Genetics, Vol. 150, 563-576, October 1998, Copyright © 1998
Histone Deacetylase Homologs Regulate Epigenetic Inheritance of Transcriptional Silencing and Chromosome Segregation in Fission Yeast
Shiv I. S. Grewala,
Michael J. Bonaducea, and
Amar J. S. Klara
a Gene Regulation and Chromosome Biology Laboratory, ABL–Basic Research Program, National Cancer Institute–Frederick Cancer Research and Development Center, Frederick, Maryland 21702-1201
Corresponding author:
Amar J. S. Klar, Gene Regulation and Chromosome Biology Laboratory, NCI–Frederick Cancer Research and Development Center, ABL–Basic Research Program, P.O. Box B, Frederick, MD 21702-1201., klarmail{at}ncifcrf.gov (E-mail).
Communicating editor: F. WINSTON
 | ABSTRACT |
|---|
Position-effect control at the silent mat2-mat3 interval and at centromeres and telomeres in fission yeast is suggested to be mediated through the assembly of heterochromatin-like structures. Therefore, trans-acting genes that affect silencing may encode either chromatin proteins, factors that modify them, or factors that affect chromatin assembly. Here, we report the identification of an essential gene, clr6 (cryptic loci regulator), which encodes a putative histone deacetylase that when mutated affects epigenetically maintained repression at the mat2-mat3 region and at centromeres and reduces the fidelity of chromosome segregation. Furthermore, we show that the Clr3 protein, when mutated, alleviates recombination block at mat region as well as silencing at donor loci and at centromeres and telomeres, also shares strong homology to known histone deacetylases. Genetic analyses indicate that silencing might be regulated by at least two overlapping histone deacetylase activities. We also found that transient inhibition of histone deacetylase activity by trichostatin A results in the increased missegregation of chromosomes in subsequent generations and, remarkably, alters the imprint at the mat locus, causing the heritable conversion of the repressed epigenetic state to the expressed state. This work supports the model that the level of histone deacetylation has a role in the assembly of repressive heterochromatin and provides insight into the mechanism of epigenetic inheritance.
PROPAGATION of stable states of gene expression is thought to be critical for the development and maintenance of differentiated cell types. Examples in higher eukaryotes include X-chromosome inactivation in female mammals (reviewed by RIGGS and PORTER 1996 
), parental imprinting (reviewed by BARLOW 1995 ; AINSCOUGH and SURANI 1996 ), and stable expression of homeotic genes in Drosophila (PARO 1993 ). The underlying mechanisms by which differentiated cells establish and maintain stable patterns of gene expression are not clear. Studies of position-effect variegation (PEV), a phenomenon in which heterochromatin variably but stably silences nearby genes, have provided a model for understanding the propagation of committed states of gene expression. It was recently suggested that polycomb-mediated stable inactivation of homeotic genes and PEV in Drosophila are related phenomena (ORLANDO and PARO 1995 ), supporting the idea that position effects are mediated through long-range control exerted by repressive chromatin structures.
Silenced genomic regions in the distantly related yeasts Saccharomyces cerevisiae and Schizosaccharomyces pombe, such as silent donor mating-type loci (reviewed by KLAR 1989 ), telomeres (GOTTSCHLING et al. 1990 ; NIMMO et al. 1994 ), and centromeres (reviewed by ALLSHIRE 1996 ), share many parallels with heterochromatic regions in higher eukaryotes. In fission yeast, the mat2 and mat3 loci, which serve as donors for mat1 alleles during mating-type interconversion, are silenced even though they contain genetic information identical to that of the expressed mat1-P or mat1-M alleles, respectively (reviewed by KLAR et al. 1998 ). Furthermore, it was recently observed that an ~11-kb recombinationally "cold" region between the donor loci, referred to as the K region, is also subject to position effects; i.e., expression of the ura4+ marker gene inserted in this region is repressed severely (THON et al. 1994 ; GREWAL and KLAR 1997 ). Mutations in six trans-acting genes [clr1, clr2, clr3, clr4 (cryptic loci regulator), swi6, and rik1] partially derepress the donor loci and alleviate the repression of ura4+ at different locations within the mat2-mat3 interval (EGEL et al. 1989 ; LORENTZ et al. 1992 ; THON and KLAR 1992 ; EKWALL and RUUSALA 1994 ; THON et al. 1994 ; GREWAL and KLAR 1996 , GREWAL and KLAR 1997 ), suggesting that these gene functions affect this region in a global fashion. These genes are also shown to prohibit meiotic and mitotic recombination in the entire mat2-mat3 region (EGEL et al. 1989 ; KLAR and BONADUCE 1991 ; LORENTZ et al. 1992 ; THON and KLAR 1992 ; THON et al. 1994 ).
All six trans-acting genes, originally identified for their role in silencing at the mating-type region, were subsequently shown to affect position-effect control of expression of a marker gene integrated at centromere 1 (cen1; ALLSHIRE et al. 1995 ), indicating their involvement in the functional organization of centromeres. A mechanistic link between centromeric and mating-type region silencing is further suggested by our recent findings that approximately one-third of the K region is homologous to the centromeric repeat sequences (GREWAL and KLAR 1997 ). Furthermore, analogous to the existence of an epigenetic component in centromere function (STEINER and CLARKE 1994 ; EKWALL et al. 1997 ), silencing, mating-type interconversion, and recombinational suppression at the mating-type region are also regulated by an epigenetic mechanism (GREWAL and KLAR 1996 ). Many studies have suggested that centromeres and the mating-type locus are assembled into a higher-order chromatin structure refractory to transcription and to recombination machinery, yet recombination required for mat1 switching is allowed. Similarly, the clr1-clr4, swi6, and rik1 gene products are believed to facilitate the assembly of a heterochromatin-like structure, which, once assembled, is capable of self-templating (GREWAL and KLAR 1996 ). Consistent with the chromatin model, Swi6 (LORENTZ et al. 1994 ) and Clr4 (IVANOVA et al. 1998 ) proteins share homology with the chromodomain motif found in heterochromatin-associated proteins in higher eukaryotes (reviewed by SINGH 1994 ), and Swi6 colocalizes with centromeres, telomeres, and the mating-type region (EKWALL et al. 1995 ). The exact mechanism of the assembly of heterochromatin-like structures is not known. However, genetic and biochemical evidence from other systems suggests that the acetylation status of conserved lysine residues in the amino-terminal tails of histones H3 and H4 plays a role in chromatin assembly and transcription regulation (see BRAUNSTEIN et al. 1993 ; BROWNELL and ALLIS 1996 ; WOLFFE 1996 ; HARTZOG and WINSTON 1997 ; PAZIN and KADONAGA 1997 ).
