Article Reference

Article

Bimodal expression of PHO84 is modulated by early termination of antisense transcription

CASTELNUOVO, Manuele, et al.

Abstract

Many Saccharomyces cerevisiae genes encode antisense transcripts, some of which are unstable and degraded by the exosome component Rrp6. Loss of Rrp6 results in the accumulation of long PHO84 antisense (AS) RNAs and repression of sense transcription through PHO84 promoter deacetylation. We used single-molecule resolution fluorescent in situ hybridization (smFISH) to investigate antisense-mediated transcription regulation. We show that PHO84 AS RNA acts as a bimodal switch, in which continuous, low-frequency antisense transcription represses sense expression within individual cells. Surprisingly, antisense RNAs do not accumulate at the PHO84 gene but are exported to the cytoplasm. Furthermore, rather than stabilizing PHO84 AS RNA, the loss of Rrp6 favors its elongation by reducing early transcription termination by Nrd1-Nab3-Sen1. These observations suggest that PHO84 silencing results from antisense transcription through the promoter rather than the static accumulation of antisense RNAs at the repressed gene.

Reference

CASTELNUOVO, Manuele, et al. Bimodal expression of PHO84 is modulated by early termination of antisense transcription. Nature Structural & Molecular Biology, 2013, vol. 20, no. 7, p. 851-858

DOI : 10.1038/nsmb.2598 PMID : 23770821

Available at: http://archive-ouverte.unige.ch/unige:36376

Disclaimer: layout of this document may differ from the published version.

1 / 1 a r t i c l e s

Bimodal expression of PHO84 is modulated by early termination of antisense transcription

Manuele Castelnuovo1,3, Samir Rahman2,3, Elisa Guffanti1,3, Valentina Infantino1, Françoise Stutz1 & Daniel Zenklusen2

Many Saccharomyces cerevisiae genes encode antisense transcripts, some of which are unstable and degraded by the exosome component Rrp6. Loss of Rrp6 results in the accumulation of long PHO84 antisense (AS) RNAs and repression of sense transcription through PHO84 promoter deacetylation. We used single-molecule resolution fluorescent in situ hybridization (smFISH) to investigate antisense-mediated transcription regulation. We show that PHO84 AS RNA acts as a bimodal switch, in which continuous, low-frequency antisense transcription represses sense expression within individual cells. Surprisingly, antisense RNAs do not accumulate at the PHO84 gene but are exported to the cytoplasm. Furthermore, rather than stabilizing PHO84 AS RNA, the loss of Rrp6 favors its elongation by reducing early transcription termination by Nrd1–Nab3–Sen1. These observations suggest that PHO84 silencing results from antisense transcription through the promoter rather than the static accumulation of antisense RNAs at the repressed gene.

Genome-wide pervasive transcription has been reported in many (refs. 20,21), URA2 (ref. 22), FLO11 (ref. 23) and IME1 (ref. 24) through eukaryotic organisms, producing hundreds of non–protein-coding various mechanisms, including co-transcriptional chromatin modifi- RNAs (ncRNAs). Even the small yeast genome encodes many inter- cations, that establish histone repositioning and a repressive chromatin genic, promoter-associated antisense transcripts, some stable and state blocking access to transcription factors. Although the RME2 AS others rapidly degraded and hence are known as cryptic unstable RNA was proposed to repress the meiotic regulator IME4 via tran- transcripts (CUTs)1–3. The degradation of these 200- to 600-base-long scription interference25,26, AS RNA transcription may also affect sense CUTs is in great part mediated by Rrp6, a 3′–5′ exonuclease belonging expression by influencing the epigenetic state of chromatin. Indeed, to the nuclear exosome4,5. Exosome-mediated degradation is assisted antisense RNA transcription originating within GAL10 and running © 2013 Nature America, Inc. All rights reserved. America, Inc. © 2013 Nature by Trf4/Trf5–Air1/Air2–Mtr4 polyadenylation (TRAMP), a surveil- into divergent GAL1 in glucose deposits histone H3 Lys4 di- and tri- lance complex containing the noncanonical poly(A) polymerase Trf4, methylation (H3K4me2/3) and Lys36 trimethylation (H3K36me3) by whereas mRNAs are polyadenylated by Pap1, resulting in stable and the Set1 and Set2 histone methyltransferases, respectively. These marks npg export-competent messenger ribonucleoprotein particles4,6–9. signal the recruitment of the Rpd3S histone deacetylase (HDAC), The Nrd1–Nab3–Sen1 (NNS) complex mediates transcription which attenuates GAL1 and GAL10 expression27,28. H3K4me3 and termination of CUTs, small nuclear (snRNA), small nucleolar RNA H3K4me2 deposited by Set1 were also implicated in causing repression (snoRNAs) and some mRNAs7,10–13. It is recruited to the 5′ end of by signaling the recruitment of the Rpd3L and Set3 HDACs, respec- most RNA polymerase II (RNAPII) transcription units through tively, to specific gene promoters24,29,30. interaction between Nrd1 and the Ser5- and/or Ser7-phosphorylated Our earlier studies focused on PHO84, which encodes a high-affinity RNAPII C-terminal domain14–16. Transcription termination by NNS phosphate transporter. PHO84 transcription is induced by the activator depends on the abundance of specific Nrd1- and Nab3-binding motifs Pho4 imported into the nucleus upon phosphate starvation31. The acti- on the nascent RNA and occurs primarily on short transcripts as the vation threshold of the PHO84 promoter depends on the nuclear con- recruitment of NNS decreases toward the 3′ end of long transcription centration of Pho4 and the accessibility of the Pho4 binding sites32,33. units. Consistent with the physical interactions between the NNS, PHO84 mRNA is weakly expressed in standard yeast medium contain- TRAMP and exosome complexes, CUT degradation has been directly ing intermediate phosphate levels. In these conditions, PHO84 also linked to NNS-mediated early termination4,7,10,11. produces two antisense transcripts (1.9 kb and 2.3 kb long) starting Genome-wide studies indicate that numerous genes produce at its 3′ end and extending into the PHO84 promoter. Loss of Rrp6 upstream tandem or antisense transcripts17,18, a fraction of which increases PHO84 AS levels, and this accumulation is paralleled by the may function in gene regulation19. Transcription of an upstream tan- recruitment of the Hda1–Hda2–Hda3 HDAC complex over the locus, dem ncRNA was proposed to interfere with the expression of SER3 histone deacetylation at the promoter and transcriptional repression.

1Department of Cell Biology and National Center of Competence in Research “Frontiers in Genetics”, University of Geneva, Geneva, Switzerland. 2Département de Biochimie, Université de Montréal, Montréal, Québec, Canada. 3These authors contributed equally to this work. Correspondence should be addressed to F.S. ([email protected]) or D.Z. ([email protected]). Received 30 November 2012; accepted 26 April 2013; published online 16 June 2013; doi:10.1038/nsmb.2598

nature structural & molecular biology VOLUME 20 NUMBER 7 JULY 2013 851 a r t i c l e s

Figure 1 PHO84 sense and antisense a b expression are anticorrelated. (a) Deletion Sense Antisense Sense Antisense of RRP6 increases expression of PHO84 PHO84 mRNA PHO84 AS antisense (AS) RNA. Northern blot (left) and W303 ∆rrp6 1.2 8 reverse transcriptase quantitative polymerase PHO84 mRNA 1.0 7 chain reaction (RT-qPCR) (right) analyses in 6

0.8 RNA relative levels wild-type (WT, W303) and ∆rrp6 cells. Error 5 Time bars reflect s.d. of an average from three 0.6 4 3 independent experiments. (b) Two possible PHO84 AS 0.4 2 models for antisense-mediated gene silencing. 0.2 1 In a graded response, gradual accumulation of

ACT1 -normalized expression 0 0 ACT1 -normalized expression PHO84 AS RNAs leads to a gradual reduction ACT1 WT WT

PHO84 expression Graded Bimodal in sense levels (top left), whereas in a bimodal ∆rrp6 ∆rrp6 response, sense and antisense expression are c Sense transcript anticorrelated (top right). Changes in PHO84 PHO84 ORF mRNA sense and antisense levels over time are Antisense transcript represented as green and red lines, respectively. Lower panels show PHO84 sense expression resulting from graded or bimodal regulation at the single-cell level. (c) Bimodal expression of PHO84 sense and antisense RNAs. smFISH WT detects PHO84 sense (green) and antisense RNAs (red) in individual WT and ∆rrp6 cells. Nuclear DNA was stained using 4′,6-diamidino- 2-phenylindole (DAPI) (blue), and cellular outlines were visualized using differential interference contrast (DIC) optics. Scale bar, 5 µm. smFISH probe positions are drawn at ∆ rrp6 the top. ORF, open reading frame. (d) Fewer cells express PHO84 sense in ∆rrp6. Frequency distribution of PHO84 sense and antisense PHO84 AS DAPI PHO84 mRNA DAPI Merge DIC expression in individual cells from c. (e) Deletion d 80 e 100 of RRP6 results in higher levels of PHO84 AS WT WT RNA in single cells. Frequency distribution of 80 60 ∆rrp6 ∆rrp6 the number of PHO84 AS RNAs per cell in WT 60 and ∆rrp6. Error bars in d and e reflect s.d. of an 40 average from three independent experiments. 40 20 Percentage of cells Percentage of cells 20 We proposed that stabilization and accumula- 0 0 tion of antisense RNAs at PHO84 might facili- Only AS Only AS and No AS or 0 1 2 3 4 5 tate Hda1 recruitment maintaining repression sense sense sense PHO84 AS RNAs per cell © 2013 Nature America, Inc. All rights reserved. America, Inc. © 2013 Nature of sense transcription34. To further elucidate the mechanism of antisense-mediated tran- Alternatively, low expression of PHO84 AS RNA in wild-type cells scription regulation, we used smFISH to detect individual sense and may continuously fine-tune PHO84 sense expression, a process that npg antisense RNAs35–37. We show that the presence of PHO84 sense and may be regulated by Rrp6. However, different levels of antisense antisense transcripts in single cells is strongly anticorrelated, suggesting expression in wild-type versus ∆rrp6 cells may also reflect different a switch-like regulation mechanism. Our data provide evidence that subclasses of cells in a population that expresses either PHO84 sense Rrp6 does not degrade full-length antisense transcripts but prevents or antisense. Different models can therefore be suggested for how antisense transcription elongation by favoring early termination by antisense-mediated silencing of PHO84 is established. The regulation NNS, whereas the H3K4 methyltransferase Set1 may antagonize this occurs either by a graded response, in which increasing antisense lev- event. These observations suggest that antisense-mediated silencing is els lead to decreasing levels of sense transcription, or by a switch-like regulated, at least in part, through transcription attenuation and that mechanism, in which low levels of antisense expression in a single cell PHO84 repression results from antisense transcription through the pro- are sufficient to repress sense transcription (Fig. 1b). moter, followed by rapid export of antisense RNA into the cytoplasm. To detect single RNA molecules, we designed smFISH probes targeted to the 5′ region of sense and antisense PHO84 transcripts. RESULTS Probes were labeled with fluorescent dyes, allowing us to distinguish Bimodal expression of PHO84 sense and antisense transcripts between sense and antisense transcripts, and hybridized to fixed We have suggested that PHO84 AS RNAs might stably associate, pos- yeast cells; this was followed by image acquisition. We first localized sibly in multiple copies, with PHO84 to facilitate efficient recruitment PHO84 transcripts in wild-type cells under conditions in which both of chromatin modifiers. Such a process would require only a single, sense and antisense RNAs are detected by northern blotting (Fig. 1a). initial burst of antisense transcription to establish silencing of sense. Although both PHO84 sense and antisense RNAs can be detected in However, results from northern blot analysis of PHO84 sense and anti- wild-type cells (Fig. 1c), they are never coexpressed (Fig. 1d), suggest- sense expression conflict with such a model, showing that low levels of ing that antisense-mediated repression of PHO84 operates through AS RNA can be detected in wild-type cells under conditions in which a switch-like rather than a graded process. Consistent with the role PHO84 sense is transcribed and indicating that very low expression of of Rrp6 in modulating sense repression through antisense RNA, the antisense might not be sufficient to repress sense transcription (Fig. 1a). fraction of cells expressing antisense increases (from 28% to 55%)

