Bases of antisense lncRNA-associated regulation of gene expression in fission yeast

Antisense (as)lncRNAs can regulate gene expression but the underlying mechanisms and the different cofactors involved remain unclear. Using Native Elongating Transcript sequencing, here we show that stabilization of antisense Exo2-sensitivite lncRNAs (XUTs) results in the attenuation, at the nascent transcription level, of a subset of highly expressed genes displaying prominent promoter-proximal nucleosome depletion and histone acetylation. Mechanistic investigations on the catalase gene ctt1 revealed that its induction following oxidative stress is impaired in Exo2-deficient cells, correlating with the accumulation of an asXUT. Interestingly, expression of this asXUT was also activated in wild-type cells upon oxidative stress, concomitant to ctt1 induction, indicating a potential attenuation feedback. This attenuation correlates with asXUT abundance, it is transcriptional, characterized by low RNAPII-ser5 phosphorylation, and it requires an histone deacetylase activity and the conserved Set2 histone methyltransferase. Finally, we identified Dicer as another RNA processing factor acting on ctt1 induction, but independently of Exo2. We propose that asXUTs could modulate the expression of their paired-sense genes when it exceeds a critical threshold, using a conserved mechanism independent of RNAi.

Several lines of evidence suggest that they are functionally important.First, lncRNAs show tissue-specific expression [7] and respond to diverse stimuli, such as oxidative stress [8], suggesting that their expression is precisely controlled.Second, several lncRNAs are misregulated in diseases including cancer and neurological disorders [9,10,11].Furthermore, there is a growing repertoire of cellular processes in which lncRNAs play important roles, including Xchromosome inactivation, imprinting, maintenance of pluripotency and transcriptional regulation [12,13].
Several classes of lncRNAs have been described [2].Among them, large intervening noncoding (linc)RNAs, which result from transcription of intergenic regions, have attracted a lot of attention as being involved in cis-and trans-regulation, mostly at the chromatin level, of genes important for development and cancer [13].
Our previous studies in budding and fission yeasts revealed that aslncRNAs are globally unstable and are mainly targeted by the cytoplasmic 5'-3' RNA decay pathway dependent on the Xrn1 and Exo2 exoribonucleases in S. cerevisiae [28,29] and S. pombe [30], respectively.Inactivation of Xrn1/Exo2 leads to the stabilization of a family of lncRNAs, referred to as Xrn1-sensititve Unstable Transcripts (XUTs), the majority of which are antisense to proteincoding genes [28,29,30].Interestingly, in S. cerevisiae, we defined among these antisense (as) XUTs a subgroup for which the sense-paired genes (referred to as class 1) undergo antisensemediated transcriptional silencing [28].However, the molecular mechanism by which asXUTs could regulate sense gene expression remains largely unknown to date, still interrogating whether sense transcription is impaired at the initiation and/or elongation and/or termination stages, whether any post-transcriptional event is in play, and whether the epigenetic landscape contributes in the regulatory determinants.In addition, such a transcriptional aslncRNAmediated regulation has not been documented yet in S. pombe.
Here we used Native Elongating Transcript sequencing (NET-Seq) to identify genomewide in fission yeast the genes attenuated at the nascent transcription level upon stabilization of their paired-asXUTs.This so-called class 1 corresponds to highly transcribed genes, displaying marks of active transcription at the chromatin level.Mechanistic investigation on a model class 1 representative, the inducible catalase-coding gene ctt1, confirmed that it is transcriptionally attenuated upon oxidative stress when its paired-asXUT is stabilized, and the level of the attenuation correlates with the abundance of the asXUT.The attenuation is characterized by low RNAPII Ser5-phosphorylation (Ser5-P) and requires histone deacetylase (HDAC) activity and the conserved Set2 histone methyltransferase (HMT).Finally, we identified Dicer as an additional regulator of ctt1 induction, acting independently of Exo2 and the asXUT.Together, our data support a model where asXUTs could modulate the expression of the paired-sense genes when it exceeds a critical threshold, using a conserved mechanism independent of RNAi.

Genome-wide identification of class 1 genes in fission yeast
In budding yeast, stabilization of asXUTs results into the attenuation, at the transcriptional level, of a subset of paired-sense genes, which are referred to as class 1 [28].We recently annotated XUTs in fission yeast [30].We observed that asXUTs accumulation correlates with down-regulation of the paired-sense mRNAs, at the RNA level [30], suggesting that the regulatory potential of asXUTs has been conserved across the yeast clade.However, whether this regulation occurs at the level of transcription in fission yeast remains unclear.
Among the 723 genes transcriptionally attenuated in exo2Δ, 175 have asXUTs (class 1).Despite the proportion of class 1 genes among the attenuated genes is limited (24.2%), it is significantly higher than expected if presence of asXUT and sense gene attenuation were independent (Chi-square test, P = 0.03), suggesting that the attenuation could depend on the stabilized asXUTs, at least in some cases.On the other hand, the transcriptional down-regulation of class 2 (no asXUT) is likely to be an indirect effect reflecting the slow growth phenotype of the exo2Δ mutant [31].Consistently, class 2 is significantly enriched for GO terms "ribosome biogenesis" (P = 1.36e -08 ) and "cellular component biogenesis (P = 1.04e -02 ), and it is known that the expression of genes involved in these biological processes directly depends on the growth rate [32].Altogether, these observations suggest that for a subgroup of genes, stabilization of the asXUT might contribute to attenuate transcription of the paired-sense gene.
Both classes 1 and 3 have asXUTs, but only class 1 is transcriptionally attenuated upon asXUTs stabilization.This suggests the existence of specificities discriminating the two classes.Indeed, in WT cells, class 1 is transcribed to higher levels than class 3 (Fig 1C We also noted that in WT cells, the nascent antisense transcription signal surrounding the TSS of class 1 genes is higher than for class 3 (S1D Fig) , suggesting that sense TSS overlap could constitute a factor for the potential regulatory activity of the XUTs antisense to class 1 genes.Together, these results show that transcriptional attenuation correlates with asXUT stabilization in fission yeast, suggesting that asXUTs might be involved in the modulation of sense genes expression.

