Sir- and Silencer-Independent Disruption of Silencing in Saccharomyces by Sas10p

Corresponding author: Jasper Rine, Division of Genetics, Department of Molecular and Cell Biology, 401 Barker Hall, University of California, Berkeley, CA 94720, jrine{at}uclink4.berkeley.edu (E-mail).

Communicating editor: F. WINSTON

	ABSTRACT

TOP ABSTRACT MATERIALS AND METHODS RESULTS DISCUSSION LITERATURE CITED

A promoter fusion library of Saccharomyces cerevisiae genes was used to exploit phenotypes associated with altered protein dosage. We identified a novel gene, SAS10, by the ability of Sas10p, when overproduced, to disrupt silencing. The predicted Sas10p was 70,200 kD and strikingly rich in charged amino acids. Sas10p was exclusively nuclear in all stages of the cell cycle. Overproduction of Sas10p caused derepression of mating type genes at both HML and HMR, as well as of URA3, TRP1, and ADE2 when inserted near a telomere or at HMR or the rDNA locus. Repressed genes not associated with silenced chromatin were unaffected. Sas10p was essential for viability, and the termination point following Sas10p depletion was as large budded cells. Remarkably, Sas10p overproduction disrupted silencing even under conditions that bypassed the requirement for Sir proteins, ORC, and Rap1p in silencing. These data implied that Sas10p function was intimately connected with the structure of silenced chromatin.

MACROMOLECULAR assemblies in biology are often sensitive to the relative dosage of different protein components. Notable examples of this principle include the assembly of T4 phage heads (FLOOR 1970 ) and the importance of balanced histone protein levels for the proper segregation of chromosomes (MEEKS-WAGNER and HARTWELL 1986 ). Although mutations in many genes encoding regulators of gene expression are fully recessive to wild type, certain classes of regulatory genes commonly exhibit gene–dosage effects. In particular, the phenotypes associated with position effect variegation in Drosophila (reviewed in WEILER and WAKIMOTO 1995 ) can be enhanced or suppressed by heterozygous mutations in a large number of genes referred to as Su(Var)s or E(Var)s, respectively. The dosage sensitivity caused by Su(Var) and E(var) mutations has fostered the view that these genes encode structural components of chromatin (REUTER and SPIERER 1992 ).

Chromatin structure plays a central role in several if not all aspects of gene regulation in Saccharomyces. One well-studied case includes the genes that regulate mating type. The mating type of haploid yeast cells is determined by the genes present at the MAT locus. MATa cells can mate with MAT cells, and vice versa, to produce a/ diploids. There are two additional copies of mating-type genes on the distal arms of chromosome III at the HML and HMR loci. In most yeast strains, HML contains an unexpressed but intact copy of the MAT allele, whereas HMR contains an unexpressed intact copy of the MATa allele. Expression of HML and HMR is blocked by a mechanism called silencing (reviewed in LOO and RINE 1995 ). In addition to HML and HMR, the telomeres of yeast chromosomes (GOTTSCHLING et al. 1990 ), as well as the rRNA genes (BRYK et al. 1997 ; SMITH and BOEKE 1997 ), are also subject to silencing.

Silencing is achieved by the concerted action of regulatory sites called silencers, proteins that bind these sites, and additional proteins that bind nucleosomes (PILLUS and GRUNSTEIN 1995 ). The best-characterized silencer is the HMR-E silencer, which contains binding sites for Rap1p (SHORE and NASMYTH 1987 ), and Abf1p (LOO et al. 1995 ). In addition, HMR-E contains an autonomous replication sequence element (ARS), which binds the origin recognition complex (ORC; FOSS et al. 1993 ). Synthetic silencers containing only these three binding sites, in place of the wild-type silencer, can efficiently repress transcription, indicating that these sites are sufficient for silencing HMR (MCNALLY and RINE 1991 ). Nevertheless, there are differences between the synthetic and wild-type HMR-E silencers primarily involving apparent redundancy of function in the wild-type silencer.

In addition to the proteins that bind the silencer, several other proteins play a role in silencing, such as the four Sir proteins (RINE and HERSKOWITZ 1987 ). Mutations in the histone genes encoding H3 and H4 also lead to derepression of the silent loci (reviewed in PILLUS and GRUNSTEIN 1995 ) pin-pointing chromatin structure as having a causal role in the mechanism of silencing.

Recent data suggest that Sir2p, Sir3p, and Sir4p form a multimeric complex (MOAZED and JOHNSON 1996 ) that may, in turn, interact with the hypoacetylated histones H3 and H4 in nucleosomes (HECHT et al. 1995 ; STRAHL-BOLSINGER et al. 1997 ), leading to the formation of a silenced chromatin domain at HMR that is inaccessible to various enzymatic probes (LOO and RINE 1994 ). Thus, it is likely that Sir proteins are structural components of the silenced domain. Like heterochromatin proteins in Drosophila, the effect of Sir3p and Sir4p appears to be partially dosage dependent (LE et al. 1997 ; MARSHALL et al. 1987 ). Increased dosage of Sir3p results in the spreading of the silent domain (RENAULD et al. 1993 ; STRAHL-BOLSINGER et al. 1997 ). In addition, cells overexpressing Sir3p have slightly elevated chromosome loss rates, while cells heterozygous for a sir3 mutation have slightly elevated mitotic recombination (PALLADINO et al. 1993 ).

To further our current understanding of the mechanism by which silenced domains are formed, we screened for genes that encode proteins whose proper dosage is essential for silencing. We identified a novel protein, Sas10p, whose overexpression derepresses gene expression at both HMR and HML, as well as at the rDNA locus and telomeres. The SAS10 gene is essential for viability, and Sas10p is a nuclear protein. Depletion experiments revealed that this protein is primarily required in the G2/M phase of the cell cycle. Finally, overexpressed Sas10p disrupted silencing, even under conditions that bypassed a requirement for Sir, Rap1p, and Orc in silencing.

	MATERIALS AND METHODS

TOP ABSTRACT MATERIALS AND METHODS RESULTS DISCUSSION LITERATURE CITED

The S. cerevisiae strains used are listed in Table 1.

