Roles for the Saccharomyces cerevisiae SDS3, CBK1 and HYM1 Genes in Transcriptional Repression by SIN3

Corresponding author: David J. Stillman, Department of Oncological Sciences, University of Utah Health Sciences Center, 50 N. Medical Dr., Rm. 5C334 SOM, Salt Lake City, UT 84132., stillman{at}hci.utah.edu (E-mail)

Communicating editor: M. HAMPSEY

	ABSTRACT

TOP ABSTRACT MATERIALS AND METHODS RESULTS DISCUSSION LITERATURE CITED

The Saccharomyces cerevisiae Sin3 transcriptional repressor is part of a large multiprotein complex that includes the Rpd3 histone deacetylase. A LexA-Sin3 fusion protein represses transcription of promoters with LexA binding sites. To identify genes involved in repression by Sin3, we conducted a screen for mutations that reduce repression by LexA-Sin3. One of the mutations identified that reduces LexA-Sin3 repression is in the RPD3 gene, consistent with the known roles of Rpd3 in transcriptional repression. Mutations in CBK1 and HYM1 reduce repression by LexA-Sin3 and also cause defects in cell separation and altered colony morphology. cbk1 and hym1 mutations affect some but not all genes regulated by SIN3 and RPD3, but the effect on transcription is much weaker. Genetic analysis suggests that CBK1 and HYM1 function in the same pathway, but this genetic pathway is separable from that of SIN3 and RPD3. The remaining gene from this screen described in this report is SDS3, previously identified in a screen for mutations that increase silencing at HML, HMR, and telomere-linked genes, a phenotype also seen in sin3 and rpd3 mutants. Genetic analysis demonstrates that SDS3 functions in the same genetic pathway as SIN3 and RPD3, and coimmunoprecipitation experiments show that Sds3 is physically present in the Sin3 complex.

EUKARYOTIC DNA is associated with histone proteins and packaged into chromatin, and transcription of specific genes can be affected by the chromatin structure at the promoter (for reviews see KINGSTON et al. 1996 ; WOLFFE and PRUSS 1996 ; KADONAGA 1998 ; STRUHL 1998 ). Each of the histones contains an evolutionarily conserved aminoterminal tail that is subject to reversible post-translational modifications such as acetylation, phosphorylation, and ubiquitination. Changes in the acetylation states of lysines on the tails of histones are correlated with gene expression, with transcriptionally active genes having hyperacetylated nucleosomes and transcriptionally inactive genes hypoacetylated nucleosomes (for reviews see GRUNSTEIN 1997 ; HAMPSEY 1997 ; STRUHL 1998 ).

Sin3 and Rpd3 are components of a transcriptional repression complex in yeast (KADOSH and STRUHL 1997 ; KASTEN et al. 1997 ) that is conserved in vertebrates (PAZIN and KADONAGA 1997 ; WOLFFE 1997 ). Sin3 cannot bind to DNA itself; however, the complex is targeted to specific promoters through interactions with sequence-specific DNA-binding proteins (ALLAND et al. 1997 ; HASSIG et al. 1997 ; HEINZEL et al. 1997 ; KADOSH and STRUHL 1997 ; LAHERTY et al. 1997 ; NAGY et al. 1997 ; PAZIN and KADONAGA 1997 ; ZHANG et al. 1997 ). The fact that RPD3 encodes a histone deacetylase (TAUNTON et al. 1996 ; KADOSH and STRUHL 1998A ) provides a mechanism for transcriptional repression, with Sin3 bringing the Rpd3 histone deacetylase to specific promoters. In vivo, the presence of the Sin3/Rpd3 complex at a promoter leads to decreased acetylation of histones H3 and H4 that is highly localized over one to two nucleosomes (KADOSH and STRUHL 1998B ; RUNDLETT et al. 1998 ). The Sap30 protein is also present in the Sin3 complex, and sap30 mutations cause similar phenotypes as sin3 and rpd3 (ZHANG et al. 1998 ; SUN and HAMPSEY 1999 ).

SIN3 was first identified as a negative regulator of HO expression (NASMYTH et al. 1987 ; STERNBERG et al. 1987 ). SIN3 has since been identified in multiple screens as a negative regulator of numerous genes, including TRK2 (VIDAL et al. 1991 ), IME2 (BOWDISH and MITCHELL 1993 ), SPO13 (STRICH et al. 1989 ), and INO1 (HUDAK et al. 1994 ). Transcriptional activation of certain genes, such as STE6 (VIDAL et al. 1991 ) and middle sporulation genes (HEPWORTH et al. 1998 ), is reduced in a sin3 mutant, although the effect may be indirect (WANG et al. 1994 ). The genes regulated by SIN3 are involved in a wide variety of biological processes and share little or no direct regulatory relationship. Regulation of repression by Sin3 must be controlled, at least in part, at the level of recruitment to promoters. However, regulation may also occur by post-translational mechanisms such as protein phosphorylation.

RPD3 was first identified as a negative regulator of the low-affinity potassium transporter TRK2 (VIDAL and GABER 1991 ). Mutations in RPD3 affect expression of the same set of genes as SIN3, and genetic analysis suggests that SIN3 and RPD3 function in the same genetic pathway (STILLMAN et al. 1994 ). We have described an assay system using a LexA-Sin3 fusion protein that represses transcription of promoters with LexA binding sites (WANG and STILLMAN 1993 ). Transcriptional repression by LexA-Sin3 is reduced in an rpd3 mutant, consistent with the proposed role for histone deacetylases in repression by Sin3 (KASTEN et al. 1997 ).

We have used the LexA-Sin3 repression system in a genetic screen to identify mutations that affect repression by Sin3. The focus of the genetic selection was to identify proteins required for repression rather than for recruitment to specific promoters. In this article we describe four genes identified in the screen, RPD3, CBK1, HYM1, and SDS3, that affect repression by Sin3. Our analysis suggests that CBK1 and HYM1 function in the same genetic pathway. We also show that mutations in these two genes do not affect all SIN3-regulated genes identically, suggesting that they may modulate Sin3 repression in a promoter-specific fashion. We show that mutations in SDS3 affect the same set of genes affected by SIN3 and RPD3, consistent with the results of VANNIER et al. 1996 . Although these workers suggested that SDS3 may function in a different pathway from SIN3 and RPD3, we show that SDS3 is in the same genetic pathway as SIN3 and RPD3. Finally, immunoprecipitation experiments show that Sds3 is present in the Sin3 complex.

