The HML mating-type cassette of Saccharomyces cerevisiae is regulated by two separate but functionally equivalent silencers

Evolution and Functional Trajectory of Sir1 in Gene Silencing

We used the budding yeasts Saccharomyces cerevisiae and Torulaspora delbrueckii to examine the evolution of Sir-based silencing, focusing on Sir1, silencers, the molecular topography of silenced chromatin, and the roles of SIR and RNA interference (RNAi) genes in T. delbrueckii. Chromatin immunoprecipitation followed by deep sequencing (ChIP-Seq) analysis of Sir proteins in T. delbrueckii revealed a different topography of chromatin at the HML and HMR loci than was observed in S. cerevisiae. S. cerevisiae Sir1, enriched at the silencers of HMLα and HMR A: , was absent from telomeres and did not repress subtelomeric genes. In contrast to S. cerevisiae SIR1's partially dispensable role in silencing, the T. delbrueckii SIR1 paralog KOS3 was essential for silencing. KOS3 was also found at telomeres with T. delbrueckii Sir2 (Td-Sir2) and Td-Sir4 and repressed subtelomeric genes. Silencer mapping in T. delbrueckii revealed single silencers at HML and HMR, bound by Td-Kos3, Td-Sir2, and Td-Sir4. The KOS3 gene mapped near HMR, and its expression was regulated by Sir-based silencing, providing feedback regulation of a silencing protein by silencing. In contrast to the prominent role of Sir proteins in silencing, T. delbrueckii RNAi genes AGO1 and DCR1 did not function in heterochromatin formation. These results highlighted the shifting role of silencing genes and the diverse chromatin architectures underlying heterochromatin.

Cell cycle regulation of silent chromatin formation

Identical genes in two different cells can stably exist in alternate transcriptional states despite the dynamic changes that will occur to chromatin at that locus throughout the cell cycle. In mammals, this is achieved through epigenetic processes that regulate key developmental transitions and ensure stable patterns of gene expression during growth and differentiation. The budding yeast Saccharomyces cerevisiae utilizes silencing to control the expression state of genes encoding key regulatory factors for determining cell-type, ribosomal RNA levels and proper telomere function. Here, we review the composition of silent chromatin in S. cerevisiae, how silent chromatin is influenced by chromatin assembly and histone modifications and highlight several observations that have contributed to our understanding of the interplay between silent chromatin formation and stability and the cell cycle. This article is part of a Special Issue entitled: Histone chaperones and Chromatin assembly.

SIR proteins and the assembly of silent chromatin in budding yeast

Saccharomyces cerevisiae provides a well-studied model system for heritable silent chromatin in which a histone-binding protein complex [the SIR (silent information regulator) complex] represses gene transcription in a sequence-independent manner by spreading along nucleosomes, much like heterochromatin in higher eukaryotes. Recent advances in the biochemistry and structural biology of the SIR-chromatin system bring us much closer to a molecular understanding of yeast silent chromatin. Simultaneously, genome-wide approaches have shed light on the biological importance of this form of epigenetic repression. Here, we integrate genetic, structural, and cell biological data into an updated overview of yeast silent chromatin assembly.

A single heterochromatin boundary element imposes position-independent antisilencing activity in Saccharomyces cerevisiae minichromosomes

Chromatin boundary elements serve as cis-acting regulatory DNA signals required to protect genes from the effects of the neighboring heterochromatin. In the yeast genome, boundary elements act by establishing barriers for heterochromatin spreading and are sufficient to protect a reporter gene from transcriptional silencing when inserted between the silencer and the reporter gene. Here we dissected functional topography of silencers and boundary elements within circular minichromosomes in Saccharomyces cerevisiae. We found that both HML-E and HML-I silencers can efficiently repress the URA3 reporter on a multi-copy yeast minichromosome and we further showed that two distinct heterochromatin boundary elements STAR and TEF2-UASrpg are able to limit the heterochromatin spreading in circular minichromosomes. In surprising contrast to what had been observed in the yeast genome, we found that in minichromosomes the heterochromatin boundary elements inhibit silencing of the reporter gene even when just one boundary element is positioned at the distal end of the URA3 reporter or upstream of the silencer elements. Thus the STAR and TEF2-UASrpg boundary elements inhibit chromatin silencing through an antisilencing activity independently of their position or orientation in S. cerevisiae minichromosomes rather than by creating a position-specific barrier as seen in the genome. We propose that the circular DNA topology facilitates interactions between the boundary and silencing elements in the minichromosomes.