Yeast transformation with a GAL1-cDNA library:
A cDNA library containing primarily full-length cDNAs cloned under the control of the GAL1 promoter in pRS316 was a gift from A. BRETSCHER (LIU et al. 1992 ) and A. STRAIGHT. JRY5322 cells were transformed by the lithium acetate method (ITO et al. 1983 ) with 2–3 µg of the GAL1p-cDNA library plasmid DNA. Approximately 30,000 transformants were plated onto glucose-casamino plates supplemented with adenine, tryptophan, and leucine, selecting for Ura+ prototrophs. Expression of genes on the transforming plasmids was induced by replica plating cells to galactose-casamino acids medium, followed by replica plating to YM-galactose plates spread with an a mating lawn.

Quantitative mating assays were performed as described (LOO et al. 1995 ). The SAS10 gene was sequenced on both strands using directed primers. The sequence obtained was 100% identical to the sequence released by the international yeast genome sequencing effort.

Disruption of SAS10:
Oligonucleotides containing sequences corresponding to the TRP1 gene flanked by sequences corresponding to the 5' and 3' ends of SAS10 were used to amplify the TRP1 gene from the vector pRS 404 (SIKORSKI and HIETER 1989 ) using Taq DNA polymerase (Roog19 = GAAGACGAAG ATGAGGAAGA AGTCTTAGCC ATGGATGAAG ATGATGAGTC AATAGATGAA CGACATTACT ATA TATATAA; Roog 20 = CCTTGTTTCT TTTAGGCGTTAG AGACCCTTGT TCTTTAAAAT TTGATAATTT AATGGCAC CT GCAGGCAAGT GCACAA). The PCR fragment generated was cotransformed with pRO14 bearing SAS10 into a trp1-1 strain, and Trp+ prototrophs were selected in which the SAS10-coding sequence on the plasmid had been replaced with TRP1 by homologous recombination. The resulting plasmid (pRO15) was digested with MluI and NdeI to release the disrupted SAS10 gene fragment that was used to transform the diploid strain JRY5503. Trp+ prototrophs were analyzed by DNA blot hybridization (HOFFMAN and WINSTON 1987 ), and the sas10-null phenotype was determined by tetrad analysis.

To isolate a haploid strain bearing sas10::TRP1, the SAS10 heterozygous diploid was first transformed with pRO14 (pRS316 with GAL1p-SAS10) before dissection of the asci resulting from sporulation of the diploid. Dissections were performed on YP-Gal plates that were top spread with 300 µl YPD.

Plasmid construction:
Subcloning of SAS10 in pRS413: Plasmid pRO14 (pRS316 with GAL1p-SAS10) was digested with PvuII, and the fragment containing the SAS10 gene under the GAL1 promoter was gel purified and cloned into the SmaI site of pRS413 to generate pRO17. The structure of the promoter fusion was confirmed by DNA sequence analysis.

GFP-SAS10 construct: pRO25 containing the glycerol-3-phosphate dehydrogenase promoter (GPDp) driving GFP-RAS2 with a PGK1 (phospho glycerokinase1) terminator (gift from J. WHISTLER) was digested with EcoRI and SalI to release the RAS2 gene fragment. The SAS10 gene fragment from pRO14 was PCR amplified with oligonucleotides containing EcoRI and SalI sites (Roog 40 = GGGTGTATAGAATTCATGGTACGCAAAGGCTC; Roog41 = CAAAAATTTT GTCGACACTAC TTGGGTAAA TAC). The PCR product was digested with EcoRI and SalI and cloned into pRO25 by a two-step process to generate plasmid pRO28, which contained green fluorescent protein (GFP) fused in frame with the SAS10-coding sequence.

Subcloning of SAS10 into YIp lac128 and integration of multiple copies of GAL1p-SAS10:
The SAS10 gene under the GAL1 promoter was isolated as a PvuII fragment from pRO14 and cloned into YIp lac128 (GIETZ and SUGINO 1988 ) digested with SmaI to generate pRO34. pRO34 was digested with AflII, and the linearized DNA fragment was used to transform W303-1A or JRY3009. Transformants were selected for Leu+ prototrophy, and the integration was checked by DNA hybridization analysis using various restriction enzymes.

Arrest phenotype:
JRY5506 contained a sas10::TRP1 disruption but could propagate because of the presence of GAL1p-SAS10::LEU2. This strain was grown overnight in galactose-YP media. Cells were then transferred to fresh medium containing either 2% galactose or glucose as a carbon source, and aliquots of cells were removed after 24 or 36 hr, washed in PBS, and fixed with formaldehyde overnight. The cells were washed repeatedly in PBS and finally in water, sonicated briefly, stained with DAPI, and visualized under the fluorescent microscope.

	RESULTS

TOP ABSTRACT MATERIALS AND METHODS RESULTS DISCUSSION LITERATURE CITED

A screen for genes that disrupted silencing when overexpressed:
Silencing of HMR depends on the proper level of a variety of proteins. Mutations in a number of genes lead to loss of silencing, as does overproduction of the proteins encoded by a subset of these genes (e.g., SIR4). We screened for new proteins that, when overproduced, lead to loss of silencing to gain a fuller description of proteins involved in silencing and, in particular, to search for new components of silenced chromatin.

These experiments used a strain that lacked functional gene information at HML (hmlap) and MAT (matap) (JRY5322). This strain carried the MAT information under the control of a synthetic silencer (HMRss) at HMR. A haploid yeast strain with no mating information at the MAT locus mates as an a cell, as long as the mating type information at HMRss is repressed. However, unlike MATa, matap is recessive to MAT. Therefore, any protein that led to derepression of HMRss would result in a phenotypic switch in the mating phenotype of this strain from a to (Figure 1A). The sensitivity of the screen was tested by overexpressing the S-phase protein kinase Cdc7p, which has been previously shown to derepress HMR upon overexpression or loss of function (AXELROD and RINE 1991 ). As predicted, overexpression of Cdc7p led to derepression of HMRss and resulted in a switch in the mating phenotype of these cells (data not shown).

View larger version (82K): In this window In a new window Download PPT slide   Figure 1. —(A) A schematic representation of the genetic screen used to isolate SAS10. Derepression of HMR by overexpressed genes in a GAL1-cDNA library resulted in a phenotypic switch in mating of the strain JRY5322 from a to . (B) Overexpression of SAS10 and SIR4 led to derepression of HMRss. Derepression of HMRss was examined in JRY5322 cells bearing either the vector alone, SIR4C, or SAS10 under the GAL1 promoter on a centromere-based plasmid. The expression of Sas10p and the Sir4p C-terminal fragment was induced by plating patches of JRY5322 transformants on casamino-galactose plates, followed by replica plating onto mating-type tester lawns.