Exo2-deficient cells are defective for ctt1 induction upon oxidative stress
To further investigate the possibility that asXUTs can regulate expression of their paired-sense gene and to get insights into the underlying molecular mechanism, we characterized a class 1 gene, ctt1, and its paired-antisense ctt1 encodes a catalase, an enzyme required for survival to oxidative stress upon exposure to H 2 O 2 [33], and it is strongly induced in this condition [34].Interestingly, we observed that exo2Δ cells displayed a slight sensitivity to H 2 O 2 in addition to the slow growth and temperature sensitivity (S2B and S2C Fig) .This suggests that ctt1 expression in exo2Δ cells might also be impaired upon oxidative stress.
We therefore analyzed ctt1 mRNA induction in WT and exo2Δ cells upon oxidative stress.Northern-blot (Fig 2B ) and RT-qPCR kinetics analyses (Fig 2C ) showed that exo2Δ exhibits a 3-fold reduction in induction rate, with a peak of induction reached 15 minutes after H 2 O 2 addition vs 10 for the WT (Fig 2C).
Strikingly, we observed that XUT0794 is also activated upon oxidative stress in a WT context (Fig 2D).Furthermore, its peak of induction is reached very rapidly (5 min), before the ctt1 mRNA peak (10 min), suggesting that it might be part of a natural attenuation mechanism (feedback loop) for ctt1 expression.
In summary, our data suggest that ctt1 induction requires Exo2 activity for maintaining a low level of XUT0794, antisense to ctt1.Upon induction, the XUT is activated and could modulate expression of ctt1, in a similar way as shown for asXUT-associated genes in S. cerevisiae, such as GAL1-10 [19,20].

ctt1 attenuation level correlates with antisense XUT0794 abundance
We designed several experiments in order to test whether ctt1 attenuation directly depends on antisense XUT0794.
Firstly, we overexpressed it in cis, in WT cells, using a regulatable P41nmt1 promoter (S3A Fig) .When the promoter is active, XUT0794 accumulates and ctt1 is not induced in response to H 2 O 2 addition (S3A Fig).This demonstrates a causal role of antisense XUT0794 expression in attenuating ctt1.However, in this particular context where XUT0794 expression is driven by the strong P41nmt1 promoter, it is difficult to draw up any conclusion about a possible role of the lncRNA itself, as the ctt1 silencing observed here probably mainly results from of genes for each class is indicated.B. Box-plot of nascent transcription (NET-Seq) signal for class 1-4 genes in WT (dark grey boxes) and exo2Δ (Δ; light grey boxes).C. Metagene view of NET-Seq signals along class 1 and 3 genes in WT cells.For each class, normalized coverage (tag/nt, log 2 ) along mRNA transcription start site (TSS) +/-1000 nt (sense) and the antisense (as) strand were piled up, in a strand-specific manner.Average signal for the sense and antisense strands was plotted for class 1 (red) and class 3 (green).The shading surrounding each line denotes the 95% confidence interval.D. Same as above in exo2Δ.E. Density-plot showing the global NET-Seq (dashed lines) and total RNA-Seq (solid lines) signals for XUTs in the WT (black) and exo2Δ (pink) strains.Total RNA-Seq data were previously described [30].F. Metagene view of H4K5/8/12/16 acetylation (H4ac) for class 1 (red), class 2 (blue), class 3 (green) and class 4 (black) genes in WT cells.ChIP-Seq libraries construction and sequencing were previously described [30].Metagene representation of signal for each class of genes was performed as above, in a strand-unspecific manner, using ratio of coverage (log 2 ) for H4ac and H3.The shading surrounding each line denotes the 95% confidence interval.
Secondly, we disrupted the XUT0794 promoter in WT cells using the ura4 gene (Fig 3A), which is controlled by a promoter much weaker than P41nmt1.Surprisingly, ura4 insertion did not abolish XUT0794 expression (Fig 3B).However, it resulted in XUT0794 levels similar to those of the exo2Δ mutant (Fig 3B).Notably, ctt1 was significantly attenuated in the ura4-XUT0794 strain, and ctt1 mRNA levels were similar to those of exo2Δ cells (Fig 3C).Hence, there is a positive correlation between ctt1 attenuation and antisense XUT0794 levels.
In a third experiment, we inserted a self-cleaving hammerhead ribozyme (RZ) at position 254/815 of XUT0794 (Fig 3D ) and integrated the construct at the ctt1 locus in WT and exo2Δ strains, without any manipulation of XUT0794 promoter.In the WT + RZ context, neither the 5' nor the 3' fragment of XUT0794 accumulated (Fig 3E), and ctt1 induction was similar to WT cells (Fig 3F).In the exo2Δ + RZ context, the 5' fragment was not detected, but the 3' fragment accumulated 5x more than in exo2Δ without RZ (Fig 3E).This imbalance between the two RNA parts indicates that RZ was efficiently cleaved, the 5' fragment being presumably degraded [35] while the 3' fragment accumulated.Importantly, the higher abundance of the In each panel, the signal corresponding to the sense (+) and antisense (-) strand is shown in blue and pink, respectively.Blue and red arrows represent the ctt1 mRNA and antisense XUT0794, respectively.NET-Seq data for the WT strain were previously described [30].The snapshot was produced using VING [59].B. Northern blot analysis of ctt1 mRNA induction in WT and exo2Δ cells.YAM2400 (WT) and YAM2402 (exo2Δ) cells were grown to mid-log phase in rich medium and collected before or after addition of 1 mM H 2 O 2 for 15 min.ctt1 mRNA and U3B snoRNA were detected from total RNA using 32 P-labelled oligonucleotides.Numbers represent the ctt1 mRNA/U3B ratio (ND: not determined).C. YAM2400 (WT) and YAM2402 (exo2Δ) cells grown in rich YES medium to mid-log phase were collected 0, 2, 5, 10, 15, 30 and 45 minutes after addition of 1 mM H 2 O 2 .ctt1 mRNA was quantified from total RNA using strand-specific RT-qPCR and normalized on the level of the U3B snoRNA.Data are presented as mean +/-standard error of the mean (SEM), calculated from three biological replicates.D. Strand-specific RT-qPCR analysis of ctt1 mRNA and XUT0794 induction in WT cells upon H 2 O 2 treatment.ctt1 mRNA (blue) and XUT0794 (red) were quantified as above.Relative level of each transcript in the non-induced condition (T0) was set to 1. Mean and SEM values were calculated from three biological replicates.https://doi.org/10.1371/journal.pgen.1007465.g002We conclude that the 255-815 fragment of XUT0794 is sufficient to attenuate ctt1 and that the level of the attenuation depends on the abundance of the asXUT, which is consistent with the hypothesis that the regulation is mediated by the asXUT but is not an indirect effect of Exo2 inactivation.

