Telomere Position Effect (TPE) is governed by strong repression signals emitted by telomeres via the Sir2/3/4 Histone Deacetylase complex. These signals are then relayed by weak proto-silencers residing in the subtelomeric core X and Y' elements. Subtelomeres also contain Sub-Telomeric Anti-silencing Regions (STARs). In this study we have prepared telomeres built of different combinations of core X, Y' and STARs and have analyzed them in strains lacking Histone-Acetyltransferase genes as well as in cdc6-1 and Δrif1 strains. We show that core X and Y' dramatically reduce both positive and negative variations in TPE, that are caused by these mutations. We also show that the deletion of Histone-Acetyltransferase genes reduce the silencing activity of an ACS proto-silencer, but also reduce the anti-silencing activity of a STAR. We postulate that core X and Y' act as epigenetic “cushioning” cis-elements.
Citation: Power P, Jeffery D, Rehman MA, Chatterji A, Yankulov K (2011) Sub-Telomeric core X and Y' Elements in S.cerevisiae Suppress Extreme Variations in Gene Silencing. PLoS ONE 6(3): e17523. doi:10.1371/journal.pone.0017523
Editor: Mary Bryk, Texas A&M University, United States of America
Received: November 9, 2010; Accepted: February 4, 2011; Published: March 17, 2011
Copyright: © 2011 Power et al. This is an open-access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.
Funding: This work was supported by the National Science and Engineering Research Council of Canada (NSERC) (University of Guelph grant # 043230). The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript.
Competing interests: The authors have declared that no competing interests exist.
Gene silencing refers to position dependent and promoter-independent repression of genes. It is characterized by local histone hypoacetylation and the formation of heterochromatin structures. In S.cerevisiae, gene silencing operates at the mating type loci HML and HMR, at the rRNA gene cluster and in the sub-telomeric regions of the chromosomes . Gene silencing at subtelomeres is referred to as Telomere Position Effect (TPE) and is governed by strong repression signals emitted by the telomere itself . These signals are relayed by weaker proto-silencers, which are positioned in the subtelomeric core X- and Y'- elements . To date, proto-silencer activity has been assigned to ARS consensus sequences (ACS) and for the binding sites for Rap1p and Abf1p , , , . The subtelomeres also contain sequences, which display anti-silencing properties and are referred to as STARs (Sub-Telomeric Anti-silencing Regions) . The antagonizing silencing and anti-silencing activities emitted by these elements confer a peculiar quasi-unstable mode of subtelomeric gene expression. Any gene residing in the subtelomeres or translocated to these loci acquires either fully silenced or fully active state. This state is maintained through many generations, however infrequent switches occur to produce expression patterns that are reminiscent to the classical variegated pigmentation in the eye of Drosophila . In all cases, the transition between the silenced and active states of expression is accompanied by histone acetylation and other post-translational histone modifications .
A Histone DeAcetylase (HDAC), Sir2p, plays a central role in the establishment and maintenance of silencing at all repressed loci. At telomeres there are two means of engaging Sir2p. The telomeric TG1-3 repeats bind Rap1p, which in turn recruits Sir3p and Sir4p to eventually recruit Sir2p . Two proteins, Rif1p and Rif2p, interfere with the interaction between Rap1p and Sir3/Sir4 thus acting as anti-silencing factors , , . At the same time the sub-telomeric ACS proto-silencers bind ORC (Origin Recognition Complex). ACS-bound Orc1 associates with Sir1p to independently recruit Sir2p to these positions . Consequently, Sir2p deacetylates the nearby nucleosome and spreads over the neighboring ones with the aid of Sir3p and Sir4p. The spreading of histone deacetylation by Sir2p is counteracted by Histone Acetyl Transferases (HAT), but the mode of their action is not understood to the extent of the SIR genes.
HATs acetylate lysines of core histones to generate events, which culminate in chromatin de-condensation. To date, nine HATs have been described in S.cerevisiae . Several studies have pointed to SAS2 as the principal SIR2-counteracting HAT at telomeres , , , , . Sas2p is responsible for the acetylation of H4-K16 in vivo, while Sir2p is deacetylating this position , . Thus, the two opposing enzymes generate a dynamic chromatin boundary at subtelomeres. Paradoxically, deletion of SAS2 very moderately increases the silencing of natural subtelomeric genes , , but dramatically reduces silencing at synthetic telomeres thus portraying SAS2 as an anti-silencing factor , , , . This stark discrepancy has not been adequately explained. On the other hand, many other lysines in H3 and H4 are hypo-acetylated in subtelomeric chromatin  suggesting that other HATs are also directly involved in anti-silencing.