To identify proteins that disrupted the silent state when overexpressed, JRY5322 was transformed with an S. cerevisiae cDNA library in which the genes were under the control of the GAL1 promoter (LIU et al. 1992 ). Cells were transformed with the plasmid library selecting for Ura+ prototrophs on minimal medium with glucose as the carbon source. Transformants were replica plated onto galactose-containing medium to induce expression of the cDNAs, and derepression of HMRss was monitored by replica-plating the colonies onto galactose-containing minimal medium plates spread with a mating lawn of MATa cells (JRY19). Of ~30,000 colonies screened, two colonies displayed a galactose-dependent, -mating phenotype, indicative of derepression of HMR (Figure 1B). Control experiments confirmed that the derepression of HMRss was plasmid, insert, and galactose dependent (data not shown).

DNA sequence analysis of one of these plasmids that conferred derepression of HMR revealed that the clone encompassed the C-terminal fragment of the SIR4 gene (SIR4c). This clone expressed the C-terminal fragment of Sir4p from amino-acid residue 1050 to the terminus under the control of the GAL1 promoter. Overexpression of Sir4p is known to derepress HMR (MARSHALL et al. 1987 ), and its isolation in this screen served as a positive control (Figure 1B).

The complete DNA sequence of the second clone was determined and revealed a novel, full-length uncharacterized gene. This gene was also identified in the course of the yeast genome sequencing project (locus YDL153c). This gene was named SAS10 for Something About Silencing. SAS10 encodes a predicted 610–amino acid protein with an apparent molecular mass of 70,200 kD. The gene lies on chromosome IV (from coordinates 183019 to 181187) and was divergently transcribed from the adjacent RPC53 gene (MANN et al. 1992 ). Derepression by SAS10 appeared to be partial because JRY5322 exhibited a bimating phenotype, with colonies of cells mating both with an a lawn and an lawn (Figure 1B). The extent of derepression at HMR in the matap HMRss strain (JRY 5322) was quantitated using MATa cells as the tester lawn. The results showed that relative to a mating efficiency of <10^-5 in wild-type cells in which the HMRss locus was silenced, cells overexpressing the C-terminal fragment of SIR4 gave a mating efficiency of 6 x 10^-2, while overproduction of SAS10 led to a mating efficiency of 3 x 10^-3 in the same strain.

The protein sequence of Sas10p (Figure 2A) was of striking composition, with long clusters of acidic residues in the N-terminal half of the protein and a high concentration of basic residues in the C-terminal half of the protein (Figure 2B). Dot matrix analysis indicated that the acidic part of the protein was composed of two repeats, suggesting that the protein may have a tripartite structure with two acidic domains and a C-terminal basic domain. A protein database search using Sas10p revealed that the protein encoded by SAS10 was homologous to mouse and human expressed sequence tags (EST) generated from embryonic cDNA libraries (Figure 2C), indicating that the C-terminal domain of SAS10 was conserved in other eukaryotes.

View larger version (44K): In this window In a new window Download PPT slide   Figure 2. —(A) Predicted amino-acid sequence of Sas10p. The putative nuclear localization signal is highlighted by a shaded box. (B) Distribution of charged residues in Sas10p. (C) Sequence alignment between Sas10p and human and mouse EST sequences in GenBank. The vertical bars represent amino-acid similarities, and the asterisks represent amino-acid identities between all three sequences.

Overexpression of Sas10p led to derepression of HML and HMR:
SAS10 was isolated from a strain in which the MAT genes were controlled by a synthetic silencer at HMR-E. To determine whether SAS10 overexpression would also derepress the wild-type alleles of both HML and HMR, a wild-type pair of strains was transformed with either the parent vector or the vector containing the SAS10 gene under control of the GAL1 promoter (GAL1p-SAS10). Overexpression of SAS10 was induced by growing cells on galactose-containing medium, and the cells were then plated on mating-type tester lawns. Sas10p overexpression or overexpression of the Sir4 C-terminal fragment both led to derepression of wild type, HML, and HMR. As observed with the synthetic silencer (Figure 1B), the SIR4 protein C-terminal fragment led to complete derepression of the silent loci, whereas overexpression of SAS10 led to only a partial derepression, as reflected by the reduced diploid formation on mating lawns (Figure 3).

View larger version (69K): In this window In a new window Download PPT slide   Figure 3. —Overexpression of Sas10p led to partial derepression of HML and HMRa. Derepression of HML and HMRa was examined in W-303 transformants bearing either the vector alone, SIR4c (pRO-135), or SAS10 (pRO14) under the GAL1 promoter on a centromere-based plasmid. The overexpression of Sas10p and Sir4p was induced by patching transformants on YM-galactose plates, followed by replica plating onto mating lawns.

Quantitative mating assays were performed to assess the degree of Sir4p- and Sas10p-mediated derepression of HMR. While the mating efficiency of wild-type cells was 0.26, overexpression of the SIR4 C-terminal fragment led to >10⁵-fold loss of mating efficiency, and overexpression of SAS10 led to a 10³-fold loss of mating efficiency.

Sas10p overexpression primarily affected the establishment of the silent state:
The partial derepression at HMR observed in cell clones caused by Sas10p overproduction could, in principle, reflect either intermediate levels of expression of HMR in all cells in a colony (an analog model) or, alternatively, a population in which some cells expressed the gene while others did not (a binary switch model; KAMAKAKA 1997 ). To distinguish between these two models, a strain with the ADE2 gene under the control of the HMR silencers was used (SUSSEL et al. 1993 ). Colonies lacking ADE2 function are red, and wild-type colonies are white. If derepression of HMR were partial in all cells, then the colony color would convert from red to pink. In contrast, if the derepression occured in only a fraction of the cells and the transcription states were mitotically stable, then the colonies should exhibit red and white sectors in which ADE2 expression is off or on, respectively.