Transcriptional attenuation of ctt1 in exo2Δ cells is characterized by partial RNAPII Ser5 phosphorylation
To determine whether the attenuation of ctt1 induction occurs at the transcriptional level, we performed RNAPII ChIP experiments in WT and exo2Δ cells.Upon oxidative stress, RNAPII occupancy in the mutant showed a significant 2-to 4-fold decrease along the ctt1 locus (Fig 4A and 4B), indicating that the attenuation is transcriptional.
Analysis of the distribution of differentially phosphorylated forms of the C-terminal domain (CTD) of Rpb1, the largest subunit of RNAPII, provided further insights into the mechanism of transcriptional attenuation.Ser5-P RNAPII is associated to the early stages of the transcription cycle and predominates in the promoter-proximal region of the gene, while Ser2-P RNAPII is associated to transcription elongation and increases along the gene core [36].Upon oxidative stress, we observed a 30% decrease of Ser5-P RNAPII in the 5' and core regions of ctt1, in the exo2Δ mutant (Fig 4C).In contrast, Ser2-P RNAPII occupancy was unaffected (Fig 4D ).
Interestingly, we noted that Ser5-P RNAPII levels remain high across the ctt1 gene body, especially in the XUT0794 overlapping region (Fig 4C , probe C).This could possibly reflect re-initiation events following collision between convergent RNAPII, in keeping with that XUT0794 expression is also activated upon oxidative stress (Fig 2D).
In summary, stabilization of XUT0794 impairs ctt1 transcription, with less RNAPII loaded on the gene in response to oxidative stress and an additional reduction of Ser5P.