	MATERIALS AND METHODS

TOP ABSTRACT MATERIALS AND METHODS RESULTS DISCUSSION LITERATURE CITED

Strains:
The yeast strains used in this study, listed in Table 1, are all isogenic in the W303 background (THOMAS and ROTHSTEIN 1989 ). Standard genetic methods were used for strain construction (GUTHRIE and FINK 1991 ). Plasmids M3737, M3780, M1436, and DV66 (VANNIER et al. 1996 ) were used to disrupt the CBK1, HYM1, SDS3, and RPD3 genes, respectively. All gene disruptions were confirmed by Southern analysis. Strains with either the CYC1-LexA-lacZ or the CYC1-LexA-HIS3 reporter integrated at the URA3 or the LYS2 loci, respectively, have been described (WANG and STILLMAN 1993 ; KASTEN et al. 1997 ). A strain with the IME2-LacZ integrated reporter was constructed by cleaving plasmid M3536 with StuI and integrating at the ADE2. Plasmids pHU10 (his3::URA3), M3927 (ura3::KanMX3), and M3926 (leu2::KanMX3) were used to convert markers (CROSS 1997 ) in disrupted alleles or in integrated reporters. A W303 strain with a trk1::HIS3 disruption was generously provided by Rick Gaber, and this marker was converted to trk1::ADE2 using pRS402 (BRACHMANN et al. 1998 ) by marker replacement (VIDAL and GABER 1994 ). The W303 strain DY5699 was made by disrupting the MET15 gene with plasmid pAD4 (BRACHMANN et al. 1998 ). Strain DY5870 with a 13 x Myc epitope tag at the C terminus of SDS3 was constructed by transforming strain DY5699 with a PCR product generated with oligonucleotides F671 (5' GAATTAACAGGTCAGCCTCCGGCTCCTTTCAGACTAAGGTCTCAGCGGATCCCCGGGTTAATTAA 3') and F672 (5'ATAATACAAAGTTAAAGTGGAAGGTTTGCAGCATAAAATAAATTAGAATTCGAGCTCGTTTAAAC 3') using plasmid pFA6a:13Myc:HIS3MX6 (LONGTINE et al. 1998 ) as template. His⁺ transformants were selected, and correct integration was verified.

Media:
Cells were grown at 30° in standard media (SHERMAN 1991 ). YEPD medium was used unless strains had plasmids, in which case cells were grown in synthetic complete medium with 2% glucose supplemented with adenine, uracil, and amino acids, as appropriate, but lacking essential components to select for plasmids. Medium lacking leucine and histidine containing 20 mM 3-aminotriazole (3-AT) was used to analyze repression of the CYC1-LexA-HIS3 reporter by LexA-Sin3. Low-potassium medium is the same as synthetic complete medium (SHERMAN 1991 ), except that sodium phosphate was substituted for potassium phosphate. High- and low-phosphate media were made as described (HAN et al. 1988 ), except that plates were made with 1.2% agarose (BRL).

Plasmids:
The plasmids used in this study are listed in Table 2. Plasmid M1436 has been described (KASTEN et al. 1997 ). Plasmids M1836, M3958, M1835, and M3957 that express LexA-Sin3 or LexA were constructed in multiple steps, and details are available upon request. Plasmid M3496 was a gift of Yi Wei Jiang. Plasmid M3295, with the CYC1-LexA-HIS3 reporter, has been described (KASTEN et al. 1997 ). Plasmid M3536 (YIp, ADE2) was constructed in several steps using the YEp-IME2-LacZ plasmid previously described (WANG and STILLMAN 1993 ). Plasmids p6HA (M1710) and pJH330 (M2022) were kindly provided by Ira Herskowitz and John Lopes, respectively. Plasmid M3365 contains a 3.5-kb EcoRI-XbaI fragment with SIN4 cloned into YIplac204 (GIETZ and SUGINO 1988 ). Plasmid M3458 contains a 1.5-kb EcoRV to AflII fragment with RPD3, cloned as a XbaI-SacI fragment (polylinker sites) into YIp352 (HILL et al. 1986 ). Plasmid M3561 expressing a Sin3-HA fusion protein was kindly provided by Kevin Struhl (KADOSH and STRUHL 1997 ). The cbk1::URA3 disruptor in plasmid M3737 removes nt -90 to +1940 (where the ATG = +1) of the CBK1 gene. The hym1::TRP1 disruptor in plasmid M3780 removes nt +253 to +1100 (where the ATG = +1) of the HYM1 gene. The rpd3::LEU2 disruptor in plasmid M1436 removes nt -556 to +1291 (where the ATG = +1) of the RPD3 gene. The sds3::HIS3 disruptor in plasmid DV66 has been described (VANNIER et al. 1996 ) and was the generous gift of David Shore. The pHU10 his3::URA3 marker converter plasmid has been described (CROSS 1997 ) and was kindly provided by Fred Cross. Plasmids M3926 and M3927, with the leu2::KanMX3 and ura3::KanMX3 marker converters, will be described elsewhere. Plasmid pFA6a:13Myc:KanMX6 (M3968) containing a 13 x Myc epitope tag and a KanMX6 selectable marker was provided by Mark Longtine (LONGTINE et al. 1998 ).

Isolation of mutants:
In the first screen, strain DY4442 with plasmid M1459 was mutagenized by treatment with UV light (to 60% viability), and cells were grown in the dark at room temperature for 1 day and then at 30° for an additional 2 days on plates lacking histidine and tryptophan with 20 mM 3-AT. From 10⁷ surviving cells, 287 colonies were obtained capable of growth. Genetic backcrosses were conducted to eliminate plasmid-based mutations and to verify that a single genetic locus was responsible for the 3-AT-resistant phenotype. A total of 13 good mutants was identified, and these fell into two complementation groups, rpd3 with five alleles and sin4 with eight. Complementing clones were obtained, with either the wild-type RPD3 or SIN4 genes, and segregation analysis demonstrated allelism of the original mutations with disruption alleles. As homozygous rpd3/rpd3 strains are sporulation defective, strains for allelism testing were sporulated with a URA3-RPD3 plasmid. After tetrad dissection, cells were cured of the plasmid before the phenotype was examined. Finally, disruption of RPD3 or SIN4 had the same effect on LexA-Sin3 repression as the UV-generated alleles.

To prevent the identification of additional alleles of sin4 and rpd3 in a second screen, strain SY161 was used that contained an additional copy of the RPD3 and SIN4 genes. After UV mutagenesis to 40% viability, 3-AT-resistant colonies were obtained as described above. Following backcrossing and elimination of weak mutants, 133 mutants were subjected to complementation analysis. There are at least 11 complementation groups, and complementation analysis continues for the other mutants. We identified two alleles of sds3, seven alleles of cbk1, and two alleles of hym1. CBK1 and HYM1 were cloned from a YCp50 library using a visual screen for complementation of the defect in colony morphology, and SDS3 was cloned by complementation of its derepression of the IME2-lacZ reporter. Homozygous mutations in cbk1, hym1, and sds3 were shown to be sporulation defective in diploids, and allelism testing was conducted as described above. Disruption alleles of cbk1, hym1, and sds3 had the same phenotypes as the UV-generated alleles, and null alleles were used for all further phenotypic analysis.

Phosphatase assays:
Phosphatase overlay assays on colonies and quantitative phosphatase assays with extracts were performed as described (TOH-E et al. 1973 ). To measure PHO5 derepression grown in liquid, cells were grown overnight in high-phosphate medium, diluted and grown to mid-log, and harvested. To measure PHO5 derepression while grown on plates, cells were grown on high-phosphate plates for 3 days at 30°, and then were scraped from the plate. Extracts were prepared by glass bead lysis. One unit of acid phosphatase is defined as the amount of enzyme that catalyzes the liberation of 1 µM of p-nitrophenol per minute at 37°. Each assay represents a minimum of three independent cultures.

