* Your assessment is very important for improving the workof artificial intelligence, which forms the content of this project
Download Lin, R., C. D. Allis and S. J. Elledge. 1996. PAT1
DNA vaccination wikipedia , lookup
Epigenetics wikipedia , lookup
Epigenetics of diabetes Type 2 wikipedia , lookup
Oncogenomics wikipedia , lookup
Designer baby wikipedia , lookup
Epigenomics wikipedia , lookup
History of genetic engineering wikipedia , lookup
Microevolution wikipedia , lookup
Therapeutic gene modulation wikipedia , lookup
Histone acetyltransferase wikipedia , lookup
Cancer epigenetics wikipedia , lookup
Epigenetics of neurodegenerative diseases wikipedia , lookup
Nutriepigenomics wikipedia , lookup
Epigenetics of human development wikipedia , lookup
Gene therapy of the human retina wikipedia , lookup
No-SCAR (Scarless Cas9 Assisted Recombineering) Genome Editing wikipedia , lookup
Mir-92 microRNA precursor family wikipedia , lookup
Site-specific recombinase technology wikipedia , lookup
Epigenetics in learning and memory wikipedia , lookup
Epigenetics in stem-cell differentiation wikipedia , lookup
Vectors in gene therapy wikipedia , lookup
Point mutation wikipedia , lookup
Artificial gene synthesis wikipedia , lookup
PAT1, an evolutionarily conserved acetyltransferase homologue, is required for multiple steps in the cell cycle Rueyling Lina, C. David Allisb and Stephen J. Elledge* Verna and Marrs McLean Department of Biochemistry, Howard Hughes Medical Institute, Baylor College of Medicine, Houston, Texas 77030, USA Communicated by : Virginia Zakian Abstract Background: Acetylation has been implicated in many biological processes. Mutations in N-terminal acetyltransferases have been shown to cause a variety of phenotypes in Saccharomyces cerevisiae including activation of heterochromatin, inability to enter G0, and lethality. Histone acetylation has been shown to play a role in transcription regulation, histone deposition and histone displacement during spermatogenesis, although no known histone acetyltransferase is essential. Results: Studies aimed at revealing a role for histone H1 in yeast have uncovered a mutation in a putative acetyltransferase, PAT1. The mutant ( pat1-1) cells can live only in the presence of vertebrate H1. PAT1 is essential for mitotic growth in S. cerevisiae; mutant cells depleted of the Pat1p show aberrant cellular Introduction The primary structural unit of chromatin in eukaryotes is the nucleosome, which consists of two copies of each core histone and 146 bp of DNA (McGhee & Felsenfeld 1980). Higher eukaryotes also contain an additional histone, H1, thought to be involved in higher order chromatin structure and possibly transcriptional control (Allan et al. 1981; Zlatanova 1990). All core histones can be divided into two functional domains: a hydrophobic carboxy-terminal domain and * Corresponding author : Fax: þ1 713 798 8515. Present addresses: aFred Hutchinson Cancer Research Center, 1100 Fairview N., Seattle, WA 98109, USA; bDepart-ment of Biology, University of Rochester, Rochester, NY 14627, USA q Blackwell Science Limited and nuclear morphology. PAT1 is required for multiple cell cycle events, including passage through START, DNA synthesis, and proper mitosis through a microtubule-mediated process. The S. pombe PAT1 gene was cloned by complementation and shown to exist as part of a larger protein, the unique portion of which is homologous to a second S. cerevisiae gene. pat1 mutants show a variety of mitotic defects including enhanced chromosome loss, accumulation of multiple nuclei, generation of giant cells, and displays classical cut phenotypes in which cytokinesis occurs in the absence of proper nuclear division and segregation. Conclusion: PAT1 controls multiple processes in cell cycle progression which suggests an essential role for the acetylation of yet unknown substrate(s). a hydrophilic amino-terminal tail (Smith 1991; Wolffe & Pruss 1996a). The hydrophobic domains of core histones are required for nucleosome assembly and cell viability. Deletions of the N-terminal domain of any of the core histones are viable (Wallis et al. 1983; Schuster et al. 1986; Kayne et al. 1988). However, deletion of the N-terminal domain of histone H4 de-represses the silent mating site locus in yeast, suggesting that this domain might play a role in the formation of heterochromatin in vivo (Kayne et al. 1988). The N-terminal domain of histone H4 contains several highly conserved lysines that are differentially subjected to reversible acetylation. Acetylation neutralizes the positive charges on lysines and is thought to regulate its interaction with negatively charged DNA (Allfrey 1977). Mutational analysis suggests that the ability to Genes to Cells (1996) 1, 923–942 923 R Lin et al. 924 Genes to Cells (1996) 1, 923–942 q Blackwell Science Limited An essential yeast acetyltransferase homologue modulate the positive charges on H4 amino terminus by acetylation is more important than the positive charge at these highly conserved positions (Megee et al. 1990; Park & Szostak 1990; Durrin et al. 1991). Acetylation is an energy intensive, dynamic phenomenon whose steady-state balance is mediated by the opposing activities of histone acetyltransferase and deacetylase enzyme systems. Recently, the non-essential yeast gene GCN5 (Georgakopoulos & Thireos 1992), was shown to be an A-type histone acetyltransferase, demonstrating a strong link between histone acetylation and regulated transcription (Brownell et al. 1996; Wolffe & Pruss 1996b), and a mammalian histone deacetylase has been related to the yeast transcriptional regulator Rpd3p (Taunton et al. 1996). Taken together, these data indicate that the steady-state balance of histone acetylation plays a direct role in the modulation of chromatin structure to create new patterns of transcription (Brownell & Allis 1996). Histone H1, unlike core histones, is highly diverged in sequence among species. However, most histone H1s have a high percentage of lysines and share a distinctive secondary structure: a central globular domain that is flanked by a short N-terminal tail and a long, highly charged C-terminal tail (Allan et al. 1980). Although histone H1 has been implicated in many biological processes, including chromatin compaction and gene regulation (Allan et al. 1981; Zlatanova 1990), its role in vivo is still unclear. This is in part due to the fact that H1 has not been identified in genetically tractable eukaryotes such as S. cerevisiae, although a number of conventional approaches have been taken (Certa et al. 1984; Srebreva et al. 1987). Electron microscopic analyses have indicated that chromatin in yeast cells can fold into a higher-order structure that appears similar to the 30-nm filaments of higher eukaryotic cells, suggesting that histone H1 or H1-like protein exists in yeast (Allan et al. 1984; Lowary & Widom 1989). In this study, we describe an attempt to identify a yeast histone H1 homologue by screening for yeast mutants whose growth is dependent on the expression of a vertebrate H1. Consistent with the results of Linder & Thoma (1994), we found that the low level expression of an exogenous histone H1 in yeast does not cause any obvious phenotype. A mutant that appears to be dependent on H1 expression was isolated by the colony-colour-sectoring assay. The wild-type gene responsible for this phenotype was cloned and characterized. It encodes a small novel protein whose function is essential for mitotic growth in Saccharomyces cerevisiae and that shares significant homology to acetyltransferases. Genetic analyses have indicated that the PAT1 gene is involved in multiple steps in the cell cycle: passage through START after an a-factor block, the progression of DNA synthesis, and mitosis through a microtubule-mediated process. Results Isolation of histone H1-dependent mutants The chicken H1 11L (chH1) genomic sequence was amplified by PCR and cloned into a centromeric expression vector pMW29. The resulting plasmid, pMW29H1, contains a GAL1-chH1 fusion. pMW29H1 was transformed into YMW1-6 and its effect on cell growth was analysed. This strain has doubling times of 81 min in SC glucose and 124 min in SC galactose media at 30 8C, which are identical to the doubling times of YMW1-6 under the same conditions. Expression of chH1 was confirmed by both mRNA Figure 1 (A) Deletion analysis defines the PAT1 genomic locus. A deletion series was generated from a complementing genomic clone and tested for their ability to complement the pat1-1 mutation; plus signs (þ) indicate clones that complement and minus signs (–) indicate a failure to complement. The complementation activity was narrowed to a 1.5 kb SacII fragment whose sequence is shown in (B). The closed arrowhead points to the first ATG and the open arrow indicates the termination codon. pat1-∆2::HIS3 is a disrupted allele in which NcoI-BalI fragment is replaced by a HIS3 cassette indicated as a hatched box. (B) The nucleotide sequence of the 1.5 kb SacII fragment and predicted amino acid sequence of Pat1p. The numbering for both nucleic acid (plain) and amino acid (bold) are shown to the left of these sequences. The region between amino acid 97 and 148 which is referred to as the ‘acetyltransferase domain’ in (C) is shaded. (C) Comparison of the Pat1p sequence with other acetyltransferases and related sequences, including Staphylococcus aureus atl gene product (STAPH, GENBANK accession number D17366), Escherichia coli streptothricin acetyltransferase (Sat3, g669115), Pseudomonas aeruginosa 6 0 -N-acetyltransferase (AAC, M29695), Mouse spermidine/spermine N1-acetyltransferase (SSAT, L10244), Saccharomyces cerevisiae L-A virus protein N-acetyltransferase (MAK3, Q03503), Saccharomyces cerevisiae hypothetical protein (SC8554.04, Z46796), Saccharomyces cerevisiae protein involved in silencing HMR (Sas2p, U14548), Saccharomyces cerevisiae hypothetical protein (YBL052C, Z35814), Saccharomyces cerevisiae N-terminal acetyltransferase (Ard1p, M11621), Saccharomyces cerevisiae histone acetyltransferase (Hat1p, U33335), and Saccharomyces cerevisiae transcriptional activator (Gcn5p, X68628). Gaps are allowed to generate the best alignment. Identical residues (to Pat1p) are indicated in black whereas conserved residues are shaded in grey. q Blackwell Science Limited Genes to Cells (1996) 1, 923–942 925 R Lin et al. analysis and by the construction of a translational fusion between chH1 and lacZ (data not shown). A genetic screen was performed to identify mutants whose growth is dependent on the expression of chH1. These mutants are expected to be defective either in the yeast histone H1 homologue