A strain that maintains the ADE2 gene inserted at HMR in a primarily repressed state (YLS409) formed uniformly dark pink colonies on rich media (Figure 4A). To study the effects of Sas10p on ADE2 gene expression at HMR, the SAS10 gene under the control of the GAL1 promoter was stably integrated at the LEU2 locus, producing two strains, one with two tandem copies (JRY5504) and one with four copies of GAL1p-SAS10 at LEU2 (JRY5505). Both strains were crossed to a strain with ADE2 integrated into HMR (YLS409) to generate haploid strains carrying ADE2 at HMR, along with either two or four SAS10 integrants (JRY5509 and JRY5510, respectively). Overexpression of SAS10 led to derepression of ADE2 at HMR, demonstrating that Sas10p-mediated derepression at the silent loci was not restricted to the mating-type genes. The strain with two copies of GAL1p-SAS10 exhibited partial derepression of ADE2 (Figure 4, compare A to B), whereas the strain with four copies of GAL1p-SAS10 exhibited complete derepression of ADE2 (Figure 4, compare A to C). Thus, the SAS10-mediated phenotype was dose dependent. The difference in the red color of the various strains was caused by the density of colonies at different regions of the plate and did not reflect changes in the extent of silencing. Further analysis revealed that the partial derepression of ADE2 manifested itself as sectored colonies (Figure 4E). In addition to the red sectors in a white colony, white sectors could be seen within red sectors. These results support a binary mode of gene regulation of ADE2 at the HMR locus.

View larger version (51K): In this window In a new window Download PPT slide   Figure 4. —Sas10p-mediated derepression was dosage dependent and leads to derepression of HMR::ADE2. (A–C) Strains YLS409, JRY5509, and JRY5510 containing either zero, two, or four integrated copies, respectively, of the SAS10 gene under a GAL1 promoter were induced by plating cells on YP-Gal plates and allowed to grow for 8–10 days to develop color before photography. (D–F) Enlarged, single colonies from strains YLS409, JRY5509, and JRY5510 demonstrating sectoring of ADE2 gene expression.

The data indicated that transcription states of ADE2 at HMR, once established, were stably inherited for several generations. Sas10p overproduction had a low but measurable probability of derepressing ADE2 at HMR during progression through the cell cycle, and once the derepressed state was established, it was stably propagated through multiple cell divisions. The red sectors within white sectors implied that silencing of ADE2 at HMR could be reestablished, even in the presence of excess Sas10p.

SAS10 overexpression lead to derepression of silencing at telomeres and the rDNA locus:
Silencing in S. cerevisiae at telomeres, HML and HMR share several factors, including dependence on SIR2, SIR3, SIR4, and RAP1 (GOTTSCHLING et al. 1990 ). It was, therefore, of interest to determine whether overexpression of SAS10 also disrupted telomeric silencing. A strain containing the URA3 and TRP1 genes integrated at the telomere on the left arm of chromosome VII was transformed with the Sas10p overexpression plasmid (pRO17). In a wild-type strain (JRY4469), both the URA3 and TRP1 genes are repressed because of telomeric silencing and, hence, the strain is phenotypically Ura- and Trp-. Overexpression of Sas10p derepressed both genes at the telomere moderately, but the 10–100-fold derepression relative to control strains containing the vector alone was observed in eight independent transformants, with each transformant analyzed in triplicate (Figure 5A).

View larger version (77K): In this window In a new window Download PPT slide   Figure 5. —Sas10p overexpression led to derepression of silencing at the telomere and the rDNA locus. (A) Strain JRY4469 containing the URA3 and TRP1 gene at the telomere of chromosome VII-L was transformed with either the vector alone or with the SAS10 gene (pRO17) under a GAL1 promoter on a centromere-based plasmid. Cells were grown selectively in minimal liquid medium with galactose as a carbon source, and 3 µl of 10-fold serial dilutions of the cells were spotted on YM-Gal plates either (A) lacking histidine (for plasmid selection) or (B) lacking histidine and tryptophan to determine derepression of the TRP1 gene. C shows serial dilutions of cells plated on YM-Gal plates lacking histidine but containing 1 mg/ml 5-FOA to also determine derepression of the URA3 gene, whereas D shows plates lacking histidine and uracil (B) to determine derepression of the URA3 gene at the telomere. (B) The parent strain (JB740) and strains containing TY1-mURA3 reporter genes integrated at either an unknown euchromatic region (R31), the NTS1 spacer (S6), or the NTS2 spacer (S3) of the rDNA locus were transformed with either the vector alone or with the SAS10 gene (pRO17) under a GAL1 promoter. These strains were analyzed as described for telomeric silencing for derepression of the URA3 gene.

In addition to the telomeres, SIR2-dependent silencing has recently been shown to occur at the rDNA locus. It was, therefore, of interest to determine whether Sas10p overproduction also affected rDNA silencing. To analyze the effects of SAS10 on rDNA silencing, strains containing URA3 integrated at various locations in the genome were used: one stain contained a single TY1-mURA3 insertion outside the rDNA in an euchromatic region of the genome (R31), a second strain contained a single TY1-mURA3 insertion in the NTS2 nontranscribed spacer of the rDNA locus (S3), and a third strain contained a single TY1-mURA3 insertion in the NTS1 nontranscribed spacer (S6). These strains were transformed with a vector containing SAS10 (pRO17) and analyzed for URA3 expression after overexpression of Sas10p (Figure 5B). In the absence of Sas10p overexpression, the URA3 gene was repressed when present at the rDNA locus (S3 and S6) but derepressed at a euchromatic locus (R31). Overexpression of Sas10p derepressed the URA3 genes at the rDNA locus in both S3 and S6 strains. Interestingly, derepression was greater in the S3 strain compared to the S6 strain, consistent with previous work (SMITH and BOEKE 1997 ) that indicated that repression was weaker at NTS2 than at NTS1.

Silencing is only one form of repression observed in yeast. There are many other genes whose expression is repressed by Sir-independent mechanisms. We tested whether derepression caused by Sas10p overproduction was specific to silencing or could act on genes repressed by other mechanisms. For example, transcriptional activation of the SUC2 and PHO5 genes has been correlated with changes in nucleosome configuration (STRAKA and HORZ 1991 ; WINSTON and CARLSON 1992 ). However, overproduction of Sas10p did not lead to derepression of genes such as PHO5, SUC2, MEV1, or ERG9, indicating that the effect of Sas10p was largely restricted to silenced chromatin domains (data not shown).