Design of a minimal silencer for the silent mating-type locus HML of Saccharomyces cerevisiae

The silent mating-type loci HML and HMR of Saccharomyces cerevisiae contain mating-type information that is permanently repressed. This silencing is mediated by flanking sequence elements, the E- and I-silencers. They contain combinations of binding sites for the proteins Rap1, Abf1 and Sum1 as well as for the origin recognition complex (ORC). Together, they recruit other silencing factors, foremost the repressive Sir2/Sir3/Sir4 complex, to establish heterochromatin-like structures at the HM loci. However, the HM silencers exhibit considerable functional redundancy, which has hampered the identification of further silencing factors. In this study, we constructed a synthetic HML-E silencer (HML-SS ΔI) that lacked this redundancy. It consisted solely of Rap1 and ORC-binding sites and the D2 element, a Sum1-binding site. All three elements were crucial for minimal HML silencing, and mutations in these elements led to a loss of Sir3 recruitment. The silencer was sensitive to a mutation in RAP1, rap1-12, but less sensitive to orc mutations or sum1Δ. Moreover, deletions of SIR1 and DOT1 lead to complete derepression of the HML-SS ΔI silencer. This fully functional, minimal HML-E silencer will therefore be useful to identify novel factors involved in HML silencing.

Isolation and characterization of conditional alleles of the yeast SIR2 gene

Sir2 is a protein deacetylase that mediates transcriptional silencing at the HM loci, telomeres, and rDNA repeats in yeast. To identify functionally significant regions of the Sir2 protein, we have characterized two types of mutations. First, we used random mutagenesis to create temperature-sensitive alleles of the SIR2 gene. Mutations conferring conditional silencing can be isolated throughout the SIR2 gene, causing both enzymatic and protein interaction defects. We used external deletions to identify regions essential for silencing in the non-conserved regions of Sir2. Deletions of the Sir2 N-terminal 89 amino acid residues caused a subtle increase in silencing, while deletions encompassing residues 110-146 caused loss of Sir2 interactions with both Sir4 and Net1. This loss of protein interaction correlates with a loss of Sir2-mediated silencing, and is consistent with a model in which Net1 and Sir4 compete for interaction with Sir2. These results indicate that recognition of the binding partners of Sir2 is a key function of non-conserved sequences.

The nuclear GTPase Gsp1p can affect proper telomeric function through the Sir4 protein inSaccharomyces cerevisiae

The small Ras-like GTPase Ran/Gsp1p is a highly conserved nuclear protein required for the nucleocytoplasmic trafficking of macromolecules. Recent findings suggest that the Ran/Gsp1p pathway may have additional roles in several aspects of nuclear structure and function, including spindle assembly, nuclear envelope formation, nuclear pore complex assembly and RNA processing. Here, we provide evidence that Gsp1p can regulate telomeric function in Saccharomyces cerevisiae. We show that overexpression of PRP20, encoding the Gsp1p GDP/GTP nuclear exchange factor, specifically weakens telomeric silencing without detectably affecting nucleocytoplasmic transport. In addition to this silencing defect, we show that Rap1p and Sir3p delocalize from their normal telomeric foci. Interestingly, Gsp1p was found to interact genetically and physically with the telomeric component Sir4p. Taken together, these results suggest that the GSP1 pathway could regulate proper telomeric function in yeast through Sir4p.

Structural analyses of Sum1-1p-dependent transcriptionally silent chromatin in Saccharomyces cerevisiae

In Saccharomyces cerevisiae, transcriptional silencing of the cryptic mating loci HML and HMR is established by the combined actions of cis-acting silencers and trans-acting proteins, including Sir2p, Sir3p and Sir4p. The Sir proteins serve as an integral part of a special silent chromatin at the HM loci. Deletion of any of the SIR2-SIR4 genes leads to a complete loss of silencing. However, the SUM1-1 mutation can restore silencing at the HM loci. Recently, it has been shown that Sum1-1p is directed to the silencers and internal regions of the HM loci, and interacts with the Hst1p histone deacetylase that is a paralog of the Sir2p histone deacetylase. Like Sir-dependent silent chromatin, Sum1-1p-dependent chromatin is hypoacetylated. These suggest that Sum1-1p and Hst1p play roles similar to those of the Sir proteins in promoting transcriptional silencing. Here, we examine whether Sum1-1p-dependent chromatin is similar to Sir-dependent silent chromatin, which is characterized by densely and precisely positioned nucleosomes. We demonstrate that Sum1-1p-dependent primary chromatin structure at HMR largely resembles, but is not identical with, Sir-dependent silent chromatin, whereas Sum1-1p-dependent HML chromatin largely resembles, but is not identical with, derepressed chromatin found in a sir- background. This correlates with the previous finding that SUM1-1 restores silencing more efficiently at HMR than at HML. We show also that DNA in Sum1-1p-dependent silent chromatin assumes a distinct topology. Moreover, we present evidence indicating that Sum1-1p can increase the stability of Sir-dependent silent chromatin, thereby providing an explanation for the finding that SUM1-1 enhances HML/HMR silencing in a SIR+ background.