Independent of any perceived molecular model of repression, we have used a genetic approach to identify trans-acting genes essential for silencing. A new locus, clr6, was identified that encodes a putative histone deacetylase. We have investigated the effects of clr6-1 mutation on silencing at the mating-type region and at centromeres and telomeres. We also examined the effects of general inhibition of histone deacetylation on epigenetic inheritance and demonstrated that transient inhibition by the drug trichostatin A (TSA) alters the imprint at the mat2-mat3 region. We also describe the molecular characterization of the previously identified clr3 gene and discuss the implications of our findings for current models of structural organization at the mating-type region and at centromeres in S. pombe, as well as for the mechanism of epigenetic inheritance in general.
| MATERIALS AND METHODS |
|---|
Strains and culture conditions:
Most of the S. pombe strains used in this study were constructed in our laboratory (Table 1). Strains FY336, FY498, FY520, and FY648 have been described by ALLSHIRE et al. 1995 . The construction of HM248 and PRZ119 was described by NIWA et al. 1986 and REYNOLDS et al. 1990 , respectively. The K
::ura4+ allele was constructed as part of an earlier study (GREWAL and KLAR 1996 ). We used standard genetic crosses to construct all other strains.
Standard conditions were used for growth, sporulation, random-spore analysis, tetrad analysis, gap repair, transformations, and construction of diploids from haploid cells (MORENO et al. 1991 ). YEA and PMA (supplemented with appropriate amino acids) were used as rich and sporulation media, respectively. Synthetic medium lacking an appropriate nutrient such as uracil (AA-URA) was used in most of the experiments. FOA medium contains 5-fluoroorotic acid (0.8 g/liter) in synthetic complete medium. Ura+ and Ura- derivatives of K::ura4+ strains were obtained by selecting for their growth on AA-URA or FOA medium, respectively. Quantitation of ura4+ expression was carried out by growing cells on synthetic Edinburgh minimal medium containing glutamate, adenine, leucine, and uracil, each at 75 mg/liter (N/S), URA- (N/S lacking uracil), and FOA (N/S supplemented with FOA). The temperature and ultraviolet radiation (UV) sensitivities were determined as described in Figure 1.
View larger version (50K):
In this window
In a new window
Download PPT slide
|
Figure 1.
The clr6-1 mutant phenotypes. (A) Cumulative effect of clr6-1 and clr1-5 mutations on mat2-P derepression. The colonies of nonswitchable M cells (mat1-Msmto) were sporulated and exposed to iodine vapors before photography. Because mat1-Msmto strains are prohibited from switching, their intensity of staining indicates the level of haploid meiosis within the colonies, qualitatively reflecting the level of mat2-P derepression. In cells of this genotype, derepression of mat3-M cannot be determined by the staining assay. The strains used and their relevant genotypes were as follows: SP1173, WT; SP1167, clr1-5; SP1240, clr6-1; and SP1195, clr1-5 clr6-1. (B) Temperature sensitivity of clr6-1. The strains previously grown under permissive growth conditions (30°) were replicated onto YEA plates and incubated overnight at 30° or 37°. clr6-1 displayed a gradual Ts- growth defect. The cells grown overnight were further replicated onto YEA and incubated at the respective temperatures for 2 days before photography. The strains used were SPG27, WT and SP1464, clr6-1. (C) UV sensitivity. Patches of the indicated strains were replicated onto YEA plates in duplicate, and one of the plates was UV irradiated (200 µJ/cm2 for 2 min). Plates were incubated in the dark at 30° for 4 days before photography. As a control, cultures grown without UV treatment are shown. The strains used were as follows: SP1302, WT; SP1251, clr1-5 clr6-1; SP1306, clr1-5 clr6-1::LEU2::clr6+ (a strain that is isogenic with SP1251 but carries clr6+ integrated next to the clr6-1 allele); SP1242, clr6-1; and PRZ119, rhp6-1 (a known UV-sensitive control strain).
|
|
Iodine-staining assay:
We used the iodine-staining assay to estimate the efficiency of mating-type switching and the level of derepression of donor loci. Individual colonies grown on sporulation medium for 3 days were exposed to iodine vapors. Synthesis of a starch-like compound by sporulating cells results in black staining of colonies after exposure to iodine vapors (BRESCH et al. 1968 ). In switching-competent cells, the intensity of iodine staining indicates the efficiency of mating-type switching; in a nonswitching background, it indicates the level of "haploid meiosis," a phenotype resulting from concurrent expression of both P and M information in a haploid cell (THON and KLAR 1992 ).
Isolation of the clr6-1 mutation strain:
SP1167 containing the nonswitchable mat1-Msmto (ENGELKE et al. 1987 ) allele and the mutant clr1-5 allele displays only partial derepression of the mat2-P locus (THON et al. 1994 ). Therefore, when grown on sporulation (PMA+) medium, SP1167 colonies showed weak haploid meiosis and, hence, stained very lightly with iodine vapors. To identify new mutants, SP1167 was mutagenized with ethylmethanesulfonate to 50% viability by following the procedure of MORENO et al. 1991 . The mutagenized cells were plated onto PMA+ medium and allowed to grow at 25° for 6 days. Several mutants showing dark iodine staining and high levels of haploid meiosis were isolated. One mutant, clr6-1, conferred a growth defect at 37° and was investigated further.
Cloning and sequencing:
To clone the clr6 gene and the previously identified clr3 (THON et al. 1994 ) gene, we transformed SP1359 and SP1195, respectively, with the S. pombe partial Sau3A genomic library cloned in the pWH5 vector (WRIGHT et al. 1986 ). The vector contains the S. cerevisiae LEU2 gene, which upon transformation, can complement the leul-32 defect in S. pombe. The screening of Leu+ transformants identified plasmids containing 8.0-kb (pAK86) and 5.0-kb (pClr6) inserts that were able to complement the recessive clr3-735 and clr6-1 alleles, respectively. Upon retransformation, both plasmids complemented mutant phenotypes of the respective strains. To confirm that cloned DNA contained wild-type copies of clr3 and clr6 rather than extragenic suppressors, the plasmids were integrated by homologous recombination and analyzed for linkage. The plasmid-borne LEU2 marker showed tight linkage to the respective loci because only the parental ditype segregation pattern was found in all the tetrads we analyzed (69 tetrads for clr3 and 44 tetrads for clr6 locus).