or other genes that are suppressible by chicken H1 protein. Cells with lethal mutations in the desired gene are unable to grow in the absence of the plasmid carrying chH1 and can be identified by screening for mutants that are unable to lose pMW29H1 in the absence of nutritional selection. The screen is based on the colony-colour-sectoring assay described previously (Kranz & Holm 1990; Zieler et al. 1995) with which the plasmid loss can be easily visualized by a colour change. This assay took advantage of the fact that ade2 yeast cells accumulate red pigments but ade2 ade3 cells do not. When an ade2 ade3 strain carries ADE3 on an autonomously replicating plasmid, it forms red colonies. White sectors form in red colonies when cells lose plasmids during colony growth (referred to as Sectþ phenotype). Three main criteria were anticipated for H1-dependent mutants; failure to survive plasmid loss resulting in homogeneously red, non-sectoring (Sect – ) colonies; inability to grow on plates containing 5-fluoro orotic acid (5-FOA), and failure to grow when chH1 expression is repressed on glucose media. To isolate H1-dependent mutants, YMW1-6 containing pMW29H1 was mutagenized and plated on YPG (galactose) plates. Forty Sect – mutants were isolated from 30 000 examined colonies. These mutants were tested for their ability to grow on SC-uracil media and sensitivity to 5-FOA to ensure that the Sect – phenotype was due to the retention of pMW29H1 rather than reasons unrelated to the plasmid. Yeast cells that contain URA3 are sensitive to 5-FOA. Among the 40 Sect – mutants, 11 were able to grow on SC-uracil plates but not 5-FOA plates. These 11 mutants were further tested for their dependency on galactose, which is indicative of the dependency on the chH1 expression. The GAL1 promoter is repressed in the presence of glucose but activated in the absence of glucose and in the presence of galactose. Only one mutant, YRL4, showed lethality on YPD plates. The other 10 Sect – mutants that showed 5-FOA sensitivity but not galactose dependency were similar to the background mutants observed in Kranz & Holm (1990). The reason most of these mutants maintain pMW29H1 was a requirement for either ADE3 or URA3 genes. Using a plasmid shuffle protocol (Rose & Fink 1987), we were able to show that these mutants can lose pMW29H1 when a 926 Genes to Cells (1996) 1, 923–942 wild-type copy of URA3 or ADE3 on a second plasmid was introduced (data not shown). They are likely to be mutations involved in nucleoside metabolic pathways. YRL4 was the only mutant that met all three criteria: Sect – , 5-FOA sensitivity, and galactose-dependency. The galactose dependency suggests that the mutation in YRL4 is suppressed by the expression of H1 and not another gene on the plasmid. However, the phenotype of YRL4 appears to be unstable; old colonies sometimes sector and grow on YEPD plates. Great efforts have been made to address the dependency of the expression of chH1 in YRL4. This includes plasmid shuffling using a second plasmid containing either wild-type or disrupted chH1 gene. Nevertheless, due to the instability of this mutation, we were unable to conclusively prove that it was the chH1 gene alone that was responsible for suppression of this mutation. We detected a fivefold increased level of chH1 message in YRL4 compared to the level of chH1 message in wildtype controls (data not shown). This may indicate that the mutant required higher levels of H1 expression than originally expressed from this plasmid, which might be an important factor complicating the dependency analysis. Isolation and characterization of the PAT1 gene The gene complementing the glucose lethality of YRL4 was isolated from genomic libraries as described in Experimental procedures. One positive clone was obtained from the centromeric-based genomic library and nine were obtained from a YEP13-based genomic library. All clones also complement both the Sect – and 5-FOA sensitive phenotypes of the mutation. Restriction mapping and Southern blotting analyses indicate that a 5 kb region is shared by all 10 clones (Fig. 1A). Further analyses revealed that the complementing activity resides within a 0.6 kb XhoI-XbaI fragment (Fig. 1A). The DNA sequence of a 1.5 kb SacII fragment (Fig. 1B) was determined and it overlaps with a sequence reported from a lambda clone from chromosome VI (accession no. Z46255). The longest open reading frame (ORF) in the 1.5 kb SacII fragment is 480 bp, starting 47 bp downstream of the XhoI site and ending 62 bp upstream of the XbaI site. When Northern blot analysis was performed using the 0.6 kb XhoI-XbaI fragment as a probe, a single band of about 700 nucleotides was detected in both wild-type and YRL4 RNA (data not shown). The predicted amino q Blackwell Science Limited An essential yeast acetyltransferase homologue Figure 2 (A) Nucleic acid and deduced amino acid sequences of the spPAT1 cDNA. The numbering for both nucleic acid (plain) and amino acid (bold) are shown to the left of these sequences. (B). The alignment of amino acid sequences among Pat1p, spPat1P and YGL021. Identical residues are shaded and conserved residues are boxed. acid sequence from this ORF shows no significant homology with histone H1. However, a domain of this protein (I97 to G152) shares significant homology with many previously identified acetyltransferases (Fig. 1C) and this gene was named PAT1 (for Putative Acetyl Transferase). The original pat1-1 allele and the parental PAT1 gene were isolated by gap repair, sequenced, and pat1-1 was found to have a D99N mutation in the homology domain indicating the relevance of this domain to Pat1p function. q Blackwell Science Limited To help establish the position of the translational initiation codon and the intron/exon structure of the PAT1 gene, we isolated cDNAs. To accomplish this, a yeast cDNA library was prepared in lYES-R (Elledge et al. 1991). The RNA source of cDNA came from the strain Y133 (MATa, ura3-52, lys2-801, ade2-101, his3-∆200, Trp1∆). We obtained 1 × 108 independent lambda clones. The 0.6 kb XhoI-XbaI fragment was used to screen this library and 10 independent cDNA clones were obtained that coincide with this ORF. Two Genes to Cells (1996) 1, 923–942 927 R Lin et al. Figure 3 Aberrant morphology of Pat1p depleted cells. Abnormal cellular (phase contrast, left-hand column) and nuclear (DAPI stain, right-hand column) morphologies becomes apparent in YRL18 (pat1-∆2::HIS3/pRL46) when shifted from YPG (a,b) to YPD (c–j) media for 24 h. Small arrows indicate aberrant nuclear segregation. Large arrows point to anucleated, large-budded cells. Bar ¼ 20 mm. of these were completely sequenced. The longer cDNA contained sequences from –13 to þ548 according to the genomic sequences in Fig. 1B where +1 is the first ATG. The shorter cDNA contained sequences from þ9 to þ551. The PAT1 gene contains no introns and most likely utilizes the first ATG in the ORF as 928 Genes to Cells (1996) 1, 923–942 the initiation codon. This would generate a protein of 159 amino acids with predicted molecular weight of 19 203 Da. To address the significance of this acetyltransferase domain and possible phylogenetic conservation of this gene we sought to isolate the S. pombe homologue by q Blackwell Science Limited An essential yeast acetyltransferase homologue complementation. An S. pombe cDNA library (Becker et al. 1991) was used to complement a pat1 null mutant (see Experimental procedures). Of 180 cDNA clones capable of suppressing pat1, all contained the same gene. The spPAT1 conceptual translation product is 356 amino acids (Fig. 2A) and the last 80 amino acids share 50% identity with the C-terminal half of the Pat1 protein (Fig. 2B). The homology extends beyond the putative acetyltransferase homology domain. Interestingly, the N-terminal third of the spPat1p is homologous to another S. cerevisiae gene YGL021 (Fig. 2B), a putative membrane protein on chromosome VII whose disruption shows no phenotype (Chen et al. 1991). No apparent membrane spanning domain is predicted from the PAT1 protein sequence. Taken together, the sequence analysis suggests that PAT1 is not the yeast homologue of histone H1. Instead, the homology with other acetyltransferases, the conservation in the putative acetyltransferase domain with its S. pombe functional homologue, and the missense mutation detected in the original pat1-1 mutation all suggest that Pat1p is likely to be an acetyltransferase. PAT1 is essential for mitotic growth A pat1 disruption allele (pat1-∆2::HIS3) was generated by one-step gene replacement in a diploid, YRL9, and used to examine whether PAT1 is required for mitotic growth. While the wild-type parental diploid gave rise to tetrads with four viable spores, YRL9 gave tetrads with only two viable spores which were always His – . The inviable spores germinated and each gave rise to an average of eight to 16 progeny cells. These data indicates that PAT1 is essential for mitotic growth. When PAT1, but not a disrupted pat1, on a LEU2 centromeric plasmid was introduced into YRL9, the resulting strain gave rise to tetrads that often contained greater than two viable spores. Every Hisþ spore was also Leuþ , indicating that PAT1 is required for the viability of spores containing the pat1-D2::HIS3 allele. The inviability of a pat1 deletion strain can also be rescued by the PAT1 cDNA under control of the GAL1 promoter. This was examined with both the longer (–13 to +548, pRL45) and shorter (+9 to +551, pRL46) cDNAs described above. The longer cDNA produces a protein of 159 amino acids, whereas the shorter cDNA produces a 148 amino acid protein lacking the first 11 amino acids. While both cDNAs rescue the spore inviability associated with the pat1 disruption allele on galactose media, they behave differently on glucose media. When streaked on YPD plates, the pat1 deletion strain that contains pRL45 q Blackwell Science Limited (YRL17) could grow, while cells containing pRL46 (YRL18) could not. Depletion of Pat1p results in multiple phenotypes To address the function of PAT1, we examined the defects in YRL18 cells when Pat1p is gradually depleted by turning off the expression of the truncated form of PAT1 cDNA in the presence of glucose. Yeast cells were grown in YPG medium to the early logarithmic phase (OD600 ¼ 0.1), washed once with water, and diluted 20-fold in YPD medium. At different intervals after transfer to YPD medium, cells were fixed and stained with the DNA specific dye DAPI. The depletion of Pat1p led to multiple phenotypes which became apparent 20 h after cells were transferred to YPD media (Fig. 3). Cell sizes varied dramatically, with a