Attenuation of ctt1 induction depends on histone deacetylation
Several studies in budding yeast have pointed out the role of HDAC, including the class II HDAC Hda1, in aslncRNA-mediated gene silencing [16,18,19].To test whether the attenuation of ctt1 induction involves an HDAC activity, WT and exo2Δ cells were treated with trichostatin A (TSA), an inhibitor of class I-II HDAC.When exposed to oxidative stress, TSAtreated exo2Δ cells accumulated ctt1 mRNA to the same level as the control (DMSO-treated) WT strain (Fig 5A).We also noted that the basal levels of ctt1 mRNA were increased in the TSA-treated WT and exo2Δ cells, indicating that ctt1 repression requires an HDAC activity.Furthermore, both ctt1 mRNA and XUT0794 levels in TSA-treated WT cells showed a 2-fold increase compared to the DMSO-treated control after H 2 O 2 addition (Fig 5A and 5B).This indicates that the XUT0794-associated feedback loop that could modulate ctt1 expression in WT cells upon exposure to H 2 O 2 is impaired when HDAC activity is inhibited.
On the basis of this observation, we predicted histone acetylation along ctt1 to decrease upon XUT0794 stabilization.ChIP experiments in ctt1 induction conditions revealed a significant 50% and 30% reduction of histone H4K5/8/12/16 acetylation and H3K14 acetylation, respectively, in the exo2Δ mutant, in the region where ctt1 gene and XUT0794 overlap (Fig 5C and S5A Fig; probe C).These data support the idea that asXUT-mediated gene attenuation (amplicon D).Data were normalized on U3B levels and are presented as mean +/-SEM, calculated from four biological replicates.ÃÃ p<0.01;ÃÃÃ p<0.001; ns, not significant upon t-test.https://doi.org/10.1371/journal.pgen.1007465.g003depends on HDAC, resulting in reduced levels of histone acetylation.Importantly, significant histone deacetylation in exo2Δ was also detected across SPAPB24D3.07c,another class 1 gene (Fig 5D ), but not across the class 2 genes ptb1 and cuf1 (S5B and S5C Fig).This indicates that histone deacetylation is not a general feature of all the genes that are transcriptionally downregulated in exo2Δ cells (classes 1-2) but is specific to those with asXUT (class 1).
In an attempt to identify the HDAC involved, we tested the effect of Clr3 (the ortholog of Hda1), Hos2 (a class I HDAC) and Clr6 (the ortholog of class I HDAC Rpd3).As Clr6 exists in at least two distinct complexes (Clr6-CI and -CII), we also tested a specific subunit for each them, namely the ING family protein Png2 (Clr6-CI) and the Sin3 family protein Pst2 (Clr6-CII), respectively [37].We used null mutants for Clr3, Hos2, Png2 and Pst2, which are non-essential.For Clr6, which is essential, we used the thermo-sensitive clr6-1 point mutation [38].Except for pst2Δ, we could successfully combine these mutations with exo2Δ.Attenuation of ctt1 was not suppressed in the exo2Δ clr3Δ, exo2Δ hos2Δ, exo2Δ png2Δ and exo2Δ clr6-1 mutants (S6A-S6D  In conclusion, XUT-mediated attenuation of ctt1 requires an HDAC activity, suggesting that mechanisms of regulation of gene expression by lncRNAs have been conserved across the yeast clade.

Role of Set2-dependent H3K36me3 in ctt1 attenuation
During transcription, elongating RNAPII recruits the histone methyltransferase (HMT) Set2, which methylates H3K36 across the gene body [39,40].Set2-mediated H3K36 trimethylation (me3) then promotes HDAC recruitment and histone deacetylation [41,42,43], in order to suppress spurious intragenic transcription initiation [44,45].The observation of decreased histone acetylation levels across the ctt1 gene body in exo2Δ cells (Fig 5C ) prompted us to test the role of Set2 and H3K36 methylation in the regulation.In WT cells, upon oxidative stress, we observed a peak of Set2 occupancy and H3K36me3 in the region overlapping XUT0794 (probe C; Fig 6A and 6B), confirming that Set2 is recruited when ctt1 expression is induced.In the exo2Δ mutant, H3K36me3 levels were significantly increased, in some positions of ctt1, including the 5' region (probe B) and the 3' extremity (probe E).Surprisingly, in the region overlapping XUT0794 (probe C), where Set2 occupancy and histone deacetylation are the most pronounced (Figs 5C and 6A), the difference with the WT control was not statistically significant (Fig 6B).Perhaps a local change of H3K36me3 does not result into histone deacetylation at that position but to the nearby region.
In parallel, we analyzed the effect of Set2 inactivation on XUT0794 expression and ctt1 mRNA induction.We could only characterize single mutants, as despite our efforts, we failed to combine set2Δ and exo2Δ, suggesting that the double mutant is lethal.Strikingly, we found that set2Δ cells accumulate XUT0794 into levels similar to the exo2Δ mutant (Fig 6C).However, ctt1 induction was found to be normal in the set2Δ context (Fig 6D).These data indicate that the XUT0794-associated regulation of ctt1 is impaired when Set2 is inactivated.
In summary, Set2 is recruited to ctt1 upon oxidative stress.In absence of Set2, XUT0794 accumulation and ctt1 attenuation are decoupled, indicating that Set2 is required for the XUT0794-associated regulation of ctt1.

Dcr1 regulates ctt1 induction independently of Exo2
The data presented above show that the Exo2-dependent RNA decay controls ctt1 induction and restricts the level of the antisense XUT0794.We asked whether other RNA processing factors could be involved in the regulation.We tested the role of Dicer (Dcr1), involved in RNAi.
As shown in Fig 7A, ctt1 attenuation is not suppressed in the exo2Δ dcr1Δ double mutant.Rather, ctt1 was found to be attenuated in the dcr1Δ single mutant, and we observed a synergic effect in the exo2Δ dcr1Δ double mutant.On the other hand, Dicer overexpression in exo2Δ cells had no impact on ctt1 attenuation (S7A-S7C Thus, Exo2 and Dcr1 regulate ctt1 induction through independent mechanisms, which is consistent with the observation that asXUTs are globally not targeted by RNAi in S. pombe [30].