The nucleotide sequence of the subcloned genomic fragments was determined by using the Prism Ready Reaction Dye Deoxy Terminator Cycle Sequencing Kit (Applied Biosystems, Foster City, CA). Samples were applied to the Applied Biosystems model 373A sequencing system. Data were analyzed by using GCG software (University of Wisconsin). To determine the location of introns, cDNA was amplified from the S. pombe cDNA library and sequenced. The nucleotide sequences of the clr6 and clr3 genes have been deposited in the GenBank database (accession numbers AFO64206 and AFO64207, respectively).
Construction of the clr6 disruption allele:
A 0.9-kb fragment containing part of the clr6 gene was subcloned into pBS/KS (Stratagene, La Jolla, CA). The ura4+ gene in a 1.8-kb HindIII fragment was then inserted into the HindIII site to construct pBS/clr6::ura4+. A 2.7-kb XhoI-PstI insert containing the disrupted clr6 open reading frame (ORF) was used to transform the clr6+/clr6+ ura4/ura4 diploid strain to uracil prototrophy. Transformants with the correct integration were identified by Southern analysis.
DNA and RNA analyses:
DNA from S. pombe cultures grown overnight was prepared as described by MORENO et al. 1991 , and RNA was prepared as described by SCHMITT et al. 1990 . Southern and Northern hybridizations were carried out as described by SAMBROOK et al. 1989 . The probes used in this study were as follows: a 1.8-kb HindIII fragment containing ura4+, a 0.9-kb fragment containing part of the clr6 gene, a 1.0-kb BclI-TaqI DNA fragment carrying part of the M cassette, and a NdeI-PstI fragment containing cdc2.
Fluctuation test:
The fluctuation test was carried out as described by LURIA and DELBRUCK 1943 . The Ura- and URA+ derivatives of SPG27 and SP1464 were diluted to a concentration of about six cells per milliliter in YEA nonselective liquid medium. Aliquots (100-µl) of cultures were dispensed into microtiter plates and allowed to grow at 30° for 24 hr before plating onto YEA plates. After 3 days at 30°, the colonies were replicated onto PMA+ sporulation and AA-URA media. Iodine staining and K::ura4+-expression phenotypes were scored after 3 days, and the rate of appearance of the variant (µ) was calculated by using the formula µ = -[ln(B/A)]/N, where N is the average number of colony-forming units per culture, B is the number of cultures without variation, and A is the total number of cultures in the experiment.
Chromosome segregation assay:
The chromosome loss experiment was carried out as described by ALLSHIRE et al. 1995 . Cells from Ade+ colonies were plated onto adenine-limited medium (YE) and incubated at 30° for 4 days. If chromosome loss occurs in the first division of a cell plated on YE, half of the resultant colony carrying Ch16 will be white, whereas the other half without Ch16 will be red. The number of half-sectored red/white colonies was determined, and the rate of chromosome loss per cell division was calculated by dividing the number of half-sectored colonies by the total number of white colonies plus half-sectored colonies. For TSA treatment, 2 µl of TSA [10 µg/ml dissolved in dimethylsulfoxide (DMSO)] was spotted onto filter paper discs (1-cm diameter) laid over a lawn of cells (HM248) spread on a plate. The plates were incubated at 33° for different time intervals, and cells growing underneath the filters were plated onto YE medium.
| RESULTS |
|---|
Isolation and characterization of the clr6-1 mutation:
Our past studies have indicated that silencing of the donor mating-type loci is regulated by redundant mechanisms and that deletion of a 1.5-kb cis-acting mat2-P-proximal region potentiates the derepressing effect of mutations in all the known trans-acting (clrl-clr4 and swi6) genes (THON et al. 1994 ). Therefore, it is likely that there exist additional trans-acting factors that are involved in silencing. To identify the hypothesized additional factors, we mutagenized a nonswitchable mat1-Msmto strain carrying the mutant clr1-5 allele (SP1167) and sought mutants exhibiting strong derepression of the mat2-P locus. We obtained several trans-acting mutations that increased haploid meiosis, a phenotype that occurs in haploid cells expressing both P and M information (THON and KLAR 1992 ) and causes increased iodine staining of colonies formed by nonswitching strains. Here we present the analysis of one of these mutations, clr6-1, which, according to our linkage analysis, was distinct from previously identified trans-acting mutations and caused increased haploid meiosis (Figure 1A). Analysis of the clr6-1 mutant cells revealed that the mutation caused a temperature-sensitive (Ts-) growth defect at 37° (Figure 1B). Interestingly, the clr6-1 mutant cells did not show the Ts- growth defect in the first five to six cell divisions after shifting to the restrictive temperature of 37°. In addition to the Ts- phenotype, we also found that the mutant cells displayed increased UV sensitivity (Figure 1C).
The observed increase in haploid meiosis could be caused by a defect in the pat1 or mei2 genes, which control meiosis and whose mutants confer this phenotype, even in the absence of donor loci (reviewed by EGEL 1994 ). Alternatively, the increase in haploid meiosis could result from derepression of donor loci. We have differentiated between these possibilities. First, genetic analysis showed that the haploid meiosis phenotype was dependent on the mat2-P and mat3-M donor loci (data not shown). Second, Northern analysis (Figure 2) revealed an increase in the mat3 transcript level in the clr1-5 clr6-1 double mutant as compared with clr1-5 alone, indicating that the clr6 gene regulates silencing of mat3. Furthermore, the clr6-1 mutation, originally isolated for its effect on mat2-P, also derepresses mat3-M in a clr1 background. To test whether the clr6-1 mutation had a cumulative effect with clr1-5 or was epistatic to clr1 function, we constructed a clr1+ clr6-1 strain. As expected for a mutation that affects only part of the redundant or overlapping silencing mechanisms, the clr6-1 allele by itself had no visible effect on the silencing of mat2 and mat3 (Figure 1A and Figure 2). We also analyzed the cumulative effect of the clr6-1 allele with mutations in clr2, clr3, clr4, and swi6, which are suggested to work in a single pathway in conjunction with clr1, as pairwise combinations do not show cumulative effect (THON et al. 1994 ). Surprisingly, the clr6-1 allele caused increased haploid meiosis only in the clr3 mutant background (SP1286; data not shown). Furthermore, the clr6-1 mutation did not show a cumulative effect when combined with mat2-P-proximal deletion. We explain these findings by suggesting that Clr6 acts in an overlapping manner with Clr1/Clr3 and that both these functions feed into a silencing pathway involving clr2, clr4, and swi6 products.