In this study we have characterized the roles of five HATs (HAT1, GCN5, SAS2, SAS3, Rtt109), of RIF1 and CDC6 on several recombinant telomeres build up of core X, Y' and STARs. These mutations produced both positive and negative effects on telomeric silencing. Unexpectedly, we have revealed that subtelomeric core X and Y' dampened down the extreme deviations of TPE caused by these mutations.
Materials and Methods
Yeast strains with deletions of HAT1, GCN5, SAS2, SAS3, YNG1, Rtt109 and RIF1 are derivatives of BY4742 and were obtained from ATCC. All other mutants are derivatives of W303. All strains used in this study are listed and referenced in Table 1.
All constructs are flanked by a portion of ADH4 and telomeric TG1-3 repeats (see Fig. 1A) and are designed for targeted integration in the left telomere of chromosome VII. URA3-tel , GF2, GF3, GF6, GF9, GF44, GF46 and GF61  were previously described. GF6ΔSTAR, GF6ΔACS and GF44ΔACS were produced by excision of the STAR element in GF6 or by site directed mutagenesis of ACS in the core X elements, respectively. All integrating constructs were produced by restriction digestion of the corresponding plasmids.
A) Telomeric reporters used in this study. Maps (not to scale) of the used constructs are shown. The positions of core X element from the IIR telomere and the Y' element from the XII-L telomere (black rectangles), the STARs from the same telomeres (grey rectangles), URA3, ADH4 and the telomeric TG1-3 repeats (black triangles) are as indicated. The position of the destroyed ACS (ARS Consensus Sequence) is depicted by an open diamond. The 5′→3′ direction of URA3 transcription is indicated in the URA3-tel construct (top) and is the same for all constructs shown. The insertions between URA3 and the telomeric repeat add 145-900 base pairs as compared to URA3-tel. B) Percentage of FOAR cells in different strains and constructs. The reporter constructs shown along the vertical axis were integrated in the strains shown on the left. Percentage of FOAR cells was measured in at least three independent experiments. Average %FOAR ± std. dev. were calculated and plotted. Data is from Table S1. C) URA3-tel recapitulates silencing effects in mutant strains. The URA3-tel construct was integrated in the strains shown along the vertical axis. The ratios of %FOAR in the mutant strains versus the %FOAR in the isogenic wild type strain were calculated and plotted. The effects of Δsas2, Δsas3, Δyng1, Δrtt109, Δhat1, Δgcn5 and Δrif1 were assessed using BY4742 as the wild type strain (Table S1). The effects of orc2-1, orc5-1, mcm5-461, cdc6-1, cdc45-1 and cdc7-1(sas1) were assessed using W303 as the wild type strain (data not shown). There is little difference in the levels of telomeric silencing between BY4742 and W303. The arrows underneath the exponential graph indicate increase or decrease of silencing.
Telomeric integration and analysis of gene silencing
Cells were transformed with integrating constructs and three single colonies were selected from SC-ura plates. To warrant for the loss of un-integrated constructs (linear DNAs lacking CEN elements), transformants were restreaked on Sc-ura and again a single colony from this SC-ura plate was streaked on both SC-ura and SC/FOA. Fluoro-orotic acid (FOA) has a selective toxicity for cells expressing URA3, hence SC/FOA selects for the repressed state of URA3 and confirms variegated expression. By the tird re-streaking the transformed cells have been grown for about 60 generations. This procedure uniformly produces cells that have integrated the test constructs (Fig. 1) in the VIIL telomere when analyzed by PCR. Finally, a single colony was taken from the third SC-ura plate and grown for about 30 generations in non-selective (YPD) medium. Serial 1∶10 dilutions were prepared for each culture and 5 µl aliquots were spotted on SC and SC/FOA plates. Colonies in two consecutive spots with less than 50 colonies (these correspond to two consecutive dilutions) were counted. The %FOAR for each independent culture was acquired as the number of colonies on SC/FOA plates divided by the number of colonies on SC plates. Finally, the average %FOAR of the counts in three independent cultures ± standard deviation were calculated and are shown in Table S1. Average values and the ratios between %FOAR in different strains and/or constructs were calculated and plotted in Microsoft Excel.