Other methods:
Assays for ß-galactosidase activity using protein extracts were performed as described (BREEDEN and NASMYTH 1987 ). Telomeric silencing was measured using strains with a URA3 gene integrated near the telomere of chromosome VII (GOTTSCHLING et al. 1990 ). Expression of the telomeric reporter was measured by plating serial dilutions of an overnight culture grown in rich media onto SC and SC-Ura plates. Immunoprecipitation and Western blotting were conducted as described (AUSUBEL et al. 1987 ) using monoclonal antibodies to the HA and Myc peptide epitopes.

	RESULTS

TOP ABSTRACT MATERIALS AND METHODS RESULTS DISCUSSION LITERATURE CITED

Identification of genes required for LexA-Sin3 repression:
We used a genetic selection to identify genes required for Sin3 to function as a transcriptional repressor. This selection uses a fusion of Sin3 to the DNA-binding domain of the bacterial LexA protein and a CYC1-LexA-HIS3 reporter construct. In the presence of the fusion, cells produce insufficient His3 protein to be able to grow in the presence of 20 mM 3-AT, a competitive inhibitor of the HIS3 gene product imidazole glycerol phosphate dehydratase. We selected for UV-light-generated mutations exhibiting growth on media containing 20 mM 3-AT but lacking histidine and leucine (to select for the LexA-Sin3 plasmid) that relieve repression by LexA-Sin3, and thus our efforts are focused on genes required for repression rather than on genes required to target the Sin3 complex to specific promoters. It is important to note that the endogenous SIN3 gene product is not required for LexA-Sin3 repression.

Genetic tests demonstrated that the mutations are recessive and that a mutation in the LexA-Sin3 plasmid was not responsible for the histidine prototrophy. A sin3 mutation derepresses the PHO5 gene under repressing conditions, as a colony-staining overlay assay shows an increase in acid phosphatase activity (VIDAL et al. 1991 ). This assay showed that all of the new mutants also had an increase in acid phosphatase activity. The mutants were backcrossed to a wild-type strain with the CYC1-LexA-HIS3 reporter, and haploid segregants were tested for growth on -Leu, -His + 3-AT plates as well as for acid phosphatase activity. Segregation analysis demonstrated that a single genetic locus was responsible for both the loss of LexA-Sin3 repression and the PHO5 derepression, except for strains with two contributing mutations that were excluded from further analysis. Complementation analysis identified at least 11 complementation groups. In this article we describe four of the mutations that we have cloned and genetically characterized. These genes are RPD3, CBK1, HYM1, and SDS3.

To demonstrate allelism of the complementing DNA with the original mutation, strains with the appropriate gene disruptions were crossed to strains with the original allele generated by UV mutagenesis. In each case, the diploids were unable to sporulate, as were diploids homozygous for mutations in SIN3, so the diploid was transformed with a URA3-based plasmid with the wild-type gene. After sporulation and tetrad dissection, cells were cured of the URA3 plasmid by growth on 5-fluoroorotic acid (5-FOA) medium before phenotypic analysis. Segregation analysis demonstrated that each gene disruption was genetically linked to the appropriate mutation and that the disruptions confer the same phenotypes as the original mutations.

Haploid strains with disruptions for each of these four genes were created, demonstrating that none of these genes is essential for viability. Figure 1 shows how these mutations reduce repression by LexA-Sin3. In the wild-type strain with the CYC1-LexA-HIS3 reporter, expression of LexA-Sin3 prevents growth on -Leu, -His plates containing 3-AT. Mutations in RPD3, CBK1, HYM1, and SDS3 allow growth on this medium presumably by reducing repression by LexA-Sin3. A mutation that affects expression or accumulation of Rpd3 or the LexA-Sin3 fusion protein would also decrease repression of the CYC1-LexA-HIS3 reporter. A Western immunoblot experiment indicated that LexA-Sin3 and Rpd3 protein levels were unaffected (data not shown). This indicates that these mutations reduce repression by affecting Sin3 function. To determine if these genes are specific to SIN3, we similarly tested whether these mutations affected repression of the unrelated SSN6/TUP1 repression complex. Experiments showed that rpd3, cbk1, hym1, and sds3 disruptions have no effect on repression by LexA-Ssn6 (data not shown). Consequently, we conclude that these mutations specifically reduce repression by LexA-Sin3. In summary, the RPD3, CBK1, HYM1, and SDS3 genes are all required for efficient repression by LexA-Sin3 and for sporulation in diploids, but are not essential for viability.

View larger version (122K): In this window In a new window Download PPT slide   Figure 1. Mutations in SDS3, CBK1, HYM1, and RPD3 result in the loss of LexA-Sin3 repression. Strains containing the integrated CYC1 UAS-LexA-HIS3 reporter and expressing LexA-Sin3 from plasmid M1836 are grown on -Leu -His + 20 mM 3-AT plates for 3 days at 30°. LexA-Sin3 represses expression of the HIS3 reporter, and wild-type strains were unable to grow. Mutations in CBK1, HYM1, RPD3, and SDS3 relieve this repression and allow growth. The strains used were SY641 (wild type), SY599 (cbk1), SY515 (hym1), SY717 (rpd3), and SY415 (sds3).

RPD3 is required for LexA-Sin3 repression:
An rpd3 mutation reduces repression by LexA-Sin3 at the CYC1-LexA-HIS3 reporter (Figure 1). This result is not surprising, as mutations in RPD3 and SIN3 have similar effects on transcriptional regulation, and rpd3 sin3 double mutants are no more severely affected than either single mutant (STILLMAN et al. 1994 ). Furthermore, biochemical analysis has shown that Rpd3 functions in a complex with Sin3 (KADOSH and STRUHL 1997 ; KASTEN et al. 1997 ). To quantitate the loss of repression, we used the CYC1-LexA-LacZ reporter, which has the same CYC1 promoter driving LacZ expression instead of HIS3. To determine repression by LexA-Sin3, we compare LacZ expression in cells expressing LexA only or the LexA-Sin3 fusion protein. As shown in Table 3, LexA-Sin3 represses transcription by 30-fold, and an rpd3 mutation reduces this repression by a factor of 7.5. The identification of rpd3 mutations as relieving repression by LexA-Sin3 demonstrates the validity of our selection strategy.

Mutations in CBK1 or HYM1 affect regulation of some SIN3-dependent genes:
The next two mutations, cbk1 and hym1, will be considered together as they cause similar phenotypes. Both cause a defect in cell separation that can be seen microscopically as large clusters of unseparated cells. The mutations also cause an abnormal colony morphology, with a rough colony surface in contrast to the smooth shimmer of a wild-type colony. In fact, the CBK1 gene (YNL161w) was given the name cell-wall biosynthesis kinase (C. HERBERT, personal communication) because cbk1 mutants display this defect in cell separation, and the protein shows homology to the AGC family of protein kinases (HUNTER and PLOWMAN 1997 ). The HYM1 gene (YKL189w) is named for its similarity to the Aspergillus nidulans gene hymA. Mutations in hymA affect conidiophore development in A. nidulans (KAROS and FISCHER 1999 ). Although it was reported that HYM1 is an essential gene in yeast (KAROS and FISCHER 1999 ), we have found that strains with a hym1 gene disruption are viable and healthy. We attribute the disparity in results to different strain backgrounds.