Regulation of transcriptional silencing in yeast by growth temperature

Increasing evidence indicates that transcriptionally silent chromatin structure is dynamic and may change its conformation in response to external or internal stimuli. We show that growth temperature affects all three forms of transcriptional silencing in Saccharomyces cerevisiae. In general, increasing the temperature within the range of 23-37 degrees C strengthens HM and telomeric silencing but reduces rDNA silencing. High temperature (37 degrees C) can suppress the silencing defects of histone H4 mutants. We demonstrate that DNA at the silent HML locus becomes more and more negatively supercoiled as temperature increases in a Sir-dependent manner, which is indicative of enhanced silent chromatin. This enhancement of silent chromatin is not dependent on silencers and therefore does not require de novo assembly of silent chromatin. We also present evidence suggesting that MAP kinase-mediated Sir3p hyperphosphorylation, which plays a role in regulating silencing in response to certain stress conditions, is not involved in high temperature-induced strengthening of silencing. In addition, Pnc1p, a positive regulator of Sir2p activity, plays no role in thermal regulation of silencing. Therefore, growth temperature regulates transcriptional silencing by a novel mechanism.

The establishment, inheritance, and function of silenced chromatin in Saccharomyces cerevisiae

Genomes are organized into active regions known as euchromatin and inactive regions known as heterochromatin, or silenced chromatin. This review describes contemporary knowledge and models for how silenced chromatin in Saccharomyces cerevisiae forms, functions, and is inherited. In S. cerevisiae, Sir proteins are the key structural components of silenced chromatin. Sir proteins interact first with silencers, which dictate which regions are silenced, and then with histone tails in nucleosomes as the Sir proteins spread from silencers along chromosomes. Importantly, the spreading of silenced chromatin requires the histone deacetylase activity of Sir2p. This requirement leads to a general model for the spreading and inheritance of silenced chromatin or other special chromatin states. Such chromatin domains are marked by modifications of the nucleosomes or DNA, and this mark is able to recruit an enzyme that makes further marks. Thus, among different organisms, multiple forms of repressive chromatin can be formed using similar strategies but completely different proteins. We also describe emerging evidence that mutations that cause global changes in the modification of histones can alter the balance between euchromatin and silenced chromatin within a cell.

A Targeted Histone Acetyltransferase Can Create a Sizable Region of Hyperacetylated Chromatin and Counteract the Propagation of Transcriptionally Silent Chromatin

Transcriptionally silent chromatin is associated with reduced histone acetylation and its propagation depends on histone hypoacetylation promoted by histone deacetylases. We show that tethered histone acetyltransferase (HAT) Esa1p or Gcn5p creates a segment of hyperacetylated chromatin that is at least 2.6 kb in size and counteracts transcriptional silencing that emanates from a silencer in yeast. Esa1p and Gcn5p counteract URA3 silencing even when they are targeted 1.7 kb downstream of the promoter and >2.0 kb from the silencer. The anti-silencing effect of a targeted HAT is strengthened by increasing the number of targeting sites, but impaired by events that enhance silencing. A tethered HAT can also counteract telomeric silencing. The anti-silencing effect of Gcn5p is abolished by a mutation that eliminated its HAT activity or by deleting the ADA2 gene encoding a structural component of Gcn5p-containing HAT complexes. These results demonstrate that a tethered HAT complex can create a sizable region of histone hyperacetylation and serve as a barrier to encroaching repressive chromatin.

Structural and dynamic functions establish chromatin domains

Drosophila and mammalian proteins protect genes from heterochromatic repression in Saccharomyces cerevisiae by two different mechanisms. Factors termed genuine boundary activities (BAs) establish a structural, unidirectional bulwark against heterochromatin. In contrast, factors termed desilencing activities (DAs) act by the formation of a bidirectional, euchromatic island that blocks spreading of heterochromatin. The Drosophila boundary protein BEAF and, unexpectedly, the mammalian factor Sp1 exhibited a robust BA in yeast. In contrast, mammalian CTCF, Drosophila GAGA factor, yeast Gcn5p, and many mammalian transcription factors, although inactive as upregulators of nonsilenced genes, work as DAs. DAs but not BAs protect telomere-linked genes from silencing, presumably due to looping of telomeres and ensuing multidirectional silencing. The data demonstrate that "genetic autonomy" of chromatin domains is established by both passive and active mechanisms.