Cloning and sequencing of clr6 reveals a similarity to histone deacetylases:
Because the clr6-1 mutant is temperature sensitive, it is likely that the lethality is caused by pleiotropic effects on diverse cellular functions. An important clue to clr6 function was provided by its nucleotide sequence. For this purpose, we isolated a clone from a partial Sau3A genomic library of S. pombe by complementing Ts- growth and the haploid meiosis defects in the mat1-Msmto clr1-5 clr6-1 (SP1195) mutant strain. A 5.0-kb insert complemented both defects. Sequencing of the genomic clr6 and its cDNA clones identified an ORF that is interrupted by 13 introns. (Figure 3A). The ORF is capable of encoding a protein of 405 amino acids (Figure 3B). Comparison of the predicted Clr6 amino acid sequence with the GenBank database revealed that Clr6 shares strong homology to members of a recently described family of histone deacetylases, such as human HDAC1 (formerly HD1; TAUNTON et al. 1996 ; 61% identity and 78% similarity) and S. cerevisiae Rpd3 (VIDAL and GABER 1991 ; 63% identity and 77% similarity; Figure 3B). Clr6 also shares similarity to the predicted amino acid sequences of two recently sequenced ORFs from S. pombe (Sanger Genome Sequencing Center), which we have designated homologs of clr6 (hoc1 and hoc2 mapping to cosmids c3G9 and SP41410, respectively). The sequence alignments revealed that Clr6 shares greater homology to Hoc1 (50% identity and 73% similarity) than to Hoc2 (26% identity and 51% similarity).
View larger version (43K):
In this window
In a new window
Download PPT slide
|
Figure 3.
The clr6 gene. (A) The clr6 ORF. Thirteen introns interrupting the ORF and the filled boxes indicating exons are shown. (B) Comparison of the predicted Clr6, S. serevisiae Rpd3, and human HDAC1 amino acid sequences. The sequences were aligned using the PILEUP program. The consensus sequence indicating identical residues in all three sequences is shown at the bottom. The asterisk indicates the residue that is changed to aspartic acid [D] in the clr6-1 mutant allele.
|
|
To map the mutation in the clr6-1 mutant, we used PCR to amplify the genomic clr6 DNA containing clr6 from wild-type and mutant cells. The sequencing of the PCR products identified a single G-to-A transition in codon 269, converting GGC (Gly) to GAC (Asp). Interestingly, the mutation maps to a region of Clr6 that is conserved in Rpd3 from S. cerevisiae and human HDAC1. The genomic location of clr6 was determined as described by HOHEISEL et al. 1993 . The results revealed that clr6 maps near the telomere of the short arm of chromosome II between markers mei3 and rad3. A disruption allele of clr6 was made by DNA-mediated transformation of a diploid strain (see MATERIALS AND METHODS). Upon dissection of the heterozygous diploid, haploid segregants carrying the clr6 null allele were found to be essentially inviable (data not shown). However, microscopic examination revealed that these haploid cells were able to divide for three to four generations before growth arrest, a phenotype similar to that of the clr6-1 Ts- strain when it was shifted to the restrictive temperature (see above). Apparently, the clr6-1 allele must cause a partial loss of function because the deletion is lethal.
Mutation in clr6 affects propagation of the epigenetic states regulating silencing, recombinational suppression, and directionality of switching:
It has been shown that in a strain with part of the K region replaced by the ura4+ marker gene (K::ura4+ allele), ura4+ expression, recombination between flanking markers, and the efficiency of mating-type interconversion are covariegated and regulated by an epigenetic mechanism (GREWAL and KLAR 1996 , GREWAL and KLAR 1997 ). More importantly, Ura- efficient switching and Ura+ inefficient switching variegated states (referred to as Ura- and Ura+ states, respectively) are inherited in cis and proposed to be the result of specific chromatin organization, with Ura+ cells being defective in assembly of the structure (GREWAL and KLAR 1996 ). These results have been subsequently confirmed (THON and FRIIS 1997 ) and support our suggestion that deletion of part of the K region compromises the establishment of a heterochromatin-like repressive structure (GREWAL and KLAR 1996 ).
Here, we determined whether a mutation in clr6, a histone deacetylase homolog, affects propagation of the epigenetic states. By genetic crosses, the clr6-1 mutation was combined with derepressed (Ura+) and repressed (Ura-) "epialleles" of K::ura4+. First, donor loci remained silent in cells carrying either the Ura+ or Ura- epistate. Second, comparison of the rate of Ura- to Ura+ transition in the clr6-1 with the rate in the clr6+ cells by dilution assay revealed that the mutation by itself did not significantly affect the mitotic stability of the Ura- state (Figure 4). We also used the fluctuation test to quantitatively measure the effect of the clr6 mutation on the stability of epigenetic states (see MATERIALS AND METHODS). Confirming the results of serial dilution (Figure 4), the clr6 mutation had only a subtle effect on the Ura- to Ura+ transition, as compared with its wild-type counterpart (Table 2). However, we observed that the clr6-1 allele caused a significant (~20-fold) increase in Ura+ to Ura- conversion. This result shows that the mutation paradoxically increases the efficiency of the establishment of a transcriptionally repressed state of K::ura4+.
The silencing function of clr6 seems to overlap with that of clr1 and clr3; therefore, we tested the effect of the clr1 clr6 and clr3 clr6 double mutations on the propagation of epigenetic states. As a control, K::ura4+ strains carrying a mutation in clr1 or clr3, which were previously shown to completely suppress variegation (GREWAL and KLAR 1996 ), were also included. Dilution analysis (Figure 4) demonstrated that like the clr1 or clr3 mutant strains, double mutants did not grow on FOA plates, indicating alleviation of ura4+ repression. More significantly, Ura+ cells with the double mutation were unable to revert to the Ura- state. Overall, these data suggest that mutation in clr6 leads to increased conversion of the transcribed state into the repressed state, presumably through assembly of a higher-order chromatin structure, and that increased repression in the clr6 mutant is dependent on the clr1 and clr3 gene products.