To quantitate the observed loss of LexA-Sin3 repression in CBK1 or HYM1 mutants (Figure 1), we used the CYC1-LexA-LacZ reporter and quantitated LacZ activity in strains expressing LexA only or the LexA-Sin3 fusion protein (Table 3). A cbk1 or a hym1 mutation results in an ~50% reduction in repression by LexA-Sin3, a much smaller effect on LexA-Sin3 repression than that observed for the rpd3 mutation. Additionally, the cbk1 rpd3 and hym1 rpd3 double mutants show no greater loss of repression than the rpd3 single mutant, and the cbk1 hym1 double mutant shows effects similar to either single mutant, suggesting that CBK1 and HYM1 function in the same genetic pathway.

Reasoning that the mutations identified in our screen should affect transcriptional regulation of genes affected by sin3 or rpd3 mutations, we therefore determined the effect of cbk1 and hym1 mutations on expression of certain SIN3-dependent genes. STE6 is an a-specific gene required in MATa cells for the production of a-factor, and expression of STE6 is sharply reduced in sin3 and rpd3 mutants (VIDAL and GABER 1991 ; VIDAL et al. 1991 ; WANG et al. 1994 ). Sin3 and Rpd3 are thought to function primarily as transcriptional repressors, and it is believed that reduced STE6 expression is an indirect effect (WANG et al. 1994 ). Isogenic strains with cbk1, hym1, and rpd3 mutations were transformed with a plasmid containing a STE6-lacZ reporter, and promoter activity was monitored by quantitating activity of the ß-galactosidase enzyme (Figure 2). There was a reduction in STE6 expression in strains with mutations in cbk1 and hym1, but not to the same extent as seen with the rpd3 mutant. No synergistic effects with this STE6-lacZ reporter were seen in these multiply mutant strains.

View larger version (13K): In this window In a new window Download PPT slide   Figure 2. Mutations in CBK1, HYM1, and RPD3 result in a loss of STE6-LacZ activation. Strains transformed with plasmid p6HA with the STE6-LacZ reporter were grown on medium lacking uracil to maintain the plasmid, and extracts were prepared for ß-galactosidase assays. The quantities represent the average of three independent transformants with standard deviations shown as error bars. Activity of STE6-LacZ was sharply reduced in rpd3 strains, and this reduction was not increased in the double and triple mutants. The strains used were DY150 (wild type), SY618 (cbk1), SY620 (hym1), SY716 (cbk1 hym1), DY1539 (rpd3), SY718 (cbk1 rpd3), SY623 (hym1 rpd3), and SY625 (cbk1 hym1 rpd3).

The PHO5 gene encodes an acid phosphatase, and this gene is induced under phosphate starvation. PHO5 is repressed in wild-type cells in high-phosphate medium, but this repression is lost in sin3 and rpd3 mutants. To measure the effects of these mutations on PHO5 expression, isogenic strains with cbk1, hym1, and sin3 mutations were grown in high-phosphate liquid media, and extracts were prepared for quantitative acid phosphatase assays. As shown in Table 4, PHO5 was not derepressed in cbk1 or hym1 mutants, but was derepressed in the sin3 mutant. We were surprised to find no increase in PHO5 expression in the cbk1 and hym1 mutants because, as noted earlier, these mutants showed an increase in acid phosphatase activity using a colony-staining overlay assay, for which solid media was used. To address this apparent discrepancy, extracts were prepared from cells grown on solid media, and acid phosphatase activity was measured. The results in Table 4 show that cbk1 and hym1 mutants have a small but significant and reproducible increase in acid phosphatase activity when cells are grown on high-phosphate plates. This derepression was not additive in the cbk1 hym1 double mutant (data not shown). It is not easy to explain the difference between the results obtained with the assays from cells grown in liquid or on plates. We do note that cells grown in patches on solid medium would result in a larger fraction of yeast that are in late-log or stationary phase, and this could affect PHO5 expression. Alternatively, there may be localized depletion of specific nutrients from the solid growth medium, and such effects would not be evident during log phase growth in liquid medium. Using cells grown in liquid medium (Table 4), we did note that there was increased derepression in the cbk1 sin3 or hym1 sin3 double mutants compared to the sin3 single mutant. This additive derepression was observed in combination with either rpd3 or sin3, and in cells grown on plates as well as in liquid (data not shown). An additive effect, cbk1 or hym1 with rpd3 or sin3, was not seen in the previous experiments looking at STE6 expression or repression by LexA-Sin3.

Meiosis-specific genes such as IME2 and SPO13 are also negatively regulated by SIN3 and RPD3 (STRICH et al. 1989 ; BOWDISH and MITCHELL 1993 ). IME2 encodes a kinase required for proper expression of meiotic genes and is expressed normally only in diploid cells preparing to undergo sporulation. Mutations in either SIN3 or RPD3 lead to IME2 expression during vegetative growth, even in the haploid state. To quantitate the level of derepression, we utilized an IME2-LacZ reporter integrated at the ADE2 locus. Haploid cells were grown in rich media to mid-log phase, and extracts were prepared for quantitative ß-galactosidase assays. Mutation in either CBK1 or HYM1 lead to a weak derepression of the IME2-LacZ reporter (Figure 3). As with PHO5, we observed a slight additive increase in IME2-LacZ expression in the cbk1 sin3 or hym1 sin3 double mutants compared to the sin3 single mutant. No additive increase was seen in the cbk1 hym1 double mutant.

View larger version (15K): In this window In a new window Download PPT slide   Figure 3. Mutations in CBK1, HYM1, and RPD3 result in derepression of IME2-lacZ. Strains containing the integrated IME2-lacZ reporter were grown on YEPD medium, and extracts were prepared for ß-galactosidase assays. The quantities represent the average of three independent transformants with standard deviations shown as error bars. IME2 is normally repressed in haploid cells. Strains lacking CBK1, HYM1, or both show a weak loss of repression. IME2-LacZ activity increased dramatically in the rpd3 strain. The double and triple mutants cbk1 rpd3, hym1 rpd3, cbk1 hym1 rpd3 have an additive loss in repression. The inset has an expanded view of the first four strains. The strains used were SY170 (wild type), SY383 (cbk1), SY482 (hym1), SY484 (cbk1 hym1), SY337 (sin3), SY389 (cbk1 sin3), SY486 (hym1 rpd3), and SY488 (cbk1 hym1 rpd3).

SIN3 and RPD3 also repress INO1 (encoding inositol-1-phosphate synthase) and TRK2 (low-affinity potassium transporter) expression. Consequently, promoter activity was determined in cbk1 and hym1 single mutants, as well as cbk1 hym1, cbk1 rpd3, and hym1 rpd3 double mutants using either an INO1-LacZ or a TRK2-LacZ reporter. An additional growth assay was used to examine mutational effects on TRK2 expression (the growth assay is described below). Our findings showed no effect of cbk1 or hym1 mutations on INO1 or TRK2 expression, either alone or when combined with a rpd3 mutation (data not shown). Furthermore, these mutations (single or in combination with an rpd3 mutation) did not affect telomeric silencing (data not shown). These observations contrast to the increase in silencing at either the silent mating type loci or at genes linked to telomeres as evidenced in sin3 or rpd3 mutations (DE RUBERTIS et al. 1996 ; RUNDLETT et al. 1996 ; VANNIER et al. 1996 ).