Roles for internal and flanking sequences in regulating the activity of mating-type-silencer-associated replication origins in Saccharomyces cerevisiae

ARS301 and ARS302 are inactive replication origins located at the left end of budding yeast (Saccharomyces cerevisiae) chromosome III, where they are associated with the HML-E and -I silencers of the HML mating type cassette. Although they function as replication origins in plasmids, they do not serve as origins in their normal chromosomal locations, because they are programmed to fire so late in S phase that they are passively replicated by the replication fork from neighboring early-firing ARS305 before they have a chance to fire on their own. We asked whether the nucleotide sequences required for plasmid origin function of these silencer-associated chromosomally inactive origins differ from the sequences needed for plasmid origin function by nonsilencer-associated chromosomally active origins. We could not detect consistent differences in sequence requirements for the two types of origins. Next, we asked whether sequences within or flanking these origins are responsible for their chromosomal inactivity. Our results demonstrate that both flanking and internal sequences contribute to chromosomal inactivity, presumably by programming these origins to fire late in S phase. In ARS301, the function of the internal sequences determining chromosomal inactivity is dependent on the checkpoint proteins Mec1p and Rad53p.

Protosilencers in Saccharomyces cerevisiae subtelomeric regions

Saccharomyces cerevisiae subtelomeric repeats contain silencing elements such as the core X sequence, which is present at all chromosome ends. When transplaced at HML, core X can enhance the action of a distant silencer without acting as a silencer on its own, thus fulfilling the functional definition of a protosilencer. Here we show that an ACS motif and an Abf1p-binding site participate in the silencing capacity of core X and that their effects are additive. In addition, in a variety of settings, core X was found to bring about substantial gene repression only when a low level of silencing was already detectable in its absence. Adjoining an X-STAR sequence, which naturally abuts core X in subtelomeric regions, did not improve the silencing capacity of core X. We propose that protosilencers play a major role in a variety of silencing phenomena, as is the case for core X, which acts as a silencing relay, prolonging silencing propagation away from telomeres.

Acetylation and chromosomal functions

Since the initial discovery of histone acetyltransferases, numerous reports have established that histone acetyltransferases and histone deacetylases regulate transcription by acetylating and deacetylating histones, respectively. Recent studies have focused on the effects of histone acetylation on gene expression regulation during development and the roles of histone hypoacetylation in the maintenance of centromeric structure, X-inactivation and genomic imprinting. Recent findings have also shown that the functions of non-histone proteins can also be regulated by acetylation. Together, these data highlight the importance of acetylation of histones and non-histone proteins in a variety of chromosomal functions.

Mating-type gene switching in Saccharomyces cerevisiae

Saccharomyces cerevisiae can change its mating type as often as every generation by a highly choreographed, site-specific recombination event that replaces one MAT allele with different DNA sequences encoding the opposite allele. The study of this process has yielded important insights into the control of cell lineage, the silencing of gene expression, and the formation of heterochromatin, as well as the molecular events of double-strand break-induced recombination. In addition, MAT switching provides a remarkable example of a small locus control region--the Recombination Enhancer--that controls recombination along an entire chromosome arm.

High frequency cDNA recombination of the saccharomyces retrotransposon Ty5: The LTR mediates formation of tandem elements

Retroelement cDNA can integrate into the genome using the element-encoded integrase or it can recombine with preexisting elements using the recombination system of the host. Recombination is a particularly important pathway for the yeast retrotransposon Ty5 and accounts for approximately 30% of the putative transposition events when a homologous substrate is carried on a plasmid and approximately 7% when the substrate is located at the chromosomal URA3 locus. Characterization of recombinants revealed that they are either simple replacements of the marker gene tandem elements. Using an assay system in which the donor element and recombination substrates are separated, we found that the long terminal repeats (LTRs) are critical for tandem element formation. LTR-containing substrates generate tandem elements at frequencies more than 10-fold higher than similarly sized internal Ty5 sequences. Internal sequences, however, facilitate tandem element formation when associated with an LTR, and there is a linear relationship between frequencies of tandem element formation and the length of LTR-containing substrates. We propose that recombination is initiated between the LTRs of the cDNA and substrate and that internal sequences promote tandem element formation by facilitating sequence alignment. Because of its location in subtelomeric regions, recombinational amplification of Ty5 may contribute to the organizations of chromosome ends.

Two new S-phase-specific genes from Saccharomyces cerevisiae

Two new yeast genes, ASF1 (Anti-Silencing Function) and ASF2, as well as a C-terminal fragment of SIR3, were identified as genes that derepressed the silent mating type loci when overexpressed. ASF2 overexpression caused a greater derepression than did ASF1. ASF1 overexpression also weakened repression of genes near telomeres, but, interestingly, ASF2 had no effect on telomeric silencing. Sequences of these two genes revealed open reading frames of 279 and 525 amino acids for ASF1 and ASF2, respectively. The ASF1 protein was evolutionarily conserved, MCB motifs, sequences commonly present upstream of genes transcribed specifically in S phase, were found in front of both genes, and, indeed, both genes were transcribed specifically in the S phase of the cell cycle. While an asf2 mutant was viable and had no obvious phenotypes, an asf1 mutant grew poorly. Neither mutant exhibited derepression of the silent mating type loci. The asf1 mutant was sensitive to methyl methane sulfonate, slightly UV-sensitive and somewhat deficient in minichromosome maintenance. It also lowered the restrictive temperature of a cdc13ts mutant. These phenotypes suggested a role for ASF1 in DNA repair and chromosome maintenance.

