Rights and permissions

Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license, and indicate if changes were made. The images or other third party material in this article are included in the article’s Creative Commons license, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons license and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this license, visit http://creativecommons.org/licenses/by/4.0/.

Reprints and permissions

About this article

Cite this article

Grzechnik, P., Szczepaniak, S.A., Dhir, S. et al. Nuclear fate of yeast snoRNA is determined by co-transcriptional Rnt1 cleavage. Nat Commun 9, 1783 (2018). https://doi.org/10.1038/s41467-018-04094-y

Download citation

Received: 02 February 2017
Accepted: 04 April 2018
Published: 03 May 2018
DOI: https://doi.org/10.1038/s41467-018-04094-y

This article is cited by

Ceg1 depletion reveals mechanisms governing degradation of non-capped RNAs in Saccharomyces cerevisiae
- Onofrio Zanin
- Matthew Eastham
- Pawel Grzechnik
Communications Biology (2023)

Comments

By submitting a comment you agree to abide by our Terms and Community Guidelines. If you find something abusive or that does not comply with our terms or guidelines please flag it as inappropriate.