In summary, strains lacking CBK1 or HYM1 show a loss of repression by LexA-Sin3, decreased repression at PHO5 and IME2, and a decrease in STE6 expression. Thus, these mutations have weak effects on several SIN3-dependent promoters, consistent with a role in the function of the Sin3/Rpd3 complex. However, the effects of cbk1 and hym1 mutations were quantitatively less severe than sin3 or rpd3, and cbk1 and hym1 fail to effect all SIN3-dependent promoters. The fact that the cbk1 hym1 double mutant has no greater effect than the single mutants implies that CBK1 and HYM1 function in the same genetic pathway. When cbk1 or hym1 are combined with rpd3 or sin3 we observe an additive effect only at PHO5 and IME2-LacZ. Finally, strains lacking either CBK1 or HYM1 display an additional phenotype, an altered colony morphology due to a defect in cell separation. This defect is not increased in the double mutant, supporting the conclusion that CBK1 and HYM1 function in the same genetic pathway. Based on the additive effects seen at PHO5 and IME2-LacZ and the failure of cbk1 or hym1 mutations to effect all SIN3-dependent genes, we suggest that CBK1 and HYM1 are in a common genetic pathway that is distinct from SIN3 and RPD3.

SDS3 is required for SIN3-dependent repression:
A mutation in the SDS3 gene reduces repression by LexA-Sin3 as shown by the histidine prototrophy in strains with the CYC1-LexA-HIS3 reporter (Figure 1). SDS3 was originally identified in a screen for mutations that restore silencing at a silencer crippled by both cis- and trans-mutations (VANNIER et al. 1996 ). This screen also identified mutations in SIN3 (SDS16) and RPD3 (SDS6) as restoring silencing. The work of VANNIER et al. 1996 suggested that SDS3 function is related to, but genetically distinct from, that of SIN3 and RPD3.

We used strains transformed with plasmids expressing either LexA or the LexA-Sin3 fusion protein and a CYC1-LexA-LacZ reporter to quantitatively measure the effects of the sds3 mutation on repression by LexA-Sin3 (Table 5). Repression was calculated as the ratio of reporter activity in cells expressing LexA only to those expressing LexA-Sin3. The strain with the sds3 mutation has a loss of LexA-Sin3 repression equivalent to that seen in the rpd3 strain. The sds3 rpd3 double mutant shows a loss of repression similar to the two single mutants, suggesting that they function in the same pathway. An sds3 mutation does not affect repression by the mechanistically distinct LexA-Ssn6 fusion protein, demonstrating specificity toward Sin3 repression. Thus, mutations in SDS3 and RPD3 cause similar phenotypes, and the double mutants are not additive.

To compare the effects of the sds3 mutation with that of rpd3 and sin3, we examined expression of a number of SIN3-dependent promoters, including STE6, PHO5, IME2, and INO1. Promoter activity was determined in sds3, sin3, and rpd3 single mutants, as well as in sds3 sin3 and sds3 rpd3 double mutants (Table 6 and data not shown). Expression of a STE6-LacZ reporter was reduced to similar extents in sds3 and rpd3 mutants (Table 6A). The PHO5 gene was derepressed in both sds3 and rpd3 mutants (Table 6B). An IME2-LacZ reporter was not expressed in vegetatively grown cells, but was derepressed in both sds3 and rpd3 mutants (Table 6C), in agreement with the previous results (BOWDISH and MITCHELL 1993 ; VANNIER et al. 1996 ). INO1, a SIN3-dependent gene, is repressed in the presence of inositol and choline. Mutations in SDS3 and RPD3 both lead to derepression of an INO1-LacZ reporter (Table 6D). There are two important results in this set of experiments on transcriptional regulation. First, an sds3 mutation has a quantitatively similar effect as sin3 or rpd3 mutations. Second, there was no increase in effect in the sds3 sin3 and sds3 rpd3 double mutants compared to the single mutants at all five transcriptional reporters.

SDS3 is in the same genetic pathway as SIN3 and RPD3:
VANNIER et al. 1996 presented evidence that SDS3 has similar functions as SIN3 and RPD3, but they also came to the conclusion that SDS3 was in a different genetic pathway than SIN3 and RPD3, based on two observations. The first was that an sds3 mutation failed to derepress a TRK2-LacZ reporter, while sin3 and rpd3 mutations caused an increase in TRK2-LacZ expression. The second observation was that sds3 sin3 and sds3 rpd3 double mutants displayed an increase in silencing compared to the single mutants, and this additive effect suggested that SDS3 functioned in a different pathway.

We performed several experiments in an attempt to resolve these apparent discrepancies about the relationship of SDS3 to SIN3 and RPD3. We first attempted to test the effect of sds3, sin3, and rpd3 on expression of the TRK2-LacZ reporter present on a multicopy plasmid. We found that this reporter failed to yield reproducible results. In some experiments there was derepression of TRK2-LacZ in an sds3 mutant, while in others this derepression was not observed. While we were always able to demonstrate significant derepression of the TRK2-LacZ reporter in sin3 and rpd3 mutants, even this was subject to significant fluctuations. Because of the lack of reproducibility with the TRK2-LacZ reporter in our hands, we abandoned this reporter in favor of the original growth assay in low-potassium medium for TRK2 expression (VIDAL et al. 1990 ).

Yeast cells contain both high- and low-affinity potassium transporters, encoded by TRK1 and TRK2, respectively. VIDAL et al. 1990 first isolated rpd3 as a suppressor mutation that allowed trk1 mutants, lacking the high-affinity potassium transporter, to grow on media with reduced potassium. The RPD gene name stands for reduced potassium dependence, and sin3 was also isolated in this screen as rpd1. Strains lacking the high-affinity potassium transporter encoded by TRK1 must rely on the low-affinity transporter, Trk2, for potassium uptake. Strains with a trk1 mutation require >5 mM potassium in the medium, and limiting the potassium concentration to 0.2 mM results in no growth. The TRK2 gene, encoding the low-affinity transporter, is normally expressed at very low levels, and mutations such as sin3 and rpd3 that increase TRK2 expression restore growth to a trk1 mutant.

We constructed isogenic trk1, trk1 sds3, trk1 sin3, and trk1 sds3 sin3 strains and determined the ability of these strains to grow on low-potassium medium (Figure 4). trk1 strains grow very poorly, with a doubling time of ~53 hr. As expected, disruption of the SIN3 (rpd1) or RPD3 gene resulted in a significant increase in growth rate under limiting potassium, to ~11 hr (Figure 4 and data not shown). The trk1 sds3 mutant grows much better than the trk1 single mutant, with a doubling time of 34 hr. Significantly, the trk1 sds3 strain does not grow as well as the trk1 sin3 mutant, suggesting that the sds3 mutation has less of an effect on TRK2 expression than sin3. Finally, the trk1 sin3 and trk1 sds3 sin3 strains grow at equivalent rates on low-potassium medium (Figure 4). Thus, the sds3, sin3, and rpd3 mutants all suppress the trk1 defect, and the mutations are not additive. These results suggest that SDS3 does regulate TRK2 and functions in the same genetic pathway as RPD3 and SIN3.