Core X and Y' curtail variations in TPE caused by deletion of HAT genes
We used the set of telomeric reporters shown in Fig. 1A to analyze the role of several non-essential HATs in TPE. These reporters contain URA3 and different combinations of subtelomeric core X, Y' and STAR elements (Fig. 1A). The ADH4-URA3-tel construct  is one of the most frequently used telomeric reporters and serves as a direct cross-reference between other studies and the current one. GF2 and GF3 contain STARs derived from the core X-IIR or Y'-XIIL elements, respectively. GF6 and GF9 contain the same STARs, but also the core X from the same telomeres, respectively. In GF44 and GF46 the core X and the Y' are positioned distal to the telomere beyond URA3. In GF61 URA3 is away from the telomere beyond two STARs, core X and TRP1. In addition, ACS and STAR were destroyed in GF6 and GF44 as indicated. The insertions between URA3 and the telomeric repeat add 145-900 base pairs in different constructs as compared to URA3-tel. Several studies have shown that the telomeric silencing for these and other constructs does not directly correlate to the distance from the telomeres , , , , . Instead, silencing is discontinuous and is strongly influenced by the nature and the positions of different regulatory elements , . Therefore, the variety of elements in these constructs allows for broad assessment of TPE in different strains.
All constructs were integrated in the left telomere of chromosome VII in BY4742 and its derivatives Δsas2, Δsas3, Δyng1, Δrtt109, Δhat1 and Δgcn5 and selected on SC-ura plates. Colonies were then streaked on SC/FOA plates, which render the URA3-expressing cells sensitive to the drug while the cells with repressed URA3 form FOAR colonies. After confirming the variegated mode of expression of the integrated reporters, three colonies were grown in non-selective medium for 30 generations to allow for the re-establishment of the silenced/active equilibrium of URA3 in these cultures. The percentage of FOAR was calculated as the number of colonies on SC/FOA plates divided by the number of colonies on SC plates. The average values ± standard deviations were calculated (Table S1) and are plotted in Figure 1B.
Next, we cross-referenced the acquired data to available data in earlier publications. URA3-tel, GF2, GF3, GF6, GF9, GF44, GF46, GF61, GF6ΔSTAR, GF6ΔACS and GF44ΔACS showed very similar levels of %FOAR in BY4742 cells as compared to the previously used W303 strain , , . In addition, the prototype URA3-tel construct recapitulated the silencing defects observed in sas2, sas3, orc2-1, orc5-1, mcm5-461, cdc6-1, cdc45-1 and cdc7-1(sas1) (Fig. 1C) , , , . Finally, we compared the magnitude of SAS2-dependent de-repression of URA3-tel in BY4742 and W303 (the only available data for direct comparison that we are aware of). The deletion of SAS2 in W303 had decreased repression in the range of 10-50 fold , , while in BY4742 we observed a reduction of 14 fold. Thus, our data is in close agreement with all earlier studies. We used the values in Table S1 to calculate the ratios of %FOAR in the mutant strains versus the %FOAR in the isogenic wild type BY4742 strain. These ratios provide quantitative assessment of the effect of each gene on the silencing of URA3 in each individual construct.
The deletion of SAS2 and SAS3 caused 10-100 fold de-repression in URA3-tel, GF2 and GF3, whereas the deletion of YNG1 ((a modulator of SAS3 activity in the NuA3 complex ) and Rtt109 caused 5-50 fold decrease of repression (Fig. 2B). In contrast, the deletion of HAT1 and GCN5 moderately (2-10 fold) increased repression (Fig. 2B). The gain in silencing in Δhat1 and Δgcn5 cells is comparable to the effect of the deletion of RIF1 (Fig. 2B), a key telomeric anti-silencing factor. We do not understand the mechanisms that lead to these somewhat surprising effects for HAT genes. However, the similarity in the magnitude of effects in Δhat1, Δgcn5 and Δrif1 cells indicates that the increase in repression in Δhat1 and Δgcn5 is significant.
The URA3-tel, GF2 and GF3 constructs (shown on top) were integrated in the strains shown on the left and the level of URA3 silencing was calculated as %FOAR cells. A) Levels of URA3 silencing (%FOAR). The 0-10% range is spread out to properly show differences at very low levels of silencing. B) Ratios of %FOAR in the mutant strains versus the %FOAR in the wild type (BY4742) strain. Data is from Table S1. The arrows underneath the exponential graph indicate increase or decrease of silencing.