852 VOLUME 20 NUMBER 7 JULY 2013 nature structural & molecular biology a r t i c l e s

a b c DNA RNAPII 100 MDN1 mRNA (WT) PHO84 AS (WT) 25 WT 80 PHO84 AS (∆rrp6)

20 Nascent mRNA 60 Nucleus 15 Cytoplasm 40 10 Percentage of cells 20

AAAAAAA Percentage of cells 5 ∆ rrp6 Cytoplasmic mRNA 0 0 0 1 2 3 4 WT rrp6 Number of nascent RNAs ∆ MDN1 mRNA DAPI PHO84 AS DAPI

Figure 2 PHO84 AS RNAs do not accumulate at the PHO84 locus. (a) PHO84 AS RNAs are exported to the cytoplasm. smFISH for MDN1 mRNA (green) and PHO84 AS RNA (red) in WT and ∆rrp6 cells. Scale bar, 5 µm. The cartoon on the left illustrates the detection of nascent and cytoplasmic RNAs. RNAPII, RNA polymerase II. (b) PHO84 AS RNAs do not accumulate at the PHO84 gene locus. Frequency distribution of the number of nascent RNAs for MDN1 (WT only) and PHO84 AS RNAs in WT and ∆rrp6. (c) Deletion of RRP6 leads to a higher frequency of cells showing nascent PHO84 AS RNAs. Frequency distribution of cells containing nascent (nuclear) PHO84 AS RNAs in WT and ∆rrp6. Error bars in b and c reflect s.d. of an average from three independent experiments.

in a ∆rrp6 strain, whereas the percentage of sense-expressing cells nascent mRNAs are detected on the long, constitutively transcribed decreases (Fig. 1c–e). MDN1; however, most nuclear PHO84 AS RNA signals show the At the single-cell level, sense expression is much higher than anti- same intensity as single cytoplasmic antisense molecules, indicating sense expression: large numbers of PHO84 mRNAs are detected within that antisense transcripts do not accumulate at PHO84 (Fig. 2a,b). individual cells, suggesting that PHO84 transcription occurs in strong Furthermore, only 13% of wild-type and 20% of ∆rrp6 cells show bursts when repression is overcome. In contrast, PHO84 AS levels are nuclear signal, inconsistent with a model in which antisense RNAs very low in individual wild-type cells, with most cells expressing no or stay associated with the gene locus for a long time (Fig. 2c). only a single antisense RNA molecule. In the ∆rrp6 strain, antisense It is likely that most nuclear antisense signals with an intensity of a levels are higher and more cells express PHO84 AS; however, most single RNA represent nascent rather than freely diffusing nucleoplas- cells still only contain 1–3 antisense RNA molecules, and a substantial mic antisense RNAs. Indeed, nuclear PHO84 AS signals, like nascent fraction (40%) show no signal (Fig. 1c–e). Double-negative cells are PHO84 mRNA signals, are always located at the nuclear periphery, not due to the inability to detect RNAs in these cells, as double stain- which indicates that the gene is located close to the nuclear periphery, ing for the constitutively expressed MDN1 RNAs shows expression of an observation consistent with the subtelomeric position of PHO84 on MDN1 in all cells (Supplementary Fig. 1a). Thus, very low antisense chromosome XIII (Supplementary Fig. 1b). Furthermore, our earlier expression appears sufficient to exert a repressive effect on PHO84 studies showed that mRNAs are rarely detected in the nucleoplasm © 2013 Nature America, Inc. All rights reserved. America, Inc. © 2013 Nature transcription in individual cells. Unexpectedly, we did not observe a except at the site of transcription, suggesting that mRNA export is significant accumulation of antisense RNA in the nucleus (Fig. 2) as fast, probably occurring within seconds after release from the site of most antisense RNAs detected in wild-type and ∆rrp6 cells are found transcription36,39. If PHO84 AS RNAs transcribed at a low frequency npg in the cytoplasm, suggesting that PHO84 AS RNAs, like mRNAs, do behave like mRNAs, detecting antisense RNAs within the nucleus is not remain associated with PHO84 but are rapidly exported. likely a rare event, except when they are nascent. Thus, nuclear PHO84 AS RNAs are likely to be nascent and behave like mRNAs that rapidly PHO84 antisense RNAs do not accumulate at the PHO84 locus dissociate from the locus after synthesis. These observations suggest The fraction of antisense RNAs detected in the nucleus can represent that antisense transcription rather than antisense RNA accumulation nascent RNAs associated with the transcription machinery, RNAs dif- at the gene may mediate PHO84 silencing. fusing in the nucleoplasm on their way to the cytoplasm or antisense RNAs associated with PHO84 in a transcription-independent manner. PHO84 antisense RNAs behave like mRNAs To distinguish between these possibilities, we further characterized To confirm that antisense transcripts behave like mRNAs, we first the nuclear PHO84 AS RNA signal. The quantitative nature of smFISH monitored antisense RNA distribution in a mutant for the poly(A) makes it possible to define how many RNAs are present in a single polymerase Pap1. mRNA cleavage and polyadenylation occurs RNA spot, and we have shown that cytoplasmic mRNA spots have co-transcriptionally and is required for nuclear export. The pap1-1 a uniform signal intensity representing single mRNAs36,38. Nuclear and pap1-1∆rrp6 temperature-sensitive strains were grown at 25 °C signals often show higher intensities because they represent sites of and shifted to 37 °C before fixation. After a 1-h heat shock, pap1-1 active transcription in which multiple nascent mRNAs are associated cells accumulate antisense RNAs in the nucleus and fewer transcripts with a gene. The frequency and number of nascent mRNAs detected are observed in the cytoplasm, a phenotype that was more pronounced for a specific gene depend on its transcription rate and length. If in pap1-1∆rrp6 (Fig. 3a,b). Antisense RNAs do not accumulate in antisense RNAs accumulate in multiple copies at PHO84, smFISH one spot but distribute throughout the nucleus, with a tendency to should detect higher-intensity nuclear signals than cytoplasmic sig- localize within the nucleolus (Supplementary Fig. 2a). The higher nals. Furthermore, if antisense RNAs stay associated at the gene for accumulation in pap1-1∆rrp6 compared with pap1-1 suggests that long periods of time, most cells with no sense expression should show antisense RNAs are degraded by Rrp6 when not polyadenylated by a nuclear antisense signal. Nuclear signals corresponding to multiple Pap1 and/or that a higher number of antisense RNAs are expressed

nature structural & molecular biology VOLUME 20 NUMBER 7 JULY 2013 853 a r t i c l e s

Figure 3 PHO84 antisense RNAs are polyadenylated by Pap1. a pap1-1 pap1-1∆rrp6 (a) Inactivation of Pap1 leads to nuclear accumulation of PHO84 AS 25 °C 37 °C 25 °C 37 °C RNAs. smFISH using probes against PHO84 AS RNAs (red) in pap1-1