TSA treatment adversely affects propagation of the epigenetic imprint at the mat2-mat3 region:
Parallel to the clr6-1 analysis, we also investigated the effect of inhibition of histone deacetylation on propagation of the Ura- state by treating cells with TSA, a specific inhibitor of histone deacetylases (YOSHIDA et al. 1995 ). A brief treatment by TSA for ~10 generations converted the majority of cells that originally displayed the Ura- state to the Ura+ state. Remarkably, subsequent plating revealed that the newly acquired Ura+ state was stable for many (>30) generations after the drug treatment (Figure 5). Measurement of the mitotic stability of the TSA-induced state indicated that these Ura+ cells reverted to the Ura- state at a frequency comparable to that of standard Ura+ cells (data not shown). Furthermore, we observed that ~30% of the cells exposed to TSA for 1 day exhibited the haploid meiosis phenotype when transferred to sporulation medium in the absence of the drug, suggesting that the drug treatment causes heritable derepression of donor loci as well. Similarly, cells displaying the Ura+ state were also exposed to TSA and, as expected, the treatment did not change them to the Ura- phenotype. Briefly, these data show that transient exposure to TSA causes conversion of the transcriptionally repressed state to the expressed state, and that the expressed state is stably propagated in the progeny for many generations in the absence of TSA.
In principle, the TSA effect (transition from Ura- to Ura+) could be caused either by stable changes in the expression of trans-acting loci that are essential for silencing or by alteration in the chromosomally inherited imprint at the mat locus (GREWAL and KLAR 1996 ). We differentiated between these possibilities by the following cross (see Figure 6). A strain carrying the Ura+ epillele (SPG27, K::ura4+ his2), obtained by TSA treatment of Ura- cells, was crossed to a Ura- (SPG51, K::ura4+ his2+) strain. The resultant diploid was sporulated and subjected to tetrad analysis. If the effect of TSA is localized to the mating-type region, two Ura- and two Ura+ segregants should be found in each tetrad, and these states should cosegregate with respective alleles of his2, a marker that is closely linked to the mating-type region. This segregation pattern was indeed observed in all 30 tetrads analyzed. Thus, inhibition of histone deacetylase activity adversely affects propagation of the stable chromosomal imprint, which determines gene activity, at the mat2-mat3 locus. Similarly, a heritable change from Ura- to Ura+ linked to the mating-type region was also observed in nonswitching mat1-Msmto K::ura4+ (SPG32) strain (data not shown).
The clr6 mutation affects ura4+ repression at three sites within cen1:
Assembly of the repressive chromatin structure is also believed to be essential for the proper functioning of fission yeast centromeres (CLARKE et al. 1993 ). As mentioned above, trans-acting elements originally identified as essential for silencing in the mating-type region also affect the silencing of markers artificially inserted into the centromeric sequences (ALLSHIRE et al. 1995 ). Therefore, it was conceivable that clr6 might also affect centromeric silencing. To test this, we combined the clr6-1 mutation with ura4+ integrated at three sites within cen1: one in the central domain, cnt1/TM1, and one in each of the outer flanking repeats, imr1R and otr1R (ALLSHIRE et al. 1995 ). Interestingly, the mutation showed contrasting effects on ura4+ integrated at different centromeric domains. Dilution analysis of strains carrying imr1R(NcoI)::ura4+ and otrR (SphI)::ura4+ insertions showed that the clr6-1 mutation significantly increased their ability to grow on URA- medium, but there was no growth difference between clr6-1 and wild-type cells on FOA medium (Figure 7, A and B). This result suggests that the level of expression of ura4+ inserted at both locations was increased, but that the mutation did not completely abolish repression, with ura4+ expression presumably remaining below the level required to cause FOA sensitivity. In contrast and paradoxically, the mutation resulted in increased repression of cnt1/TM1(NcoI)::ura4+ insertion at the central domain of cen1, as indicated by the impaired ability of cells to form colonies on URA- medium and the corresponding increased growth on FOA plates (Figure 7C).
View larger version (203K):
In this window
In a new window
Download PPT slide
|
Figure 7.
The effect of clr6 mutation on silencing of ura4+ inserted at three different locations within cen1 and adjacent to a telomere. Cultures with imr1R(NcoI)::ura4+ (A), otr1R(SphI)::ura4+ (B), cnt1/TM1::ura4+ (C), or Ch16m23-ura4+-TEL (D) in N/S medium were suspended in water. Fivefold serial dilutions of cell suspensions were spotted onto N/S, URA-, and FOA media, and the plates were incubated at 30° for 3–4 days before photography. A strain containing the functional ura4+ gene at the wild-type chromosomal location (SP412) and another with the ura4 deletion (SP837) were used as controls. The strains used were as follows: (A) FY498, WT; SPG141, clr6-1; SPG142, clr1-5; SPG143, clr1-5 clr6-1; SPG158, clr3-735; and SPG159, clr3-735 clr6-1; (B) FY648, WT; SPG144, clr6-1; SPG145, clr1-5; SPG146, clr1-5 clr6-1; SPG160, clr3-735; and SPG161, clr3-735 clr6-1; (C) FY336, WT; SPG138, clr6-1; SPG139, clr1-5; SPG140, clr1-5 clr6-1; SPG162, clr3-735; and SPG163, clr3-735 clr6-1; and (D) FY520, WT; SPG147, clr6-1; SPG148, clr1-5; and SPG149, clr1-5 clr6-1. All strains containing ura4+ at cen1 or a telomere carried a mutation of the indigenous ura4. WT indicates the clr+ genotype.
|
|
We also measured the expression of cen1 ura4+ constructs in the clr1-5 clr6-1 and clr3-735 clr6-1 double-mutant backgrounds. As expected, this double-mutant combination resulted in additive derepression of ura4+ in the imr1R(NcoI)::ura4+ and otr1R(SphI)::ura4+ strains (Figure 7A and Figure B). Additionally, clr1- promoted derepression was reduced by the clr6 mutation at the cnt1/TM1(NcoI)::ura4+ insertion (Figure 7C). We also tested the clr6 mutation for its effect on telomere-mediated repression of ura4+ (Ch16m23-ura4+-TEL), but found that it had no visible effect alone or in combination with clr1-5 (Figure 7D).
In summary, these results indicate that there are distinct domains within cen1 and that clr6 differentially affects these domains, presumably by participating in their assembly into heterochromatin-like structures. Furthermore, the observed additive derepression in double mutants further suggests that Clr6 might perform an overlapping function with Clr1 plus Clr3 activity.
The clr6 mutation reduces the fidelity of chromosome segregation:
It was previously shown that mutations affecting centromeric silencing also disrupt chromosome segregation (ALLSHIRE et al. 1995 ). Therefore, it was formally possible that the clr6-1 allele could also impair centromere function. We tested the effect of the clr6-1 mutation on the mitotic stability of the minichromosome Ch16, a 530-kb derivative of chromosome III (NIWA et al. 1986 ), as described in MATERIALS AND METHODS. As shown in Table 3, strains bearing a lesion at clr6 missegregated the minichromosome at a 26-fold higher rate compared with their wild-type counterpart. This result further indicated that clr6 gene function might also be involved in the formation of fully functional centromeres.