Sas10p localized to the nucleus:
The subcellular localization of Sas10p in S. cerevisiae was determined with a chimeric protein containing the GFP fused in frame to the N terminus of SAS10 under the control of the GPDp. This fusion was able to complement a sas10 allele. Fluorescence microscopy indicated that the GFP-Sas10 fusion protein was exclusively nuclear (Figure 6). The GFP fluorescence profile was completely superimposable with the fluorescence of DAPI staining, and the GFP-Sas10p remained nuclear even during mitosis (data not shown). The primary amino-acid sequence of Sas10p revealed putative nuclear localization signals (similar to the SV40 and the bipartite nuclear localization signals) in the C-terminal half of the protein (see Figure 2) which may be involved in its nuclear import. Because Sas10p was nuclear, the phenotype caused by Sas10p overproduction on silencing was most likely a direct effect on nuclear gene expression.

View larger version (49K): In this window In a new window Download PPT slide   Figure 6. —Sas10p localized to the yeast nucleus. Cells carrying a GFP-Sas10 fusion protein under the control of the GPD promoter on a 2µ-based plasmid (pRO28) were grown to log phase, stained with DAPI, and directly visualized under a fluorescent microscope using the FITC filter for GFP. (A) A representation of typical cells viewed under Nomarski optics. (B) The GFP fluorescence of the cells in A. (C) The nuclei of the same cells in the DAPI channel. (D) An overlay of the GFP and DAPI images.

SAS10 depletion caused a cell cycle arrest:
A null allele of SAS10 was created to analyze the effects of SAS10 deletion on cell growth and silencing. To construct a disruption of the SAS10 gene, the coding sequence from amino acids 55–542 was replaced with the TRP1 gene, and the null allele was used to knock out one wild-type allele in a diploid strain that was homozygous trp1-1. Subsequently, the diploid was sporulated and tetrads were dissected. This analysis revealed that SAS10 was an essential gene because all 15 tetrads examined had two viable Trp- spores and two inviable spores. Microscopic analysis of dissected tetrads revealed that SAS10 did not have a role in germination of spores because sas10 spores germinated and gave rise to microcolonies of eight to 60 cells that appeared to arrest with large buds.

In Drosophila, mutations in many genes affecting position effect variegation produce a mutant phenotype when heterozygous, reflecting the haplo-insuffiency of many su(Var) and e(Var) proteins. We tested whether a 50% reduction in the dose of Sas10p had any effects on silencing. To perform this analysis, we compared the mating phenotype of a diploid strain with the genotype matap/matap hmlap/hmlaDp HMRss/HMRss (ROY461) to the mating phenotype of the diploid strain matap/matap hmlap/hmlap HMRss/HMRss sas10/SAS10 (ROY463). Both strains had identical mating behavior, indicating that a 50% reduction in Sas10p level did not result in derepression of HMR (data not shown).

Because SAS10 was an essential gene, the terminal phenotype of a sas10 null allele was determined by using a strain with a sas10::TRP1 allele that also carried an integrated, single wild-type copy of the SAS10-coding sequence under the GAL1 promoter (JRY5506). These cells grew on medium containing galactose but were unable to grow on medium containing glucose, and they ceased to grow after 12–15 hr after a shift from galactose- to glucose-containing medium.

The terminal phenotype of Sas10p depletion was further analyzed by growing cells into log phase in medium containing galactose and then shifting them to glucose-containing medium. Aliquots were removed, fixed with formaldehyde, and examined under the microscope. After the SAS10 shutoff, the majority (70–80%) of the cells were uniformly arrested as large-budded cells (Figure 7, compare B to D). DAPI staining of nuclear DNA showed weak DAPI staining accompanied by the loss of a well-defined nucleus (Figure 7, compare A to C). The arrest phenotype (large-budded cells) associated with loss of Sas10p function is suggestive of cells arrested in late S or G2/M phases of the cell cycle.

View larger version (39K): In this window In a new window Download PPT slide   Figure 7. —(A–D) Sas10p-depleted cells accumulated as large-budded cells with aberrant nuclei. Strain JRY5506 containing sas10::TRP1 with a single integrated copy of GAL1p-SAS10::LEU2 was grown to log phase in YP-galactose and diluted into either YP-glucose or YP-galactose, and samples were removed every 12 hr, fixed with formaldehyde, followed by DAPI staining before visualization under the Axioscope (Carl Zeiss, Thornwood, NY) fluorescent microscope. (C and D) Representative JRY5506 cells grown in glucose for 24 hr and observed either with phase contrast (D) or fluorescence (C). (A and B) Representative JRY5506 cells grown in galactose and viewed either with phase contrast (B) or fluorescence (A).

Sas10p-mediated derepression was independent of Orc, Rap1p, or the Sir proteins:
If the Sas10p-mediated derepression was a consequence of sequestration of silencing factors or disruption of a multimeric silencing complex, then identification of the factors that interact with Sas10p would elucidate the mechanism by which this protein functions. To test whether Sas10p mediated its effect through any of the proteins known to bind the silencer DNA itself, we determined whether Sas10p overexpression could disrupt silencing that was independent of the silencer-binding proteins. For this purpose, we used three different isogenic strains in which the wild-type silencer was replaced with a synthetic silencer. In the first strain (JRY4473), the synthetic silencer contained binding sites for ORC, Rap1p, and Abf1p. In the other two strains (JRY5507 and JRY5508), the synthetic silencer was engineered such that either the ORC- and Rap1p-binding site was replaced by binding sites for Gal4p, respectively. In these strains, repression at HMR was maintained by a Gal4-Sir1 fusion protein, which bypasses the requirement for ORC and Rap1p, respectively (FOX et al. 1997 ). We then tested whether overexpressed SAS10 could derepress HMR in these strains. If Sas10p caused derepression by sequestering either ORC or Rap1p, then it would be unable to exert its effect in a strain in which these proteins are not required for silencing. However, overproduction of Sas10p was able to derepress HMR in the absence of Rap1p- or ORC-binding sites (Figure 8). Thus, ORC and Rap1p were unlikely to be the targets of Sas10p overproduction.