Discussion
Previous studies in different eukaryotic models have shown that aslncRNAs can regulate sense gene expression [14].However, the molecular bases for aslncRNAs-mediated regulation remain largely unknown.In budding and fission yeasts, aslncRNAs are actively degraded by the Xrn1/Exo2-dependent cytoplasmic 5'-3' RNA decay pathway [28,29,30].These Xrn1/ Exo2-sensitive aslncRNAs are named XUTs [28].In budding yeast, asXUTs stabilization was shown to result into transcriptional attenuation of a subset of genes, referred to as class 1 [28].Whether such an asXUT-associated regulation is conserved in other organisms was unknown.
Here, we used NET-Seq to identify genes showing transcriptional attenuation upon stabilization of their paired-asXUT (class 1) in S. pombe.Importantly, asXUT presence and sense gene attenuation in exo2Δ are not independent, supporting the idea that the regulation is mediated by the stabilized asXUTs and is not a side effect of Exo2 inactivation.However, additional mechanistic analyses are required to confirm this hypothesis.
In a previous study, we reported that the asXUT-associated genes are globally less transcribed than the 'solo' ones (without asXUT), displaying an hypoacetylated promoter and hyperacetylation across the gene body [30].Here we show that the asXUT-associated genes can be separated in two distinct subgroups, namely class 1 (attenuated upon asXUT stabilization in exo2Δ) and class 3 (unchanged in exo2Δ).Class 1 corresponds to highly transcribed genes showing prominent nucleosome depletion and high histone acetylation levels at the promoter.In addition, class 1 displays high TSS-proximal antisense transcription, suggesting that the TSS region could be a possible determinant for aslncRNA-mediated regulation.In contrast, class 3 is weakly transcribed, and displays poor promoter-proximal nucleosome depletion and low histone acetylation.Upon stabilization of asXUTs, transcription of class 1 drops down to the basal levels of class 3.This suggests the existence of a regulatory threshold, ie asXUTs would modulate expression of their associated sense genes, only if expression is above this threshold.This hypothesis contrasts with a previous model based on the analysis of senseantisense RNA levels in budding yeast, which proposed that antisense-mediated repression would be restricted to low sense expression [46].
Our data suggests that a subset of asXUTs could regulate gene expression at the transcriptional level, reducing sense transcription, as previously shown in S. cerevisiae [18,28].XUTs could also act at other steps of the gene expression process, especially at the post-transcriptional level.In this regard, aslncRNAs have been shown to modulate protein production in response to osmotic stress in S. pombe [23].In S. cerevisiae, disruption of several aslncRNAs results into increased protein synthesis from their paired-sense mRNAs, indicating a role of these aslncRNAs in the control of protein abundance [47].Future investigations will be required to explore the regulatory potential of asXUTs and to determine the step(s) of the gene expression process they act on.
To get insights into the mechanism by which asXUTs could attenuate gene expression, we selected a class 1 representative, the catalase-coding gene ctt1, for further characterization.Induction of ctt1 in response to oxidative stress was attenuated in exo2Δ cells, correlating with antisense XUT0794 accumulation.Our data indicate that ctt1 attenuation in Exo2-deficient cells occurs at the transcriptional level and is mediated by HDAC activity.Our attempts to identify the HDAC involved in the attenuation mechanism were unsuccessful, most likely due to redundancy of HDAC activities, considering the results obtained upon TSA treatment.On the other hand, we show that the Clr6CI-II complexes (the homolog of Rpd3L and Rpd3S, respectively) are required for ctt1 induction (S6C-S6E Fig) .The mechanism of HDAC recruitment also remains to be determined.Although we cannot formally exclude a direct recruitment by the asXUT itself, the HDAC is probably recruited through the Set2-dependent H3K36me3 marks.Consistent with this hypothesis, the most hypoacetylated region of ctt1 corresponds to the peak of Set2 occupancy and H3K36me3.Furthermore, at the RNA level, the loss of HDAC activity and the inactivation of Set2 have a similar effect, decoupling XUT0794 accumulation and ctt1 regulation.
Whether ctt1 attenuation in exo2Δ cells depends on the stabilized asXUT per se and/or on the act of antisense transcription remains unsolved to date.On one hand, the ribozyme experiment shows that ctt1 attenuation level positively correlates with XUT0794 abundance (Fig 3E and 3F), in a context where the promoter of the XUT (and presumably the level of antisense nascent transcription) remains unchanged, which is consistent with a regulation mediated by the RNA.On the other hand, the local increase of H3K36me3 across the ctt1 gene body in exo2Δ cells (Fig 6B ) suggests that ctt1 regulation could also depend on transcriptional interference.In fission yeast, gene repression by transcriptional interference requires Set2 and the Clr6CII complex [48].Here, we show that Set2 and Clr6CII have different effects on ctt1 induction: it is normal in Set2-deficient cells ( The model of a RNA-mediated regulation raises a key mechanistic question: how could an asXUT, which in all likelihood accumulates in the cytoplasm in exo2Δ cells, act in the nucleus to regulate the transcription of their paired-sense genes?This remains unknown to date.One possibility is that the stabilized XUTs could shuttle between the cytoplasm to the nucleus, as shown for tRNAs in budding yeast [49,50] and also in mammalian cells [51].
Most of the effects we described have been observed in mutant cells.But does antisense XUT0794 play any role in WT cells?Interestingly, we observed that XUT0794 is also rapidly induced in WT cells after H 2 O 2 addition, suggesting that it could participate in the modulation of ctt1 induction.In this respect, a recent study in human fibroblasts identified a class of stress-induced aslncRNAs, which are activated upon oxidative stress [8], suggesting that aslncRNAs induction might be part of a conserved response to oxidative stress in Eukaryotes.Demonstrating that XUT0794 plays a direct role in the modulation of ctt1 during oxidative stress relies, among others, on the ability to block its expression.Unfortunately, none of the strategies we used in this work succeeded into blocking XUT0794.Additional work will be required to implement in fission yeast other techniques developed in S. cerevisiae to strand-specifically block aslncRNA synthesis [47,52], which remains technically challenging yet.For instance, at some loci, the CRISPR interference approach is not strand-specific and results in the production of novel isoforms of the targeted aslncRNA [53].
Efficient induction of ctt1 upon oxidative stress depends on multiple factors [54].In addition to Exo2, we showed that Dcr1 also contributes to ctt1 induction.Our data indicate that Dcr1 and Exo2 regulate ctt1 through distinct mechanisms, which is consistent with the observations that asXUTs are not targeted by Dicer in fission yeast [30].Interestingly, Dcr1 was recently shown to promote efficient termination of a set of highly transcribed genes, corresponding to sites of replication stress and DNA damage [55], and ctt1 belongs to this set of Dcr1-terminated genes.Thus, one possibility could be that Dcr1 regulates ctt1 induction at the level of transcription termination.However, our ChIP data did not reveal any significant change of RNAPII occupancy at the 3' extremity of ctt1 in dcr1Δ cells, upon oxidative stress (Fig 7C).Additional analyses are therefore required to decipher the mechanism by which Dcr1 regulates ctt1 induction.
In conclusion, our work in budding and fission yeasts shows that the cytoplasmic 5'-end RNA decay plays a key role in controlling aslncRNAs endowed with regulatory potential.Given the high conservation of Xrn1 in Eukaryotes, it is tempting to speculate that asXUTs and their regulatory activity are conserved in higher eukaryotes, contributing in buffering genome expression, and adding another layer to the complexity of gene regulation.