and pap1-1∆rrp6 cells grown at 25 °C and either directly fixed or shifted DAPI to 37 °C for 1 h before fixation. Nuclear DNA was stained using DAPI (blue), and cellular outlines were visualized using DIC optics. Scale bar, 5 µm. (b) pap1-1∆rrp6 cells accumulate high numbers of PHO84 AS RNA in the nucleus. Frequency distribution of the number of PHO84 PHO84 AS AS RNAs detected by smFISH in pap1-1 and pap1-1∆rrp6 cells after 1 h shift to 37 °C. (c) PHO84 AS RNAs polyadenylation requires Pap1. Northern blot membranes with oligo(dT)-purified total RNA were hybridized with PHO84 AS–specific probes. Strains were exponentially DIC grown in SC medium with 2% glucose (Glu; lanes 1–4) or 2% galactose (Gal; lane 5) followed by 20 h in 2% glucose (Glu; lane 6) to deplete Trf5 as indicated. ACT1 and TRF4 mRNA–specific probes were used to control for loading and TRF4 deletion. b c 50 pap1-1 pap1-1∆rrp6 rrp6 trf4∆rrp6trf4 GAL–TRF5 Gal 40 WT ∆ ∆trf4 ∆ ∆ ∆trf4 GAL–TRF5 Glu in a pap1-1∆rrp6 background (Fig. 3b). Loss of the noncanonical 30 PHO84 AS poly(A) polymerases Trf4 and Trf5 did not reduce the amount of 20 ACT1 polyadenylated PHO84 AS RNAs, confirming their polyadenylation 10 Percentage of cells by Pap1 (Fig. 3c). Notably, shifting pap1-1∆rrp6 double-mutant cells, 0 TRF4 but not pap1-1 single-mutant cells, to 37 °C results in the accumula- 0 1 2 3 4 5 6 7 8 9 10 11 12 13 1 2 3 4 5 6 PHO84 AS RNAs per cell tion of an elongated polyadenylated antisense RNA (Supplementary Poly(A)+RNA Fig. 2b). Together, these analyses suggest that when Pap1 is inactive, a single, long antisense transcript is produced that remains in the in the mex67-5∆rrp6 conditional mutants when shifted to 37 °C nucleus and is degraded by Rrp6, presumably following polyade- (Supplementary Fig. 3a,b). Moreover, the number of cytoplasmic nylation by the noncanonical Trf4 and/or Trf5 poly(A) polymerase PHO84 AS RNAs greatly increases in ∆xrn1 cells, indicating that, as a result of nuclear surveillance40. Thus, the classical cleavage like mRNAs, they undergo 5′-to-3′ exonucleolytic degradation in this and polyadenylation machinery is required for 3′-end processing compartment (Supplementary Fig. 3c). and export of PHO84 AS RNA, confirming that these long ncRNAs behave like mRNAs. Accordingly, their nuclear export is mediated by Antisense RNA at PHO84 requires active transcription the general mRNA export receptor Mex67 because PHO84 AS A feature of bona fide mRNAs is their rapid dissociation from the transcripts accumulate in the nuclei of the mex67-5 and even more gene after transcription termination; nascent mRNA detection therefore requires ongoing transcription. To 0 min 5 min/37 °C 10 min/37 °C 20 min/37 °C define whether detection of nuclear antisense RNAs requires transcription, we determined PHO84 AS localization and abundance in the © 2013 Nature America, Inc. All rights reserved. America, Inc. © 2013 Nature rpb1-1 strain, which contains a temperature- sensitive mutation in the major RNAPII sub- unit41. To test the efficiency of transcription npg shutoff, we simultaneously monitored MDN1 PHO84 AS DAPI mRNA distribution. Figure 4 shows that after 5 min at 37 °C, most cells have lost nuclear MDN1 signal, and mRNA abundance declines further over time, consistent with transcription shutoff. Similarly, nuclear PHO84 AS signal is quickly lost, and cytoplasmic RNA

rpb1-1 numbers subsequently decrease. Thus, ongo- MDN1 mRNA DAPI ing transcription is required to detect nuclear antisense RNA, further indicating that PHO84 AS RNA does not stay associated with PHO84. The observations that the number of cells with antisense RNA increases in ∆rrp6 PHO84 AS DAPI Figure 4 PHO84 antisense nuclear detection needs ongoing transcription. smFISH detecting MDN1 mRNA (green) and PHO84 AS RNA (red) in rpb1-1 and rpb1-1∆rrp6 cells grown at 25 °C and shifted to 37 °C for 5, 10 and 20 min before fixation. Nuclear DNA was stained using DAPI (blue). rpb1-1 ∆ rrp6 MDN1 mRNA DAPI Scale bar, 5 µm.

854 VOLUME 20 NUMBER 7 JULY 2013 nature structural & molecular biology a r t i c l e s

a b c 100 5′ M 3′ 5′ M 3′ WT 15 PHO84 rrp6 PHO84 ACT1 80 ∆ AS ∆xrn1 1.5 60 10 PHO84 ACT1 1.5 40

expression 5 1.0 1.0

20 SCR1 -normalized normalized to SCR1 Percent RNA remaining 0 0 0.5 0 10 20 30 0.5 WT xrn1 t (min) after phenanthroline addition ∆rrp6∆ H3K4me3/H3 relative fold

0 H3K4me3/H3 relative fold 0 5′ M 3′ 5′ M 3′ Figure 5 Effect of ∆rrp6 on antisense RNA half-life and transcription. ∆pho4 ∆pho4∆rrp6 ∆pho4∆set1 ∆pho4∆rrp6∆set1 (a) Deletion of RRP6 does not alter PHO84 AS RNA half-life. 1.5 PHO84 ACT1 RT-qPCR analysis measuring PHO84 AS RNA decay rates in WT, 1.5 ∆rrp6 and ∆xrn1 after transcription shutoff by adding 100 µg/ml * 1,10-phenanthroline to the medium. PHO84 AS RNA levels were 1.0 1.0 normalized to SCR1 RNA, stable at 30 min. Data are expressed as a percentage of the amounts present before inhibitor addition. 0.5 Error bars represent s.d. of an average from three independent 0.5 experiments. (b) PHO84 AS RNA levels are increased in ∆rrp6 H3K4me2/H3 relative fold

and ∆xrn1. Antisense RNA levels were measured by RT-qPCR 0 H3K4me2/H3 relative fold 0 and expressed relative to the levels in WT, which were set to 1. 5′ M 3′ 5′ M 3′ Error bars reflect s.d. of an average obtained from three independent experiments. (c) Higher levels of H3K4me2/3 at the 3′ end of PHO84 in ∆rrp6. Chromatin immunoprecipitation (ChIP) analysis of H3K4me3 (top) and H3K4me2 (bottom) at the PHO84 locus. ChIP with anti-H3K4me3, anti-H3K4me2 or anti-H3 antibodies from ∆pho4, ∆pho4∆rrp6, ∆pho4∆set1 and ∆pho4∆rrp6∆set1 strains. Immunoprecipitated DNA was quantified by qPCR with primers specific for the 5′, middle (M) and 3′ regions of PHO84 and ACT1 (as indicated on top). H3K4me2/3 values were normalized to H3 values, and the highest value was arbitrarily set to 1. Error bars reflect s.d. of an average from three independent experiments. Comparison of the mean differences was analyzed using Student’s t-test. *P < 0.05.

(Fig. 1d) and that antisense transcription rather than accumulation In this setup, H3K4 methylation derives only from antisense tran- is required to mediate sense silencing (Fig. 1 and 2) suggest that loss scription. Interestingly, we observe that the H3K4me3 and H3K4me2 of Rrp6 does not primarily affect antisense RNA stability but may peaks at the 3′ end and middle regions, respectively, of PHO84 are influence its transcription. substantially increased upon loss of Rrp6. As a control, ACT1 showed To compare PHO84 AS RNA turnover in wild-type and ∆rrp6 cells, the expected high level of H3K4me3 at its 5′ end, with no enrichment we measured antisense levels at various times following inhibition of at the 3′ end, consistent with the absence of antisense transcription RNAPII transcription with phenanthroline (Fig. 5a)42. Surprisingly, on this gene. Because of the low antisense transcription frequency, PHO84 AS RNA decays at a similar rate in both strains with a half- RNAPII is barely detectable at the 3′end of PHO84 in a ∆pho4 strain, © 2013 Nature America, Inc. All rights reserved. America, Inc. © 2013 Nature life of 11.4 min in wild-type and 12 min in ∆rrp6 cells (see Online yet the levels slightly increase in ∆rrp6∆pho4 (data not shown). The Methods). In contrast, the half-life increased to 27.3 min in the ∆xrn1 more efficient detection of H3K4 methylation suggests persistence of strain, confirming 5′-to-3′ antisense RNA degradation in the cytoplasm this histone mark between transcription events. These observations npg as revealed by smFISH (Supplementary Fig. 3c). Because loss of Rrp6 support the view that loss of Rrp6 increases antisense transcription. does not substantially increase PHO84 AS RNA half-life, these results indicate that the higher levels of antisense RNA in ∆rrp6 (Fig. 5b) are PHO84 antisense elongation is regulated by the NNS complex due to increased antisense RNA production rather than stability. To investigate how loss of Rrp6 may increase transcription, we explored the physical and functional links between Rrp6 and the NNS Loss of Rrp6 increases antisense transcription and TRAMP complexes4,7,11. Transcription termination by NNS is Increased PHO84 AS transcription in ∆rrp6 predicts a higher number stimulated by Nrd1 and Nab3 binding motifs on the nascent RNA. of nascent antisense RNAs in this strain as compared to the wild-type Interestingly, several potential Nrd1–Nab3 binding sites are present strain. Indeed, besides an increased number of both antisense-producing within the 5′ end of PHO84 AS RNA (Fig. 6a and Supplementary cells and antisense RNA molecules per cell (Fig. 1e), more ∆rrp6 cells Fig. 4). Furthermore, transcriptome-wide analyses of Nrd1– (20%) than wild-type cells (13%) show nascent antisense RNAs, con- Nab3-bound RNA sequences revealed association with the 5′ end sistent with higher transcription frequency in ∆rrp6 (Fig. 2c). of many antisense transcripts, including PHO84 AS RNA, suggest- One hallmark of active transcription is Lys4 methylation on histone ing that these ncRNAs undergo early transcription termination45,46. H3 by Set1, the only yeast H3K4 histone methyltransferase recruited Accordingly, depletion of the essential Nrd1 protein using the to the 5′ end of transcription units43,44. Most active genes show peaks glucose-repressible GAL1 promoter leads to increased PHO84 AS of H3K4 trimethylation at the 5′ end, dimethylation in the middle and levels in wild-type cells, and this effect is even more pronounced monomethylation at the 3′ end. We postulated that if loss of Rrp6 in ∆rrp6 (Fig. 6b). Moreover, a modified PHO84 gene in which a increases antisense transcription, Set1-dependent H3K4me3 should number of putative Nrd1–Nab3 binding sites at the 5′ end of the increase over the PHO84 3′ end in ∆rrp6 versus wild-type cells. We antisense RNA have been mutagenized produces more antisense performed chromatin immunoprecipitation (ChIP) of H3K4me3 and transcripts both in wild-type and ∆rrp6 cells. The relatively mod- H3K4me2 in wild-type and ∆rrp6 cells also devoid of the transcrip- est effect of the cis mutations may be due to only partial removal of tion factor Pho4, completely abrogating sense transcription (Fig. 5c). potential NNS binding sites to keep the PHO84 open reading frame

nature structural & molecular biology VOLUME 20 NUMBER 7 JULY 2013 855 a r t i c l e s