Hence, at telomeres lacking core X or Y' elements different HATs operate by different mechanisms and can produce both positive and negative effects on TPE. As expected, the addition of STARs in GF2 and GF3 further reduced the level of silencing in Δsas2, Δsas3, rtt109 and Δyng1 cells. Surprisingly, the calculations for Δhat1 and Δgcn5 cells showed that the addition of STARs generated modest, but consistent increase in telomeric silencing. It is conceivable that STAR activity is diminished in these mutants. Alternatively, the overall increase of telomeric silencing in them can over-compensate for the anti-silencing effect of STARs. We deal with this ambiguity in Fig. 6.
The calculations of %FOAR in the mutant strains versus %FOAR in the wild type strain in GF6, GF9, GF44, GF46 and GF61 revealed that the silencing of these reporters was marginally influenced by the deletions of individual HAT genes (Fig. 3B). All these reporters contain a single copy of core X or Y' (black rectangles in the graphs shown on top of Figure 3). Hence, the strong repression or anti-repression effects, which were observed in URA3-tel, GF2 and GF3 (Fig. 2B) were dramatically reduced by the addition of core X or Y' regardless of the position of these elements relative to URA3 and the telomere. The consistent decrease of silencing abbearations in all mutants and constructs strongly suggests that the subtelomeric core X and Y' curtail variations in TPE and maintain the epigenetic plasticity of these loci.
The GF6, GF9, GF44, GF46 and GF61 constructs (shown on top) were integrated in the strains shown on the left and the level of URA3 silencing was calculated as %FOAR cells. A) Levels of URA3 silencing (%FOAR). B) Ratios of %FOAR in the mutant strains versus the %FOAR in the wild type (BY4742) strain. Data is from Table S1. The arrows underneath the exponential graph indicate increase or decrease of silencing.
Core X and Y' curtail variations in TPE in cdc6-1 and Δrif1 cells
We tested if the observed “cushioning” behavior of X and Y' is similar in non-HAT mutants. For these analyses we selected cdc6-1 and Δrif1 cells. Rif1p counteracts the association of Sir3p/4p with the telomere-bound Rap1p , . Consequently, the deletion of RIF1 boosts telomeric silencing . On the other hand, the cdc6-1 mutation dramatically reduces telomeric silencing independently of the ACS proto-silencers positioned in the core X and Y' elements . Hence, these two mutations provide two opposing effects on TPE that are not directly mediated by core X and Y'. In Fig. 4B we show the analysis of telomeric silencing in these two mutants. As expected, cdc6-1 and Δrif1 significantly decreased or increased the silencing of URA3 in the constructs lacking core X and Y' (URA3-tel, GF2, GF3). These effects were not seen in the constructs with core X and Y' (GF6, GF9, GF46). In conclusions, we observed that core X and Y' can curtail both positive and negative effects on TPE in diverse mutants.
The URA3-tel, GF2, GF3, GF6, GF9 and GF46 constructs (shown on top) were integrated in the strains shown on the left and the level of URA3 silencing was calculated as %FOAR cells. A) Levels of URA3 silencing (%FOAR). The 0-10% range is spread out to properly show differences at very low levels of silencing. B) Ratios of %FOAR in the mutant strains versus the %FOAR in the wild type strain. Wild type depicts BY474 for Δrif1 and W303 (not shown) for cdc6-1. The arrows underneath the exponential graph indicate increase or decrease of silencing.
ACS and STAR confer opposing activities upon deletion of GCN5 and Rtt109
Subtelomeric ACS function as weak silencers , which relay the silencing signals emitted by the telomere. Recently we have demonstrated that in several strains, which harbor mutations in replication factor genes, ACS convert to weak anti-silencers . Is it then possible that the cushioning effect of core X and Y' is linked to similar conversions of these ACS? We tested this possibility by destroying the ACS in two of the constructs to produce GF6ΔACS and GF44ΔACS. We introduced these constructs in HAT-deletion mutants and then calculated the ratios %FOARGF6ΔACS/%FOARGF6 and %FOARGF44ΔACS/%FOARGF44. The results are shown in Figure 5. The deletion of ACS in both GF6 and GF44 reduced the silencing in BY4742, Δsas2, Δsas3, Δyng1 and Δhat1 cells. In contrast, the destruction of ACS had very little effect in Δgcn5 and Δrtt109 cells. This observation suggests that GCN5 and Rtt109 directly or indirectly stimulate the silencing activity of subtelomeric ACS. At this point we can not explain the mechanism of their action. We also noticed that the deletions of SAS2, SAS3, YNG1 and HAT1 did not alter the ACS-dependent silencing in GF6 relative to wild type cells, while in GF44 there was about two-fold reduction in these mutants. The differences between GF6 and GF44 are obviously caused by the different position of core X, but at present we cannot explain the nature of this specific effect.