View larger version (48K): In this window In a new window Download PPT slide   Figure 8. —Sas10p-mediated derepression occurs in the absence of bound ORC or Rap1p at the silencer. JRY4473, JRY5507, or JRY5508 were either cotransformed with plasmids bearing GAL1p-SAS10 and ADH1p-GAL4SIR1 or the appropriate control vectors. All three strains carry variations of the synthetic silencer at HMR. JRY4473 has an ARS element and binding sites for Rap1p and Abf1p, whereas JRY5507 has a Gal4-binding site in place of the ARS element, and JRY5508 has a Gal4 binding site in place of the Rap1-binding site. Sas10p-mediated derepression was monitored as described in the legend for Figure 3.

Finally, to test whether Sir2p, Sir3p, or Sir4p could be the targets of Sas10 overproduction, a bypass suppressor of SIR gene function was used. SUM1-1 is a dominant mutation that restores silencing in the absence of any of the Sir proteins (LAURENSON and RINE 1991 ). If Sas10p overproduction interfered with the function of Sir proteins, then overexpression of Sas10p in a sir SUM1-1 strain should not result in derepression of HMR. However, if Sas10p caused derepression of HMR by interacting with other components of silent chromatin, then overexpression should result in derepression, even in the absence of the Sir proteins. SAS10 was overexpressed in strains deleted for the SIR genes and carrying a SUM1-1 mutation that maintained silencing at HMR (Figure 9). Significantly, Sas10p overexpression led to significant derepression in this strain, implying that the Sir proteins themselves were either not the targets of Sas10p or that they were not the only targets of Sas10p.

View larger version (73K): In this window In a new window Download PPT slide   Figure 9. —Sas10p derepressed HMRa in sir2, sir3, and sir4 SUM1-1 cells. The strains JRY2456 (SUM1-1), JRY2466 (sir2::HIS3 SUM1-1), JRY2816 (sir3::LYS2 SUM1-1), and JRY2467 (sir4::HIS3 SUM1-1) are all MAT and were transformed with either the GAL1p-SAS10 plasmid or vector. Sas10p-mediated derepression was monitored as described in the legend for Figure 3.

	DISCUSSION

TOP ABSTRACT MATERIALS AND METHODS RESULTS DISCUSSION LITERATURE CITED

Recent studies on Drosophila strains bearing mutations in components of heterochromatin have led to the model that heterochromatin formation and position effect variegation are mediated by a multimeric complex that is sensitive to the proper balance of its constituent components (HENIKOFF 1996 ). Similar conclusions have been drawn from genetic and biochemical studies on silencing in the yeast S. cerevisiae (LOO and RINE 1995 ; PILLUS and GRUNSTEIN 1995 ). We therefore undertook an altered dosage approach to identify novel proteins that, when overproduced, affect silencing at the HMR locus in S. cerevisiae. Of ~30,000 transformants screened, two clones that affected silencing were isolated. One of these clones encoded the C-terminal fragment of SIR4. Previous data established that overexpressed Sir4p could lead to derepression of HMR (MARSHALL et al. 1987 ). In addition to the Sir4p C-terminal fragment, a second previously uncharacterized gene was isolated, which we have named SAS10.

The predicted sequence of Sas10p was remarkably rich in charged amino acids, which were clustered into one basic and two acidic domains. Putative nuclear localization signals were also evident, consistent with the nuclear localization of Sas10p throughout the cell cycle. The Sas10 protein was essential for the viability of the cell. Because silencing is not essential for viability, Sas10p must have multiple roles in the cell, including a role in silencing. Several other proteins such as Rap1p and Orc have essential functions and also function in silencing (FOSS et al. 1993 ; SHORE and NASMYTH 1987 ).

Haploid sas10 spores from a SAS10/sas10 diploid continued to divide for four to six generations, indicating that Sas10p was present in excess to that needed for a cell, and that it did not need to be synthesized anew in each cell cycle. Similar conclusions were reached from promoter shut-off experiments, which also established that the termination point of cells lacking SAS10 function was in late S or G2/M phases of the cell cycle. These arrested cells exhibited an aberrant nuclear morphology with weak DAPI staining accompanied by the loss of a well-defined nucleus. These results suggest that Sas10p may have a role in the generation of higher-order chromatin structures or a role in the completion of chromatin replication.

Overproduction of Sas10p led to derepression of the mating-type genes at both HMR and HML. Sas10p-mediated derepression was not specific to the mating-type genes because the URA3 and TRP1 gene inserted near a telomere, the URA3 gene inserted at the rDNA locus, and the ADE2 gene inserted at HMR were also derepressed. However, this derepression was not broadly pleiotropic because PHO5, SUC2, MEV1, and ERG9 were not significantly derepressed by overproduction of SAS10. These data suggest that derepression was specific to genes in silenced chromatin. Such a gene-independent derepression of transcription at the silent loci has also been observed for other proteins involved in silencing, such as the Sir proteins. Thus, Sas10p may have a direct role in silencing, but as noted above, it must have a separate and essential role as well.

The different silent loci require different proteins for repression. HMR and HML require Sir1p, Sir2p, Sir3p, and Sir4p for efficient repression, whereas silencing at telomeres does not require Sir1p, and silencing at the rDNA requires only Sir2p. Because Sas10p overexpression derepressed all four loci, the target of Sas10p was neither Sir1p, Sir3p, nor Sir4p. In addition, simple genetic experiments demonstrating that Sas10p-mediated disruption of silencing at HMR occurred in strains in which neither Sir protein, ORC, nor Rap1p were required for silencing severely restricted the number of mechanisms by which Sas10p overproduction could lead to loss of silencing. These data placed Sas10p function, as measured by disruption of silencing, at the heart of silenced chromatin structure, more central even than the Sir proteins. Derepression by Sas10p could result from specific interactions between Sas10p and the histones present at the silent loci, perhaps through electrostatic interactions between oppositely charged surfaces. Biochemical analyses have suggested a role for histone deacetylation in silencing, and particularly of histones H3 and H4 (BRAUNSTEIN et al. 1993 ). If deacetylation of histones at HML and HMR is causal for silencing, then Sas10p might exert its role by altering the patterns of acetylation or deacetylation. Alternatively, Sas10p overproduction may affect higher-order chromosome functions that impinge on silencing and other essential aspects of chromatin structure. Whatever its mechanism, it appears that Sas10p overproduction causes derepression by influencing either the probability that the silenced state is established in all cell divisions or the probability that it is established in cells that have somehow lost silencing, based on the clonal inheritance of red and white sectors induced by Sas10p overexpression in cells with ADE2 at HMR.