Redistribution of silencing proteins from telomeres to the nucleolus is associated with extension of life span in S. cerevisiae

A prior genetic study indicated that activity of Sir silencing proteins at a hypothetical AGE locus is essential for long life span. In this model, the SIR4-42 mutation would direct the Sir protein complex to the AGE locus, giving rise to a long life span. We show by indirect immunofluorescence that Sir3p and Sir4p are redirected to the nucleolus in the SIR4-42 mutant. Furthermore, this relocalization is dependent on both UTH4 a novel yeast gene that extends life span, and its homologue YGL023. Strikingly, the Sir complex is relocalized from telomeres to the nucleolus in old wild-type cells. We propose that the rDNA is the AGE locus and that nucleolar function is compromised in old yeast cells in a way that may be mitigated by targeting of Sir proteins to the nucleolus.

Evidence that the transcriptional regulators SIN3 and RPD3, and a novel gene (SDS3) with similar functions, are involved in transcriptional silencing in S. cerevisiae

In a screen for extragenic suppressors of a silencing defective rap 1s hmr delta A strain, recessive mutations in 21 different genes were found that restored repression to HMR. We describe the characterization of three of these SDS (suppressors of defective silencing) genes. SDS16 and SDS6 are known transcriptional modifiers, SIN3(RPD1/UME4/SDI1/GAM2) and RPD3(SDI2), respectively, while the third is a novel gene, SDS3. SDS3 shares the meiotic functions of SIN3 and RPD3 in that it represses IME2 in haploid cells and is necessary for sporulation in diploid cells. However, sds3 mutations differ from sin3 and rpd3 mutations in that they do not derepress TRK2. These sds mutations suppress a variety of cis- and trans-defects, which impair the establishment of silencing at HMR. Any one of the sds mutations slightly increases telomere position effect while a striking synergistic increase in repression is observed in a rap 1s background. Epistasis studies suggest that SDS3 works in a different pathway from RPD3 and SIN3 to affect silencing at HMR. Together these results show that defects in certain general transcriptional modifiers can have a pronounced influence on position-effect gene silencing in yeast. Mechanisms for this increase in position effect are discussed.

Genetic analysis of Rap1p/Sir3p interactions in telomeric and HML silencing in Saccharomyces cerevisiae

We have identified three SIR3 suppressors of the telomeric silencing defects conferred by missense mutations within the Rap1p C-terminal tail domain (aa 800-827). Each SIR3 suppressor was also capable of suppressing a rap1 allele (rap1-21), which deletes the 28 aa C-terminal tail domain, but none of the suppressors restored telometric silencing to a 165 amino acid truncation allele. These data suggest a Rap1p site for Sir3p association between the two truncation points (aa 664-799). In SIR3 suppressor strains lacking the Rap1p C-terminal tail domain, the presence of a second intragenic mutation within the rap1s domain (aa 727-747), enhanced silencing 30-300-fold. These data suggest a competition between Sir3p and factors that interfere with silencing for association in the rap1s domain. Rap1-21 strains containing both wild-type Sir3p and either of the Sir3 suppressor proteins displayed a 400-4000-fold increase in telomeric silencing over rap1-21 strains carrying either Sir3p suppressor in the absence of wild-type Sir3p. We propose that this telomere-specific synergism is mediated in part through stabilization of Rap1p/Sir3p telometric complexes by Sir3p-Sir3p interactions.

Cooperation at a distance between silencers and proto-silencers at the yeast HML locus

Transcriptional repression at the silent yeast mating type loci is achieved through the formation of a particular nucleoprotein complex at specific cis-acting elements called silencers. This complex in turn appears to initiate the spreading of a histone binding protein complex into the surrounding chromatin, which restricts accessibility of the region to the transcription machinery. We have investigated long-range, cooperative effects between silencers by studying the repression of a reporter gene integrated at the HML locus flanked by various combinations of wild-type and mutated silencer sequences. Two silencers can cooperate over >4000 bp to repress transcription efficiently. More importantly, a single binding site for either the repressor activator protein 1 (Rap1), the autonomous replicating sequence (ARS) binding factor 1 (Abf1) or the origin recognition complex (ORC) can enhance the action of a distant silencer without acting as a silencer on its own. Functional cooperativity is demonstrated using a quantitative assay for repression, and varies with the affinity of the binding sites used. Since the repression mechanism is Sir dependent, the Rap1, ORC and/or Abf1 proteins bound to distant DNA elements may interact to create an interface of sufficiently high affinity such that Sir-containing complexes bind, nucleating the silent chromatin state.