View larger version (19K): In this window In a new window Download PPT slide   Figure 4. SDS3 regulates TRK2 expression and is not additive with sin3. Cells lacking the high-affinity potassium transporter TRK1 must have increased expression of the TRK2 low-affinity potassium transporter in order to survive on low potassium medium. Strains were pregrown in rich medium (replete potassium) and then diluted to a low density in synthetic complete medium supplemented with 0.2 mM potassium (limiting potassium) and grown at 30°. Cell growth was monitored over time by optical density at 660 nm for three independent cultures. Growth rates are plotted with standard deviations shown as error bars for each time point. The strains used were SY562 (trk1), SY564 (trk1 sds3), SY566 (trk1 sin3), and SY568 (trk1 sds3 sin3).

VANNIER et al. 1996 suggested that SDS3 is functionally different from RPD3 and SIN3, based on an additive effect in silencing with the hmrAE::TRP1 reporter. We have compared the sds3 single mutant to sds3 sin3 and sds3 rpd3 double mutants with a variety of transcriptional reporters, and we have not seen any additive effects. As a silencing assay to examine whether sds3 is additive with sin3 or rpd3, we constructed isogenic strains with a URA3 reporter integrated near the telomere of chromosome VII. This telomeric reporter does not require any specific mutations at the HMR-E or RAP1 loci and gives a significantly stronger signal than the hmrAE::TRP1 reporter. Mechanistically, the URA3 telomere-silencing assay is thought to be similar to the HMR-silencing assay, as both are dependent upon the SIR2, SIR3, SIR4, and RAP1 genes (APARICIO et al. 1991 ).

The results of the telomere-silencing assay are shown in Figure 5. In this assay, one measures the fraction of cells with a transcriptionally inactive telomere-linked URA3 gene by determining the fraction of cells incapable of growth on medium lacking uracil. It has been shown that DNA near telomeres is assembled into a heterochromatic state that represses transcription (GRUNSTEIN 1998 ), with the efficiency of this silencing decreasing with distance from the telomere (RENAULD et al. 1993 ). With the URA3 reporter placed 15 kb from the telomere, silencing is quite inefficient in wild-type strains, with nearly 100% of cells growing without added uracil. For the sds3 and rpd3 mutant strains, only ~10% of cells grow on the plate lacking uracil, showing that the two mutations cause a quantitatively similar increase in silencing. Importantly, telomeric silencing in the sds3 rpd3 double mutant was the same as in the sds3 and rpd3 single mutants. Thus, there is no additive effect between sds3 and rpd3, and we believe that SDS3 functions in the same genetic pathway as RPD3 and SIN3.

View larger version (106K): In this window In a new window Download PPT slide   Figure 5. sds3 is not additive with rpd3 in telomeric silencing. A URA3 gene located 15 kb from the telomere on chromosome VII weakly silenced by telomeric heterochromatin, but mutations that increase telomeric silencing will result in decrease plating efficiency on media lacking uracil. Mutations in SDS3 and RPD3 both increase telomeric silencing, but the effect was not additive in sds3 rpd3 double mutants. Serial dilutions (10-fold) of each culture were spotted to medium lacking uracil or to synthetic complete medium, as a control. The strains used were DY5888 (wild type), DY5892 (sds3), DY5894 (rpd3), and DY5900 (sds3 rpd3).

Sds3 is present in the Sin3 complex:
Sin3 and Rpd3 are physically associated (KADOSH and STRUHL 1997 ; KASTEN et al. 1997 ). We used immunoprecipitation methods to determine whether Sds3 is present in the Sin3 complex. Strains were constructed that expressed epitope-tagged versions of Sin3 and Sds3. Sin3 contained a 3 x HA epitope tag, and Sds3 contains a 13 x Myc tag, both as C-terminal fusions. When extracts were prepared from strains and immunoprecipitated with anti-HA antibody, the Sin3-HA fusion protein was detected in Western blots (Figure 6, lanes 1 and 2). The Western blot signal was absent from strains not expressing the Sin3-HA fusion (Figure 6, lanes 4 and 5), and the signal was also abolished by addition of blocking peptide (Figure 6, lane 3). Sds3-Myc coprecipitates with Sin3-HA (Figure 6, lane 1), and this Western blot signal was absent from strains lacking the Myc-tagged Sds3 (Figure 6, lane 2). These experiments show clearly that Sds3 is present in the Sin3 complex.

View larger version (38K): In this window In a new window Download PPT slide   Figure 6. Sds3 and Sin3 coimmunoprecipitate. Extracts were prepared from strains expressing Sin3-HA and/or Sds3-Myc, as indicated, precipitated with antibody to HA, and the immunoprecipitates were probed in Western blots for Sin3-HA and Sds3-Myc. An excess of blocking peptide was added to the sample in lane 3. Strains DY5699 (wild type) and DY5870 (Sds3-Myc) and plasmid M3561 (Sin3-HA) were used.

	DISCUSSION

TOP ABSTRACT MATERIALS AND METHODS RESULTS DISCUSSION LITERATURE CITED

Sin3 is a transcriptional repressor that is targeted to specific promoters by interacting with DNA-binding proteins. Sin3 is present in a large multiprotein complex that includes the Rpd3 histone deacetylase, and thus Sin3 functions as a repressor, at least in part, by altering the acetylation state of chromatin. We set up a genetic screen to identify other genes that play a role in Sin3-mediated repression. The screen used a LexA-Sin3 fusion protein that represses transcription of promoters with LexA binding sites, and a number of mutations that reduced this repression were isolated. In this article we describe four genes, RPD3, SDS3, CBK1, and HYM1, that play a role in Sin3 function.

The fact that we obtained RPD3 in the screen validates our selection strategy. In addition to the fact that Rpd3 is physically associated with Sin3, rpd3 mutations have similar effects on transcriptional regulation as sin3 mutants. Additionally, the phenotype of a sin3 rpd3 double mutant is similar to the single mutants, suggesting that Sin3 and Rpd3 function together (STILLMAN et al. 1994 ).

A variety of experimental observations suggest that SDS3 functions in the same genetic pathway as SIN3 and RPD3. We have shown that sds3 mutations have a similar effect on transcriptional regulation as sin3 and rpd3 mutations. Eight different SIN3-responsive transcription units (CYC1-LexA-HIS3, CYC1-LexA-LacZ, STE6-LacZ, PHO5, IME2-LacZ, INO1-LacZ, TRK2, and telomeric silencing) were used to determine the effect of an sds3 mutation on gene expression. In every case, sds3 has the same effect as sin3 and rpd3. Moreover, in quantitative assays, sds3 was similar to sin3 and rpd3, except in the TRK2 assay for suppression of the poor growth in low-potassium medium due to a trk1 gene deletion. The Swi/Snf complex provides an example where mutations in different components all have related phenotypes, but there can be differences in the phenotypic severity (CAIRNS et al. 1996 ). We conclude that an sds3 mutation causes the same phenotypes as sin3 and rpd3.

