DegrAAAded into Silence
2007; Cell Press; Volume: 129; Issue: 4 Linguagem: Inglês
10.1016/j.cell.2007.05.004
ISSN1097-4172
AutoresElizabeth H. Bayne, Sharon A. White, Robin C. Allshire,
Tópico(s)Chromosomal and Genetic Variations
ResumoIn fission yeast, RNA interference (RNAi)-dependent heterochromatin formation silences transgenes inserted at centromeres. In this issue, Bühler et al., 2007Bühler M. Haas W. Gygi S.P. Moazed D. Cell. 2007; (this issue)PubMed Google Scholar demonstrate that the RNAi machinery directly targets transgene transcripts. Furthermore, they link transgene silencing to a protein complex resembling the TRAMP complex of budding yeast, which promotes transcript degradation via the exosome. Thus, RNAi-independent transcript degradation may also contribute to heterochromatin gene silencing. In fission yeast, RNA interference (RNAi)-dependent heterochromatin formation silences transgenes inserted at centromeres. In this issue, Bühler et al., 2007Bühler M. Haas W. Gygi S.P. Moazed D. Cell. 2007; (this issue)PubMed Google Scholar demonstrate that the RNAi machinery directly targets transgene transcripts. Furthermore, they link transgene silencing to a protein complex resembling the TRAMP complex of budding yeast, which promotes transcript degradation via the exosome. Thus, RNAi-independent transcript degradation may also contribute to heterochromatin gene silencing. The packaging of chromosomal DNA into heterochromatin is important for cellular processes such as regulation of gene expression and accurate chromosome segregation. In the fission yeast, Schizosaccharomyces pombe, heterochromatin is found at the mating-type locus, telomeres, and centromeres. Regions of heterochromatin are generally associated with transcriptional repression, and consistent with this finding, marker genes inserted into fission yeast heterochromatin are silenced. Heterochromatin assembly involves an ordered series of events in which lysine 9 on histone H3 becomes methylated by the histone methyltransferase Clr4 (equivalent to metazoan Suv39), creating a binding site for chromodomain proteins such as Swi6, Chp1, and Chp2 (HP1-related proteins). RNAi is required to establish and maintain heterochromatin at centromeres but is dispensable for maintenance of heterochromatin at the mating-type locus (Grewal and Jia, 2007Grewal S.I. Jia S. Nat. Rev. Genet. 2007; 8: 35-46Crossref PubMed Scopus (932) Google Scholar). In mutants of the RNAi pathway centromeric small interfering (si)RNA production is defective and homologous centromeric repeat transcripts accumulate. This has led to a model whereby siRNAs generated from centromere transcripts are required to target chromatin-modifying machinery to the centromere, resulting in transcriptional repression (Grewal and Jia, 2007Grewal S.I. Jia S. Nat. Rev. Genet. 2007; 8: 35-46Crossref PubMed Scopus (932) Google Scholar). However, it is paradoxical that transcriptional “silencing” should require transcription itself. Recent evidence suggests that levels of transcription of centromere repeats are largely unaffected by their assembly into “silent” heterochromatin (Buhler et al., 2006Buhler M. Verdel A. Moazed D. Cell. 2006; 125: 873-886Abstract Full Text Full Text PDF PubMed Scopus (303) Google Scholar, Volpe et al., 2002Volpe T.A. Kidner C. Hall I.M. Teng G. Grewal S.I. Martienssen R.A. Science. 2002; 297: 1833-1837Crossref PubMed Scopus (1563) Google Scholar). In this issue of Cell, Bühler et al., 2007Bühler M. Haas W. Gygi S.P. Moazed D. Cell. 2007; (this issue)PubMed Google Scholar shed new light on the mechanism of heterochromatin silencing in fission yeast, presenting evidence for posttranscriptional silencing by an RNAi-independent pathway. Bühler et al., 2007Bühler M. Haas W. Gygi S.P. Moazed D. Cell. 2007; (this issue)PubMed Google Scholar set out to address an outstanding question: How do marker genes inserted into regions that surround heterochromatin become silenced? Although siRNAs corresponding to centromere