Mutation in the silencing gene SIR4 can delay aging in S. cerevisiae

Aging in S. cerevisiae is exemplified by the fixed number of cell divisions that mother cells undergo (termed their life span). We have exploited a correlation between life span and stress resistance to identify mutations in four genes that extend life span. One of these, SIR4, encodes a component of the silencing apparatus at HM loci and telomeres. The sir4-42 mutation extends life span by more than 30% and is semidominant. Our findings suggest that sir4-42 extends life span by preventing recruitment of the SIR proteins to HM loci and telomeres, thereby increasing their concentration at other chromosomal regions. Maintaining silencing at these other regions may be critical in preventing aging. Consistent with this view, expression of only the carboxyl terminus of SIR4 interferes with silencing at HM loci and telomeres, which also extends life span. Possible links among silencing, telomere maintenance, and aging in other organisms are discussed.

DNA structure-dependent requirements for yeast RAD genes in gene conversion

In Saccharomyces cerevisiae, HO endonuclease-induced mating-type (MAT) switching is a specialized mitotic recombination event in which MAT sequences are replaced by those copied from a distant, unexpressed donor (HML or HMR). The donors have a chromatin structure inaccessible for both transcription and HO cleavage. Here we use physical monitoring of DNA to show that MAT switching is completely blocked at an early step in recombination in strains deleted for the DNA repair genes RAD51, RAD52, RAD54, RAD55 or RAD57. We find, however, that only RAD52 is required when the donor sequence is simultaneously not silenced and located on a plasmid. RAD51, RAD54, RAD55 and RAD57 are still required when the same transcribed donor is on the chromosome. We conclude that recombination in vivo occurs between DNA molecules in chromatin, whose structure significantly influences the outcome. RAD51, RAD54, RAD55 and RAD57 are all required to facilitate strand invasion into otherwise inaccessible donor sequences.

Mutational analysis defines a C-terminal tail domain of RAP1 essential for Telomeric silencing in Saccharomyces cerevisiae

Alleles specifically defective in telomeric silencing were generated by in vitro mutagenesis of the yeast RAP1 gene. The most severe phenotypes occur with three mutations in the C-terminal 28 amino acids. Two of the alleles are nonsense mutations resulting in truncated repressor/activator protein 1 (RAP1) species lacking the C-terminal 25-28 amino acids; the third allele is a missense mutation within this region. These alleles define a novel 28-amino acid region, termed the C-terminal tail domain, that is essential for telomeric and HML silencing. Using site-directed mutagenesis, an 8-amino acid region (amino acids 818-825) that is essential for telomeric silencing has been localized within this domain. Further characterization of these alleles has indicated that the C-terminal tail domain also plays a role in telomere size control. The function of the C-terminal tail in telomere maintenance is not mediated through the RAP1 interacting factor RIF1: rap1 alleles defective in both the C-terminal tail and RIF1 interaction domains have additive effects on telomere length. Overproduction of SIR3, a dose-dependent enhancer of telomeric silencing, suppresses the telomeric silencing, but not length, phenotypes of a subset of C-terminal tail alleles. In contrast, an allele that truncates the terminal 28 amino acids of RAP1 is refractory to SIR3 overproduction. These results indicate that the C-terminal tail domain is required for SIR3-dependent enhancement of telomeric silencing. These data also suggest a distinct set of C-terminal requirements for telomere size control and telomeric silencing.

Three additional linkage groups that repress transcription and meiotic recombination in the mating-type region of Schizosaccharomyces pombe

The mating-type genes of Schizosaccharomyces pombe are found at three locations in the same chromosomal region. These genes are in an active configuration at the mat1 locus and in an inactive configuration at the mat2 and mat3 loci. The mechanism that represses transcription of mat2 and mat3 also inactivates other promoters introduced nearby and is accompanied by a block to meiotic recombination in the mat2-mat3 interval, suggesting that this mechanism involves a particular chromatin structure. We present evidence that the transcription and recombination blocks require three newly defined trans-acting loci, clr2, clr3 and clr4, in addition to the previously identified clr1, rik1 and swi6 loci. We also investigated the role of mat2 cis-acting sequences in silencing. Four cis-acting elements that repress mat2 in a plasmid context were previously identified. Deletion of two of these elements proved to have little effect in a chromosomal context. However, when combined with mutations in trans-acting genes, deletion of the same two elements greatly enhanced mat2 expression. The observed cumulative effects suggest a redundancy in the silencing mechanism.