GF6, GF6ΔACS, GF44 and GF6ΔACS constructs (shown on the left) were integrated in the strains shown along the vertical axis and the level of URA3 silencing was calculated as %FOAR cells. A) Levels of URA3 silencing (%FOAR). The levels of silencing of ACS-containing (black bars) and ACS-less (grey bars) constructs are shown side by side. B) Ratios of %FOAR in ACS-less versus ACS-containing constructs. The ratios %FOARGF6ΔACS/%FOARGF6 and %FOARGF44ΔACS/%FOARGF44 were calculated and plotted. The arrows underneath the exponential graph indicate increase or decrease of silencing.
Another set of experiments was conducted to directly assess the effects of STARs within the mutant strains by comparing the levels of silencing in STAR-less (URA3-tel and GF6ΔSTAR) and STAR containing (GF2 and GF6) constructs. Our calculations showed that the STAR in GF2 was 2-3 fold more efficient in Δsas2, Δsas3 and Δyng1 cells relative to wild type cells, but 4-6 fold less efficient in Δrtt109, Δhat1 and Δgcn5 cells. The STAR in the core X-containing GF6 operates at marginal efficiency. These observations demonstrated that core X can dominantly suppress the contribution of STARs to the overall level of gene silencing and that STARs probably function through the joint activity of Rtt109, HAT1 and GCN5. More importantly, the deletions of Rtt109 and GCN5, which have reduced the anti-silencing activity of the tested STAR (Fig. 6B) have also reduced the silencing activity of the ACSs proto-silencers in core X (Fig. 5B). These observations provide a plausible mechanism for the chromatin modulating activity of core X and Y'.
The URA3-tel, GF2, GF6 and GF6ΔACS constructs (shown on top) were integrated in the strains shown along the vertical axis and the level of URA3 silencing was calculated as %FOAR cells. A) Levels of URA3 silencing (%FOAR). The 0-10% range in the upper graph is spread out to properly show differences at very low levels of silencing. The levels of silencing of STAR-containing (grey bars) and STAR-less (black bars) constructs are shown side by side. B) Ratios of %FOAR in STAR-less versus STAR-containing constructs. The ratios %FOARURA3-tel/%FOARGF2 and %FOARGF6ΔSTAR/%FOARGF6 were calculated and plotted. The arrows underneath the exponential graph indicate increase or decrease of silencing.
The comparison of eight recombinant telomeres in eight mutant strains has clearly demonstrated that core X and Y' elements curtail extreme changes in TPE. We show that telomeres without core X and Y' elements are subject to significant shifts towards de-repression or repression upon deletion of HAT genes (Figures 2, 4). In contrast, TPE remains largely undisturbed in core X- and Y'- containing telomeres (Figures 3, 4). In an earlier study we have also observed that the anti-silencing caused by mutations in DNA replication factors is also reduced by core X- and Y' . Whereas the precise mechanism of the effects of each individual HAT or replication factor mutation remains unknown, it is apparent that core X- and Y' moderate all these effects. We also need to point out that the synthetic core X- and Y'- containing telomeres display moderate deviations in TPE that compare in magnitude the effects observed at natural telomeres , , , , .
It has been previously shown that core X and Y' contribute to gene repression, and that subtelomeres contain anti-silencing modules such as the STARs , , , . The opposing signals emitted by these elements have been implicated in the variegated nature of subtelomeric gene expression . An important feature of TPE at individual telomeres is that despite the seemingly random conversion between active and repressed state, the proportion of cells with active/repressed genes remains stable. The mechanisms that sustain this meta-stable balance are not so well understood. Here we propose the subtelomeric core X and Y' could play a significant and unexpected role in the dynamic meta-stability of telomeric gene expression. Previous studies have provided extensive evidence in support of their ability to reconstitute telomeric gene repression when silencing is decreased , , , . For this reason, core X and Y' are generally viewed as proto-silencers. Our data show that these elements can also reduce telomeric gene repression when silencing increases.