repeat sequences are abundant, siRNAs from a ura4+ marker gene inserted within centromere repeats have previously been undetectable. Consequently, it has been unclear whether the RNAi-dependent silencing machinery is recruited directly to the ura4 gene or whether heterochromatin simply spreads into this gene from the surrounding centromeric sequence. To enrich for siRNAs, Bühler et al., 2007Bühler M. Haas W. Gygi S.P. Moazed D. Cell. 2007; (this issue)PubMed Google Scholar exploit the siRNA-binding activity of the RNAi component Ago1 (Argonaute) and also delete the ribonuclease eri1—which is normally required to suppress high levels of siRNAs. In this way the authors could detect a low level of siRNAs corresponding to centromeric ura4+. This result demonstrates that transcripts from this ura4+ gene are processed into siRNAs and therefore that the transgene could be a direct target of the RNAi machinery. The authors hypothesized that these ura4 siRNAs alone are insufficient to direct silencing of the transgene for two reasons: (1) the very low abundance of siRNAs from ura4 relative to those from centromeric sequences and (2) the siRNAs detected are predominantly of sense orientation and therefore unable to target Ago1 to the ura4 mRNA. These observations led the authors to investigate the role of an additional RNA degradation pathway in heterochromatin silencing. In the budding yeast Saccharomyces cerevisiae, polyadenylation can stimulate RNA degradation by the exosome. This polyadenylation is mediated by the TRAMP complex, which contains the poly(A) polymerases, Trf4 and Trf5 (LaCava et al., 2005LaCava J. Houseley J. Saveanu C. Petfalski E. Thompson E. Jacquier A. Tollervey D. Cell. 2005; 121: 713-724Abstract Full Text Full Text PDF PubMed Scopus (666) Google Scholar). Fission yeast has six members of the Cid1 family of noncanonical poly(A) polymerases. One of these, Cid12, has previously been implicated in heterochromatin silencing (Motamedi et al., 2004Motamedi M.R. Verdel A. Colmenares S.U. Gerber S.A. Gygi S.P. Moazed D. Cell. 2004; 119: 789-802Abstract Full Text Full Text PDF PubMed Scopus (416) Google Scholar, Stevenson and Norbury, 2006Stevenson A.L. Norbury C.J. Yeast. 2006; 23: 991-1000Crossref PubMed Scopus (49) Google Scholar). Cid12 is found in the RNAi complex RDRC (RNA-dependent RNA polymerase complex) along with Rdp1 and Hrr1. Moreover, cid12 mutants alleviate silencing in a similar way to other RNAi mutants, although the exact role of Cid12 in heterochromatin silencing has yet to be elucidated (Motamedi et al., 2004Motamedi M.R. Verdel A. Colmenares S.U. Gerber S.A. Gygi S.P. Moazed D. Cell. 2004; 119: 789-802Abstract Full Text Full Text PDF PubMed Scopus (416) Google Scholar). In their new work, Bühler et al., 2007Bühler M. Haas W. Gygi S.P. Moazed D. Cell. 2007; (this issue)PubMed Google Scholar investigate the role of a second member of the family, Cid14, which is the S. pombe functional homolog of S. cerevisiae Trf4/Trf5. The authors find that functional Cid14 is required for intact silencing at centromeres and for the generation of centromeric siRNAs. Unlike RNAi mutants, deletion of the cid14+ gene also alleviates silencing at the mating-type locus. Curiously, the cid14 mutant shows a less marked effect on the levels of H3K9 methylation and Swi6 associated with heterochromatin than do RNAi mutants. Bühler et al., 2007Bühler M. Haas W. Gygi S.P. Moazed D. Cell. 2007; (this issue)PubMed Google Scholar also confirm that Cid14 has poly(A) polymerase activity in vitro and that mutations in the catalytic residues of the enzyme alleviate silencing in vivo. Thus, the authors propose a model in which Cid14-mediated polyadenylation of heterochromatin transcripts is required for silencing by the RNAi machinery, the exosome, or both (Figure 1). Biochemical purification of Cid14 did not reveal any association with known RNAi or heterochromatin components. Instead Cid14 associates with fission yeast homologs of other TRAMP complex