It was recently demonstrated that mutations in clr4, rik1, and swi6 also cause elevated rates of chromosome loss, whereas clr1, clr2, and clr3 mutations have a negligible effect on chromosome segregation (ALLSHIRE et al. 1995 ). Because Clr6 acts in an overlapping manner with Clr1 and Clr3 (this study), we investigated chromosome loss in the clr1 clr6 and clr3 clr6 double mutants. The data presented in Table 3 suggest that the clr6 mutation further stimulates clr1- and clr3-promoted chromosome loss. Likewise, we also studied the effect of TSA treatment on the fidelity of chromosome segregation. Remarkably, transient exposure of minichromosome-containing cells to TSA resulted in elevated levels of missegregation in subsequent generations (Table 4). The effect of transient exposure to TSA clearly persists over many generations.
Clr3 is related to histone deacetylases:
The partially redundant function of the putative histone deacetylase Clr6 and proteins encoded by the clr1 and clr3 genes suggests that the latter might also encode components of the histone deacetylase machinery. In this regard, Clr1 was found to contain the DNA-binding zinc finger motifs (G. THON and A. KLAR, unpublished results), but Clr3 remains to be characterized. To define the mechanism of action of Clr3, we cloned the gene by transforming a h90 mat3-M::ura4+ clr3-735 strain (SP1359) with a partial Sau3A library. The transformants were screened for repression of mat3-M::ura4+ by selecting for their growth on FOA-containing medium. An 8.0-kb insert resulted in complementation of the ura4+ expression phenotype and of the sporulation defect caused by derepression of donor loci in the clr3-735 background. Subcloning and sequencing analysis identified an ORF that is capable of encoding a protein of 687 amino acids.
The encoded Clr3 protein sequence was compared with the sequences in the GenBank database. The best scores were obtained for a group of proteins that include known histone deacetylases, such as Hda1 (RUNDLETT et al. 1996 ) and Rpd3 (VIDAL and GABER 1991 ) from S. cerevisiae and HDAC1 (TAUNTON et al. 1996 ) from humans. Furthermore, we noticed that similarity between Clr3 and Hda1 (41% identity, 65% similarity; Figure 8) was greater than that between Clr3 and S. cerevisiae Rpd3 (27% identity, 49% similarity) or human HDAC1 (26% identity, 49% similarity). In addition to eukaryotic sequences, Clr3 also shares a high degree of similarity (41–52%) to acetylpolyamine hydrolase proteins from Synechocystis, Mycoplasma ramosa, and Methanococcus janaschii. Further comparison of Clr3 to other S. pombe sequences revealed that it shares only 26% (51% similarity) and 28% (52% similarity) identity to Clr6 and Hoc1 (see above) sequences, respectively, but was found to be homologous to the subsequently reported Hoc2 protein (see above) identified by Sanger Genome Sequencing Center (cosmid SP41410). Physical mapping of the clr3 gene confirmed that it is the same gene as hoc2, which maps to the short arm of chromosome II, which is located centromere distal to rad11. Cloning and sequencing analysis of the clr3-735 mutant allele identified a single G-to-A transition in codon 232, resulting in the D232N change (Figure 8).
View larger version (51K):
In this window
In a new window
Download PPT slide
|
Figure 8.
Alignment of the predicted Clr3 and S. cerevisiae Hda1 amino acid sequences. The sequences were aligned using the GAP program. Identical amino acids are marked by vertical lines, and similar amino acids are indicated by colons. The asterisk indicates the residue that is changed to asparagine [N] in the clr3-735 mutant allele.
|
|
These data suggest that the putative Clr3 protein belongs to the family of histone deacetylases. Unlike Clr6, which is more similar to S. cerevisiae Rpd3 and its human homolog, HDAC1, Clr3 shares greater similarity to S. cerevisiae Hda1. Therefore, it is possible that Clr6 and Clr3 proteins are fission yeast homologs of Rpd3 and Hda1, respectively.
| DISCUSSION |
|---|
In fission yeast, there are three known domains that exhibit position-effect control, namely the silent mating-type region, centromeres, and telomeres. It has been suggested that repression at these loci is mediated through assembly of a heterochromatin-like structure. Consistent with this model, we report the identification of an essential gene, clr6, encoding a putative histone deacetylase, which, when mutated, affects centromeric as well as donor loci silencing and reduces the fidelity of chromosome segregation. Furthermore, the clr6 mutation affects propagation of the epigenetic states regulating expression of a marker gene integrated in the silent mating-type region, recombination suppression, and directionality of mating-type switching.
Genetic evidence suggests that Clr6 function is partially redundant to the function of Clr1/Clr3. In this respect, we find that the putative Clr3 protein also shares similarity to histone deacetylases. Our earlier study indicated that clr1 and clr3 work together, either in the same pathway or as a complex (THON et al. 1994 ). Taken together, these data suggest that silencing at above-mentioned loci might be regulated by two distinct yet functionally overlapping histone deacetylation mechanisms: Clr1 and Clr3 might be components of one such system, whereas Clr6 is a component of a second system. Likewise, the budding yeast proteins Hda1 and Rpd3, which share high similarity to Clr3 and Clr6, respectively, have been found to be associated with the distinct histone deacetylase complexes HDA and HDB, respectively (RUNDLETT et al. 1996 ).