Yeast strains, plasmids and media
All the strains used in this study are listed in S5 Table .Mutant strains were constructed by meiotic cross or transformation, and verified by PCR on genomic DNA and/or RT-qPCR.Plasmid pAM353 for expression of XUT0794 in trans was constructed by cloning XUT0794 in the SalI site of pREP41 (ars1 LEU2 P41nmt1).Sanger sequencing confirmed the correct orientation of the insert and the absence of mutation.Hammerhead ribozyme [56] was inserted in XUT0794 by two-step PCR, giving a 3.2 Kb final product corresponding to ctt1 mRNA coordinates +/-500 bp that was cloned in pREP41.After verification of absence of additional mutations by Sanger sequencing, the ribozyme-containing construct was excised and transformed in the YAM2534 strain (ctt1::ura4).Transformants were selected on 5-FOA plates and analyzed by PCR on genomic DNA.Deletion of exo2 was performed subsequently.
Strains were grown at 32˚C to mid-log phase (OD 595 0,5) in YES or EMM-L medium.For ctt1 induction, 1 mM H 2 O 2 was added for 15 minutes [34], or different time points for analysis of kinetics of induction.Expression from P41nmt1 was repressed by growing cells in EMM-L + 15 μM thiamine for 24 hours.

NET-Seq
NET-Seq libraries were constructed from biological duplicates of YAM2507 (exo2Δ rpb3-flag) cells and sequenced as previously described [30].Libraries for the WT strain YAM2492 (rpb3-flag) were described in the same previous report [30].
After removal of the 5'-adapter sequence, reads were uniquely mapped to the reference genome (ASM294v2.30)using version 0.12.8 of Bowtie [57], with a tolerance of 2 mismatches.
Differential analysis was performed between the IP samples from WT and exo2Δ using DESeq [58].Genes showing significant decrease (P-value <0.05, adjusted for multiple testing with the Benjamini-Hochberg procedure) in the mutant were defined as class 1 & 2.
Raw sequences have been deposited to the NCBI Gene Expression Omnibus (accession number GEO: GSE106649).A genome browser for visualization of NET-Seq processed data is accessible at http://vm-gb.curie.fr/mw3.

Total RNA extraction
Total RNA was extracted from exponentially growing cells using standard hot phenol procedure, resuspended in nuclease-free H 2 O (Ambion) and quantified using a NanoDrop 2000c spectrophotometer.

Northern blot
10 μg of total RNA were loaded on denaturing 1.2% agarose gel and transferred to Hybond-XL nylon membrane (GE Healthcare).ctt1 mRNA and U3B were detected using AMO2063 and AMO2081 oligonucleotides, respectively (see S6 Table ). 32P-labelled probes were hybridized overnight at 42˚C in ULTRAhyb-Oligo hybridization buffer (Ambion).Quantitation used a Typhoon Trio PhosphorImager and the ImageQuant TL v5.2 sofware (GE Healthcare).

Strand-specific RT-qPCR
Strand-specific reverse transcription (RT) reactions were performed from at least three biological replicates, using 1 μg of total RNA and the SuperScript II Reverse Transcriptase kit (Invitrogen), in the presence of 6,25 μg/ml actinomycin D. For each sample, a control without RT was included.Subsequent quantitative real-time PCR were performed on technical duplicates, using a LightCycler 480 instrument (Roche).Oligonucleotides used are listed in S6 Table.