Analysis of double mutant strains also suggests that SDS3 functions with SIN3 and RPD3. We have examined expression from all of the SIN3-dependent reporters in sds3 rpd3 or sds3 sin3 double mutants. In all cases, we fail to see any additive effects in the double mutants, compared to the rpd3 or sin3 single mutants. Importantly, this is also true in the TRK2 bioassay, where the sds3 mutant is a less effective suppressor, showing that they are not functioning in different pathways. Of course, an sds3 mutation could reduce repression by reducing expression of either SIN3 or RPD3. However, an sds3 mutation affects repression by both endogenous Sin3 and the LexA-Sin3 fusion protein that is expressed by a different promoter, and Western immunoblots showed that Rpd3 and LexA-Sin3 levels are unaffected.

SDS3 was identified in a screen for mutations that cause increased silencing of a crippled HMR silencer (VANNIER et al. 1996 ). This screen also identified SIN3 and RPD3, suggesting a connection between SDS3 and SIN3/RPD3. These authors concluded that SDS3 was in a different genetic pathway based upon a lack of derepression of a TRK2-lacZ plasmid reporter and additive effects in double mutants observed with a TRP1 gene present at a crippled silencer. However, we found that an sds3 mutation does affect TRK2 expression, based on a bioassay involving growth on low potassium, and we do not see any additive effects on silencing in sds3 rpd3 and sds3 rpd3 double mutant strains. We also conducted double mutant analysis for all of our transcription assays, and we observed no additive effects with sds3. Finally, coimmunoprecipitation experiments show that Sds3 is physically present in the Sin3 complex. We conclude that SDS3 functions in the same genetic pathway as RPD3 and SIN3.

The screen that originally identified sds3 used strains sensitized for silencing, with either the hmrA::ADE2 or hmrA::TRP1 reporter in a rap1^s strain (SUSSEL et al. 1995 ; VANNIER et al. 1996 ). There are two mutations that increase the sensitivity for changes in silencing, one the A mutation in the ORC binding site in the HMR-E silencer, and the other the rap1^s mutation in the gene encoding the Rap1 protein that binds to silencers. Quantitative analysis showed that the sds3 mutation caused a 1000-fold increase in silencing (SUSSEL et al. 1995 ; VANNIER et al. 1996 ). For the experiments showing an additive effect with sds3, the authors used a hmrAE::TRP1 reporter, which has mutations in both the ORC and Rap1 binding sites in the HMR-E silencer (VANNIER et al. 1996 ). With this reporter system, 95% of wild-type cells are Trp⁺, 70% of sds3 and sin3 mutants are Trp⁺, and 36% of sds3 sin3 double mutants are Trp⁺. We feel that the dynamic range of this hmrAE::TRP1 assay is quite different from the 1000-fold seen with the hmrA::TRP1 rap1^s assay system, and thus an apparent additive effect was observed with this assay system for the sds3 sin3 double mutant.

CBK1 and HYM1 appear to function in the same genetic pathway. In addition to reducing repression by LexA-Sin3, cbk1 and hym1 mutations both cause defects in cell separation and altered colony morphology. Similar effects on expression of SIN3-dependent genes are seen in both cbk1 and hym1 mutants. The effects of cbk1 and hym1 are specific to SIN3, as these mutations have no effect on repression by a LexA-Ssn6 fusion protein. Of the seven transcriptional effects analyzed, three (INO1-LacZ, TRK2, and telomeric silencing) were unaffected by these mutations, while cbk1 and hym1 mutations have weak effects on four (CYC1-LexA-LacZ, STE6-LacZ, IME2-LacZ, and PHO5) SIN3-dependent genes. Finally, cbk1 hym1 double mutants show no additive effects, compared to single mutants, in any assay, suggesting that CBK1 and HYM1 function together.

The transcriptional effects of cbk1 and hym1 mutations can be separated genetically from SIN3 and RPD3. These mutations cause extremely weak derepression at IME2-LacZ and PHO5, and the derepression at PHO5 is affected by growth on solid vs. liquid medium. Importantly, the effect is additive when cbk1 or hym1 mutations are combined with sin3 or rpd3. This result suggests that CBK1 and HYM1 are in a separate genetic pathway from SIN3 and RPD3, in terms of regulation of IME2-LacZ and PHO5. The increased derepression observed at PHO5, which is not dependent upon growth conditions, is greater than the sum of the effects of the two mutations and may represent a synergistic effect. Thus, at IME2:LacZ and PHO5, CBK1 and HYM1 function through a mechanism that is independent of Sin3 and Rpd3, suggesting that CBK1/HYM1 are in a different pathway from SIN3/RPD3.

However, double mutant analysis using CYC1-LexA-LacZ and STE6-LacZ indicates that at these two promoters cbk1 and hym1 have modest effects that are genetically inseparable from the effect of the Sin3/Rpd3 complex. At these two promoters, the function of CBK1 and HYM1 appears to be through the Sin3/Rpd3 complex, since the transcriptional effect in strains lacking both CBK1 and RPD3 is no more severe than in the rpd3 single mutant. Thus, the function of CBK1 and HYM1 at these two promoters differs from that seen at IME2:LacZ and PHO5.

We conclude from our analysis that cbk1 and hym1 mutations have different effects at distinct promoters regulated by SIN3. Some promoters, such as CYC1-LexA-LacZ and STE6-LacZ, show modest effects in cbk1 and hym1 mutants and do not show additive effects when combined with rpd3 mutations. At IME2-LacZ and PHO5, there are weak or conditional effects that are additive with rpd3 or sin3. Given that Cbk1 shows strong homology to serine/threonine kinases, it seems likely that its function must be the phosphorylation of some substrate. Although HYM1 has high homology (e-⁴⁰ to e-⁷²) to genes in mouse, Drosophila melanogaster, Caenorhabditis elegans, Arabidopsis thaliana, and A. nidulans, the predicted amino acid sequence provides no clues as to function. Perhaps Hym1 is a subunit of the Cbk1 kinase complex, providing a substrate recognition function. The additive effects between mutations in CBK1/HYM1 and SIN3/RPD3, at least at some promoters, suggest that Sin3 and Rpd3 are not the relevant substrates of the Cbk1 kinase. We propose that the Cbk1 kinase may phosphorylate a chromatin protein. Phosphorylation of such a protein could have different consequences at different promoters, consistent with the different effects of cbk1 and hym1 mutations at different genes.

	ACKNOWLEDGMENTS

We thank members of the Stillman Lab for many helpful suggestions, Janet Shaw for advice on UV mutagenesis, Brad Cairns for advice on immunoprecipitation, Jarmilla Janatova for providing antibodies, and Karen Freedman for making the trk1::ADE2 strain. We thank Jef Boeke, Fred Cross, Rick Gaber, Ira Herskowitz, Yi Wei Jiang, Mark Longtine, John Lopes, David Shore, and Kevin Struhl for providing plasmids and strains. This work was supported by grants from the National Institutes of Health awarded to D.J.S.

Manuscript received August 4, 1999; Accepted for publication October 12, 1999.