high percentage (30%) of giant cells whose volumes enlarged by < 100–200-fold (Fig. 3c–f). The majority of cells were still attached to each other, generating long chains and multiply budded cells. DAPI staining detected defects in nuclear segregation which gave rise to cells with either multiple nuclei (Fig. 3d,f,i) or lacking nuclei (Fig. 3j). These enlarged cells and multi-nucleated phenotypes have been observed in mutations which affect bud emergence (Bender & Pringle 1991). It was surprising to note that cytokinesis seemed to be proceeding in anucleated cells; anucleated cells with large buds were observed (large arrows in Fig. 3h,j), although we cannot rule out the possibility that these cells once had DNA, which was degraded prior to our analysis. In addition, it appears that cytokinesis occurred prior to the completion of nuclear migration and segregation in some cells, causing uneven nuclear division (see small arrows in Fig. 3e,f,g,i). However, in many cases observed here, cells seemed unable to complete cytokinesis when the nuclei were caught in the neck portion of two cells. Therefore, many cells were still connected, forming long chains. Whether this is an arrested point or a part of a dynamic process is still unclear. But the results obtained from a temperature sensitive mutation described later strongly suggest that this is an arrest point. The original pat1-1 mutant also exhibits the phenotype described here when it is transferred to glucose media, but less severely in galactose media. These results suggest that Pat1p may play a role in mitosis, concerning chromosome segregation and/or nuclear division. The phenotypic heterogeneity in Pat1p-depleted cells indicates that Pat1p is either involved in multiple processes or a process that is essential throughout the cell cycle. Genes to Cells (1996) 1, 923–942 929 R Lin et al. Figure 4 Terminal phenotypes of the pat1-3 mutation (pat1-3/ pat1-∆2::HIS3, YRL34) at either the permissive (23 8C, a,b) or nonpermissive temperatures (37 8C, c–h). Phase contrast micrographs are shown in the left-hand column and the DAPI staining are shown in the right-hand column. Multi-nucleated cells are indicated by open arrowheads and anucleated cells are pointed by close arrowheads. Small arrows point to the ‘cut-like’ nuclear division and large arrows indicate cells with diffused or fragmented nuclei. Bar ¼ 20 mm. Isolation of a temperature sensitive allele of the PAT1 gene Phenotypes resulting from depletion experiments, although informative, lack the resolution that can be obtained by rapidly inactivating the activity of a protein using conditional mutants. Therefore, efforts were made to isolate Ts – PAT1 alleles by in vitro mutagenizing a plasmid containing a wild-type PAT1 gene followed by the plasmid shuffle method, as described in Rose & Fink (1987). One Ts – allele (pat1-3) was isolated from 20 000 clones screened. The plasmid pRL71, which contains pat1-3, can confer the Ts – phenotype on cells whenever it is reintroduced into a pat1-∆2::HIS3 mutant strain. The pat1-3 mutant strain (YRL34) had slightly longer generation times (132 min) than wild-type (CRY1, 90 min) or phenotypically 930 Genes to Cells (1996) 1, 923–942 wild-type (YRL37, 111 min) control strains at 23 8C, suggesting that pat1-3 mutants are partially defective, even at permissive temperatures. The growth of the pat1-3 strain upon temperature shift to 37 8C was examined and found to arrest growth prior to one complete cell doubling (data not shown). The survival of the wild-type and mutant cells at 37 8C as a function of time was also examined. Wild-type cells tolerated the non-permissive temperature as expected, but cells expressing the pat1-3 allele lost viability at 37 8C with a half-life of < 3 h (data not shown). The terminal phenotypes of the pat1-3 mutants at 37 8C for 3.5 h were also examined (Fig. 4) and found to show great similarity to those obtained in the depletion experiment, but also revealed several unique phenotypes. For instance, the abnormality of chromatin structure and nuclear segregation were apparent in both cases, whereas giant cells and anucleated cells with large buds were not observed in the pat1-3 mutant. At 37 8C, pat1-3 cells were not arrested at a defined cell cycle stage, although an increase of large-budded cells (G2-M phase, 42%) and a decrease of small-budded cells (S phase, 17%) were observed compared to wildtype controls (23% and 25%, respectively) grown at 37 8C. Many cells were still attached to each other, forming groups of three, four, or even more cells, most of which had abnormal nuclei that were either fragmented or had a threaded appearance (large arrows in Fig. 4c,d). Nuclear segregation also appears to be affected and a high percentage of nuclei were either caught in the junction of two cells or improperly segregated, resulting in multiple nucleated (open arrowheads in Fig. 4e,g) or anucleated cells (closed arrowheads in Fig. 4f,h). It was evident that cytokinesis occurred prior to the completion of nuclear migration and segregation, resulting in cut-like phenotypes (small arrows in Fig. 4c,d,e,g; Hirano et al. 1986; Samejima et al. 1993). However, cytokinesis of the pat1-3 mutant was often incomplete, with the mother and daughter cells connected by a bridge of nuclear material. The pat1-3 mutant is sensitive to benomyl Since the above data indicate that Pat1p might be involved in nuclear segregation, a function in which tubulin is involved, pat1 mutants were tested for its sensitivity to an anti-microtubule drug, benomyl. It has been shown that microtubule-mediated processes, such as nuclear division, nuclear migration and nuclear fusion, are inhibited in the presence of benomyl (Quinlan et al. 1980; Delgado & Conde 1984; Jacobs et al. 1988) and that many tubulin or tubulin-related q Blackwell Science Limited An essential yeast acetyltransferase homologue Figure 5 Benomyl sensitivity of pat1 mutants. YRL34 (pat1-3/pat1-itDelta2::HIS3, right half of the plate) and YRL37 (PAT1/pat1∆2::HIS3, left half of the plate) were streaked on YPD plates containing various amounts (0, 8, 16, 20 mg/mL) of the anti-tubulin drug, benomyl. Plates were incubated at either the permissive (23 8C) or semi-permissive (31.5 8C) temperatures. mutants are supersensitive to benomyl in yeast (Umesono et al. 1983; Huffaker et al. 1988; Matsuzaki et al. 1988; Schatz et al. 1988; Stearns et al. 1990). The level of benomyl sensitivity is temperature dependent; yeast cells are more sensitive to benomyl at lower temperatures (Stearns et al. 1990). If PAT1 is involved in a microtubule-mediated process, the sensitivity to benomyl in pat1 mutant cells might be exacerbated at conditions that are suboptimal for growth. In fact, the original pat1-1 mutant (YRL4) was unable to grow on YPG plates containing 20 mg/mL benomyl at 30 8C. However, its parental strain YMW1-6/pMW29H1 can grow in the presence of up to 30 mg/mL benomyl at 30 8C. Introducing a wild-type PAT1 gene to YRL4 restored its benomyl resistance to the wild-type level (data not shown), indicating that the sensitivity to benomyl in YRL4 is associated with the pat1-1 mutation. We also examined the benomyl sensitivity of the Ts – pat1-3 strain (YRL34) at a semi-permissive temperature (31.5 8C, Fig. 5). At the permissive temperature (23 8C), both YRL34 and the Pat1+ control YRL37 show similar sensitivity to benomyl; neither strain grows in the presence of 16 mg/mL benomyl. At 31.5 8C, YRL34 grows very poorly at 16 mg/mL and dies at 20 mg/mL of benomyl, whereas YRL37 grows well at 16 mg/mL and poorly at 20 mg/mL of benomyl. This result suggests that under conditions where pat1 mutants are stressed but still viable, they show increased benomyl sensitivity compared to wild-type cells. The sensitivity to benomyl for pat1 is comparable to that for mutants which are defective in processes involved in nuclear q Blackwell Science Limited division (Hoyt et al. 1991; Li & Murray 1991; Ursic & Culbertson 1991) but is lower than that for tubulin mutants (Matsuzaki et al. 1988; Schatz et al. 1988). This further suggests that Pat1p functions in a process where tubulin or related proteins are also involved. pat1 mutants have defects in spindle organization Immunofluorescence was carried out with anti-tubulin antibodies to visualize microtubules in the pat1-3 mutant. Figure 6 shows microtubule staining of the pat1-3 strain at either permissive (23 8C) or nonpermissive (37 8C) temperatures. At 23 8C, the distribution of microtubules in pat1-3 (Fig. 6a–c) is indistinguishable from that in the wild-type, in which < 30% cells have elongated spindles (Pringle & Hartwell 1981; Kilmartin & Adams 1984). After a shift to 37 8C for 4 h, YRL34 (pat1-3) has very few (< 3%) cells with elongated spindles; the majority of cells have short spindles (small arrows in Fig. 6d–i) which were similar to those observed in cells treated with hydroxyurea, a DNA synthesis inhibitor (Byers & Goetsch 1975; Hartwell 1976). However, unlike spindles seen in hydroxyurea-treated cells that are usually located between mother cells and their large buds, spindles in YRL34 are seen even in unbudded and small budded cells (Fig. 6d–f ). Further examination of those short spindles revealed that some were in fact two closely located spindle pole bodies (SPB, see closed arrowheads in Fig. 6g–i,j,l,n). The intranuclear spindles Genes to Cells (1996) 1, 923–942 931 R Lin et al. Figure 6 Aberrant microtubular morphology in pat1-3 mutant cells. YRL34 (pat1-3/pat1-∆2::HIS3) incubated at either 23 8C (a,b,c) or 37 8C (d-o) were prepared for immunofluorescence using anti-tubulin antibodies (YOL1/34) and FITC-conjugated second antibodies. Cells were examined by either phase contrast (a,d,g, j,k), FITC-fluorescence (c,f,i,n,o) or DAPI (b,e,h,l,m) fluorescence microscopy. The small arrows point to the short spindles observed in the pat1-3 mutation (d–i) at the nonpermissive temperature. This is different from the elongated spindles seen in wild-type or pat1-3 of the same stage at permissive temperatures (large arrows in a–c). The closed arrowheads indicate cells with two separated SPBs (g–i, j,l,n). The open arrowhead points to a mis-orientated spindle. Panels k, m, and o show an example of a monopolar spindle in a dividing nucleus. Bar ¼ 20 mm. typically seen were either missing or greatly reduced here. Particularly revealing is the example shown in Fig. 6, panels j, l and n, in which nuclei have already separated and two SPBs are separated, yet are still at the proximal ends of the two nuclei. This is distinct from the wild-type cells at the same stage in which a long spindle connected by