On the basis of recent studies (BRAUNSTEIN et al. 1993 ; BROWNELL and ALLIS 1996 ; WOLFFE 1996 ; HARTZOG and WINSTON 1997 ; PAZIN and KADONAGA 1997 ), it has been suggested that histone deacetylation leads to assembly of "compact" chromatin, which prevents access to transcription factors, whereas acetylation makes it "open," allowing access to the transcription machinery. In contradiction to this model, mutations in RPD3 and related genes were recently shown to enhance silencing in the heterochromatic regions of both budding yeast and Drosophila (DERUBERTIS et al. 1996 ; RUNDLETT et al. 1996 ; VANNIER et al. 1996 ). Similarly, mutation in clr6 causes a significant (20-fold) increase in Ura+ to Ura- transition of K::ura4+ epialleles and silencing at the central domain (TM1) of cen1. However, the effects of clr6-1 are not uniform across all sites tested, and the marker gene integrated at either of two outer flanking repeats (imr1 and otr1) of cen1 was weakly derepressed (this study). In contrast to increased repression and consistent with the proposed role of histone deacetylases, a mutation in clr3 adversely affects silencing both at the silent mating-type region (THON et al. 1994 ; GREWAL and KLAR 1997 ) and at centromeres (ALLSHIRE et al. 1995 ). Also, functional clr3+ is required for propagation of the repressed (Ura-) epigenetic state (GREWAL and KLAR 1996 ). To accommodate these seemingly contradictory results, we propose a hypothetical "competition model." We suggest that, as in budding yeast, where the subtelomeric region and silent mating-type loci compete for silencing proteins (BUCKS and SHORE 1995 ; MAILLET et al. 1996 ), different heterochomatin domains in fission yeast might compete for a defined pool of heterochromatic proteins (HPs) such as Swi6. Furthermore, the affinity of a particular chromosomal domain for HPs may depend on its state of histone acetylation. If, for example, HPs preferentially bind to deacetylated histone H3 and H4, regional differences in acetylation state could ensure their targeting to specific chromosomal domains. In turn, differences in acetylation pattern could depend on region-specific preferential recruitment of different combinations of Clr3, Clr6, Hoc1, or other deacetylating proteins. In this way, mutation in a particular deacetylase might result in relocalization of HPs from one chromosomal domain to another. For example, mutation in clr6 might delocalize HPs from the imr and otr regions of centromeres. The free HPs could then be recruited to the mating-type and central centromeric regions, leading to contrasting effects on corresponding loci. Consistent with this competition model, we see that the increase in repression at the mat2-mat3 region observed in clr6 mutants is dependent on the presence of functional Clr1 and Clr3 (this study). Also, the clr6-1-mediated increased silencing is sensitive to TSA treatment (our unpublished data), indicating the probable involvement of other deacetylases.
As an alternative to the competition model, we imagine that the clr6 mutation might affect silencing indirectly by altering the regulation of trans-acting genes, including clr1-clr4, swi6, and rik1. It has been suggested that enhanced repression in mutant cells might be caused by a specific pattern of histone acetylation (such as increased acetylation of H4 lysine 12; RUNDLETT et al. 1996 ) or by the assembly of a specialized chromatin structure (DERUBERTIS et al. 1996 ). Supporting the latter argument, it was previously observed that the central centromeric domain in fission yeast is packaged into an unusual chromatin structure (POLIZZI and CLARKE 1991 ; TAKAHASHI et al. 1992 ). These three models are not mutually exclusive, and further work is needed to differentiate among them.
During development and differentiation, propagation of the stable states of gene expression through numerous cell divisions is essential (FELSENFELD 1992 ). It is possible that the histone acetylation pattern of chromosomal domains acts as a "cell memory" and helps to maintain the respective states of gene expression. Consistent with this idea, we observed that transient inhibition of histone deacetylation by TSA causes a remarkable heritable change in the expression of a marker gene integrated in the silent mating-type region, converting the epigenetically repressed state to the expressed state in the majority of cells. The expressed state is inherited in cis as a marker linked to the mating-type locus and, as observed previously (GREWAL and KLAR 1996 ), correlates with reduced efficiency of mating-type interconversion. In our chromatin-replication model, we previously suggested that silencing, efficiency of mating-type switching, and recombination suppression in the mat2-mat3 interval are regulated by a heterochromatin structure that, once assembled, is capable of self-replicating for many generations through cooperative binding of HPs (GREWAL and KLAR 1996 , GREWAL and KLAR 1997 ). It is possible that inhibition of histone deacetylation compromises the assembly of the proposed chromatin structure. It remains to be determined whether the change in imprint is caused by direct alteration of the acetylation state at the silent mating-type region or by an indirect transitory change in the expression of trans-acting factors required for silencing, which, in turn, cause a change in the imprint. There is also a possibility that TSA treatment affects the timing of replication and/or nuclear localization of the mat region, causing changes in the structural and functional imprint. Regardless of whether the TSA effect is direct or indirect, the final effect is shown to cause a heritable change in the imprint at the mat2-mat3 region. Interestingly, EKWALL et al. 1997 recently showed that transient TSA treatment induces a heritable hyperacetylated state in centromeric chromatin, which correlates with functionally defective centromeres, suggesting a link between histone acetylation and the assembly of functional centromeres.
Unlike histone deacetylase from S. cerevisiae (VIDAL and GABER 1991 ), S. pombe clr6 is an essential gene. We believe that lethality is at least partly related to a reduced fidelity of chromosome segregation in mutant strains. In addition, the possible disruption of chromatin structure throughout the genome in the clr6 mutant may result in deleterious deregulation of gene expression. We note that clr6-1 mutant is UV sensitive, a phenotype common in mutants of genes controlling the assembly of chromatin structure (KAUFMAN et al. 1997 ) and the DNA repair process (REYNOLDS et al. 1990 ). In conclusion, this study implicates putative histone deacetylases in the assembly of heterochromatin-like structures at the silent mating-type region and at centromeres. Supporting our chromatin replication model, we demonstrate that inhibition of histone deacetylase activity can compromise the propagation of the stable chromosomal imprint at the mat locus.
| ACKNOWLEDGMENTS |
|---|
We thank M. Yanagida and R. Allshire for providing strains and D. Beach and P. Young for the S. pombe library. We also thank J. Sabl, N. Trun, and A. Arthur for carefully reading the manuscript. This work was sponsored by the National Cancer Institute Department of Health and Human Services (DHHS) under contract with ABL. The contents of this publication do not necessarily reflect the views or policies of the DHHS, nor does mention of trade names, commercial products, or organizations imply endorsement by the U.S. Government.
Manuscript received February 24, 1998; Accepted for publication June 10, 1998.
| LITERATURE CITED |
|---|
AINSCOUGH, J. F., and A. SURANI, 1996 Organization and control of imprinted genes: the common features, pp. 173–194 in Epigenetic Mechanisms of Gene Regulation, edited by V. E. A. RUSSO, R. A. MARTIENSSEN and A. D. RIGGS. Cold Spring Harbor Laboratory Press, Cold Spring Harbor, NY.
ALLSHIRE, R. C., 1996 Transcriptional silencing in the fission yeast: a manifestation of higher order chromosome structure and functions, pp. 443–466 in Epigenetic Mechanisms of Gene Regulation, edited by V. E. A. RUSSO, R. A. MARTIENSSEN and A. D. RIGGS. Cold Spring Harbor Laboratory Press, Cold Spring Harbor, NY.