Figure 6 PHO84 antisense transcription is a Rrp6 attenuated by Nrd1–Nab3–Sen1 (NNS). Nrd1 Nab3 (a) NNS terminates short AS transcripts. TRAMP PHO84 Sen1 RNAPII Cartoon illustrates the role of NNS in PHO84 PHO84 Nrd1–Nab3 binding sites AS transcription. Short antisense RNAs (red Export Pap1 AAA AAA Nrd1-terminated AS transcripts line) previously shown to be polyadenylated by AAA AAA CPF–Pap1-terminated AS transcripts Trf4 and degraded by Rrp6 (ref. 1) are proposed here to be terminated at Nrd1–Nab3 motifs b PHO84 AS c PHO84 3′ (blue bars) by the NNS complex, whereas 5 Gal ** ∆pho4 set1 rrp6 the long read-through antisense transcripts ∆ ∆ ** Glu 7 h 0.5 ∆pho4∆set1 (green lines) are subjected to 3 -end cleavage 4 pho4 pho4 ′ ∆pho4∆ ∆ ∆pho4∆rrp6 and polyadenylation by Pap1 before export 0.4 3 Nrd1–HA into the cytoplasm. (b) Depletion of Nrd1 0.3 increases PHO84 AS RNA levels. PHO84 AS 2 Ssn6 ** 0.2 RNA levels were measured using RT-qPCR on 1 GAL–Nrd1 and GAL–Nrd1∆rrp6 strains grown Percent of input 0.1

in SC medium containing 2% galactose (Gal) or ACT1-normalized expression 0 0 shifted for 7 h in 2% glucose (Glu) to deplete Nrd1–HA No Ab rrp6 Nrd1. Error bars reflect s.d. of an average from three independent experiments. Comparison GAL–Nrd1 of the mean differences was analyzed using GAL–Nrd1∆ Student’s t-test. **P < 0.01. The value of GAL–Nrd1 (Glu 7 h) was arbitrarily set to 1. (c) Deletion of RRP6 reduces Nrd1 recruitment. ChIP analysis of Nrd1–hemagglutinin binding at the PHO84 3′ end quantified by qPCR and expressed as percent of input. Three biological and two technical repeats were analyzed; error bars reflect s.e.m. Comparison of the mean differences was analyzed using Student’s t-test. **P < 0.01. The small panel shows western blot analysis of Nrd1 protein levels in the strains used for the ChIP. Ssn6 was used as normalization control.

intact (Supplementary Fig. 4). These observations confirm the role level indicate that low-frequency antisense transcription, but not the of NNS in PHO84 AS transcription attenuation. antisense RNA itself, contributes to PHO84 repression. Our earlier studies showed that an extra copy of PHO84 induces Rrp6 and Set1 have opposite effects on early termination repression of both the transgene and the endogenous copy, and they To address whether the absence of Rrp6 might increase antisense suggested that PHO84 AS RNAs may participate in a still poorly transcription elongation by affecting optimal NNS function, we defined mechanism of silencing in trans that is independent of the monitored Nrd1 association with the 3′ end of PHO84 by ChIP Hda1 HDAC complex and therefore distinct from silencing in cis48. in wild-type or ∆rrp6 cells (Fig. 6c). Although loss of Rrp6 does Based on the rapid export of antisense RNAs revealed by smFISH, not affect Nrd1 protein levels, we observed a large decrease in Nrd1 it seems unlikely that antisense RNAs act in trans by diffusing from binding at the PHO84 3′ end in ∆rrp6, suggesting that loss of Rrp6 one gene copy to the other, unless the two genes undergo pairing. The may affect early termination by lowering the association of NNS. primarily cytoplasmic localization of PHO84 AS RNAs suggests that The additive effect on antisense RNA production of ∆rrp6 and Nrd1 they are more likely to act in trans through an indirect mechanism. © 2013 Nature America, Inc. All rights reserved. America, Inc. © 2013 Nature depletion or Nrd1–Nab3 binding site mutagenesis (Fig. 6b and These possibilities should be investigated in the future. Supplementary Fig. 4b), situations that weaken but do not elimi- nate NNS function, supports the notion that NNS and Rrp6 act in smFISH reveals distinct sense and antisense expression modes npg the same pathway. The single-molecule microscopy approach revealed critical param- Interestingly, Nrd1 association with the PHO84 3′ end was slightly eters on PHO84 regulation that could not be obtained using classical enhanced in ∆set1, suggesting that, in contrast to ∆rrp6, loss of Set1 ensemble measurements (Fig. 1). First, we showed that antisense- may increase early termination (Fig. 6c). A recent study similarly mediated regulation does not generate a gradual decrease of sense reported elevated Nrd1 binding in ∆set1 and correlated this phenotype transcription but modulates the threshold of the PHO84 activation with increased phosphorylation of Ser5 in the RNAPII C-terminal switch. Second, smFISH revealed that sense and antisense expression domain, the mark implicated in NNS recruitment16,47. This is also in are achieved through different modes: PHO84 mRNA is transcribed agreement with our earlier data showing reduced PHO84 AS RNA in bursts that lead to a strong accumulation in a fraction of cells, and production in ∆set1 (ref. 48). Accordingly, smFISH analyses indicate antisense RNA is transcribed constantly at a very low rate in most cells reduced antisense expression in ∆set1 and restoration of antisense not expressing PHO84 mRNA. Third, the ability to localize individual RNA levels in ∆set1∆rrp6 (Supplementary Fig. 5). Taken together, RNAs within different cellular compartments showed that PHO84 the data suggest that Rrp6 and Set1 have antagonistic effects in the AS RNA behaves like an mRNA that dissociates from the gene locus regulation of antisense RNA production by respectively facilitating after polyadenylation by Pap1, leaves the nucleus using the canonical and interfering with early transcription termination by NNS. Mex67-dependent mRNA export pathway and is eliminated by the cytoplasmic, Xrn1-dependent RNA degradation machinery. DISCUSSION Expanding on an extensive list of cis- and trans-acting factors, recent Loss of Rrp6 favors antisense transcription elongation studies have established ncRNAs as additional players in controlling Consistent with the increased levels of antisense RNA observed in the regulated expression of protein-coding genes. Transcription regu- ∆rrp6 as compared to wild-type cells through classical RNA analyses lation by ncRNAs is achieved in multiple ways; however, in-depth (Fig. 1a), smFISH revealed more antisense RNA molecules per ∆rrp6 mechanistic understanding is still missing. Our detailed analyses of cell as well as a higher proportion of ∆rrp6 cells with antisense RNA PHO84, cis-acting antisense RNAs at a single-cell and single-molecule (Fig. 1d,e). Our observations indicate that loss of Rrp6 does not result