SPBs at distal ends of two nuclei is usually apparent (see large arrows in Fig. 6a–c). In panels k, m and o, we show a large-budded cell whose 932 Genes to Cells (1996) 1, 923–942 nucleus begins to migrate toward the daughter cell; only one SPB (or two but unseparated SPBs) was observed. Usually in wild-type cells, one SPB is orientated toward the bud site causing the spindle to orientate itself parallel to the long axis of the cell. In the pat1-3 mutant, misoriented spindles are often observed (open arrowheads in Fig. 6d–f ). This result suggests that the spindle apparatus is affected in the pat1-3 mutant at 37 8C and is consistent with other observations indicating that q Blackwell Science Limited An essential yeast acetyltransferase homologue pat1-3 is abnormal in nuclear segregation and is more sensitive to benomyl. A requirement for Pat1p function in G1 and M phases The pat1-3 Ts – mutants do not arrest at a defined point of the cell cycle when shifted to nonpermissive temperatures. However, there is an increase in the large-budded cells and a decrease in the small-budded cells, suggesting that PAT1 might be required for the S phase in addition to mitosis. We performed FACS analyses to examine the DNA contents of the pat1-3 mutant arrested at a nonpermissive temperature. When an asynchronous pat1-3 culture is shifted to 37 8C for 3.5 h, the majority of cells have DNA contents either greater than 2C or less than 1C, suggesting the occurrence of aberrant chromosome segregation (data not shown). This was consistent with the observation of ‘cut’ cells. To more precisely determine the defects at particular cell cycle stages, this analysis was also performed with cells synchronized with a-factor at 23 8C. After release from a-factor arrest, cells were either transferred immediately or allowed to recover at 23 8C for different intervals before transfer to 37 8C and incubation for 3.5 h. Control strains proceeded through the cell cycle synchronously after they were released from the a-factor arrest but gradually lost the synchrony after the second cycle. Figure 7 shows the FACS profiles of a wild-type control arrested with a-factor (a) and released from a-factor to the 37 8C for 3.5 h (b). It was clear that cells are arrested with 1C DNA content in the presence of a-factor but lost the synchrony (the majority of cells are still at the G2 of the third cycle) after release from a-factor and incubated for 3.5 h at 37 8C. When pat1-3 cells were transferred to 37 8C immediately after release from the a-factor arrest, their DNA remained unreplicated (Fig. 7c,d) and cells were still schmooed after 3.5 h (data not shown) suggesting that PAT1 is required for exiting the a-factor block, passage through START, or the initiation of DNA synthesis. When cells were allowed to recover at 23 8C for 75 min (they appeared to be small budded and have initiated DNA synthesis at this point), they became arrested as large-budded cells, the majority of which have a G2 DNA content after 3.5 h at 37 8C (Fig. 7n), suggesting that PAT1 is also required for mitosis. If the recovery time at 23 8C was shorter than 75 min (Fig. 7e,g,i,k), the majority of cells showed DNA contents intermediate between 1C and 2C after 3.5 h at 37 8C (Fig. 7f,h,j,l). This result suggests that PAT1 is also required for the progression of DNA q Blackwell Science Limited Figure 7 FACS analyses of a-factor-synchronized pat1-3 mutants. The pat1-3 mutant cells (pat1-3/pat1-∆2::HIS3, YRL34) were synchronized with a-factor at 23 8C. After release from a-factor, cells were either transferred immediately to 37 8C (00 ) or allowed to recover for various periods of time (150 , 300 , 450 , 600 and 750 ) at 23 8C before transfer to 37 8C (a,c,e,g,i,k,m). After transfer to 37 8C, cells were incubated for 3.5 h (b,d,f,h, j,l,n) and prepared for FACS analyses. y-axis indicates cell count. x-axis indicates fluorescence obtained from propidium iodide. At least 10 000 cells were measured in each sample. synthesis. The FACS analysis of pat1 Ts – mutants released from a-factor arrest point is consistent with other analyses suggesting that PAT1 is required for multiple points of the cell cycle: to exit the a-factor block, progression of DNA synthesis, and mitosis. Genes to Cells (1996) 1, 923–942 933 R Lin et al. Isolation of cold sensitive suppressors for the pat1-3 Ts – mutation We have isolated three spontaneous cold sensitive (Cs – ) suppressors to the Ts – pat1-3 mutation (see Experimental procedures). One is a dominant suppressor that is recessive for the Cs – phenotype. The other two (YRL40 and YRL41) belong to the same complementation group and are recessive for both Cs – and suppression phenotypes. We isolate genomic clones that complement the Cs – phenotype of YRL40 and show that the gene responsible for the complementation is a previously identified gene SAC1 (suppressor of actin, Novick et al. 1989). SAC1 encodes an integral membrane protein that localizes to the yeast Golgi complex and ER (Whitters et al. 1993). It was shown by gene disruption that the function of SAC1 is only essential at temperatures below 17 8C (Novick et al. 1989). Although sac1 was originally isolated as an allelespecific Cs – suppressor to act1-1 Ts – mutation it was also later shown to be capable of suppressing three mutations that are in the secretory pathway, sec14, sec6 and sec9 (Cleves et al. 1989). We have generated a sac1 deletion allele and shown that it is cold sensitive as described by Novick et al. (1989), and that it fails to suppress the pat1-3 temperature sensitivity at 37 8C (Experimental procedures). sac1 null mutations also fail to suppress act1-1 mutants. The isolation of sac1 as a suppressor of pat1 potentially links Pat1p to actin and to secretory-mediated processes. Discussion In this study, we have isolated a yeast mutation in the gene PAT1 using a colony-colour-sectoring assay for mutations whose growth depends on the expression of a vertebrate histone H1. pat1-1 can be partially suppressed by the presence of a plasmid expressing H1, although we cannot explain how this suppression occurs. We have shown that PAT1 is essential for mitotic growth and that it encodes a small novel protein which shares homology with several acetyltransferases. Mutant cells depleted of Pat1p show aberrant cellular and nuclear morphology. We have also shown using a temperature sensitive mutation (pat1-3) that PAT1 functions at multiple steps in the cell cycle. PAT1 is required at multiple steps in the cell cycle Our analysis of pat1 mutants suggests that PAT1 is 934 Genes to Cells (1996) 1, 923–942 required at multiple points in the cell cycle: to exit the a-factor block, the progression of DNA synthesis, and proper mitosis. pat1-3 mutant cells do not undergo DNA replication and remain schmooed if transferred immediately to 37 8C after release from a-factor. DNA replication can occur and be completed, however, if the pat1-3 mutant cells are allowed to recover at permissive temperatures for 75 min before being shifted to 37 8C. At 75 min, DNA synthesis is already initiated. Nevertheless, the initiation of DNA synthesis is not sufficient, because pat1-3 mutant cells recovered at permissive temperatures for 60 min show a similar extent of DNA synthesis but fail to complete DNA synthesis when transferred to 37 8C. Both microscopic examination using DAPI and FACS analyses suggest that Pat1p is also required at mitosis for chromosome segregation and nuclear division. The pleiotrophic phenotype of the pat1 mutant is superficially similar to the phenotypes caused by defects in protein synthesis (Burke & Church 1991). Treatment with cycloheximide suggests that protein synthesis is required at multiple steps in the cell cycle, including completion of G1, initiation of DNA synthesis, progression from the end of DNA synthesis to nuclear division, completion of cytokinesis and cell separation. However, the nuclear catastrophe observed during mitosis and chromosome segregation defects indicated by FACS analyses in the pat1-3 mutant are not observed in cycloheximide treated yeast cells. Therefore, although PAT1 is involved in multiple steps in the cell cycle, it is unlikely that PAT1 functions by affecting protein synthesis. Several mutants in S. pombe show a ‘cut’ phenotype like that observed in pat1-3 Ts – mutants. These include those defective in sister chromatid separation such as topoisomerase II mutants (Goto & Wang 1985; Uemura & Yanagida 1986) and those that produce premature or inappropriate mitosis such as severe wee mutants (Enoch & Nurse 1990) and certain checkpoint defective mutants when DNA replication is delayed (Enoch et al. 1991; Al-khodairy & Carr 1992; Rowley et al. 1992). Presumably, S. cerevisiae topoisomerases activate a checkpoint that normally prevents cut phenotypes. pat1 mutants on the other hand are defective in some aspect of coordination of chromosome separation and nuclear division with cytokinesis. While some checkpoints may be defective, the S phase checkpoint appears to be intact because pat1 mutants are not sensitive to hydroxyurea which blocks DNA replication. It is curious that in a subset of cells observed using a Pat1p depletion protocol there appears to be reduced cytokinesis, with groups of cells connected to each other while others in the Ts – pat1 mutants show this q Blackwell Science Limited An essential yeast acetyltransferase homologue Cut – phenotype. Clearly this points to a very complicated set of defects in pat1 mutants. The overall phenotypes of the pat1 mutation resemble the phenotype of the double mutants in type I (topI) and type II (topII) topoisomerases in yeast. The complete defect in controlling the superhelical density can only be observed in the topI-topII double mutants whose phenotype is strikingly different from those of either of the single mutant (Uemura & Yanagida 1984; Goto & Wang 1985). At nonpermissive temperatures, the Ts – topI-topII double mutant cells have diminished DNA, RNA and protein synthesis and are quickly arrested at various stages of the cell cycle with defects in DNA synthesis, mitosis, nuclear division and chromosome segregation (Uemura & Yanagida 1984, 1986). On the basis of these similarities between phenotypes of pat1 mutants and topI-topII double mutants, it is plausible to suggest that the PAT1 gene may somehow be involved in the same pathway as topoisomerase I and II to influence DNA superhelicity. Does PAT1 encode an acetyltransferase? PAT1 shows no resemblance to histone H1 in either amino acid composition or secondary structure. However, we can not rule out the possibility that PAT1 is indirectly performing a role similar