We propose that these elements contain not only individual proto-silencers such as ACS and binding sites Rap1p and Abf1p , but also some unidentified anti-silencers. These anti-silencers are independent of the previously characterized STARs. Ultimately, the multiplicity of individual weak proto-silencers and anti-silencers in core X and Y'44 build up “buffering” cis-elements, which suppress extreme variations in TPE. Such individual elements can acquire opposing activities upon changes of environment or in different genetic contexts. Indeed, we show that the deletion of GCN5 or Rtt109 reduces both the anti-silencing activity of a STAR and the silencing activity of an ACS (Fig. 4). Consequently, the net effect of the deletions of these two genes on the tested core X- and Y'- containing telomeres is minimal.
What are the STARs?
STARs have been characterized as anti-silencing modules residing in proximity of core X and Y' elements . Independently of the core X and Y', STARs reduce silencing when introduced in a modified HMR mating type locus . The mechanism of action of STARs is largely unknown. They contain binding sites for Tbf1p and Reb1p thus implicating these two proteins in STAR activity , , but additional details are missing. Here show that GCN5, RTT109 and HAT1 affect the strength of STAR activity (Fig. 6B). It is therefore possible that Tbf1p and Reb1p promote the activity of these HATs. Finally, STARs significantly reduce the silencing only at telomeres, which do not contain core X or Y' (Fig. 4). Hence, core X and Y' activity seems dominant relative to STARs.
Technical issues in studies on TPE
Several earlier studies have pointed out significant discrepancies in the silencing at natural telomeres and at synthetic telomeres on truncated chromosomes. For example, the deletion of SAS2 had caused 10-50 fold reduction of silencing of the simple truncated URA3-tel reporter , . Yet, RT-PCR or microarray analyses of natural subtelomeric genes had shown very moderate (two fold) alteration in expression in Δsas2 cells , , .
In this study we show that synthetic telomeres, which contain core X and Y' elements, closely recapitulate the modest effects of the deletion of SAS2 at natural telomeres. The same moderate effects apply for all other HATs tested. Hence, analyses of telomeric reporters, which contain core X/Y' elements, present a solid alternative to the analyses at natural telomeres.
On the other hand, “complex” synthetic telomeres can muffle weak effects on TPE. For example, studies on SAS3 have been said to be hampered by the lack of readily detectable phenotypes . Here we demonstrate a readily detectable effect of the deletion of SAS3. Indeed, the deletion of SAS3 reduces telomeric silencing as strongly as the deletion of SAS2 (Fig. 2). Therefore, “simple” synthetic telomeres need to be used for the analysis of weak silencing effects.
Role of different HATs in TPE
This study has been initiated as a screen for the effects of different HATs on TPE before it has refocused on the consistent effects of core X and Y'. Consequently, we provide abundant data on the effects of HAT deletions on TPE. Whereas none of these effects is guaranteed to be direct, two points of potential significance need to be raised.
The first point is the modest but consistent reduction in the efficiency of STARs in Δrtt109, Δhat1 and Δgcn5 (Fig. 4). As mentioned, very little is known about the mode of operation of these cis-elements. It is premature to suggest that STARs recruit these HATs. The weak effects of Rtt109, HAT1 and GCN5 corroborate this notion. It is more likely that these subtelomeric regions somehow confer access to HATs, which can passively act to disrupt the spreading of heterochromatin. This hypothesis should be tested by focused mechanistic studies in single and double mutants in these genes.
The other point of discussion is the similarity in the effects of SAS2, SAS3, YNG1 and Rtt109 on simple telomeres. SAS2 counteracts the deacetylation of H4-K16 by Sir2p , , , . Hence, in these meticulous studies SAS2 is acting as an anti-silencing factor. However, at simple telomeres or modified mating type loci the deletion of SAS2 causes dramatic loss of repression therefore portraying SAS2 as a silencing factor (Fig. 2 and , , , ). It is possible that loss of boundary activity and/or the redistribution of a limiting silencing factor such as Sir3p , ,  could indirectly produce these effects. If so, SAS3 and Rtt109 could also act to limit the indiscriminate association of silencing factors to chromatin away from the telomere as is the case with SAS2 , . The possible role of these HATs in boundary formation should also be considered. In summary, the present study provides clues for the possible roles of HATs in TPE. The actual mechanism of their action will be addressed in future studies.
Levels of gene silencing in different mutants.
We thank Genevieve Fourel for generous donation of telomeric reporters; J. Rine and B. Stillman for yeast strains.
Conceived and designed the experiments: MAR KY. Performed the experiments: PP DJ MAR AC. Analyzed the data: MAR KY. Contributed reagents/materials/analysis tools: MAR DJ PP AC. Wrote the paper: KY.