856 VOLUME 20 NUMBER 7 JULY 2013 nature structural & molecular biology a r t i c l e s

in nuclear stabilization of full-length antisense RNAs but rather pro- Because the activity of both Nrd1 and Rrp6 is regulated in different motes antisense transcription elongation followed by rapid export. physiological conditions52,53, genes such as PHO84 may be controlled First, although the number of cells showing nascent transcripts is in part through modulation of antisense transcription elongation. increased in ∆rrp6 cells, it is rare that more than one antisense RNA molecule is observed at the transcription site. Moreover, in both wild- A new view on antisense-mediated gene repression type and ∆rrp6 cells, this nuclear signal is strictly dependent on ongo- Our data show that PHO84 transcription is regulated by a sensitive ing transcription, indicating that once made, antisense transcripts on-off switch through which sense transcription is either completely do not remain at the gene (Figs. 2–4). Second, the antisense RNA turned off or strongly induced once the repression is overcome. The turnover rate is comparable in wild-type and ∆rrp6 strains, support- activation threshold of Pho4-regulated genes is defined by both the ing the view that the increased steady-state levels in ∆rrp6 are due to nuclear concentration of the Pho4 transcription factor and the acces- increased antisense RNA production (Fig. 5a,b). Third, H3K4me3 and sibility of Pho4 binding sites32. Antisense transcription may ensure H3K4me2 at the 3′ end and middle region of PHO84 are more abun- that PHO84 sense expression is activated only in the presence of a dant in the absence of Rrp6 and consistent with increased antisense strong enough stimulus either by reducing Pho4 accessibility through transcription (Fig. 5c). Combining mean transcript values and half- promoter nucleosome rearrangement and/or, as shown previously, by life data (Figs. 1 and 5) indicates PHO84 AS RNA transcription fre- placing repressive histone marks34. Antisense transcription is not able quencies of only 1 and 3 RNAs per hour in wild-type and ∆rrp6 cells, to establish stable repressive marks, as cells rapidly induce PHO84 respectively (Supplementary Table 1). These numbers are consistent sense expression when shifted from high-phosphate medium, a condi- with the incidence of nascent transcripts, another measure for tran- tion in which antisense RNA is abundant, to low-phosphate medium scription frequency. In ∆rrp6, 20% of cells show a nuclear PHO84 AS (Supplementary Fig. 6). Antisense transcription might therefore act signal (Fig. 2c), suggesting that a cell contains a nascent mRNA 20% as a buffer, protecting cells from responding to weak signals. of the time—that is, for 12 min every hour. Assuming that transcrip- The H3K4me3 and H3K4me2 marks deposited by Set1 have been tion of antisense RNA occurs at a rate similar to that for other low- implicated in promoting antisense-mediated gene repression49,54 frequency transcribed genes (0.8kb/min) and that termination and/or by recruiting, respectively, the HDACs Rpd3L and Set3 at promoter transcript release is a rate-limiting step, as has been suggested for regions24,29,55. We observed that besides the Hda1 complex, PHO84 mRNAs, transcription of the 2.3-kb antisense RNA would take almost antisense–dependent repression also depends on Set1 and Rpd3 4 min to complete36. This fits well with a transcription frequency of (J. Zaugg, M.C., N. Luscombe and F.S., personal communication). 3 PHO84 AS RNAs per hour, as a nascent antisense signal would be Thus, in addition to promoting antisense production, H3K4 meth- detected 3 times per hour during 4 min. Consistently, pap1-1∆rrp6 ylation deposited by Set1 during antisense transcription may also cells accumulate, on average, 3.7 PHO84 AS RNAs after 1-h heat shock contribute to PHO84 repression by enhancing HDAC recruitment (Fig. 3b). These data indicate that continuous but low-frequency anti- to the sense promoter. sense RNA transcription occurs in cells not expressing sense. Recent global studies show that many chromatin regulators, including Set1, barely affect steady-state gene expression but are required Rrp6 and Set1 influence antisense early termination by Nrd1 for rapid transcriptional responses to environmental stresses. Many Antisense RNA transcription frequency is greater in ∆rrp6 than in of these highly regulated genes are associated with distal or antisense wild-type strains and is accompanied by a higher fraction of cells with ncRNA transcription29,30,49. Consistently, our large-scale search for a repressed PHO84 gene. Regulating antisense transcription frequency PHO84-like genes—that is, those repressed by antisense transcrip- © 2013 Nature America, Inc. All rights reserved. America, Inc. © 2013 Nature could therefore be a way to modulate the strength of repression. tion in ∆rrp6 in a process dependent on Set1 and the HDACs Rpd3 Transcription frequency of PHO84 AS RNA appears to be controlled and Hda1—identified highly regulated TATA box–containing genes both at the level of initiation and through the regulation of elongation (J. Zaugg, M.C. and F.S., personal communication). These genes npg and termination efficiency of a short transcript by the NNS complex. It are frequently expressed in transcription bursts, and their promot- is unclear what controls initiation; the presence of a nucleosome-free ers undergo important chromatin rearrangements upon activation region in the 3′ untranslated region of PHO84 may be sufficient to or repression, as described for PHO84 (refs. 32,33). Thus, a larger allow low-frequency transcription of antisense RNA18. This ‘default’ picture emerges that suggests the role of noncoding transcription antisense transcription may be further controlled by the NNS termina- may be to reinforce the rapid on-off switch of highly regulated genes tion pathway. Indeed, mutagenesis of Nrd1- and Nab3-binding motifs by promoting the formation of repressive chromatin. This process or depletion of Nrd1 results in increased antisense levels. Moreover, occurs in wild-type cells and is enhanced in ∆rrp6. Further studies the association of Nrd1 with the PHO84 3′ end is strongly reduced in will address how, following a sense transcription burst, low-frequency ∆rrp6, suggesting that Rrp6 may contribute to antisense early termina- antisense transcription contributes to efficient nucleosome reassem- tion by favoring stable NNS complex association (Fig. 6). Notably, as bly at the promoter, thereby preventing inappropriate transcription recently observed47, Set1 has opposite effects because its loss increases factor binding and firing of sense transcription. Nrd1 binding (Fig. 6c), suggesting that Set1 and/or H3K4 methylation may interfere with early termination efficiency. These observations are Methods consistent with the positive effect of Set1 and H3K4me3 on antisense Methods and any associated references are available in the online RNA production at PHO84 and other antisense-producing genes48,49 version of the paper. (J. Zaugg, M.C. and F.S., personal communication). Interestingly, both gene-specific and genome-wide studies suggest that TRAMP and exo- Note: Supplementary information is available in the online version of the paper. some components are required for snRNA and/or snoRNA transcription termination by Nrd1, and loss of Trf4 was shown to reduce Nrd1 Acknowledgments 50,51 We thank T.H. Jensen (Aarhus University, Denmark), D. Libri (Centre National binding to snRNA genes . Together with our results, these observa- de la Recherche Scientifique, Gif-sur-Yvette, France), A. Morillon (Curie Institute, tions support the view that both TRAMP and Rrp6 may more gener- Paris) and D. Tollervey (University of Edinburgh, UK) for strains; K. Weis ally contribute to efficient NNS-dependent transcription termination. (University of California, Berkeley) for communicating data before publication;

nature structural & molecular biology VOLUME 20 NUMBER 7 JULY 2013 857 a r t i c l e s