to histone H1 in the regulation of higher order chromatin organization. In support of a chromatin role, cells in the Pat1p depletion experiments had chromatin with an unusual ‘threaded’ appearance, indicating a possible role for Pat1p in higher-order chromatin structure; such a role could explain the pleotrophic phenotypes associated with pat1 mutants through modulation of gene expression in a chromatin environment. Several lines of evidence suggest that Pat1p is an acetyltransferase. First, it has significant homology with the conserved domain of several known acetyltransferases such as Hat1p (Kleff et al. 1995), Ard1p (Mullen et al. 1989; Lee et al. 1989), Mak3p (Tercero & Wickner 1992), and several bacterial acetyltransferases. Secondly, the original pat1-1 mutation has a D99N change within this conserved domain, suggesting that this domain is important for Pat1p function. Finally, an S. pombe gene that was isolated by its ability to complement the S. cerevisiae pat1 deletion mutant also contains a high conservation of the putative acetyltransferase domain. However, the spPat1p might have additional functions or might function in a different cellular compartment, since its N-terminal third is homologous to another S. cerevisiae protein, a putative membrane protein (Chen et al. 1991). Efforts have q Blackwell Science Limited been made to detect the acetyltransferase activity with HA-tagged Pat1p. While the anti-HA antibody was able to immunoprecipitate HA-Pat1p, no activity was so far detected using either histones or tubulin as substrates (data not shown). However, this analysis suffers from an absence of knowledge of the proper substrate for this protein and whether the antibody recognizes or inhibits the active form of Pat1p. Acetylation has been implicated in a wide variety of biological processes. Acetylation of the N-terminal NH2 occurs in almost all proteins co-translationally and this modification appears to be essential for viability in S. cerevisiae (Kulkarni & Sherman 1994). Three N-terminal acetyltransferases have been identified in S. cerevisiae, ARD1/NAT1, NAT2 and MAK3 (Lee et al. 1989; Mullen et al. 1989; Park & Szostak 1992; Tercero & Wickner 1992; Kulkarni & Sherman 1994). Mutations in nat2 cause lethality whereas mutations in either nat1 or ard1 activate the silent mating type locus (HML) and prevent cells from entering into G0 (Whiteway & Szostak 1985; Lee et al. 1989; Mullen et al. 1989; Kulkarni & Sherman 1994). In addition, N-terminal acetylation has been shown to be important for the processing of actin protein and the polymerization of striated alpha tropomyosin as well as its ability to bind to F-actin (Hitchcock-DeGregori & Heald 1987; Cook et al. 1991; Urbancikova & Hitchcock-DeGregori 1994). A second kind of acetylation occurs posttranslationally on the NH2 group of internal lysines. One of the better studied examples of internal acetylation occurs on the N-terminal domains of all core histones. Strong correlations have been made between histone acetylation and histone deposition (Allis et al. 1985; Csordas 1990), transcriptional activity of certain genes (Hebbes et al. 1988; Csordas 1990; Wolffe 1994; Brownell et al. 1996; Wolffe & Pruss 1996b), and histone displacement during mammalian spermatogenesis (Christensen & Dixon 1982). The lysines (K5, K8, K12, K16) that can be subjected to acetylation on the N-terminal domain of histone H4 are highly conserved in evolution. Cells with substitution of asparagine for all lysines in this domain are not viable (Megee et al. 1990), whereas substitution of arginine for these lysines results in either very slow growing (Durrin et al. 1991) or lethal (Megee et al. 1990) phenotypes. Substituting glutamine for all four lysines is not lethal, but confers several phenotypes, including prolonged S and G2+M periods of the division cycle, mating sterility and temperature sensitive growth (Megee et al. 1990). Recently, it was shown that certain patterns of acetylation are essential for the Genes to Cells (1996) 1, 923–942 935 R Lin et al. efficient silencing of transcription at the mating type loci: K16 must be deacetylated and at least K5 or K12 must be acetylated (Park & Szostak 1990; Braunstein et al. 1996). In addition, heterochromatin in yeast and flies is enriched in histone H4 that is acetylated at a specific lysine residue, K12 (Braunstein et al. 1996; Turner et al. 1992). Taken together, these results suggest that while the positive charge contained in the N-terminal domains of H4 (Johnson et al. 1990; Megee et al. 1990; Park & Szostak 1990) and H3 (Thompson et al. 1994) contributes to the formation and/or stability of heterochromatin in yeast, acetylation of specific lysine residues (K12) also plays an important role. Presumably acetylation influences heterochromatin formation through protein–protein interactions that are only beginning to be defined (Hecht et al. 1995). Although we have not shown that Pat1p exhibits acetyltransferase activity, it is possible that Pat1p is a histone acetyltransferase in S. cerevisiae. To date, none of the current histone acetyltransferases that have been cloned are essential. Given the increasingly clear involvement of histone acetylation/deacetylation in the activation/repression of at least some yeast genes (Brownell et al. 1996; Taunton et al. 1996), it is not difficult to imagine that mutations in histone acetyltransferases can indirectly affect multiple cellular functions through changes in gene expression. We do not know why the expression of a vertebrate histone H1 suppresses the original pat1-1 (but not pat1-∆2::HIS3 nor pat1-3) mutation, although a link to chromatin is suggested. Both histone H1 and acetylation on the core histone amino-terminal domains are thought to modulate the higher order chromatin organization, and several studies have shown that H1-mediated reactions are affected by the core histone aminoterminal domains and their acetylation (Ridsdale et al. 1990; Perry & Annunziato 1991; Juan et al. 1994). An intriguing possibility is that Pat1p functions in the acetylation of lysine 12, which appears to be associated with heterochromatin in yeast (Braunstein et al. 1996). Inability to acetylate lysine 12 and establish proper heterochromatin formation might result in the ‘threaded’ and ‘diffused’ chromatin observed in the pat1 mutants. In this scenario, expression of a vertebrate H1 might help to counteract the compromised organization of heterochromatin caused by the lack of acetylation at lysine 12 in H4. Suppression of the pat1-3 by sac1 mutations Our suppressor analysis on the pat1-3 Ts – mutation has identified two sac1 Cs – mutations. sac1 was initially 936 Genes to Cells (1996) 1, 923–942 identified as an allele-specific suppressor to the act1-1 Ts – mutation, and is distinguishable from all other sac mutations in its capability to suppress some secretion mutants in addition to act1-1 (Cleves et al. 1989; Novick et al. 1989). It is an integral membrane protein in Golgi and ER (Whitters et al. 1993). We do not at present understand why mutations in SAC1 suppress the pat1-3 mutation, but would like to speculate on two possibilities. It was proposed that SAC1 might play a role in biological events whereby secretory and cytoskeletal activities are coordinated. Although we have not detected abnormalities in the actin network in the pat1-3 mutants, it does affect the microtubule organization and benomyl sensitivity. It has been shown in many cases that the actin and microtubule machinery interact (Schmit & Lambert 1988; Lillie & Brown 1992; Pedrotti et al. 1994; Ursic et al. 1994), therefore it is possible that sac1 also suppress some microtubulerelated defects. It has been suggested that acetylation increases the stability of microtubule and rabbit skeletal troponin C and also increases the interaction of the latter with other troponin components (Maruta et al. 1986; Grabarek et al. 1995). Perhaps pat1 mutants are affecting the ability of these two networks to properly interact. Intriguingly, Smy1 is a microtubule-based motor protein which was isolated as a multi-copy suppressor of a mutation in an actin-based motor protein, Myo2, whose function has been implicated in the process of polarized secretion in yeast (Lillie & Brown 1992). A second possibility is that although Pat1p does not contain an obvious transmembrane domain, it might associate with membrane through interaction with a membrane protein. The interaction might allow Pat1p to function in a particular cellular compartment or facilitate the assembly of Pat1p into a complex. Mutations in pat1-3 might affect the association with this hypothetical membrane protein and disable Pat1p from functioning in the proper cellular compartment. Since mutations in sac1 can suppress—and in some cases bypass—several mutations in the secretory pathway, they might suppress pat1-3 by allowing the interaction between Pat1p and the proper membrane protein or by allowing Pat1p to be located in the proper cellular compartment. This notion is supported by the interesting observation that spPat1p consists of two separate domains, each of which is homologous to a different S. cerevisiae protein. The N-terminal third is homologous to a putative transmembrane protein, YGL021, and the C-terminal third is to Pat1p. Nevertheless, YGL021 deletion does not result in any detectable phenotype (Chen et al. 1991). If the function of q Blackwell Science Limited An essential yeast acetyltransferase homologue Table 1 Strains used in this study Strain YMW1-6 CRY1 CRY1U6 CRY1L CRY3 YRL4 YRL9 YRL17 YRL18 YRL26 YRL13 YRL30 YRL34 YRL37 YRL40 YRL41 YRL98 YRL101 YRL102 YRL127 YRL130 Plasmid pUN105 pMW29H1 pRL45 pRL46 pRL57 pRL71 pRL71 pRL58 pUN105 pUN105 pUN105 Genotype MATa ade2-1 ade3D his3-11,15 leu2-3, 112 ura3-1 MATa can1-100 ade2-1 his3-11, 15 leu2-3, 112 trp1-1 ura3-1 MATa can1-100 ade2-1 his3-11, 15 leu2-3, 112 trp1-1 URA3 MATa can1-100 ade2-1 his3-11, 15 leu2-3, 112 trp1-1 ura3–1 (LEU2) MATa/MATa can1-100 ade2-1 his3-11, 15 leu2-3, 112 trp1-1 ura3-1 MATa YMW1-6 pat1-1 (URA3 ADE3) MATa/MATa CRY3 PAT1/pat1-∆2::HIS3 As CRY1 but MATa pat1-∆2::HIS3 (URA3 GAL1-PAT1 †) As CRY1 but MATa pat1-∆2::HIS3 (URA3 GAL1-PAT1‡) As CRY1 but MATa pat1-∆2::HIS3 (URA3 PAT1) As CRY3 but PAT1/PAT1::HIS3 As CRY1 but MATa pat1-∆2::HIS3 (LEU2 pat1-3) As CRY1 but pat1-∆2::HIS3 (LEU2 pat1-3) As CRY1 but pat1-∆2::HIS3 (LEU2 PAT1) As CRY1 but MATa sac1 pat1-∆2::HIS3 (LEU2 pat1-3) As CRY1 but MATa sac1∆2::HIS3 (LEU2 pat1-3) MATa/MATa can1-100 ade2-1 his3-11, 15 leu2-3, 112/LEU2 trp1-1 ura3-1/URA3 CS/cs MATa/MATa can1-100 ade2-1 his3-11, 15 leu2-3, 112/LEU2 trp1-1 ura3-1/URA3 CS/cs::sac1∆ MATa/MATa can1-100 ade2-1 his3-11, 15 leu2-3, 112/LEU2 trp1-1 ura3-1/URA3 CS/cs::sac1∆ As CRY1 but MATa pat1-∆2::HIS3, LEU2::GAL1-PAT1‡ As CRY1 but sac1::URA3 * Zieler et al. 1995. † PAT1 cDNA. ‡ Shorter version of PAT1 cDNA. Markers in parentheses reside on the plasmid. Pat1p was dependent on interaction with a membrane protein, it would have to be a protein other than YGL021. Experimental procedures Strains, media and microbial techniques Strains used in this study are listed in Table 1. Yeast media were prepared as in Sherman et al. (1986). When appropriate, 2% galactose was used in place of 2% glucose. Unless otherwise stated, all yeast strains were grown at 30 8C. Benomyl was dissolved in dimethylsulphoxide (DMSO) as a 30 mg/mL stock. Yeast transformations were performed by the lithium acetate method (Ito et al. 1983). JM107 was used as a host for all plasmid constructions and amplification. XL1Blue was used for the preparation of single stranded DNA, and LE392 was used as a host for the l phage (Elledge et al. 1991). To isolate the pat1-1 allele, YMW1-6 containing plasmid pMW29H1 was mutagenized with EMS to 60% survival, as described in Sherman et al. (1986), resuspended in YPG medium to a density of 2 × 107 cells/mL, incubated for 4 h and plated on YPG plates at a density of < 700 colonies/150 mm plate, and q Blackwell Science Limited allowed to grow at 30 8C for 5–7 days before identifying homogeneously red colonies. The pat1-∆2::HIS3 allele was generated by transforming a pat1 disruption construct, pRL9 (see below) cleaved with SacII, into the diploid CRY3 and selecting for histidine prototrophy. Homologous recombinants were verified by Southern analysis. The resulting strain YRL9 is heterozygous for the disrupted pat1 locus (pat1-∆2::HIS3/PAT1) and was sporulated and examined for the viability of its progeny by tetrad analyses. The pat1-3 Ts – allele was isolated by the in vitro mutagenesis of pRL58 with hydroxylamine and detected using a plasmid shuffle protocol according to procedures described previously (Rose & Fink 1987; Sikorski & Boeke 1990). Mutagenized pRL58 was transformed into YRL26 (pat1-D2::HIS3) carrying PAT1 on a centromeric URA3 plasmid (pRL57) treated as described (Sikorski & Boeke 1990). Only one temperature sensitive clone, pat1-3, was obtained from 20 000 colonies. The plasmid (pRL71) carrying the pat1-3 mutation was recovered from yeast and provides temperature sensitivity when reintroduced into a PAT1 disrupted strain. Plasmid constructions and DNA manipulation Plasmids related to this study are listed in Table 2. DNA Genes to Cells (1996) 1, 923–942 937 R Lin et al. Table 2 Plasmids used in this study Plasmids Yeast marker Relevant features and comments pMW29H1 pUN105 pRL45 pRL46 pSS1.5 pRL57 pRL58 pRL55 pRL56 pRL71 pRL9 pRL131 pRL149 URA3 LEU2 URA3 URA3 LEU2 URA3 LEU2 URA3 URA3 LEU2 LEU2 URA3 LEU2 GAL1-chH1, ADE3, CEN4* CEN4† GAL1-PAT1‡, CEN4 GAL1-PAT1§, CEN4 1.5 kb SacII fragment containing PAT1, CEN4 0.6 kb XhoI-XbaI fragment containing PAT1, CEN4 0.6 kb XhoI-XbaI fragment containing PAT1, CEN4 GAL-chH1-lacZ, CEN4 similar to pRL55, lacZ was fused out of frame with chH1, CEN4 pat1-3, CEN4 pat1-∆2::HIS3, CEN4 2.4 kb genomic fragment containing SAC1, CEN4 an integration plasmid containing GAL1-PAT1 * † ‡ § Provided by Mark Walberg. Elledge & Davis 1988. Longer PAT1 cDNA. Shorter PAT1 cDNA. manipulations are essentially as described in Maniatis et al. (1982). The chicken histone H1 gene 11 L (chH1) contains no intron and was amplified from chicken genomic DNA by polymerase chain reaction (PCR) with the following two oligonucleotides: TCTCCTCGAGCAGCTCCGACATGTC and GCGGGTC AAGGGAATTCTCCCACAAG (Coles et al. 1987). This 711 bp fragment was digested with XhoI and EcoRI, whose recognition sites are underlined in the 50 and 30 oligonucleotides, respectively, and subcloned into XhoI-EcoRI cleaved Bluescript pBSKSIIþ. The chH1 clone was sequenced to ensure that no errors were generated by PCR. The chH1 gene was then digested with XhoI, filled in with dNTPs and Klenow, digested with XbaI, and cloned into the SmaI-XbaI digested pMW29 (kindly provided by Mark Walberg, VT Southwestern, Dallas, Zieler et al. 1995) to make pMW29H1. The molecular characterization of genomic DNA containing the PAT1 locus was performed with a genomic clone isolated from the centromeric-based library. The 1.5 kb SacII fragment was subcloned into pUN105 (Elledge & Davis 1988) giving rise to pSS1.5. pRL58 was created by an internal deletion of the XbaI fragment from pSS1.5 which removes the sequence downstream of the XbaI site to a XbaI site in the vector. The SacII-XbaI fragment was subcloned from pRL58 into SacII and XbaI digested pUN95 (Elledge & Davis 1988), giving pRL57. The deletion construct, pRL9, was generated by replacing the NcoIBalI fragment on pSS1.5 with a 2.5 kb HIS3 cassette from pJA50 containing the HIS3 and the Tn 5 neo genes (Allen & Elledge 1994). PAT1 cDNA clones from the yeast cDNA library were subcloned as XhoI fragments into XhoI-cleaved pBSKSIIþ for sequencing or XhoI-cleaved pSE936 to place them under GAL1 control. 938 Genes to Cells (1996) 1, 923–942 Sequencing was performed by the generation of a series of nested deletion clones generated by DNaseI as described in Maniatis et al. (1982). Single-stranded DNAs were prepared following the methods of Zagursky & Berman (1984), using the helper phage R408 (Russel et al. 1986). All sequencing was performed on both strands by the dideoxy chain termination method (Sanger et al. 1977). The GAL1-PAT1 cDNA was purified as a ScaI-XbaI fragment from pRL46 (lacking the first three amino acids of Pat1p) and subcloned onto the SmaI and XbaI digested pIClac28, an integration vector (Gietz & Sugino 1988). The resulting plasmid pRL149 was digested with EcoRV and transformed the yeast strain YRL26. The yeast transformant was streaked on SC-Leu galactose plates, followed by 5-FOA galactose plates to remove the plasmid pRL57. The resulting strain YRL127 contains a disrupted pat1 gene (pat1-∆2::HIS3) and a GAL1PAT1 cDNA integrated to the LEU2 locus. Isolation of the PAT1 gene YRL4 (pat1-1) was transformed with two LEU2 genomic libraries, one in p366, LEU2 CEN, or YEP13 LEU2 2m (Nasmyth & Reed 1980) yeast shuttle vectors. Transformants were selected on SC-leucine glucose plates. Approximately 100 000 transformants were screened from each library. One PAT1 clone from the p366-based library and nine from the YEP13 library were isolated, all contained the same gene and were able to complement all the pat1 phenotypes. To ensure the phenotype of YRL4 is indeed due to defects in the PAT1 gene we PCR amplified the PAT1 locus from YRL4 and showed that it fails to complement a pat1 deletion strain to the wild-type level. The PAT1 locus from either a wild-type q Blackwell Science Limited An essential yeast acetyltransferase homologue strain (YMW1-6) or YRL4 was amplified using oligonucleotides flanking the PAT1 ORF. These two PCR products were separately digested with SacI and XbaI and subcloned to SacI XbaI digested pUN105. Five independent clones were obtained from each PCR and transformed into YRL26 (pat1-∆2::HIS3, pRL57). pRL58 and pUN105 were served as positive and negative controls, respectively. Transformants were selected on SC-Leu-Ura plates and, from each clone, four transformants were streaked on 5-FOA plates to test whether the new plasmid can replace pRL57 and complement pat1 deletion. While all clones grow equally well on the SC-Leu-Ura plates, only clones from wild-type DNA and pRL58 can fully substitute pRL57 and grow well on 5-FOA plates. All clones from YRL4 DNA only partially substitute pRL57 and grow poorly on 5-FOA plates; both the number and size of colonies on 5-FOA were greatly reduced. This result indicates that YRL4 has defects in the PAT1 locus instead of having another mutation that is suppressible by the PAT1 gene. Isolation of Schizosaccharomyces pombe PAT1 homologue by functional complementation YRL127 contains pat1-D2::HIS3 and survives due to the GAL1PAT1 cDNA integrated in LEU2 locus. YRL127 is unable to grow on glucose plates but grows well on galactose plates. An S. pombe cDNA library under ADH promoter control in pDB20 (URA3, 2m, Becker et al. 1991) was transformed in YRL127 and the transformant was selected on SC-Ura glucose plates. From over 200 000 transformants screened, 180 positive clones were obtained that suppress the inviability of YRL127 on glucose in a 5-FOA-dependent manner. Hybridization analysis showed that all the 185 clones belonged to the same gene. Further restriction mapping was carried out on nine clones and showed that they all contain a cDNA insert of < 1.3 kb and all have an internal HindIII site that is 1 kb from one end. Only one cDNA clone was sequenced. The sequence was confirmed by the S. pombe sequencing project from the Sanger Center. pat1-3 suppressor screen The spontaneous reversion rate for the pat1-3 Ts – allele (YRL30) at 37 8C is < 3.5 × 10 –7. Two hundred independent YRL30 colonies grown at 23 8C were inoculated into 200 individual 3 mL liquid YPD media to a density of around 1 × 107/mL. From each individual culture, 1 × 107 cells were spun down and plated on to a 10 mm YPD plate. After 2 days incubation at 37 8C, colonies grown were replica-plated on to two YPD plates, and grown at 23 and 37 8C, respectively. Colonies that grow at 37 8C, but not 23 8C contain candidate Cs – suppressors of the pat1-3 allele. Only one clone from each original plate was analysed further. Three clones that belong to two complementation groups were isolated. The first group is a dominant suppressor but is recessive for the Cs – phenotype and was not further characterized. The second group contains two members and are named YRL40 and YRL41. q Blackwell Science Limited Characterization of the SAC clones Genomic clones complementing the Cs – lethality of YRL40 were isolated from a yeast centromeric genomic library. A 2.4 kb XbaI fragment was cloned to XbaI-cleaved pUN70 (Elledge & Davis 1988) resulting in pRL131 which was shown to be sufficient for the complementation. Partial sequencing analysis revealed that this genomic clone contains a previously identified gene, SAC1 (Whitters et al. 1993). A frame shift introduced to the SAC1 ORF on pRL131 eliminated its ability to complement YRL40. The sac1 disruption allele was generated by replacing the 1.3 kb XhoI-BclI fragment with a 2.4 kb BamHI-XhoI URA3 cassette from pJA53 (Allen & Elledge 1994). A sac1 deletion mutant (YRL130) is cold sensitive at temperatures below 17 8C, consistent with results described by Novick et al. (1989). To analyse whether a sac1 deletion