- 1. Rusche LN, Kirchmaier AL, Rine J (2003) The establishment, inheritance, and function of silenced chromatin in Saccharomyces cerevisiae. Annu Rev Biochem 72: 481–516.
- 2. Fourel G, Lebrun E, Gilson E (2002) Protosilencers as building blocks for heterochromatin. Bioessays 24: 828–835.
- 3. Pryde FE, Louis EJ (1999) Limitations of silencing at native yeast telomeres. Embo J 18: 2538–2550.
- 4. Fourel G, Revardel E, Koering CE, Gilson E (1999) Cohabitation of insulators and silencing elements in yeast subtelomeric regions. Embo J 18: 2522–2537.
- 5. Lebrun E, Revardel E, Boscheron C, Li R, Gilson E, et al. (2001) Protosilencers in Saccharomyces cerevisiae subtelomeric regions. Genetics 158: 167–176.
- 6. Boscheron C, Maillet L, Marcand S, Tsai-Pflugfelder M, Gasser SM, et al. (1996) Cooperation at a distance between silencers and proto-silencers at the yeast HML locus. Embo J 15: 2184–2195.
- 7. Fourel G, Magdinier F, Gilson E (2004) Insulator dynamics and the setting of chromatin domains. Bioessays 26: 523–532.
- 8. Muller HJ, Altenburg E (1930) The Frequency of Translocations Produced by X-Rays in Drosophila. Genetics 15: 283–311.
- 9. Hardy CF, Sussel L, Shore D (1992) A RAP1-interacting protein involved in transcriptional silencing and telomere length regulation. Genes Dev 6: 801–814.
- 10. Wotton D, Shore D (1997) A novel Rap1p-interacting factor, Rif2p, cooperates with Rif1p to regulate telomere length in Saccharomyces cerevisiae. Genes Dev 11: 748–760.
- 11. Marcand S, Wotton D, Gilson E, Shore D (1997) Rap1p and telomere length regulation in yeast. Ciba Found Symp 211: 76–93;discussion 93-103.
- 12. Allis CD, Berger SL, Cote J, Dent S, Jenuwien T, et al. (2007) New nomenclature for chromatin-modifying enzymes. Cell 131: 633–636.
- 13. Dang W, Steffen KK, Perry R, Dorsey JA, Johnson FB, et al. (2009) Histone H4 lysine 16 acetylation regulates cellular lifespan. Nature 459: 802–807.
- 14. Kimura A, Umehara T, Horikoshi M (2002) Chromosomal gradient of histone acetylation established by Sas2p and Sir2p functions as a shield against gene silencing. Nat Genet 32: 370–377.
- 15. Suka N, Luo K, Grunstein M (2002) Sir2p and Sas2p opposingly regulate acetylation of yeast histone H4 lysine16 and spreading of heterochromatin. Nat Genet 32: 378–383.
- 16. Kozak ML, Chavez A, Dang W, Berger SL, Ashok A, et al. (2010) Inactivation of the Sas2 histone acetyltransferase delays senescence driven by telomere dysfunction. EMBO J 29: 158–170.
- 17. Ehrentraut S, Weber JM, Dybowski JN, Hoffmann D, Ehrenhofer-Murray AE (2010) Rpd3-dependent boundary formation at telomeres by removal of Sir2 substrate. Proc Natl Acad Sci U S A 107: 5522–5527.
- 18. Axelrod A, Rine J (1991) A role for CDC7 in repression of transcription at the silent mating-type locus HMR in Saccharomyces cerevisiae. Mol Cell Biol 11: 1080–1091.
- 19. Reifsnyder C, Lowell J, Clarke A, Pillus L (1996) Yeast SAS silencing genes and human genes associated with AML and HIV-1 Tat interactions are homologous with acetyltransferases. Nat Genet 14: 42–49.
- 20. Ehrenhofer-Murray AE, Rivier DH, Rine J (1997) The role of Sas2, an acetyltransferase homologue of Saccharomyces cerevisiae, in silencing and ORC function. Genetics 145: 923–934.
- 21. Xu EY, Kim S, Rivier DH (1999) SAS4 and SAS5 are locus-specific regulators of silencing in Saccharomyces cerevisiae. Genetics 153: 25–33.
- 22. Suka N, Suka Y, Carmen AA, Wu J, Grunstein M (2001) Highly specific antibodies determine histone acetylation site usage in yeast heterochromatin and euchromatin. Mol Cell 8: 473–479.