	FOOTNOTES

¹ Present address: Laboratory of Molecular Embryology, National Institutes of Child Health and Human Development, 18 Library Dr., Bethesda, MD 20892.

	ACKNOWLEDGMENTS

We thank A. STRAIGHT and A. BRETSCHER for providing the yeast GAL1p-cDNA library, and to D. SHORE, J. BOEKE, C. FOX, P. HERMAN and J. WHISTLER for generously providing various strains and plasmids. We would also like to thank A. STRUNNIKOV for help with the microscopy, D. ROLAND WALKER for help with sequence alignments and the members of the Rine laboratory for experimental ideas and suggestions and to N. DHILLON, A. DILLIN, A. EHRENHOFER-MURRAY, P. WADE and J. STROBOULIS for comments on the manuscript. This research was supported by National Institutes of Health (NIH) grant to J. RINE and NIH grant ZO1HD01904-01 to R.T. KAMAKAKA.

Manuscript received November 17, 1997; Accepted for publication March 17, 1998.

	LITERATURE CITED

TOP ABSTRACT MATERIALS AND METHODS RESULTS DISCUSSION LITERATURE CITED

AXELROD, A. and J. RINE, 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[Abstract/Free Full Text].

BRAUNSTEIN, M., A. B. ROSE, S. G. HOLMES, C. D. ALLIS, and J. R. BROACH, 1993  Transcriptional silencing in yeast is associated with reduced nucleosome acetylation. Genes Dev. 7:592-604[Abstract/Free Full Text].

BRYK, M., M. BANERJEE, M. MURPHY, K. E. KNUDSEN, and D. J. GARFINKEL et al., 1997  Transcriptional silencing of Ty1 elements in the RDN1 locus of yeast. Genes Dev. 11:255-269[Abstract/Free Full Text].

FLOOR, E., 1970  Interaction of morphogenetic genes of bacteriophage T4. J. Mol. Biol. 47:293-306[Medline].

FOSS, M., F. J. MCNALLY, P. LAURENSON, and J. RINE, 1993  Origin recognition complex (ORC) in transcriptional silencing and DNA replication in S. cerevisiae.. Science 262:1838-1844[Abstract/Free Full Text].

FOX, C. A., A. E. EHRENHOFER-MURRAY, S. LOO, and J. RINE, 1997  ORC, SIR1 and the S-phase requirement for silencing. Science 276:1547-1551[Abstract/Free Full Text].

GIETZ, R. D. and A. SUGINO, 1988  New yeast-Escherichia coli shuttle vectors constructed with in vitro mutagenized yeast genes lacking six-base pair restriction sites. Gene 74:527-534[Medline].

GOTTSCHLING, D. E., O. M. APARICIO, B. L. BILLINGTON, and V. A. ZAKIAN, 1990  Position effect at S. cerevisiae telomeres: reversible repression of Pol II transcription. Cell 63:751-762[Medline].

HECHT, A., T. LAROCHE, B. S. STRAHL, S. M. GASSER, and M. GRUNSTEIN, 1995  Histone H3 and H4 N-termini interact with SIR3 and SIR4 proteins: a molecular model for the formation of heterochromatin in yeast. Cell 80:583-592[Medline].

HENIKOFF, S., 1996 Position-Effect Variegation in Drosophila: Recent Progress. Cold Spring Harbor Laboratory Press, Cold Spring Harbor, NY.

HOFFMAN, C. S. and F. WINSTON, 1987  A ten minute DNA preparation from yeast efficiently releases autonomous plasmids for transformation of Escherichia coli. Gene 57:267-272[Medline].

ITO, H., Y. FUKUDA, K. MURATA, and A. KIMURA, 1983  Transformation of intact yeast cells treated with alkali cations. J. Bacteriol. 153:163-168[Abstract/Free Full Text].

KAMAKAKA, R. T., 1997  Silencers and locus control regions: opposite sides of the same coin. Trends Biochem. Sci. 22:124-128[Medline].

LAURENSON, P. and J. RINE, 1991  SUM1-1: a suppressor of silencing defects in Saccharomyces cerevisiae.. Genetics 129:685-696[Abstract].

LE, S., C. DAVIS, J. B. KONOPKA, and R. STERNGLANZ, 1997  Two new S-phase-specific genes from Saccharomyces cerevisiae.. Yeast 13:1029-1042[Medline].

LIU, H., J. KRIZEK, and A. BRETSCHER, 1992  Construction of a GAL1-regulated yeast cDNA expression library and its application to the identification of genes whose overexpression causes lethality in yeast. Genetics 132:665-673[Abstract].

LOO, S. and J. RINE, 1994  Silencers and domains of generalized repression. Science 264:1768-1771[Abstract/Free Full Text].

LOO, S. and J. RINE, 1995  Silencing and heritable domains of gene expression. Annu. Rev. Cell Dev. Biol. 11:519-548[Medline].

LOO, S., P. LAURENSON, M. FOSS, A. DILLIN, and J. RINE, 1995  Roles of ABF1, NPL3, and YCL54 in silencing in Saccharomyces cerevisiae.. Genetics 141:889-902[Abstract].

MANN, C., J.-Y. MICOUIN, N. CHIANNILKULCHAI, I. TREICH, and J.-M. BUHLER et al., 1992  RPC53 encodes a subunit of Saccharomyces cerevisiae RNA polymerase C(III) whose inactivation leads to a predominantly G1 arrest. Mol. Cell. Biol. 12:4314-4326[Abstract/Free Full Text].

MARSHALL, M., D. MAHONEY, A. ROSE, J. B. HICKS, and J. R. BROACH, 1987  Functional domains of SIR4, a gene required for position effect regulation in Saccharomyces cerevisiae.. Mol. Cell. Biol. 7:4441-4452[Abstract/Free Full Text].

MCNALLY, F. J. and J. RINE, 1991  A synthetic silencer mediates SIR-dependent functions in Saccharomyces cerevisiae.. Mol. Cell. Biol. 11:5648-5659[Abstract/Free Full Text].