ALLSHIRE, R. C., E. R. NIMMO, K. EKWALL, J.-P. JAVERZAT, and G. CRANSTON, 1995 Mutations derepressing silent centromeric domains in fission yeast disrupt chromosome segregation. Genes Dev 9:218-233[Abstract/Free Full Text].
ARCANGIOLI, B. and A. J. S. KLAR, 1991 A novel switch-activating site (SASI) and its cognate binding factor (SAP1) required for efficient mat1 switching in Schizosaccharomyces pombe. EMBO J. 10:3025-3032[Medline].
BARLOW, D. P., 1995 Gametic imprinting in mammals. Science 270:1610-1613[Abstract/Free Full Text].
BRAUNSTEIN, M., A. B. ROSE, S. G. HOLMES, C. D. ALLIS, and J. R. BROACH, 1993 Transcriptional silencing in yeast is associated with reduced nucleosome acetylation. Genes Dev. 7:592-604[Abstract/Free Full Text].
BRESCH, C., G. MULLER, and R. EGEL, 1968 Genes involved in meiosis and sporulation of a yeast. Mol. Gen. Genet. 102:301-306[Medline].
BROWNELL, J. E. and C. D. ALLIS, 1996 Special HATs for special occasions; linking histone acetylation to chromatin assembly and gene activation. Curr. Opin. Genet. Dev. 6:176-184[Medline].
BUCKS, S. W. and D. SHORE, 1995 Action of a RAP1 carboxy-terminal silencing domain reveals an underlying competition between HMR and telomeres in yeast. Genes Dev. 9:370-384[Abstract/Free Full Text].
CLARKE, L., M. BAUM, L. G. MARSCHALL, V. K. NGAN, and N. C. STEINER, 1993 Structure and function of Schizosaccharomyces pombe centromeres. Cold Spring Harbor Symp. Quant. Biol. 58:687-695[Medline].
DERUBERTIS, 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].
EGEL, R., 1994 Regulation of meiosis and sporulation in Schizosaccharomyces pombe, pp. 251–265 in The Mycota I: Growth, Differentiation and Sexuality, edited by F. WESSELS and J. G. H. MEINHARDT. Springer-Verlag, Berlin.
EGEL, R., M. WILLER, and O. NEILSEN, 1989 Unblocking of meiotic crossing-over between the silent mating-type cassettes of fission yeast, conditioned by the recessive, pleioropic mutant rik1. Curr. Genet. 15:407-410.
EKWALL, K. and T. RUUSALA, 1994 Mutationis in rik1, clr2, clr3, and clr4 genes asymmetrically derepress the silent mating-type loci in fission yeast. Genetics 136:53-64[Abstract].
EKWALL, K., J.-P. JAVERZAT, A. LORENTZ, H. SCHMIDT, and G. CRANSTON et al., 1995 The chromodomain protein Swi6: a key component at fission yeast centromeres. Science 269:1429-1431[Abstract/Free Full Text].
EKWALL, K., T. OLSSON, B. M. TURNER, G. CRANSTON, and R. ALLSHIRE, 1997 Transient inhibition of histone deacetylation alters the structural and functional imprint at fission yeast centreomeres. Cell 91:1021-1032[Medline].
ENGELKE, U., L. GRABOWSKI, H. GUTZ, L. HEIM, and H. SCHMIDT, 1987 Molecular characterization of h- mutants of Schizosaccharomyces pombe. Curr. Genet. 12:535-542.
FELSENFELD, G., 1992 Chromatin as an essential part of the transcriptional mechanism. Nature 355:219-224[Medline].
GOTTSCHLING, D. E., O. M. APARICIO, B. L. BILLINGTON, and V. A. ZAKIAN, 1990 Position effect at S. cerevisiae telomeres: reversable repression of Pol II transcription. Cell 63:751-762[Medline].
GREWAL, S. I. S. and A. J. S. KLAR, 1996 Chromosomal inheritance of epigenetic states in fission yeast during mitosis and meiosis. Cell 86:95-101[Medline].
GREWAL, S. I. S. and A. J. S. KLAR, 1997 A recombinationally repressed region between mat2 and mat3 loci shares homology to centromeric repeats and regulates directionality of mating-type switching in fission yeast. Genetics 146:1221-1238[Abstract].
HARTZOG, G. A. and F. WINSTON, 1997 Nucleosomes and transcription: recent lessions from genetics. Curr. Opin. Genet. Dev. 7:192-198[Medline].
HOHEISEL, J. D., E. MAIER, R. MOTT, L. MCCARTHY, and A. V. GRIGORIEV et al., 1993 High resolution cosmid and P1 maps spanning the 14 Mb genome of the fission yeast S. pombe. Cell 73:109-120[Medline].
IVANOVA, A. V., M. J. DONADUCE, S. V. IVANOV, and A. J. S. KLAR, 1998 The chromo and SET domains of the Clr4 protein are essential for silencing in fission yeast. Nat. Genet. 19:192-195[Medline].
KAUFMAN, P. D., R. KOBAYASHI, and B. STILLMAN, 1997 Ultraviolet radiation sensitivity and reduction of telomeric silencing in Saccharomyces cerevisiae cells lacking chromatin assembly factor-I. Genes Dev. 11:345-357[Abstract/Free Full Text].
KLAR, A. J. S., 1989 The interconversion of yeast mating type: Saccharomyces cerevisiae and Schizosaccharomyces pombe, pp. 671–691 in Mobile DNA, edited by D. E. BERG and M. W. HOWE. American Society for Microbiology, Washington, DC.
KLAR, A. J. S. and M. J. BONADUCE, 1991 swi6, a gene required for mating-type switching, prohibits meioic recombination in the mat2-mat3 "cold spot" of fission yeast. Genetics 129:1033-1042[Abstract].
KLAR, A. J. S., A. V. IVANOVA, J. Z. DALGAARD, M. J. BONADUCE and S. I. S. GREWAL, 1998 Multiple epigenetic events regulate mating-type switching of fusion yeast, pp. 87–103 in Epigenetics, edited by G. CARDEW and D. J. CHADWICK. Wiley, Chichester.
LORENTZ, A., L. HEIM, and H. SCHMIDT, 1992 The switching gene swi6 affects recombination and gene expression in the mating-type region of Schizosaccharomyces pombe. Mol. Gen. Genet. 233:436-442[Medline].
LORENTZ, A., K. OSTERMANN, O. FLECK, and H.
TOP
ABSTRACT
MATERIALS AND METHODS