D. Larson (National Cancer Institute, Bethesda, Maryland) for updates of 24. van Werven, F.J. et al. Transcription of two long noncoding RNAs mediates mating- image analysis software; and M. Oeffinger, F. Robert, O. Gahura, A. Maffioletti, type control of gametogenesis in budding yeast. Cell 150, 1170–1181 (2012). C. Dargemont, D. Libri and V. Géli for discussions and critical reading of the 25. Gelfand, B. et al. Regulated antisense transcription controls expression of cell- type-specific genes in yeast. Mol. Cell Biol. 31, 1701–1709 (2011). manuscript. This work was supported by SystemsX and Novartis fellowships 26. Hongay, C.F., Grisafi, P.L., Galitski, T. & Fink, G.R. Antisense transcription controls (M.C.); an EMBO fellowship (E.G.); the Swiss National Science Foundation (grant cell fate in Saccharomyces cerevisiae. Cell 127, 735–745 (2006). no. 31003A_130292) and National Center of Competence in Research “Frontiers 27. Houseley, J., Rubbi, L., Grunstein, M., Tollervey, D. & Vogelauer, M. A ncRNA in Genetics,” IGE3 and the Canton of Geneva (F.S.); and the Canadian Institutes modulates histone modification and mRNA induction in the yeast GAL gene cluster. of Health Research (MOP-BMB-232642), the Canadian Foundation for Mol. Cell 32, 685–695 (2008). Innovation and the ‘Fonds de recherche du Québec–Santé’ (Chercheur 28. Pinskaya, M., Gourvennec, S. & Morillon, A. H3 lysine 4 di- and tri-methylation Boursier Junior I) (D.Z.). deposited by cryptic transcription attenuates promoter activation. EMBO J. 28, 1697–1707 (2009). AUTHOR CONTRIBUTIONS 29. Kim, T., Xu, Z., Clauder-Munster, S., Steinmetz, L.M. & Buratowski, S. Set3 HDAC mediates effects of overlapping noncoding transcription on gene induction kinetics. M.C. performed ChIP and RNA analyses; S.R. performed the smFISH experiments; Cell 150, 1158–1169 (2012). E.G. prepared strains and performed RNA analyses; V.I. made mutants and RNA 30. Weiner, A. et al. Systematic dissection of roles for chromatin regulators in a yeast analyses. M.C., S.R., D.Z. and F.S. analyzed that data; D.Z. and F.S. supervised the stress response. PLoS Biol. 10, e1001369 (2012). project and wrote the manuscript. 31. Komeili, A. & O’Shea, E.K. Roles of phosphorylation sites in regulating activity of the transcription factor Pho4. Science 284, 977–980 (1999). COMPETING FINANCIAL INTERESTS 32. Lam, F.H., Steger, D.J. & O’Shea, E.K. Chromatin decouples promoter threshold The authors declare no competing financial interests. from dynamic range. Nature 453, 246–250 (2008). 33. Wippo, C.J. et al. Differential cofactor requirements for histone eviction from two Reprints and permissions information is available online at http://www.nature.com/ nucleosomes at the yeast PHO84 promoter are determined by intrinsic nucleosome reprints/index.html. stability. Mol. Cell Biol. 29, 2960–2981 (2009). 34. Camblong, J., Iglesias, N., Fickentscher, C., Dieppois, G. & Stutz, F. Antisense RNA stabilization induces transcriptional gene silencing via histone deacetylation in 1. Neil, H. et al. Widespread bidirectional promoters are the major source of cryptic S. cerevisiae. Cell 131, 706–717 (2007). transcripts in yeast. Nature 457, 1038–1042 (2009). 35. Femino, A.M., Fay, F.S., Fogarty, K. & Singer, R.H. Visualization of single RNA 2. David, L. et al. A high-resolution map of transcription in the yeast genome. transcripts in situ. Science 280, 585–590 (1998). Proc. Natl. Acad. Sci. USA 103, 5320–5325 (2006). 36. Zenklusen, D., Larson, D.R. & Singer, R.H. Single-RNA counting reveals alternative 3. Xu, Z. et al. Bidirectional promoters generate pervasive transcription in yeast. modes of gene expression in yeast. Nat. Struct. Mol. Biol. 15, 1263–1271 (2008). Nature 457, 1033–1037 (2009). 37. Raj, A., van den Bogaard, P., Rifkin, S.A., van Oudenaarden, A. & Tyagi, S. Imaging 4. Jacquier, A. The complex eukaryotic transcriptome: unexpected pervasive individual mRNA molecules using multiple singly labeled probes. Nat. Methods transcription and novel small RNAs. Nat. Rev. Genet. 10, 833–844 (2009). 5, 877–879 (2008). 5. Houseley, J. & Tollervey, D. The many pathways of RNA degradation. Cell 136, 38. Zenklusen, D. & Singer, R.H. Analyzing mRNA expression using single mRNA 763–776 (2009). resolution fluorescent in situ hybridization. Methods Enzymol. 470, 641–659 6. LaCava, J. et al. RNA degradation by the exosome is promoted by a nuclear (2010). polyadenylation complex. Cell 121, 713–724 (2005). 39. Oeffinger, M. & Zenklusen, D. To the pore and through the pore: a story of mRNA 7. Thiebaut, M., Kisseleva-Romanova, E., Rougemaille, M., Boulay, J. & Libri, D. export kinetics. Biochim. Biophys. Acta 1819, 494–506 (2012). Transcription termination and nuclear degradation of cryptic unstable transcripts: 40. Rougemaille, M. et al. Dissecting mechanisms of nuclear mRNA surveillance in a role for the nrd1-nab3 pathway in genome surveillance. Mol. Cell 23, 853–864 THO/sub2 complex mutants. EMBO J. 26, 2317–2326 (2007). (2006). 41. Nonet, M., Scafe, C., Sexton, J. & Young, R. Eucaryotic RNA polymerase conditional 8. Vanácová, S. et al. A new yeast poly(A) polymerase complex involved in RNA quality mutant that rapidly ceases mRNA synthesis. Mol. Cell Biol. 7, 1602–1611 control. PLoS Biol. 3, e189 (2005). (1987). 9. Wyers, F. et al. Cryptic pol II transcripts are degraded by a nuclear quality control 42. Grigull, J., Mnaimneh, S., Pootoolal, J., Robinson, M.D. & Hughes, T.R. pathway involving a new poly(A) polymerase. Cell 121, 725–737 (2005). Genome-wide analysis of mRNA stability using transcription inhibitors and 10. Arigo, J.T., Eyler, D.E., Carroll, K.L. & Corden, J.L. Termination of cryptic unstable microarrays reveals posttranscriptional control of ribosome biogenesis factors. transcripts is directed by yeast RNA-binding proteins Nrd1 and Nab3. Mol. Cell Mol. Cell Biol. 24, 5534–5547 (2004). 23, 841–851 (2006). 43. Krogan, N.J. et al. The Paf1 complex is required for histone H3 methylation by 11. Vasiljeva, L. & Buratowski, S. Nrd1 interacts with the nuclear exosome for 3′ COMPASS and Dot1p: linking transcriptional elongation to histone methylation. © 2013 Nature America, Inc. All rights reserved. America, Inc. © 2013 Nature processing of RNA polymerase II transcripts. Mol. Cell 21, 239–248 (2006). Mol. Cell 11, 721–729 (2003). 12. Carroll, K.L., Ghirlando, R., Ames, J.M. & Corden, J.L. Interaction of yeast RNA- 44. Ng, H.H., Robert, F., Young, R.A. & Struhl, K. Targeted recruitment of Set1 histone binding proteins Nrd1 and Nab3 with RNA polymerase II terminator elements. methylase by elongating Pol II provides a localized mark and memory of recent RNA 13, 361–373 (2007). transcriptional activity. Mol. Cell 11, 709–719 (2003). npg 13. Steinmetz, E.J., Conrad, N.K., Brow, D.A. & Corden, J.L. RNA-binding protein Nrd1 45. Creamer, T.J. et al. Transcriptome-wide binding sites for components of the directs poly(A)-independent 3′-end formation of RNA polymerase II transcripts. Saccharomyces cerevisiae non-poly(A) termination pathway: Nrd1, Nab3, and Sen1. Nature 413, 327–331 (2001). PLoS Genet. 7, e1002329 (2011). 14. Gudipati, R.K., Villa, T., Boulay, J. & Libri, D. Phosphorylation of the RNA 46. Wlotzka, W., Kudla, G., Granneman, S. & Tollervey, D. The nuclear RNA polymerase II polymerase II C-terminal domain dictates transcription termination choice. surveillance system targets polymerase III transcripts. EMBO J. 30, 1790–1803 Nat. Struct. Mol. Biol. 15, 786–794 (2008). (2011). 15. Kim, H. et al. Gene-specific RNA polymerase II phosphorylation and the CTD code. 47. Soares, L.M. & Buratowski, S. Yeast Swd2 is essential because of antagonism Nat. Struct. Mol. Biol. 17, 1279–1286 (2010). between Set1 histone methyltransferase complex and APT (associated with Pta1) 16. Vasiljeva, L., Kim, M., Mutschler, H., Buratowski, S. & Meinhart, A. The Nrd1– termination factor. J. Biol. Chem. 287, 15219–15231 (2012). Nab3–Sen1 termination complex interacts with the Ser5-phosphorylated RNA 48. Camblong, J. et al. Trans-acting antisense RNAs mediate transcriptional gene polymerase II C-terminal domain. Nat. Struct. Mol. Biol. 15, 795–804 (2008). cosuppression in S. cerevisiae. Genes Dev. 23, 1534–1545 (2009). 17. Xu, Z. et al. Antisense expression increases gene expression variability and locus 49. Margaritis, T. et al. Two distinct repressive mechanisms for histone 3 lysine 4 interdependency. Mol. Syst. Biol. 7, 468 (2011). methylation through promoting 3′-end antisense transcription. PLoS Genet. 8, 18. Murray, S.C. et al. A pre-initiation complex at the 3′-end of genes drives antisense e1002952 (2012). transcription independent of divergent sense transcription. Nucleic Acids Res. 40, 50. Grzechnik, P. & Kufel, J. Polyadenylation linked to transcription termination directs 2432–2444 (2012). the processing of snoRNA precursors in yeast. Mol. Cell 32, 247–258 (2008). 19. Tisseur, M., Kwapisz, M. & Morillon, A. Pervasive transcription—Lessons from yeast. 51. Gudipati, R.K. et al. Extensive degradation of RNA precursors by the exosome in Biochimie 93, 1889–1896 (2011). wild-type cells. Mol. Cell 48, 409–421 (2012). 20. Hainer, S.J., Pruneski, J.A., Mitchell, R.D., Monteverde, R.M. & Martens, J.A. 52. Darby, M.M., Serebreni, L., Pan, X., Boeke, J.D. & Corden, J.L. The S. cerevisiae Intergenic transcription causes repression by directing nucleosome assembly. Genes Nrd1-Nab3 transcription termination pathway acts in opposition to Ras signaling Dev. 25, 29–40 (2011). and mediates response to nutrient depletion. Mol. Cell Biol. 32, 1762–1775 21. Martens, J.A., Laprade, L. & Winston, F. Intergenic transcription is required to (2012). repress the Saccharomyces cerevisiae SER3 gene. Nature 429, 571–574 (2004). 53. Lardenois, A. et al. Execution of the meiotic noncoding RNA expression program 22. Thiebaut, M. et al. Futile cycle of transcription initiation and termination modulates and the onset of gametogenesis in yeast require the conserved exosome subunit the response to nucleotide shortage in S. cerevisiae. Mol. Cell 31, 671–682 Rrp6. Proc. Natl. Acad. Sci. USA 108, 1058–1063 (2011). (2008). 54. van Dijk, E.L. et al. XUTs are a class of Xrn1-sensitive antisense regulatory non- 23. Bumgarner, S.L. et al. Single-cell analysis reveals that noncoding RNAs coding RNA in yeast. Nature 475, 114–117 (2011). contribute to clonal heterogeneity by modulating transcription factor recruitment. 55. Buratowski, S. & Kim, T. The role of cotranscriptional histone methylations. Mol. Cell 45, 470–482 (2012). Cold Spring Harb. Symp. Quant. Biol. 75, 95–102 (2010).

858 VOLUME 20 NUMBER 7 JULY 2013 nature structural & molecular biology ONLINE METHODS isothiocyanate (FITC)), SP-102v1 (Cy3/DyLight550), SP-103v1 (Cy3.5/ Strains, media and culture conditions. The yeast strains used in this study are DyLight594) and CP-104 (Cy5/DyLight650) (Chroma Technology)) corre- listed in Supplementary Table 2. Yeast strains were streaked on YEPD plates at sponding to the excitation and emission spectra of the smFISH probes used. 25 °C. Liquid cultures were inoculated with cells taken from plates and grown at Three-dimensional image data sets were acquired, with 200 mn z stacks covering 25 °C for 16–24 h under exponential growth conditions (OD600 < 0.8) in YEPD the entire depth of cells. The z stacks were projected onto a two-dimensional plane or synthetic complete (SC) minimal medium. by applying a maximum projection using ImageJ. RNA signals were detected and quantified using a spot-detection algorithm fitting a two-dimensional Gaussian Fluorescent in situ hybridization. Fluorescent in situ hybridization procedure. mask implemented with custom-made software for the IDL platform (ITT Visual Twenty-nucleotide-long DNA oligonucleotides containing a single 3′ amine were Information Solutions); cell and nuclear segmentation as well as quantification of labeled post-synthesis with amine-reactive fluorescent dyes and hybridized to nascent RNAs were performed all as described in ref. 36. For all quantifications, paraformaldehyde-fixed yeast cells as described in refs. 36 and 37. Images were data from at least three different experiments were analyzed, each containing acquired using epifluorescence microscopy, and three-dimensional data sets >100 cells. were reduced to two-dimensional data sets for image analysis. Cell segmentation, single-RNA counting and quantification of nascent transcripts were done Plasmid constructions. The PHO84 plasmid with the mutated Nrd1 and Nab3 as described in ref. 36. motifs was obtained by first cloning the PHO84 wild-type gene (from −1,000 bp Probe design and labeling. Twenty-nucleotide-long DNA oligonucleotide to +350 bp) as a SalI fragment into pUC18 to create pFS3594. A BglII-Nde1 DNA probes were designed using the online software Stellaris Probe Designer version fragment spanning the PHO84 3′ end and downstream vector sequences was 2.0 at the Biosearch Technologies website. Probes typically have a 50% GC con- synthesized by mutagenizing the putative Nrd1 and Nab3 binding motifs encoded tent; however, GC content can range from 40% to 55% (for probe sequences see within the PHO84 3′ end on the antisense strand. The wild-type BglII-Nde1 Supplementary Table 2). Probes were synthetized containing a single 3′ amine fragment of pFS3521 was replaced by the synthetic mutant fragment to create that can be coupled to an amine-reactive fluorescent dye. For a typical labeling pFS3644. Both the wild-type and mutant PHO84 were subcloned into YCpLac111 reaction, 20 µg of pooled probes (31 for PHO84 antisense, 30 for PHO84 sense as SalI fragments to generate pFS3521 and pFS3625, respectively. and 48 probes for MDN1) were lyophilized, resuspended in labeling buffer (0.1 M sodium bicarbonate, pH 9.0) and mixed with a single reactive dye pack of amine- Northern blot analysis and RT-qPCR. Total RNA was prepared and analyzed reactive dye (DyLight amine-reactive dyes: DyLight 550 (#62263), DyLight 594 by northern blotting using standard methods as described in ref. 34. For RT- (#46413) and DyLight 650 (#62266) (Thermo Scientific)). The reaction was car- qPCR quantifications, total RNA was treated with DNase (Ambion) to remove ried out overnight in the dark at room temperature. Labeled probes were puri- genomic DNA contamination. cDNAs of sense or antisense RNAs were generated fied using the Quiagen QIAquick Nucleotide Removal columns (Qiagen #28304) by SuperScript II reverse transcriptase (Invitrogen) with 1 µg of DNase-treated according to the manufacturer’s instructions. Probe concentration and labeling total RNA using gene- and strand-specific primers. cDNAs were quantified by efficiency were measured using NanoPhotometer Pearl (Implen) and calculated RT-qPCR (Bio-Rad). The same amplicon was used to quantify sense and antisense as described in ref. 38. Probes were stored at −20 °C in the dark. cDNA. The sequences of all the primers are listed in Supplementary Table 2. Cell fixation, preparation, storage and hybridization. Cells were grown in SD complete and 2% glucose at 25 °C overnight to mid-log phase (OD600 = 0.6–0.8) Chromatin immunoprecipitation. ChIPs were performed essentially as 34 and fixed by adding paraformaldehyde (Electron Microscopy Science #15714) described previously . Yeast strains were grown to OD600 = 0.8 in YEPD medium to a final concentration of 4% for 45 min at room temperature. Cells were sub- at 25 °C and cross-linked for 10 min by the addition of formaldehyde to a final sequently washed 3× with 10 ml of Buffer B (1.2 M sorbitol, 100 mM KHPO4, concentration of 1.2%. Cross-linked and sonicated chromatin extracts from pH 7.5) and stored overnight at 4 °C in Buffer B. Cell walls were then digested with 1.5 mg of Bradford-quantified proteins were immunoprecipitated overnight in lyticase (Sigma #L2524, dissolved in 1× PBS to 25,000 U/ml. Stored at −20 °C). the presence of protein G Sepharose (Amersham, Pharmacia) with 5 µl of anti- Digested cells were plated on poly-l-lysine–treated coverslips and stored in 70% body against H3K4me3 (Abcam 8580), H3K4me2 (Abcam 32356), H3 (Abcam