Supporting information
), the latter actually showing the lowest transcription levels among the four classes (S1B Fig).In exo2Δ, transcription of class 1 falls to the low, basal level of class 3 (Fig 1D).Notably, transcription of XUTs antisense to class 1 and 3 genes is globally unaffected in the exo2Δ mutant (S1C Fig), indicating that XUTs accumulation in this context is due to the inactivation of their decay and not to a global increase of their synthesis (Fig 1E).

Fig 1 .
Fig 1.Genome-wide identification of class 1 genes in S. pombe.A. Transcriptional attenuation in exo2Δ cells.NET-Seq analysis was performed from biological duplicates of WT and exo2Δ cells.Data for the WT strain were previously described[30].After sequencing, differential analysis discriminated genes showing significant (P<0.05)reduction of transcription (classes 1-2) or not (classes 3-4).Among them, classes 1 and 3 have asXUT.The number

Fig 2 .
Fig 2. ctt1 induction upon oxidative stress is impaired in Exo2-deficient cells.A. Snapshot of nascent transcription (NET-Seq) signals along the ctt1 gene in WT (upper panel) an exo2Δ (lower panel) cells.In each panel, the signal corresponding to the sense (+) and antisense (-) strand is shown in blue and pink, respectively.Blue and red arrows represent the ctt1 mRNA and antisense XUT0794, respectively.NET-Seq data for the WT strain were previously described[30].The snapshot was produced using VING[59].B. Northern blot analysis of ctt1 mRNA induction in WT and exo2Δ cells.YAM2400 (WT) and YAM2402 (exo2Δ) cells were grown to mid-log phase in rich medium and collected before or after addition of 1 mM H 2 O 2 for 15 min.ctt1 mRNA and U3B snoRNA were detected from total RNA using 32 P-labelled oligonucleotides.Numbers represent the ctt1 mRNA/U3B ratio (ND: not determined).C. YAM2400 (WT) and YAM2402 (exo2Δ) cells grown in rich YES medium to mid-log phase were collected 0, 2, 5, 10, 15, 30 and 45 minutes after addition of 1 mM H 2 O 2 .ctt1 mRNA was quantified from total RNA using strand-specific RT-qPCR and normalized on the level of the U3B snoRNA.Data are presented as mean +/-standard error of the mean (SEM), calculated from three biological replicates.D. Strand-specific RT-qPCR analysis of ctt1 mRNA and XUT0794 induction in WT cells upon H 2 O 2 treatment.ctt1 mRNA (blue) and XUT0794 (red) were quantified as above.Relative level of each transcript in the non-induced condition (T0) was set to 1. Mean and SEM values were calculated from three biological replicates.

Fig 3 .
Fig 3. ctt1 attenuation level correlates with XUT0794 abundance.A. Schematic representation of ura4 insertion upstream from XUT0794.B. Analysis of XUT0794 level upon ura4 insertion.Strains YAM2400 (WT), YAM2402 (exo2Δ) and YAM2817 (ura4-XUT0794) cells were grown to midlog phase in rich YES medium.XUT0794 level was determined by strand-specific RT-qPCR from total RNA as described in Fig 2C.Data are presented as mean +/-SEM, calculated from three biological replicates.ÃÃ p < 0.01; ns, not significant upon t-test.C. Analysis of ctt1 mRNA level upon ura4 insertion.Strains YAM2400 (WT), YAM2402 (exo2Δ) and YAM2817 (ura4-XUT0794) cells grown to mid-log phase in rich medium were collected before or after addition of 1 mM H 2 O 2 for 15 min.ctt1 mRNA level was determined by strand-specific RT-qPCR as described in Fig 2C.Mean and SEM are calculated from three biological replicates.ÃÃÃ p < 0.001; ns, not significant upon t-test.D. Schematic representation of hammerhead ribozyme (RZ) inserted within XUT0794.Self-cleaving RZ was inserted at position 254 of XUT0794.Position of qPCR amplicons C and D is indicated.E. Analysis of XUT0794 upstream from and downstream to RZ insertion site.YAM2400 (WT), YAM2402 (exo2Δ), YAM2565 (WT + RZ) and YAM2567 (exo2Δ + RZ) were grown as described in Fig 3B.Strand-specific RT on XUT0794 was performed from total RNA using oligonucleotide AMO2069.Oligonucleotides AMO2535-6 (amplicon D) and AMO2069-70 (amplicon C) were used for qPCR detection of XUT0794 5' and 3' fragment, respectively.Data were normalized on U3B snoRNA.For each amplicon, the normalized level of XUT0794 in exo2Δ was then set to 1. Results are presented as mean +/-SEM, calculated from four biological replicates.F. Analysis of ctt1 mRNA levels.Strains and cultures were as in Fig 3E and Fig 3C, respectively strand-specific RT on ctt1 mRNA was performed using oligonucleotide AMO2535; qPCR was performed using AMO2535-6 Fig), indicating that none of the four tested factor is involved in the attenuation mechanism.In contrast, the png2Δ, pst2Δ and clr6-1 single mutants exhibited a strong defect of ctt1 induction (S6C-S6E Fig).In addition, the png2Δ and clr6-1 mutations were synergic with exo2Δ (S6C and S6D Fig).This indicates that Exo2 and the Clr6 HDAC complexes are required for efficient ctt1 induction, but act independently.