MEEKS-WAGNER, D. and L. H. HARTWELL, 1986  Normal stoichiometry of histone dimer sets is necessary for high fidelity of mitotic chromosome transmission. Cell 44:43-52[Medline].

MOAZED, D. and A. D. JOHNSON, 1996  A deubiquitinating enzyme interacts with SIR4 and regulates silencing in S. cerevisiae.. Cell 86:667-677[Medline].

PALLADINO, F., T. LAROCHE, E. GILSON, A. AXELROD, and L. PILLUS et al., 1993  SIR3 and SIR4 proteins are required for the positioning and integrity of yeast telomeres. Cell 75:543-555[Medline].

PILLUS, L., and M. GRUNSTEIN, 1995 Chromatin Structure and Epigenetic Regulation in Yeast. IRL Press, Oxford.

RENAULD, H., O. M. APARICIO, P. D. ZIERATH, B. L. BILLINGTON, and S. K. CHHABLANI et al., 1993  Silent domains are assembled continuously from the telomere and are defined by promoter distance and strength, and by SIR3 dosage. Genes Dev. 7:1133-1145[Abstract/Free Full Text].

REUTER, G. and P. SPIERER, 1992  Position effect variegation and chromatin proteins. BioEssays 14:605-612[Medline].

RINE, J. and I. HERSKOWITZ, 1987  Four genes responsible for a position effect on expression from HML and HMR in Saccharomyces cerevisiae.. Genetics 116:9-22[Abstract/Free Full Text].

SHORE, D. and K. NASMYTH, 1987  Purification and cloning of a DNA binding protein from yeast that binds to both silencer and activator elements. Cell 51:721-732[Medline].

SIKORSKI, R. S. and P. HIETER, 1989  A system of shuttle vectors and yeast host strains designed for efficient manipulation of DNA in Saccharomyces cerevisiae.. Genetics 122:19-27[Abstract/Free Full Text].

SMITH, J. S. and J. D. BOEKE, 1997  An unusual form of transcriptional silencing in yeast ribosomal DNA. Genes Dev. 11:241-254[Abstract/Free Full Text].

STRAHL-BOLSINGER, S., A. HECHT, K. LUO, and M. GRUNSTEIN, 1997  SIR2 and SIR4 interactions differ in core and extended telomeric heterochromatin in yeast. Genes Dev. 11:83-93[Abstract/Free Full Text].

STRAKA, C. and W. HORZ, 1991  A functional role for nucleosomes in the repression of a yeast promoter. EMBO J. 10:361-368[Medline].

SUSSEL, L., D. VANNIER, and D. SHORE, 1993  Epigenetic switching of transcriptional states: cis- and trans-acting factors affecting establishment of silencing at the HMR locus in Saccharomyces cerevisiae.. Mol. Cell. Biol. 13:3919-3928[Abstract/Free Full Text].

WEILER, K. S. and B. T. WAKIMOTO, 1995  Heterochromatin and gene expression in Drosophila. Annu. Rev. Genetics 29:577-605[Medline].

WINSTON, F. and M. CARLSON, 1992  Yeast SNF/SWI transcriptional activators and the SPT/SIN chromatin connection. Trends Genet. 8:387-391[Medline].

This article has been cited by other articles:

				C.-R. Sihn, Y.-S. Lee, J.-S. Jeong, K. Park, and S. H. Kim CANu1, a novel nucleolar protein, accumulated on centromere in response to DNA damage Genes Cells, August 1, 2008; 13(8): 787 - 796. [Abstract] [Full Text] [PDF]

				S. Tu, E. M. M. Bulloch, L. Yang, C. Ren, W.-C. Huang, P.-H. Hsu, C.-H. Chen, C.-L. Liao, H.-M. Yu, W.-S. Lo, et al. Identification of Histone Demethylases in Saccharomyces cerevisiae J. Biol. Chem., May 11, 2007; 282(19): 14262 - 14271. [Abstract] [Full Text] [PDF]

				J.-H. Lim, S.-J. Cho, S.-K. Park, J. Kim, D. Cho, W. J. Lee, and C.-J. Kang Stage-Specific Expression of Two Neighboring Crlz1 and IgJ Genes during B Cell Development Is Regulated by Their Chromatin Accessibility and Histone Acetylation J. Immunol., October 15, 2006; 177(8): 5420 - 5429. [Abstract] [Full Text] [PDF]

				M. Kimura and A. Ishihama Tfg3, a subunit of the general transcription factor TFIIF in Schizosaccharomyces pombe, functions under stress conditions Nucleic Acids Res., December 22, 2004; 32(22): 6706 - 6715. [Abstract] [Full Text] [PDF]

				M. Oki, L. Valenzuela, T. Chiba, T. Ito, and R. T. Kamakaka Barrier Proteins Remodel and Modify Chromatin To Restrict Silenced Domains Mol. Cell. Biol., March 1, 2004; 24(5): 1956 - 1967. [Abstract] [Full Text] [PDF]

				N. T. PETERS, J. A. ROHRBACH, B. A. ZALEWSKI, C. M. BYRKETT, and J. C. VAUGHN RNA editing and regulation of Drosophila 4f-rnp expression by sas-10 antisense readthrough mRNA transcripts RNA, June 1, 2003; 9(6): 698 - 710. [Abstract] [Full Text] [PDF]

				J. S. Smith, E. Caputo, and J. D. Boeke A Genetic Screen for Ribosomal DNA Silencing Defects Identifies Multiple DNA Replication and Chromatin-Modulating Factors Mol. Cell. Biol., April 1, 1999; 19(4): 3184 - 3197. [Abstract] [Full Text] [PDF]

				D. Donze, C. R. Adams, J. Rine, and R. T. Kamakaka The boundaries of the silenced HMR domain in Saccharomyces cerevisiae Genes & Dev., March 15, 1999; 13(6): 698 - 708. [Abstract] [Full Text]

				K. A. Gardner, J. Rine, and C. A. Fox A Region of the Sir1 Protein Dedicated to Recognition of a Silencer and Required for Interaction with the Orc1 Protein in Saccharomyces cerevisiae Genetics, January 1, 1999; 151(1): 31 - 44. [Abstract] [Full Text]