© 2013 Nature America, Inc. All rights reserved. America, Inc. © 2013 Nature ethanol at −20 °C in 12-well cell culture plates. Cells can be stored in 70% ethanol 1791, clone Y47) or HA epitope (Covance monoclonal antibody HA.11, clone for several months before hybridization. For hybridization, cells were removed 16B12) for the Nrd1–HA tagged strains. All immunoprecipitations were repeated from 70% ethanol, washed twice with 2× saline sodium citrate (SSC) and hydrated at least three times with different chromatin extracts from independent cultures. in 10% formamide/2× SSC. Labeled probes were resuspended in 10% (v/v) for- Immunoprecipitated DNA was purified and quantified by qPCR with the primers npg mamide, 2× SSC, 1 mg ml–1 BSA, 10 mM ribonucleoside vanadyl complex (NEB listed in Supplementary Table 2 and expressed as the percent of input DNA or –1 –1 #S1402S), 5 mM NaHPO4, pH7.5, 0.5 mg ml Escherichia coli tRNA and 0.5 mg ml percent of input DNA normalized to H3. single-stranded DNA and hybridized overnight at 37 °C. Cells were then washed in 10% formamide/2× SSC at 37 °C for 1 h, followed by a quick wash in Determination of decay rates. Cells were grown to OD600 = 0.8 in YEPD 1× PBS at room temperature. The coverslips were quickly dried in 100% ethanol medium. At T = 0, 100 µg/ml of 1,10-phenanthroline (Sigma) was added to and mounted on glass slides using Prolong Gold with DAPI mounting medium the culture (as described in ref. 42 and references therein). Samples were taken (Invitrogen #P36935). For a more detailed protocol, see refs. 36 and 38. at different time points and analyzed for their RNA expression by RT-qPCR Image acquisition and analysis. Images were acquired using an epifluores- as described above. cence microscope, either a Nikon E800 upright microscope equipped with a Half-lives were calculated by the equation t1/2 = 0.693/k, in which k is the Photometrics CoolSNAP HQ (CCD) camera or a Zeiss Axio Observer Z1 inverse rate constant for mRNA decay. Values of each time point are normalized for microscope with a Zeiss AxioCam MRm camera, using 100× oil-objective internal variations with SCR1 RNA, a control that is still stable at the 30-min and specific filter cubes (Chroma Filters 31000 (DAPI), 41001 (fluorescein time point.

doi:10.1038/nsmb.2598 nature structural & molecular biology Supplementary information for:

Bimodal expression of PHO84 is modulated by early termination of antisense transcription

Manuele Castelnuovo1,3, Samir Rahman2,3, Elisa Guffanti1,3, Valentina Infantino1, Françoise Stutz1 & Daniel Zenklusen2

3These authors contributed equally to this work.

Correspondence should be addressed to F.S. ([email protected]) or D.Z. ([email protected]).

This supplemental information includes: Supplementary Figures 1 to 6 Supplementary Tables 1 and 2 Supplementary References

Nature Structural & Molecular Biology: doi:10.1038/nsmb.2598 Nature Structural & Molecular Biology: doi:10.1038/nsmb.2598 Nature Structural & Molecular Biology: doi:10.1038/nsmb.2598 Nature Structural & Molecular Biology: doi:10.1038/nsmb.2598 Supplementary Figure 4 a

CTTGTTGATCCCTGAGACAAAACGTAAAACGTTGGAGGAGATCAATGAGCTATACCACGATGAAATCGATCCTGCCACTCTAAATTTTCGTAATAAAAACAACGATATCGAATCTTCCAGCCCATCTCAGTTGCAGCACGAGGCTTGAAAGCCTCAAAGATGCACTAAAACTTGTAAACTAG mutant |||||||||||||||#||#||##|##|#||##|#||#||#||#||#||#||||||||||||||||||||||||||#||#|||||#||##|#||#||#||#||#||#||#||||||||||||||||||||##|#||#||#||#||#|#||||||||||||||||||||||||||||||||||| CTTGTTGATCCCAGAAACTAAGAGAAAGACTCTAGAAGAAATTAACGAGCTATACCACGATGAAATCGATCCTGCTACGCTAAACTTCAGAAACAAGAATAATGACATTGAATCTTCCAGCCCATCTCAACTTCAACATGAAGCATAAAAGCCTCAAAGATGCACTAAAACTTGTAAACTAG WT STOP

Nab3 binding sites UAS PHO84 Nrd1 binding sites ATG STOP Nrd1-Nab3 terminated AS RNAs PAR-CLiP signal CPF-Pap1 terminated AS RNAs Pho4 binding sites

b 10 PHO84 AS sio n 8 pre s

6 pAS-PHO84 WT 4 pAS-PHO84 mut.

ex 1-normalized 2

ACT 0

rrp6 Δ pho84 Δ pho84 Δ Supplementary Figure 4: Nrd1 and Nab3 binding site mutations increase AS RNA production. (a) Schematic view of the mutagenesis of Nrd1 and Nab3 binding sites at the PHO84 3’end. The Nrd1 PAR-CLiP signal detected over the same region (Creamer et al., 2011) 45 is indicated above. Nrd1 and Nab3 recognition motifs were defined based on the PAR-CliP binding data and the consensus binding sites derived from these analyses (Creamer et al., 2011) 45. The WT and mutated PHO84 coding strands are shown. The mutations introduced maintain the PHO84 open reading frame. The sequence of theAS RNA corresponds to the reverse and complement of the indicated sequences. The 3’ ends of early terminated PHO84 AS transcripts have been described (Neil et al., 2009) 1. (b) Analysis of PHO84 AS RNA levels in a Δpho84 strain transformed with a PHO84 plasmid containing the wild-type (pAS-PHO84 WT) or the mutant (pAS-PHO84 mut.) sequence, in which Nrd1 and Nab3 motifs on the AS orientation are mutagenized. Plasmid encoded PHO84 AS Nature Structural & Molecular Biology: doi:10.1038/nsmb.2598 transcripts were measured by RT-qPCR with specific primers as described in Methods. CPF: cleavage and polyadenlytion factor. Nature Structural & Molecular Biology: doi:10.1038/nsmb.2598 Nature Structural & Molecular Biology: doi:10.1038/nsmb.2598

Mean number of total Half-life (min) measured by PHO84 AS RNAs 1,10 Phenanthroline induced Transcription frequency per cell transcription inhibition (RNAs/hour)

Wild type 0.36±0.098 11.4±3.9 1.31

rrp6 0.82±0.18 12±2.8 2.84

Supplementary Table 1: Transcription frequency in wild-type and rrp6 cells. mRNA half-life was calculated using mean PHO84 AS expression levels in wild-type and rrp6 strains measured by smFISH and decay rates measured by qRT-PCR after transcription shutoff by 1,10 Phenanthroline. Assuming mRNA decay follows first-order kinetics, transcription frequency can be calculated using (Holstege et al. 1998; Wang et al. 2002) 56,57 :

Transcription frequency = Ln2 x (steady state RNA level)/half life (min)

Nature Structural & Molecular Biology: doi:10.1038/nsmb.2598 Supplementary Table 2