Fig 4 .
Fig 4. asXUT-mediated attenuation of ctt1 is transcriptional.A. Schematic map of the ctt1 locus, with positions of the qPCR oligonucleotide pairs.B. ChIP analysis of RNAPII occupancy across ctt1.Strains YAM2400 (WT) and YAM2402 (exo2Δ) were grown as in Fig 2B.After cross-linking, chromatin extraction and sonication, RNAPII was immunoprecipitated using antibody against the CTD of its largest subunit Rpb1.Co-precipitated DNA was purified and quantified by qPCR.Data were normalized on the act1 signal, which is not controlled to an asXUT (see S4 Fig).The dashed line indicates the background signal observed on an intergenic region of chromosome I used as negative control.Data are presented as mean +/-SEM, from three biological replicates.Ã p<0.05;ÃÃÃ p<0.001; ns, not significant upon t-test.C-D.ChIP analysis of Ser5-P and Ser2-P occupancy along ctt1.RNAPII was immunoprecipitated from the same chromatin extracts as above, using antibody against the Ser5-P (C) or the Ser2-P (D) form of Rpb1 CTD.Data normalization was as above.For each position of ctt1, the ratio between Ser5-P or Ser2-P and total RNAPII is shown.Mean and SEM values were calculated from three biological replicates.Ã p<0.05;ÃÃ p<0.01; ns, not significant upon t-test.https://doi.org/10.1371/journal.pgen.1007465.g004

Fig 5 .
Fig 5. asXUT-mediated attenuation of ctt1 depends on HDAC.A-B.Strand-specific RT-qPCR analysis of ctt1 mRNA and XUT0794 levels in the presence of HDAC inhibitor.Strains YAM2400 (WT) and YAM2402 (exo2Δ) were grown in rich medium to mid-log phase before addition of 40 μg/ ml TSA or equivalent volume of DMSO for 2 hours.TSA-treated and control cells were then collected prior or after addition of H 2 O 2 for 15 min.ctt1 mRNA (A) and XUT0794 (B) were quantified as described above.Data are presented as mean +/-SEM, calculated from three biological replicates.C. ChIP analysis of H4K5/8/12/16 acetylation along ctt1.Culture, cross-linking and chromatin extraction were as described in Fig 4B.For each position, data were first normalized on act1, then on the level of histone H3, immunoprecipitated from the same chromatin.Data are presented as above.Ã p<0.05;ÃÃÃ p<0.001; ns, not significant upon t-test.D. ChIP analysis of H4K5/8/12/16 acetylation along the SPAPB24D3.07cgene (class 1) in WT and exo2Δ cells.YAM2400 (WT) and YAM2402 (exo2Δ) cells were grown to mid-log phase in rich medium.Cross-linking, chromatin extraction, data analysis and presentation were as described above.Ã p<0.05;ÃÃ p<0.01; ns, not significant upon t-test.https://doi.org/10.1371/journal.pgen.1007465.g005

Fig 6 .
Fig 6.Role of Set2-dependent H3K36me3 in ctt1 attenuation.A. Set2 is recruited to ctt1 upon oxidative stress.YAM2816 (Set2-13Myc [60]) cells were grown as in Fig 2B.Cross-linking and chromatin extraction were as described in Fig 4. Data are presented as mean +/-SEM, calculated from three biological replicates.B. ChIP analysis of H3K36 trimethylation (me3) along ctt1 in exo2Δ cells.Strains YAM2400 (WT) and YAM2402 (exo2Δ) were grown as in Fig 2B.Cross-linking and chromatin extraction were as described in Fig 4. Data analysis was performed as described in Fig 5C.Average values and SEM were calculated from three biological replicates.Ã p<0.05; ns, not significant upon t-test.C. Analysis of XUT0794 level in set2Δ cells.Strains YAM2400 (WT), YAM2402 (exo2Δ) and YAM2797 (set2Δ) were grown to mid-log phase in rich medium.XUT0794 level was quantified from total RNA as described in Fig 2C.Data are presented as mean +/-SEM, calculated from three biological replicates.ÃÃ p < 0.01; ÃÃÃ p < 0.001; ns, not significant upon t-test.D. Analysis of ctt1 mRNA induction in set2Δ cells.Strains YAM2400 (WT), YAM2402 (exo2Δ) and YAM2797 (set2Δ) were grown as in Fig 2B.ctt1 mRNA was quantified by strand-specific RT-qPCR as described in Fig 2C.Data are presented as above.ÃÃÃ p < 0.001; ns, not significant upon t-test.https://doi.org/10.1371/journal.pgen.1007465.g006 Fig).These data indicate that Exo2 and Dcr1 control ctt1 induction through distinct mechanisms.This is further supported by the observation that XUT0794 levels are unchanged in exo2Δ dcr1Δ cells compared to exo2Δ cells (Fig 7B) and by ChIP experiments showing that in oxidative stress conditions, RNAPII (Fig 7C) and H3K36me3 (S7D Fig) levels are normal in dcr1Δ cells while histone H4K5/8/12/16 acetylation is strongly reduced across the whole ctt1 locus, including the promoter region (Fig 7D).
Fig. Antisense XUT stabilization induces transcriptional attenuation of a class of highly expressed genes. A. Global RNAPII transcription in WT and exo2Δ cells.Density plot of exo2Δ/WT NET-Seq signal ratio for mRNAs (blue), sn(o)RNAs (black) and XUTs (red).B. Metagene view of NET-Seq signals along class 1-4 genes in WT cells.