components, Mtr4 and Air1, as well as ribosome synthesis factors, consistent with the known role of Cid14 in rRNA polyadenylation (Win et al., 2006aWin T.Z. Draper S. Read R.L. Pearce J. Norbury C.J. Wang S.W. Mol. Cell. Biol. 2006; 26: 1710-1721Crossref PubMed Scopus (60) Google Scholar). These findings suggest that cid14 might act as part of a fission yeast TRAMP complex (spTRAMP) to target heterochromatin transcripts for degradation. Deletion of air1+ shows no effect on heterochromatin silencing, whereas mutation of mtr4 alleviates silencing at the mating-type locus but not at centromeres, indicating that the components of spTRAMP play varying roles in heterochromatin silencing. To further investigate the possibility that Cid14 directs degradation of heterochromatin transcripts by the exosome, the authors also tested heterochromatin silencing in the absence of a component of the nuclear exosome, Rrp6. Deletion of rrp6 alleviates silencing both at centromeres and at the mating-type locus. This is consistent with recent findings that Dis3, an exosome-associated ribonuclease, is also required for silencing at centromeres and the mating-type locus (Murakami et al., 2007Murakami H. Goto D.B. Toda T. Chen E.S. Grewal S.I. Martienssen R.A. Yanagida M. PLoS ONE. 2007; 2: e317Crossref PubMed Scopus (45) Google Scholar). Unlike cid14, neither of these exosome components is required to generate siRNAs that are homologous to heterochromatin regions, suggesting that the role of cid14 may be more complex than simply exosome recruitment. The findings of Bühler et al., 2007Bühler M. Haas W. Gygi S.P. Moazed D. Cell. 2007; (this issue)PubMed Google Scholar strongly suggest that Cid14 is involved in targeting centromere transcripts for degradation. However, it remains to be determined whether heterochromatin transcripts are bona fide substrates for Cid14 polyadenylation. Centromere transcripts are known to have poly(A) tails, and these tails are unchanged in cells lacking Cid12, so it might be revealing to check their status in a cid14 mutant (Win et al., 2006bWin T.Z. Stevenson A.L. Wang S.W. Mol. Cell. Biol. 2006; 26: 4435-4447Crossref PubMed Scopus (17) Google Scholar). It would also be informative to examine whether Cid14 or the whole TRAMP complex associates with centromeric transcripts. Another outstanding question is the relationship between Cid14 and Cid12. Based on their observation of a large RNA species associated with Ago1 in cells lacking Cid14, the authors suggest that Cid14 may be required to convert single-stranded precursor RNA into dsRNA. This is a role also proposed for the RDRC complex raising the possibility that Cid12 and Cid14 may have some functional redundancy, analogous to Trf4 and Trf5. Such an effect might explain how Cid14 can be intimately associated with the RNAi pathway despite having a distinct mutant phenotype. Clearly much remains to be revealed about the mechanisms underlying RNAi-directed heterochromatin formation and silencing. Nevertheless, the analyses by Bühler et al., 2007Bühler M. Haas W. Gygi S.P. Moazed D. Cell. 2007; (this issue)PubMed Google Scholar reveal that siRNAs are made from transgene insertions at centromeres, and expose intriguing connections between heterochromatin silencing and general RNA turnover mechanisms. RNAi-Dependent and -Independent RNA Turnover Mechanisms Contribute to Heterochromatic Gene SilencingBühler et al.CellMay 18, 2007In BriefIn fission yeast, the RNAi pathway is required for heterochromatin-dependent silencing of transgene insertions at centromeric repeats and acts together with other pathways to silence transgenes at the silent mating-type locus. Here, we show that transgene transcripts at centromeric repeats are processed into siRNAs and are therefore direct targets of RNAi. Furthermore, we show that Cid14, a member of the Trf4/5 family of poly(A) polymerases, has poly(A) polymerase activity that is required for heterochromatic gene silencing. Full-Text PDF Open Archive
Referência(s)