- 23. Gottschling DE, Aparicio OM, Billington BL, Zakian VA (1990) Position effect at S. cerevisiae telomeres: reversible repression of Pol II transcription. Cell 63: 751–762.
- 24. Rehman MA, Fourel G, Mathews A, Ramdin D, Espinosa M, et al. (2006) Differential requirement of DNA replication factors for subtelomeric ARS consensus sequence protosilencers in Saccharomyces cerevisiae. Genetics 174: 1801–1810.
- 25. Rehman MA, Wang D, Fourel G, Gilson E, Yankulov K (2009) Subtelomeric ACS-containing proto-silencers act as antisilencers in replication factors mutants in Saccharomyces cerevisiae. Mol Biol Cell 20: 631–641.
- 26. Rehman MA, Yankulov K (2009) The dual role of autonomously replicating sequences as origins of replication and as silencers. Curr Genet 55: 357–363.
- 27. Fox CA, Loo S, Dillin A, Rine J (1995) The origin recognition complex has essential functions in transcriptional silencing and chromosomal replication. Genes Dev 9: 911–924.
- 28. Micklem G, Rowley A, Harwood J, Nasmyth K, Diffley JF (1993) Yeast origin recognition complex is involved in DNA replication and transcriptional silencing. Nature 366: 87–89.
- 29. Reifsnyder C, Lowell J, Clarke A, Pillus L (1996) Yeast sas silencing genes and human genes associated with aml and hiv-1 tat interactions are homologous with acetyltransferases. Nat Genet 14: 42–49.
- 30. Howe L, Kusch T, Muster N, Chaterji R, Yates3rd JR, et al. (2002) Yng1p modulates the activity of Sas3p as a component of the yeast NuA3 Hhistone acetyltransferase complex. Mol Cell Biol 22: 5047–5053.
- 31. Fourel G, Miyake T, Defossez PA, Li R, Gilson E (2002) General regulatory factors (GRFs) as genome partitioners. J Biol Chem 277: 41736–41743.
- 32. Koering CE, Fourel G, Binet-Brasselet E, Laroche T, Klein F, et al. (2000) Identification of high affinity Tbf1p-binding sites within the budding yeast genome. Nucleic Acids Res 28: 2519–2526.
- 33. Shia WJ, Li B, Workman JL (2006) SAS-mediated acetylation of histone H4 Lys 16 is required for H2A.Z incorporation at subtelomeric regions in Saccharomyces cerevisiae. Genes Dev 20: 2507–2512.
- 34. Lafon A, Chang CS, Scott EM, Jacobson SJ, Pillus L (2007) MYST opportunities for growth control: yeast genes illuminate human cancer gene functions. Oncogene 26: 5373–5384.
- 35. Wiley EA, Zakian VA (1995) Extra telomeres, but not internal tracts of telomeric DNA, reduce transcriptional repression at Saccharomyces telomeres. Genetics 139: 67–79.
- 36. Maillet L, Boscheron C, Gotta M, Marcand S, Gilson E, et al. (1996) Evidence for silencing compartments within the yeast nucleus: a role for telomere proximity and Sir protein concentration in silencer-mediated repression. Genes Dev 10: 1796–1811.
- 37. Roy N, Runge KW (2000) Two paralogs involved in transcriptional silencing that antagonistically control yeast life span. Curr Biol 10: 111–114.
- 38. Liang C, Stillman B (1997) Persistent initiation of DNA replication and chromatin-bound MCM proteins during the cell cycle in cdc6 mutants. Genes Dev 11: 3375–3386.
- 39. Ehrenhofer-Murray AE, Gossen M, Pak DT, Botchan MR, Rine J (1995) Separation of origin recognition complex functions by cross-species complementation. Science 270: 1671–1674.
- 40. Dillin A, Rine J (1997) Separable functions of ORC5 in replication initiation and silencing in Saccharomyces cerevisiae. Genetics 147: 1053–1062.
- 41. Pasero P, Duncker BP, Schwob E, Gasser SM (1999) A role for the Cdc7 kinase regulatory subunit Dbf4p in the formation of initiation-competent origins of replication. Genes Dev 13: 2159–2176.
- 42. Dziak R, Leishman D, Radovic M, Tye BK, Yankulov K (2003) Evidence for a role of MCM (mini-chromosome maintenance)5 in transcriptional repression of sub-telomeric and Ty-proximal genes in Saccharomyces cerevisiae. J Biol Chem 278: 27372–27381.