SIR3 and SIR4 proteins are required for the positioning and integrity of yeast telomeres

Heritable inactivation of genes occurs in specific chromosomal domains located at the silent mating type loci and at telomeres of S. cerevisiae. The SIR genes (for silent information regulators) are trans-acting factors required for this repression mechanism. We show here that the SIR3 and SIR4 gene products have a sub-nuclear localization similar to the telomere-associated RAP1 protein, which is found primarily in foci at the nuclear periphery of fixed yeast spheroplasts. In strains deficient for either SIR3 or SIR4, telomeres lose their perinuclear localization, as monitored by RAP1 immunofluorescence. The length of the telomeric repeat shortens in sir3 and sir4 mutant strains, and the mitotic stability of chromosome V is reduced. These data suggest that SIR3 and SIR4 are required for both the integrity and subnuclear localization of yeast telomeres, the loss of which correlates with loss of telomere-associated gene repression.

Donor locus selection during Saccharomyces cerevisiae mating type interconversion responds to distant regulatory signals

Mating type interconversion in homothallic strains of the yeast Saccharomyces cerevisiae results from directed transposition of a mating type allele from one of the two silent donor loci, HML and HMR, to the expressing locus, MAT. Cell type regulates the selection of the particular donor locus to be utilized during mating type interconversion: MATa cells preferentially select HML alpha and MAT alpha cells preferentially select HMRa. Such preferential selection indicates that the cell is able to distinguish between HML and HMR during mating type interconversion. Accordingly, we designed experiments to identify those features perceived by the cell to discriminate HML and HMR. We demonstrate that discrimination does not derive from the different structures of the HML and HMR loci, from the unique sequences flanking each donor locus nor from any of the DNA distal to the HM loci on chromosome III. Moreover, we find that the sequences flanking the MAT locus do not function in the preferential selection of one donor locus over the other. We propose that the positions of the donor loci on the left and right arms of chromosome III is the characteristic utilized by the cell to distinguish HML and HMR. This positional information is not generated by either CEN3 or the MAT locus, but probably derives from differences in the chromatin structure, chromosome folding or intranuclear localization of the two ends of chromosome III.

Disruption of a silencer domain by a retrotransposon

A galactose-inducible Ty element carrying the HIS3 gene has been used as an insertional mutagen to generate alpha-factor resistant mutants. This collection of Ty-induced mutations includes insertions into the gene for the alpha-factor receptor (STE2), several nonspecific STE genes, and mutations that lead to the expression of the normally silent HML alpha locus. The hml alpha "on" mutations fall into two classes, those that disrupt trans-acting regulators involved in silencing HML alpha and a novel class of mutations that activate HML alpha by insertion at that locus. The hml alpha::Ty "on" mutations illustrate the unusual ability of these retrotransposons to activate genes by overcoming gene silencing mechanisms. The hml alpha::Ty "on" mutations include examples of multimeric Ty arrays. Single Ty and solo delta insertion derivatives of these Ty multimers restore the ability of the silencing mechanism to repress HML alpha.

The clr1 locus regulates the expression of the cryptic mating-type loci of fission yeast

The mat2-P and mat3-M loci of fission yeast contain respectively the plus (P) and minus (M) mating-type information in a transcriptionally silent state. That information is transposed from the mat2 or mat3 donor locus via recombination into the expressed mating-type locus (mat1) resulting in switching of the cellular mating type. We have identified a gene, named clr1 (for cryptic loci regulator), whose mutations allow expression of the mat2 and mat3 loci. clr1 mutants undergo aberrant haploid meiosis, indicative of transcription of the silent genes. Production of mRNA from mat3 is detectable in clr1 mutants. Furthermore, the ura4 gene inserted near mat3, weakly expressed in wild-type cells, is derepressed in clr1 mutants. The clr1 mutations also permit meiotic recombination in the 15-kb mat2-mat3 interval, where recombination is normally inhibited. The clr1 locus is in the right arm of chromosome II. We suggest that clr1 regulates silencing of the mat2 and mat3 loci, and participates in establishing the "cold spot" for recombination by organizing the chromatin structure of the mating-type region.

An acquired state: epigenetic mechanisms in transcription

Even when we know the primary sequences and binding specificities of every transcriptional activator and repressor, our understanding of transcriptional regulation will be rudimentary. This is partly because major aspects of gene expression are governed by epigenetic mechanisms. These mechanisms may be responsible for apparently identical sequences being read by the transcriptional machinery in two different but heritable ways: active or repressed. Epigenetic transcriptional states have been encountered in a number of recent experiments.