mutation can suppress the temperature sensitivity of pat1-3, YRL130 was mated with YRL30 (pat1-∆2::HIS3/pat1-3) and diploids were selected on SC-Leu-Ura plates. These diploids were sporulated at 23 8C and the viable spores that are Uraþ Leuþ Hisþ are tested for their viability at 37 8C. Of 10 such spores analysed, none were able to grow at 37 8C, suggesting that a sac1 deletion mutation cannot suppress the pat1-3 temperature sensitivity. Tetrad analyses were also performed to ensure that the Cs – phenotype in YRL40 is linked to the SAC1 locus. YRL40 was mated to CRY1U6 and the diploid was sporulated. Cs – spores that were His – Leu – Uraþ were mated to CRY1 L to generate YRL98 (Csþ /Cs – ). YRL98 was then transformed with a sac1 disruption allele which was generated by replacing the 1.3 kb XhoI-BclI fragment with a 2.5 kb BamHI-SalI HIS3 cassette from pJA50 (Allen & Elledge 1994). His+ transformants were selected and sporulated. Two Hisþ transformants, YRL101 and YRL102, were sporulated for tetrad analysis. Since it is not known whether the sac1 disruption occurs on the wild-type or sac1-2 allele, there were two possible outcomes of sporulation. If the original Cs – mutant is in the SAC1 gene, we would obtain either 4:0 or 2:2 Cs – /Csþ segregation in each tetrad dissected. If the Cs – phenotype is due to a mutation unlinked to SAC1, then we would observe both DT, NPT and TT. Twenty tetrads for each strain were analysed from YRL101 and YRL102 and they all have only two Cs – spores, which are also always Hisþ , indicating that the Cs – phenotype in YRL40 is linked to the SAC1 locus. Immunofluorescence and microscopy The immunofluorescence procedure described by Kilmartin & Adams (1984) was followed with the exception that cells were fixed in formaldehyde for 2 h at 30 8C and cell walls were subsequently removed by treatment with 50 mg/mL Zymolyase 100T for 1 h at 37 8C. Rat monoclonal anti-tubulin antibody YOL 1/34 and FITC-conjugated goat anti-rat IgG antiserum were obtained from Accurate Chemical & Scientific Corporation (Westbury, NY). After incubation with the secondary antibody, cells were stained with 1 mg/mL DAPI for 5 min. In cases where only DAPI staining was required, cells were fixed in 95% ethanol, incubated 1 mg/mL DAPI for 5 min at room temperature. Cells were then washed five times with water, resuspended in 100% Genes to Cells (1996) 1, 923–942 939 R Lin et al. methanol, and applied to poly-lysine coated slides. Preparations were viewed on a Zeiss Axioscope equipped for epifluorescence microscopy (Carl Zeiss, Germany). Cell-cycle synchronization and f low cytometry Cells were grown in YPD at pH 3.9 to an OD600 of 0.2 at 23 8C. a-factor was added to a final concentration of 10 mM/mL. Cells were incubated for 3 h, during which time an additional 10 mM/mL of a-factor was added every hour. When < 90% of the cells appeared unbudded and schmooed, cells were pelleted, washed twice with YPD of normal pH, and resuspended in YPD to a final density of 1 × 106 cells/mL. Cells were then either shifted to 37 8C immediately or allowed to recover at 23 8C for various lengths of time (every 15 min up to 75 min) before shifting to 37 8C. Cells were incubated for 3.5 h at 37 8C and prepared for cytological examination or for flow cytometry analyses. Yeast cells were prepared for flow cytometry as described by Hutter & Eipel (1979). The flow cytometer used was a Coulter Epics 753 (Coulter Electronics, Hialeah, FL). The excitation wavelength was 488 nm. A barrier filter was used to remove emissions above 590 nm. Acknowledgements We thank Mark Walberg for providing materials for our genetic screen and P. Hieter and F. Spencer, Johns Hopkins University, for the genomic library. We also thank Z. Chen, R. Gibbs and W. Schober for advice. This work was supported by grants from the National Institutes of Health, nos NIH GM44664 to S.J.E. and NIH GM53512 to C.D.A. S.J.E. is a PEW Scholar in the Biomedical Sciences and an Investigator of the Howard Hughes Medical Institute. References Al-khodairy, F. & Carr, A.M. (1992) DNA repair mutants defining G2 checkpoint pathways in Schizosaccharomyces pombe. EMBO J. 11, 1343–1350. Allan, J., Cattini, P., Cowling, G.J., Craigie, R., Harborne, N. & Gould, H. (1981) Regulation of the higher-order structure of chromatin by histone H1 and H5. J. Cell. Biol. 90, 279–288. Allan, J., Hartman, P.G., Crane-Robinson, C. & Aviles, F.X. (1980) The structure of histone H1 and its location in the chromatin. Nature 288, 675–679. Allan, J., Rau, D.C., Harborne, N. & Gould, H. (1984) Higher order structure in a short repeat length chromatin. J. Cell. Biol. 98, 1320–1327. Allen, J.B. & Elledge, S.J. (1994) A family of vectors that facilitate transposon and insertional mutagenesis of cloned genes in yeast. Yeast 10, 1267–1272. Allfrey, V.G. (1977) Post-synthetic modifications of histone structure: A mechanism for the control of chromosome structure by the modulation of histone–DNA interactions. In: 940 Genes to Cells (1996) 1, 923–942 Chromatin and Chromosome Structure (eds J. Li & R. Eckhadt), pp. 167–191. New York: Academic Press Inc. Allis, C.D., Chicoine, L.G., Richman, R. & Schulman, I.G. (1985) Deposition-related histone acetylation in macronuclei of conjugating Tetrahymena. Proc. Natl. Acad. Sci. USA 82, 8048–8052. Becker, D.M., Fikes, J.D. & Guarente, L. (1991) A cDNA encoding a human CCAAT-binding protein cloned by functional complementation in yeast. Proc. Natl. Acad. Sci. USA 88, 1968–1972. Bender, A. & Pringle, J.R. (1991) Use of a screen for synthetic lethal and multicopy suppressee mutants to identify two new genes involved in morphogenesis in Saccharomyces cerevisiae. Mol. Cell. Biol. 11, 1295–1301. Braunstein, M., Sobel, R.E., Allis, C.D., Turner, B.M. & Broach, J.R. (1996) Transcriptional silencing in yeast requires a heterochromatin histone acetylation pattern. Submitted. Brownell, J. & Allis, C.D. (1996) Special HATs for special occasions: Linking histone acetylation to chromatin assembly and gene activation. Curr. Opin. Genet. Dev. 6, 176–184. Brownell, J.E., Zhou, J. & Ranalli, T., et al. (1996) Tetrahymena histone acetyltransferase A: a transcriptional co-activator linking gene expression to histone acetylation. Cell 84, 843– 851. Burke, D.J. & Church, D. (1991) Protein synthesis requirements for nuclear division, cytokinesis, and cell separation in Saccharomyces cerevisiae. Mol. Cell. Biol. 11, 3691–3698. Byers, B. & Goetsch, L. (1975) Behavior of spindles and spindle plaques in the cell cycle and conjugation of Saccharomyces cerevisiae. J. Bacteriol. 124, 511–523. Certa, U., Colavito-Shepanski, M. & Grunstein, M. (1984) Yeast may not contain histone H1: the only known ‘histone H1-like’ protein in Saccharomyces cerevisiae is a mitochondrial protein. Nucl. Acids Res. 12, 7975–7985. Chen, W., Capieaux, E., Balzi, E. & Goffeau, A. (1991) The YGL021 gene encodes a putative membrane protein with a putative leucine zipper motif. Yeast 7, 301–303. Christensen, M.E. & Dixon, G.H. (1982) Hyperacetylation of histone H4 correlates with the terminal, transcriptionally inactive stages of spermatogenesis in rainbow trout. Dev. Biol. 93, 404–415. Cleves, A.E., Novick, P.J. & Bankaitis, V.A. (1989) Mutations in the SAC1 gene suppress defects in yeast Golgi and yeast actin function. J. Cell. Biol. 109, 2939–2950. Coles, L.S., Robins, A.J., Madley, L.K. & Wells, J.R.E. (1987) Characterization of the chicken H1 gene complement: Generation of a complete set of vertebrate H1 protein sequences. J. Biol. Chem. 262, 9656–9663. Cook, R.K., Sheff, D.R. & Rubenstein, P.A. (1991) Unusual metabolism of the yeast actin amino terminus. J. Biol. Chem. 266, 16825–16833. Csordas, A. (1990) On the biological role of histone acetylation. Biochem. J. 265, 23–38. Delgado, M.A. & Conde, J. (1984) Benomyl prevents nuclear fusion in Saccharomyces cerevisiae. Mol. Gen. Genet. 193, 188– 189. Durrin, L.K., Mann, R.K., Kayne, P.S. & Grunstein, M. (1991) Yeast histone H4 N-terminal sequence is required for promoter activation in vivo. Cell 65, 1023–1031. Elledge, S.J. & Davis, R.W. (1988) A family of versatile centromeric vectors designed for use in the sectoring-shuffle mutagenesis assay in Saccharomyces cerevisiae. Gene 70, 303–312. q Blackwell Science Limited An essential yeast acetyltransferase homologue Elledge, S.J., Mulligan, J.T., Ramer, S.W., Spottswood, M. & Davis, R.W. (1991) l YES: A multifunctional cDNA expression vector for the isolation of genes by complementation of Yeast and E. coli mutations. Proc. Natl. Acad. Sci. USA 88, 1731–1735. Enoch, T., Gould, K.L. & Nurse, P. (1991) Mitotic checkpoint control in fission yeast. Cold Spring Harb. Symp. Quant. Biol. 56, 409–416. Enoch, T. & Nurse, P. (1990) Mutation of fission yeast cell cycle control genes abolishes dependence of mitosis on DNA replication. Cell 60, 665–673. Georgakopoulos, T. & Thireos, G. (1992) Two distinct yeast transcriptional activators require the function of the GCN5 protein to promote normal levels of transcription. EMBO J. 11, 4145–4152. Gietz, R.D. & Sugino, A. (1988) New yeast-Escherichia coli shuttle vectors constructed with in vitro mutagenized yeast genes lacking six-base pair restriction sites. Gene 74, 527–534. Goto, T. & Wang, J.C. (1985) Cloning of yeast TOP1, the gene encoding DNA topoisomerase I, and construction of mutants defective in both DNA topoisomerase I and DNA topoisomerase II. Proc. Natl. Acad. Sci. USA 82, 7178–7182. Grabarek, Z., Mabuchi, Y. & Gergely, J. (1995) Properties of troponin C acetylated at lysine residues. Biochemistry 34, 11872–11881. Hartwell, L.H. (1976) Sequential function of gene products relative to DNA synthesis in the yeast cell cycle. J. Mol. Biol. 104, 803–817. Hebbes,T.R., Thorne,A.W. & Crane-Robinson, C. (1988) A direct link between core histone acetylation and transcriptionally active chromatin. EMBO J. 7, 1395–1402. Hecht, A., Laroche, T., Strahl-Bolsinger, S., Gasser, S.M. & Grunstein, M. (1995) Histone H3 and H4 N-termini interact with SIR3 and SIR4 proteins: a molecular model for the formation of heteochromatin in yeast. Hirano, T., Funahashi, S., Uemura, T. & Yanagida, M. (1986) Isolation and characterization of Schizosaccharomyces pombe cut mutants that block nuclear division but not cytokinesis. EMBO J. 5, 2973–2979. Hitchcock-DeGregori, S.E. & Heald, R.W. (1987) Altered actin and troponin binding of amino-terminal variants of chicken striated muscle a-tropomyosin expressed in Escherichia coli. J. Biol. Chem. 262, 9730–9735. Hoyt, M.A., Totis, L. & Roberts, B.T. (1991) S. cerevisiae genes required for cell cycle arrest in response to loss of microtubule function. Cell 66, 507–517. Huffaker, T.C., Thomas, J.H. & Botstein, D. (1988) Diverse effects of b-tubulin mutations on microtubule formation and function. J. Cell. Biol. 106, 1997–2010. Hutter, K.J. & Eipel, H.E. (1979) Microbial determinations by flow cytometry. J. Gen. Microbiol. 