	LITERATURE CITED

TOP ABSTRACT MATERIALS AND METHODS RESULTS DISCUSSION LITERATURE CITED

ALLAND, L., R. MUHLE, H. HOU, JR., J. POTES, and L. CHIN et al., 1997  Role for N-CoR and histone deacetylase in Sin3-mediated transcriptional repression. Nature 387:49-55[Medline].

APARICIO, O. M., B. L. BILLINGTON, and D. E. GOTTSCHLING, 1991  Modifiers of position effect are shared between telomeric and silent mating-type loci in S. cerevisiae.. Cell 66:1279-1287[Medline].

AUSUBEL, F. M., R. BRENT, R. E. KINGSTON, D. E. MOORE, J. G. SEIDMAN et al., 1987 Current Protocols in Molecular Biology. John Wiley & Sons, New York.

BOWDISH, K. S. and A. P. MITCHELL, 1993  Bipartite structure of an early meiotic upstream activation sequence from Saccharomyces cerevisiae.. Mol. Cell. Biol. 13:2172-2181[Abstract/Free Full Text].

BRACHMANN, C. B., A. DAVIES, G. J. COST, E. CAPUTO, and J. LI et al., 1998  Designer deletions strains derived from Saccharomyces cerevisiae S288C: a useful set of strains and plasmids for PCR-mediated gene disruption and other applications. Yeast 14:115-132[Medline].

BREEDEN, L. and K. NASMYTH, 1987  Cell cycle control of the yeast HO gene: cis- and trans-acting regulators. Cell 48:389-397[Medline].

CAIRNS, B. R., R. S. LEVINSON, K. R. YAMAMOTO, and R. D. KORNBERG, 1996  Essential role of Swp73p in the function of yeast Swi/Snf complex. Genes Dev. 10:2131-2144[Abstract/Free Full Text].

CROSS, F. R., 1997  Marker swap' plasmids: convenient tools for budding yeast molecular genetics. Yeast 13:647-653[Medline].

DE RUBERTIS, F., D. KADOSH, S. HENCHOZ, D. PAULI, and G. REUTER et al., 1996  The histone deacetylase RPD3 counteracts genomic silencing in Drosophila and yeast. Nature 384:589-591[Medline].

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].

GRUNSTEIN, M., 1997  Histone acetylation in chromatin structure and transcription. Nature 389:349-352[Medline].

GRUNSTEIN, M., 1998  Yeast heterochromatin: regulation of its assembly and inheritance by histones. Cell 93:325-328[Medline].

GUTHRIE, C., and G. R. FINK (Editors), 1991 Guide to Yeast Genetics and Molecular Biology. Academic Press, San Diego.

HAMPSEY, M., 1997  A SAGA of histone acetylation and gene expression. Trends Genet. 13:427-429[Medline].

HAN, M., U. J. KIM, P. KAYNE, and M. GRUNSTEIN, 1988  Depletion of histone H4 and nucleosomes activates the PHO5 gene in Saccharomyces cerevisiae.. EMBO J. 7:2221-2228[Medline].

HASSIG, C. A., T. C. FLEISCHER, A. N. BILLIN, S. L. SCHREIBER, and D. E. AYER, 1997  Histone deacetylase activity is required for full transcriptional repression by mSin3A. Cell 89:341-347[Medline].

HEINZEL, T., R. M. LAVINSKY, T. M. MULLEN, M. SODERSTROM, and C. D. LAHERTY et al., 1997  A complex containing N-CoR, mSin3 and histone deacetylase mediates transcriptional repression. Nature 387:43-48[Medline].

HEPWORTH, S. R., H. FRIESEN, and J. SEGALL, 1998  NDT80 and the meiotic recombination checkpoint regulate expression of middle sporulation-specific genes in Saccharomyces cerevisiae.. Mol. Cell. Biol. 18:5750-5761[Abstract/Free Full Text].

HILL, J. E., A. M. MYERS, T. J. KOERNER, and A. TZAGOLOFF, 1986  Yeast/E. coli shuttle vectors with multiple unique restriction sites. Yeast 2:163-167[Medline].

HUDAK, K., J. LOPES, and S. HENRY, 1994  A pleiotropic phospholipid biosynthetic regulatory mutation in Saccharomyces cerevisiae is allelic to sin3 (sdi1, ume4, rpd1). Genetics 136:475-483[Abstract].

HUNTER, T. and G. D. PLOWMAN, 1997  The protein kinases of budding yeast: six score and more. Trends Genet. 22:18-22.

KADONAGA, J. T., 1998  Eukaryotic transcription: an interlaced network of transcription factors and chromatin-modifying machines. Cell 92:307-313[Medline].

KADOSH, D. and K. STRUHL, 1997  Repression by Ume6 involves recruitment of a complex containing Sin3 corepressor and Rpd3 histone deacetylase to target promoters. Cell 89:365-371[Medline].

KADOSH, D. and K. STRUHL, 1998a  Histone deacetylase activity of Rpd3 is important for transcriptional repression in vivo. Genes Dev. 12:797-805[Abstract/Free Full Text].

KADOSH, D. and K. STRUHL, 1998b  Targeted recruitment of the Sin3-Rpd3 histone deacetylase complex generates a highly localized domain of repressed chromatin in vivo. Mol. Cell. Biol. 18:5121-5127[Abstract/Free Full Text].

KAROS, M. and R. FISCHER, 1999  Molecular characterization of HymA, an evolutionarily highly conserved and highly expressed protein of Aspergillus nidulans.. Mol. Gen. Genet. 260:510-521[Medline].

KASTEN, M. M., S. DORLAND, and D. J. STILLMAN, 1997  A large protein complex containing the Sin3p and Rpd3p transcriptional regulators. Mol. Cell. Biol. 16:4215-4221[Abstract].

KINGSTON, R. E., C. A. BUNKER, and A. N. IMBALZANO, 1996  Repression and activation by multiprotein complexes that alter chromatin structure. Genes Dev. 10:905-920[Abstract/Free Full Text].

LAHERTY, C., W. M. YANG, J. M. SUN, J. R. DAVIE, and E. SETO et al., 1997  Histone deacetylases associated with the mSIN3 corepressor mediate Mad:Max transcriptional repression. Cell 89:349-356[Medline].

LONGTINE, M. S., A. MCKENZIE, III, D. J. DEMARINI, N. G. SHAH, and A. WACH et al., 1998  Additional modules for versatile and economical PCR-based gene deletion and modification in Saccharomyces cerevisiae.. Yeast 14:953-961[Medline].

NAGY, L., H. Y. KAO, D. CHAKRAVARTI, R. J. LIN, and C. A. HASSIG et al., 1997  Nuclear receptor repression mediated by a complex containing SMRT, mSin3A, and histone deacetylase. Cell 89:373-380[Medline].

NASMYTH, K., D. STILLMAN, and D. KIPLING, 1987  Both positive and negative regulators of HO transcription are required for mother-cell-specific mating-type switching in yeast. Cell 48:579-587[Medline].

PAZIN, M. J. and J. T. KADONAGA, 1997  What's up and down with histone deacetylation and transcription? Cell 89:325-328[Medline].

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].

RUNDLETT, S. E., A. A.