Silencing: the establishment and inheritance of stable, repressed transcription states

Silencing refers to a particular type of transcriptional repression characterized by the formation of a genetically heritable, repressed transcriptional state. Examples of silencing include position-effect variegation, X-chromosome inactivation, and the repression of the silent mating-type gene loci in yeast. Recent discoveries suggest that silencing in yeast, like silencing in larger eukaryotes, results from a particular chromatin structure that defines a chromosomal domain. In addition, a chromosomal origin of DNA replication is required for silencing in yeast, suggesting that DNA replication plays a role in forming functional chromosomal domains.

Repression of a mating type cassette in the fission yeast by four DNA elements

The fission yeast, Schizosaccharomyces pombe, expresses one of two alternative mating types. They are specified by one of two determinants (M or P) present at the mat1 locus. In addition, silent copies of M and P are present on the same chromosome. In the present work we demonstrate that the difference between the active and the silent stage of the P determinant is controlled by four repressive elements that are located at the silent locus. There are two elements to the left and two to the right of the mating type cassette. Both elements to the left and either one of the two elements to the right are required for an effective blockage of transcription. When they are combined, the four elements define a highly efficient silencer functionally similar to the HMRE and HMLE and HMLI silencers in Saccharomyces cerevisiae. In addition, the DNA surrounding the silent P locus confers symmetric partitioning in mitosis to Schizosaccharomyces pombe ars plasmids.

Modifiers of position effect are shared between telomeric and silent mating-type loci in S. cerevisiae

Genes placed near telomeres in S. cerevisiae succumb to position-effect variegation. SIR2, SIR3, SIR4, NAT1, ARD1, and HHF2 (histone H4) were identified as modifiers of the position effect at telomeres, since transcriptional repression near telomeres was no longer observed when any of the modifier genes were mutated. These genes, in addition to SIR1, have previously been shown to repress transcription at the silent mating loci, HML and HMR. However, there were differences between transcriptional silencing at telomeres and the HM loci, as demonstrated by suppressor analysis and the lack of involvement of SIR1 in telomeric silencing. These findings provide insights into telomeric structure and function that are likely to apply to many eukaryotes. In addition, the distinctions between telomeres and the HM loci suggest a hierarchy of chromosomal silencing in S. cerevisiae.

Position effect at S. cerevisiae telomeres: reversible repression of Pol II transcription

S. cerevisiae chromosomes end with the telomeric repeat (TG1-3)n. When any of four Pol II genes was placed immediately adjacent to the telomeric repeats, expression of the gene was reversibly repressed as demonstrated by phenotype and mRNA analyses. For example, cells bearing a telomere-linked copy of ADE2 produced predominantly red colonies (a phenotype characteristic of ade2- cells) containing white sectors (characteristic of ADE2+ cells). Repression was due to proximity to the telomere itself since an 81 bp tract of (TG1-3)n positioned downstream of URA3 when URA3 was approximately 20 kb from the end of chromosome VII did not alter expression of the gene. However, this internal tract of (TG1-3)n could spontaneously become telomeric, in which case expression of the URA3 gene was repressed. These data demonstrate that yeast telomeres exert a position effect on the transcription of nearby genes, an effect that is under epigenetic control.

Promoter function and in situ protein/DNA interactions upstream of the yeast HSP90 heat shock genes

We have mapped in vivo protein/DNA interactions within the upstream regulatory regions of the two yeast HSP90 genes, and have begun mutagenizing footprinted sequences in an effort to identify the cis-acting determinants of heat shock transcription. Genomic footprinting of the HSP82 promotor using chemical and enzymatic nucleases reveals that irrespective of transcriptional state, the most proximal of three heat shock elements, HSE1, is occupied along both sugar-phosphate backbones as well as within its major groove, while the TATA box is bound along both sugar-phosphate backbones. Distorted DNA structure is associated with each constitutively bound factor: protein binding to HSE1 appears to induce a local A-form-like helical conformation, whereas occupancy of the TATA box is associated with strand-specific nuclease hypersensitivity of an adjacent polypurine tract. In situ mutagenesis experiments indicate that HSE1 is absolutely required for both basal and induced expression, and that basal transcription can be preferentially abolished by point mutations within this sequence. In contrast, point mutations within the TATA element have the reverse effect, as induced transcription is more significantly affected. Similar to HSE1 point mutants, we have found that basal transcription is preferentially repressed by an HMRE silencer element when it is transplaced approximately 1 kb upstream of the HSP82 start site. Finally, a complementary footprinting analysis of the upstream region of the constitutively expressed HSC82 gene reveals the presence of three discrete protein complexes. These map to the TATA box, the promotor-distal heat shock element, C.HSE1, and a novel sequence upstream of C. HSE1, suggesting that the 10-fold higher basal transcription of HSC82 stems, at least in part, from a non-HSE-binding factor.