STRAINS USED IN THIS STUDY

Code Name Genotype Reference

W303 background FSY1742 WT MATa ade2 his3 leu2 trp1 ura3 FSY3117 Δrrp6 MATa ade2 his3 leu2 trp1 ura3 Δrrp6::Kanr Camblong et al., 2007 34 34 FSY3518 Δhda2 MATa ade2 his3 leu2 trp1 ura3 Δhda2::TRP1 Camblong et al., 2007 FSY3018 Δhda2Δrrp6 MATa ade2 his3 leu2 trp1 ura3 Δrrp6::Kanr Δhda2::TRP1 this study FSY3517 Δset1 MATa ade2 his3 leu2 trp1 ura3 Δset1::TRP1 this study FSY3833 Δset1Δrrp6 MATa ade2 his3 leu2 trp1 ura3 Δrrp6::Kanr Δset1::TRP1 this study FSY1982 mex67-5 MATa ade2 his3 leu2 trp1 ura3 mex67-5 integrated Jimeno et al., 2002 58 FSY1985 mex67-5 Δrrp6 MATa ade2 his3 leu2 trp1 ura3 mex67-5 integrated Δrrp6::Kanr this study FSY2078 Δxrn1 MATa ade2 his3 leu2 trp1 ura3 Δxrn1::TRP1 Gift from F. Lacroute FSY1968 pap1-1 MATa ade2 his3 leu2 trp1 ura3 pap1-1 Dower K. and Rosbash M., 2002 59 9 FSY1988 pap1-1 Δrrp6 MATa ade2 his3 leu2 trp1 ura3 pap1-1 Δrrp6::Kanr Wyers et al., 2005 FSY4838 pap1-1 Δtrf4 (DL705) MATa ade2 his3 leu2 trp1 ura3 pap1-1 Δtrf4::URA3 Gift from D. Libri FSY4275 GAL-NRD1 MATa ade2 his3 leu2 trp1 ura3 HisMX6-pGAL-NRD1 Thiebaut et al., 2006 7 7 FSY4282 GAL-NRD1 Δrrp6 MATa ade2 his3 leu2 trp1 ura3 HisMX6-pGAL-NRD1 Δrrp6::Kanr Thiebaut et al., 2006 FSY2527 Δpho4 MATa ade2 his3 leu2 trp1 ura3 Δpho4::Kanr this study FSY3313 Δpho4Δrrp6 MATa ade2 his3 leu2 trp1 ura3 Δpho4::Kanr Δrrp6::TRP1 this study FSY4265 Δpho4Δset1 MATa ade2 his3 leu2 trp1 ura3 Δpho4::Kanr Δset1::HIS3 this study FSY4266 Δpho4Δset1Δrrp6 MATa ade2 his3 leu2 trp1 ura3 Δpho4::Kanr Δset1::HIS3 Δrrp6::TRP1 this study FSY4911 NRD1-HA Δpho4 MATa ade2 his3 leu2 trp1 ura3 NRD1-HA-HIS3 Δpho4::Kanr this study FSY4918 NRD1-HA Δpho4 Δrrp6 MATa ade2 his3 leu2 trp1 ura3 NRD1-HA-HIS3 Δpho4::Kanr Δrrp6::TRP1 this study FSY4912 NRD1-HA Δpho4 Δset1 MATa ade2 his3 leu2 trp1 ura3 NRD1-HA-HIS3 Δpho4::Kanr Δset1::TRP1 this study FSY4841 rpb1-1 MATa ade2 his3 leu2 trp1 ura3 rpb1-1 Wyers et al., 2005 9 FSY4842 rpb1-1 Δrrp6 (DL 713) MATa ade2 his3 leu2 trp1 ura3 rpb1-1 Δrrp6::URA3 Gift from D. Libri 48 FSY3799 Δpho84 MATa ade2 his3 leu2 trp1 ura3 Δpho84::Kanr Camblong et al., 2009 48 FSY3811 Δpho84 Δrrp6 MATa ade2 his3 leu2 trp1 ura3 Δpho4::Kanr Δrrp6::TRP1 Camblong et al., 2009

BY4741 background FSY3158 WT MATa his3 leu2 lys2 ura3 Euroscarf 60 FSY4747 Δrrp6 MATa his3 leu2 lys2 ura3 Δrrp6::NatMX Houseley J. & Tollervey D., 2006 FSY3182 Δtrf4 MATa ade2 his3 leu2 ura3 Δtrf4::Kanr " FSY4748 Δtrf4Δrrp6 MATa ade2 his3 leu2 ura3 Δtrf4::Kanr Δrrp6::NatMX " FSY4749 Δtrf4 GAL-TRF5 MATa ade2 his3 leu2 ura3 Δtrf4::Kanr HisMX6-pGAL-3HA::trf5 "

PRIMERS USED IN THIS STUDY

Code Name Sequence

OFS1741 ACT1 F Mid 5'-TTCCAGCCTTCTACGTTTCCATC-3' OFS1742 ACT1 R Mid 5'-CGTGAGGTAGAGAGAAACCAGC-3' OFS735 ACT1 F 3' 5'-TACTCCGTCTGGATTGGTGGTT-3' OFS736 ACT1 R 3' 5'-GGTGAACGATAGATGGACCACTT-3' OFS737 ACT1 F 5' 5'-TGGTATGTTCTAGCGCTTGCAC-3' OFS738 ACT1 R 5' 5'-GTCAATATAGGAGGTTATGGGAGAGTG-3' OFS1077 PHO84 F 3' 5'-GAAATTAACGAGCTATACCACGATGAAATC-3' OFS1078 PHO84 R 3' 5'-CATGTTGAAGTTGAGATGGGCTGG-3' OFS1158 PHO84 F M 5'-CTGCCGCACAAGAACAAGATGG-3' OFS1159 PHO84 F M 5'-TTTGGAGGATGATTGTCAAGAGATTCG-3' OFS1075 PHO84 F 5' 5'-CCGTCAATAAAGATACTATTCATGTTGCTG-3' OFS1076 PHO84 F 5' 5'-AAAATCATTCAAATGGTTGTGGAAGGC-3' OFS1717 SCR1 F 5'-AACCGTCTTTCCTCCGTCGTAA-3' OFS1718 SCR1 R 5'-CTACCTTGCCGCACCAGACA-3' OFS1249 PHO84 -1000 SalI F 5'-GGGGGGGGGGGTCGACCGAGAGTGATAAAGAAGAGGCGGT-3' OFS1363 PHO84 +355 Sal1 R 5'-GGGGGGGGGGGTCGACGTCTCAAGTCGCTTGCTTAGTCGA-3'

smFISH PROBES USED IN THIS STUDY

PHO84 mRNA, PHO84 AS and MDN1 probes were labeled in a single position with eigther Dylight 549 or DyLight 594 36 ITS2-1 probe was labeled in 3 positions with cy5 (Zenklusen et al., 2008) . Labeled positions are shown in bold.

PHO84 sense PHO84 AS ITS2-1 MDN1

ccttcggtaaggtgttcttt cccatctcaacttcaacatg GAT ATG CTT AAG TTC AGC GGG TAC TCC TAC CTG ATT TGA GGT C cagagggaaaagcagaattg aaatggttgtggaaggccat ctgctacgctaaacttcaga ttgttgctaaagtggaaggg cttctttccagaggatcttc gctataccacgatgaaatcg cggctatgtagtatagttcc aaaccttcgtcatcgatgga ccttgttgatcccagaaact gaaaacatgaccagtgatgg ggtcttaacttgttgccaac gccttattcatgttgttggg caacgtgccactatctctaa aaaccaacaccagcaatgga gttacctcacgtcatggaaa ggcattgcaacgggaaatat ggacatcatagtgataccca aactgtgctagagacggtaa gaatccttttgtgtggatgg atactaccgtgccagtaaac tttgggtactctaatcgacc ggttgaatgaagcgtgcaaa acaaggtttgacttggacct gtgccattattgcacaaacc ccaacaatgaatcgcgtgat ccaacagaagtggaaacctt tttctgctgcatctggtaag tccagaatcgccctcaaaat gcagaatggtacagacaatc tacagatctactgctcatgg gttcgaggaagctgtaatca cccatgacaatacggtagaa aacttcggtccaaacacaac gcacatctcttagcttcgtt ggtagtggcaaattcagaag catcatcttgaccgctttgt gaacactgggtagtttagag ttagcaaagacagcacccat tcattacctggttactgggt cctgaggctcaataatgaag tgataccaccggagatttga ctgctgtcggtaatctgatt attgacgcgaccttagtact tttgcgtattctagttcgcc cgggttgagtttaaacagtg tttgtcgtggatagtgtgga aaacatgccaaccctagaac gctggttcatggtttacctt ggacaaaggttaatgggtag ccaacttagcgttaacatcc gtcaatggaagtacggtaag cgaagagaggaaaccgtttt cactggtgtcgtgaattttc ccaaaggcttcgttcaaaga ccgcttttccaatgagcatt attgtcaagagattcgacgg ggtttggaaagagcttctac gagtcatggcaacccatata cagcagtacctagcaaaatc cagtgatgaagacatggcaa ggcatcagtttgttcaccta cccgtagaaagcaacatcta cacaagaacaagatggcgaa caccagaggtataagtacca ggtttgcagaataacagcac taagttggaacttgctgctg acaccagctctccattcaaa tttttggaaccggcataacc atccttattgggttgggtac tacccatctcccttctttga atcgacagtgaagacggata aaggcttgtgaccaaatgtg cgcgcttttctaaaagcgat accgatgacacagaacaaag gctgaatgtgatgctagatg tcctctggatggaatggtta caccaagtttatggtatgcg ctcttatcttggttgctgct catttgcagcctttacagtc taaagagccaacagaccatg aaatggagaggtgccatcat cgctaggttcctctaattca ccaggaacaataaaggtggt gtgactacccactatcttct ggttggtcaaaatgggaaac gtaacgagttgggaaacact cctgctattaacttcgttgc cccccttgttcaatgaaatg gcaaaccactgttgctcatt ctaacctttcgcacagctta caggtttgttgatgccattg cgtagacagaagactggatt tgaattctccaatagcgcca ctcgccgattgcttgtataa ccttgaggaagcaatgtcta ggcacatgttgggtcaaaaa gcgaacaatgtgctgttcat ccgactagtaaaacaggttc tgaacgactgtagttttccc ccaagaagatcaccagtttc gcatcttgtgaaacttctcg gtacgcttcgttccaaagtt cagcccatttgtcaagtaac cctcaaacttcttcactgag ccctcgacaaaattgaagac gccctgatagtctttaccaa ttcatcgagcaatagccact Nature Structural & Molecular Biology: doi:10.1038/nsmb.2598 Supplementary References

56. Holstege, F.C. et al. Dissecting the regulatory circuitry of a eukaryotic genome. Cell 95, 717-28 (1998). 57. Wang, Y. et al. Precision and functional specificity in mRNA decay. Proc Natl Acad Sci U S A 99, 5860-5 (2002). 58. Jimeno, S., Rondon, A.G., Luna, R. & Aguilera, A. The yeast THO complex and mRNA export factors link RNA metabolism with transcription and genome instability. Embo J 21, 3526-35 (2002). 59. Dower, K. & Rosbash, M. T7 RNA polymerase-directed transcripts are processed in yeast and link 3' end formation to mRNA nuclear export. Rna 8, 686-97 (2002). 60. Houseley, J. & Tollervey, D. Yeast Trf5p is a nuclear poly(A) polymerase. EMBO Rep 7, 205-11 (2006).

Nature Structural & Molecular Biology: doi:10.1038/nsmb.2598