113, 369–375. Ito, H., Jukuda, Y., Murata, K. & Kimura, A. (1983) Transformation of intact yeast cells treated with alkali cations. J. Bacteriol. 153, 163–168. Jacobs, C.W., Adams ,A.E.M., Staniszlo, P.J. & Pringle, J.R. (1988) Functions of microtubules in the Saccharomyces cerevisiae cell cycle. J. Cell. Biol. 107, 1409–1426. Johnson, L.M., Fisher-Adams, G. & Grunstein, M. (1990) Genetic evidence for an interaction between SIR3 and histone H4 in the repression of the silent mating loci in Saccharomyces cerevisiae. Proc. Natl. Acad. Sci. USA 87, 6286–6290. Juan, L.J., Utley, R.T., Adams, C.C., Vettese-Dadey, M. & Workman, J.L. (1994) Differential repression of transcription q Blackwell Science Limited factor binding by histone H1 is regulated by the core histone amino termini. EMBO J. 13, 6031–40. Kayne, P.S., Kim, U.-J., Han, M., Mullen, J., Yoshizaki, F. & Grunstein, M. (1988) Extremely conserved histone H4 N-terminus is dispensable for growth but essential for repressing the silent mating loci in yeast. Cell 55, 27–39. Kilmartin, J.V. & Adams, A.E.M. (1984) Structural rearrangements of tubulin and actin during the cell cycle of the yeast Saccharomyces. J. Cell. Biol. 98, 922–933. Kleff, S., Andrulis, E.D., Anderson, C.W. & Sternglanz, R. (1995) Identification of a gene encoding a yeast histone H4 acetyltransferase. J. Cell. Biol. 270, 24674–24677. Kranz, J.E. & Holm, C. (1990) Cloning by function: an alternative approach for identifying yeast homologs of genes from other organisms. Proc. Natl. Acad. Sci. USA 87, 6629– 6633. Kulkarni, M.S. & Sherman, F. (1994) NAT2, an essential gene encoding methionine N alpha-acetyltransferase in the yeast Saccharomyces cerevisiae. J. Biol. Chem. 269, 13141–13147. Lee, F.J., Lin, L.W. & Smith, J.A. (1989) Molecular cloning and sequencing of a cDNA encoding N alpha-acetyltransferase from Saccharomyces cerevisiae. J. Biol. Chem. 264, 12339–12343. Li, R. & Murray, A.W. (1991) Feedback control of mitosis in budding yeast. Cell 66, 519–531. Lillie, S.H. & Brown, S.S. (1992) Suppression of a myosin defect by a kinesin-related gene. Nature 356, 358–361. Linder, C. & Thoma, F. (1994) Histone H1 expressed in Saccharomyces cerevisiae binds to chromatin and affects survival, growth, transcription, and plasmid stability but does not change nucleosomal spacing. Mol. Cell Biol. 14, 2822–2835. Lowary, P.T. & Widom, J. (1989) Higher-order structure of Saccharomyces cerevisiae chromatin. Proc. Natl. Acad. Sci. USA 86, 8266–8270. Maniatis, T., Fritsch, E.F. & Sambrook, J. (1982) Molecular Cloning: a Laboratory Manual. Cold Spring Harbor, NY: Cold Spring Harbor Laboratory Press. Maruta, H., Greer, K. & Rosenbaum, J.L. (1986) The acetylation of a-tubulin and its relationship to the assembly and disassembly of microtubules. J. Cell. Biol. 103, 571–579. Matsuzaki, F., Matsumoto, S. & Yahara, I. (1988) Truncation of the carboxy-terminal domain of yeast b-tubulin causes temperature-sensitive growth and hypersensitivity to antimitotic drugs. J. Cell. Biol. 107, 1427–1435. McGhee, J.D. & Felsenfeld, G. (1980) Nucleosome structure. Annu. Rev. Biochem. 49, 1115–1156. Megee, P.C., Morgan, B.A., Mittman, B.A. & Smith, M.M. (1990) Genetic analysis of histone H4: essential role of lysines subject to reversible acetylation. Science 247, 841–845. Mullen, J.R., Kayne, P.S. & Moerschell, R.P., et al. (1989) Identification and characterization of genes and mutants for an Nterminal acetyltransferase from yeast. EMBO J. 8, 2067–2075. Nasmyth, K.A. & Reed, S.I. (1980) Isolation of genes by complementation in yeast. Proc. Natl. Acad. Sci. USA 77, 2119–2123. Novick, P., Osmond, B.C. & Botstein, D. (1989) Suppressors of yeast actin mutations. Genetics 121, 659–674. Park, E.C. & Szostak, J.W. (1990) Tl Point mutations in the yeast histone H4 gene prevent silencing of the silent mating type locus HML. Mol. Cell. Biol. 10, 4932–4934. Park, E.C. & Szostak, J.W. (1992) ARD1 and NAT1 proteins form a complex that has N-terminal acetyltransferase activity. EMBO J. 11, 2087–2093. Genes to Cells (1996) 1, 923–942 941 R Lin et al. Pedrotti, B., Colombo, R. & Islam, K. (1994) Microtubule associated protein MAP1A is an actin-binding and crosslinking protein. Cell Motil. Cytoskeleton 29, 110–116. Perry, C.A. & Annunziato, A.T. (1991) Histone acetylation reduces H1-mediated nucleosome interactions during chromatin assembly. Exp. Cell Res. 196, 337–345. Pringle, J.R. & Hartwell, L.H. (1981) The Saccharomyces cerevisiae cell cycle. In: The Molecular Biology of the Yeast Saccharomyces, Life Cycle and Inheritance (eds J.N. Strathern, E.W. Jones & J.R. Broach), pp. 97–143. Cold Spring Harbor, NY: Cold Spring Harbor Laboratory Press. Quinlan, R.A., Pogson, C.I. & Gull, K. (1980) The influence of the microtubule inhibitor, methyl benzimidazol-2-ylcarbamate (MBC) on nuclear division and the cell cycle in Saccharomyces cerevisiae. J. Cell Sci. 46, 341–352. Ridsdale, J.A., Hendzel, M.J., Delcuve, G.P. & Davie, J.R. (1990) Histone acetylation alters the capacity of the H1 histones to condense transcriptionally active/competent chromatin. J. Biol. Chem. 265, 5150–5156. Rose, M.D. & Fink, G.R. (1987) KAR1, a gene required for function of both intranuclear and extranuclear microtubules in yeast. Cell 48, 1047–1060. Rowley, R., Subramani, S. & Young, P.G. (1992) Checkpoint controls in Schizosaccharomyces pombe: rad1. EMBO J. 11, 1335– 1342. Russel, M., Kidd, S. & Kelly, M.R. (1986) An improved filamentous helper phage for generating single-stranded plasmid DNA. Gene 45, 333–338. Samejima, I., Beach, D., Matsumoto, T., Yanagida, M. & Nakaseko, Y. (1993) Identification of seven new cut genes involved in Schizosaccharomyces pombe mitosis. J. Cell Sci. 105, 135–143. Sanger, F., Nicklen, S. & Coulson, A.R. (1977) DNA sequencing with chain-terminating inhibitors. Proc. Natl. Acad. Sci. USA 74, 5463–5467. Schatz, P.J., Solomon, F. & Botstein, D. (1988) Isolation and characterization of conditional-lethal mutations in the TUB1 a-tubulin gene of the yeast Saccharomyces cerevisiae. Genetics 120, 681–695. Schmit, A.C. & Lambert, A.M. (1988) Plant actin filament and microtubule interactions during anaphase–telophase transition: effects of antagonist drugs. Biol. Cell 64, 309–319. Schuster, T., Han, M. & Grunstein, M. (1986) Yeast histone H2A and H2B amino-termini have interchangeable functions. Cell 45, 445–451. Sherman, F., Fink, G.R. & Hick, J.B. (1986) Methods in Yeast Genetics. Cold Spring Harbor, NY: Cold Spring Harbor Laboratory Press. Sikorski, R.S. & Boeke, J.D. (1990) In vitro mutagenesis and plasmid shuffling: from cloned gene to mutant yeast. Meth. Enzymol. 194, 302–318. Smith, M.M. (1991) Histone structure and function. Curr. Opin. Cell Biol. 3, 429–437. Srebreva, L., Zlatanova, J., Miloshev, G. & Tsanev, R. (1987) Immunological evidence for the existence of H1-like histone in yeast. Eur. J. Biochem. 165, 449–454. Stearns, T., Hoyt, M.A. & Botstein, D. (1990) Yeast mutants sensitive to antimicrotubule drug define three genes that affect microtubule function. Genetics 124, 251–262. Taunton, J., Hassug, C.A. & Schreiber, S.L. (1996) A mammalian histone deacetylase related to the yeast transcriptional regulator Rpd3p. Science 272, 408–272. 942 Genes to Cells (1996) 1, 923–942 Tercero, J.C. & Wickner, R.B. (1992) MAK3 encodes an N-acetyltransferase whose modification of the L-A gag NH2 terminus is necessary for virus particle assembly. J. Biol. Chem. 267, 20277–20281. Thompson, J.S., Ling, X. & Grunstein, M. (1994) Histone H3 amino terminus is required for telomeric and silent mating locus repression in yeast. Nature 369, 245–247. Turner, B.M., Birley, A.J. & Lavender, J. (1992) Histone H4 isoforms acetylated at specific lysine residues define individual chromosomes and chromatin domains in Drosophila polytene nuclei. Cell 69, 375–384. Uemura, T. & Yanagida, M. (1984) Isolation of type I and II DNA polymerase mutants from fission yeast: single and double mutants show different phenotypes in cell growth and chromatin organization. EMBO J. 3, 1737–1744. Uemura, T. & Yanagida, M. (1986) Mitotic spindle pulls but fails to separate chromosomes in type II DNA topoisomerase mutants: uncoordinated mitosis. EMBO J. 5, 1003–1010. Umesono, K.T., Toda, T., Hayashi, S. & Yanagida, M. (1983) Two cell division cycle genes NDA2 and NDA3 of the fission yeast Schizosaccharomyces pombe control microtubular organization and sensitivity to anti-mitotic benzimidazol compounds. J. Mol. Biol. 168, 271–284. Urbancikova, M. & Hitchcock-DeGregori, S.E. (1994) Requirement of amino-terminal modification for striated muscle a-tropomyosin function. J. Biol. Chem. 269, 24310–24315. Ursic, D. & Culbertson, M.R. (1991) The yeast homolog to mouse TOP-1 affects microtubule-mediated processes. Mol. Cell. Biol. 11, 2629–2640. Ursic, D., Sedbrook, J.C., Himmel, K.L. & Culbertson, M.R. (1994) The essential yeast Tcp1 protein affects actin and microtubules. Mol. Biol. Cell 5, 1065–1080. Wallis, J.W., Rykowski, M. & Grunstein, M. (1983) Yeast histone H2B containing large amino terminus deletions can function in vivo. Cell 35, 711–719. Whiteway, M. & Szostak, J.W. (1985) The ARD1 gene of yeast functions in the switch between the mitotic cell cycle and alternative developmental pathways. Cell 43, 483–492. Whitters, E.A., Cleves, A.E., McGee, T.P., Skinner, H.B. & Bankaitis V.A. (1993) Sac1p is an integral membrane protein that influences the cellular requirement for phospholipid transfer protein function and inositol in yeast. J. Cell. Biol. 122, 79–94. Wolffe, A.P. (1994) Nucleosome positioning and modification: chromatin structures that potentiate transcription. Trends Biol. Sci. 19, 240–244. Wolffe, A.P. & Pruss, D. (1996a) Deviant nucleosomes: the functional specialization of chromatin. Trends Genet. 12, 58–62. Wolffe, A.P. & Pruss, D. (1996b) Targeting chromatin disruption: Transcription regulators that acetylate histones. Cell 84, 817–819. Zagursky, R.J. & Berman, M.L. (1984) Cloning vectors that yield high levels of single-stranded DNA for rapid sequencing. Gene 27, 183–191. Zieler, H.A., Walberg, M. & Berg, P. (1995) Suppression of mutations in two Saccharomyces cerevisiae genes by the adenovirus E1A protein. Mol. Cell Biol. 15, 3227–3237. Zlatanova, J. (1990) Histone H1 and the regulation of transcription of eukaryotic genes. TIBS 15, 273–276. Received: 6 June 1996 Accepted: 11 October 1996 q Blackwell Science Limited