A new role for MCM1 in yeast: cell cycle regulation of SW15 transcription

In the yeast Saccharomyces cerevisiae cell cycle-regulated SW15 transcription is essential for ensuring that mother and not daughter cells switch mating type. We have identified a 55-bp promoter sequence that appears to be responsible for restricting transcription to the late S, G2, and M phases of the cell cycle. Two proteins, MCM1, a transcription factor described previously, and SFF (SWI five factor, a newly identified factor) bind this sequence in vitro. MCM1 binds the DNA tightly on its own, but SFF will only bind as part of a ternary complex with MCM1. We observe a strong correlation between the ability of mutated SWI5 promoter sequences to form a ternary MCM1-SFF-containing complex in vitro and to activate transcription in vivo, which suggests that efficient transcription requires that both proteins bind DNA. Through its interactions with cell type-specific coactivators and corepressors, MCM1 controls cell type-specific expression of pheromone and receptor genes. By analogy, we propose that SFF enables MCM1 to function as a part of a cell cycle-regulated transcription complex.

Changes in a SWI4,6-DNA-binding complex occur at the time of HO gene activation in yeast

The yeast HO gene is transcribed transiently during G1 as cells undergo START. START-specific HO activation requires two proteins, SWI4 and SWI6, which act via a motif (CACGA4) repeated up to 10 times within the URS2 region of the HO promoter. We identified a DNA-binding activity containing SWI4 and SWI6 that recognizes the CACGA4 sequences within URS2. Two forms of SWI4,6-DNA complexes called L and U can be distinguished by their electrophoretic mobility. L complexes can be detected at all stages of the cell cycle, but U complexes are only detected in cells that have undergone START. The formation of U complexes may be the trigger of HO activation. The SWI6 protein is concentrated in the nucleus throughout G1, but at some point in S or G2 significant amounts accumulate in the cytoplasm. This change in cellular location of the SWI6 protein might contribute to the turnoff of HO transcription after cells have undergone START.

The role of SWI4 and SWI6 in the activity of G1 cyclins in yeast

Entry into the mitotic cycle (START) requires a protein kinase encoded by the CDC28 gene and one of three redundant G1-specific cyclins encoded by CLN1, -2, and -3. SWI4 and SWI6 are transcription factors required for the START-dependent activation of the HO endonuclease gene. They also fulfill an overlapping but essential role for cell division since cells deleted for both genes are inviable. We show that the essential role of SWI4 and SWI6 is to ensure the activity of G1-specific cyclin genes. SWI4 and SWI6 appear necessary for the transcription of CLN1 and CLN2, but not for that of CLN3. CLN3 function is, however, also dependent on SWI4 and SWI6.

The role of phosphorylation and the CDC28 protein kinase in cell cycle-regulated nuclear import of the S. cerevisiae transcription factor SWI5

The intracellular localization of the S. cerevisiae transcription factor SWI5 is cell cycle dependent. The protein is nuclear in G1 cells but cytoplasmic in S, G2, and M phase cells. We have identified SWI5's nuclear localization signal (NLS) and show that it can confer cell cycle-dependent nuclear entry to a heterologous protein. Located within or close to the NLS are three serine residues, mutation of which results in constitutive nuclear entry. These residues are phosphorylated in a cell cycle-dependent manner in vivo, being phosphorylated when SWI5 is in the cytoplasm and dephosphorylated when it is in the nucleus. As all three serines are phosphorylated by purified CDC28-dependent H1 kinase activity in vitro, we propose a model in which the CDC28 kinase acts directly to control nuclear entry of SWI5.

Positive feedback in the activation of G1 cyclins in yeast

Yeast cells become committed to the mitotic cell cycle at a stage during G1 called Start. To enter Start, cells must grow to a critical size. They also require the CDC28 protein kinase and at least one of three G1-specific cyclins encoded by CLN1, 2, and 3. It is thought that Start is triggered by the accumulation of G1 cyclins that bind to the CDC28 kinase and activate it. So what determines the accumulation of G1 cyclins? For CLN1 and CLN2, transcriptional activation could be involved because their RNAs appear transiently during the cell cycle as cells undergo Start. Here we report that the appearance of CLN1 and CLN2 RNAs depends on an active CDC28 kinase and is stimulated by CLN3 activity. We propose that CDC28 kinase activity due to CLN1 and CLN2 proteins arises through a positive feedback loop which allows CLN proteins to promote their own synthesis.

A general approach to the isolation of cell cycle-regulated genes in the budding yeast, Saccharomyces cerevisiae

We describe a general approach to the isolation of cell cycle-dependently regulated transcripts in Saccharomyces cerevisiae. This approach is based on the physical identification of cell cycle-regulated transcripts by Northern hybridization using as probes yeast DNA isolated from an ordered S. cerevisiae genomic library. The purpose of this is twofold; first, to assess the importance of transcriptional regulation in cell cycle control; and second, to identify novel genes that may have important roles in the eukaryotic cell cycle. We report the isolation of two previously uncharacterized genes that are transcribed at points in the cell cycle to which specific transcriptional activation has not been assigned: namely, mitosis and early G1 phase. It is argued that these transcripts serve as important landmarks for cell cycle events that are not readily distinguished by either morphological or cytological criteria. The cell cycle-dependent transcription of the RNR1 and CLN1 genes is also described and the implications for cell cycle control, in G1, are discussed with reference to these two genes.

The role of CDC28 and cyclins during mitosis in the budding yeast S. cerevisiae

cdc28-1N is a conditional allele that has normal G1 (Start) function but confers a mitotic defect. We have isolated seven genes that in high dosage suppress the growth defect of cdc28-1N cells but not of Start-defective cdc28-4 cells. Three of these (CLB1, CLB2, and CLB4) encode proteins strongly homologous to G2-specific B-type cyclins. Another gene, CLB3, was cloned using PCR, CLB1 and CLB2 encode a pair of closely related proteins; CLB3 and CLB4 encode a second pair. Neither CLB1 nor CLB2 is essential; however, disruption of both is lethal and causes a mitotic defect. Furthermore, the double mutant cdc28-1N clb2::LEU2 is nonviable, whereas cdc28-4 clb2::LEU2 is viable, suggesting that the cdc28-1N protein may be defective in its interaction with B-type cyclins. Our results are consistent with CDC28 function being required in both G1 and mitosis. Its mitotic role, we believe, involves interaction with a family of at least four G2-specific cyclins.

The identification of a second cell cycle control on the HO promoter in yeast: cell cycle regulation of SW15 nuclear entry

HO encodes a site-specific endonuclease that initiates mating type switching in S. cerevisiae. It is expressed only transiently during the cell cycle of mother cells, as they undergo Start, but not in daughter cells. Since SWI5 appears to be the only HO transcription factor missing when daughter cells undergo Start, we were interested in the intracellular distribution of SWI5 at cell division. We discovered that SWI5 is found equally concentrated in the nuclei of both mother and daughter cells at the end of anaphase, suggesting that its subsequent fate must somehow differ. Prior to the end of anaphase, SWI5 accumulates in the cytoplasm and only moves into the nucleus when cells enter G1. A version of the HO promoter that has lost its dependence on Start is nevertheless still strongly cell cycle regulated and is activated when SWI5 moves into the nucleus.

Mating-type control in Saccharomyces cerevisiae: isolation and characterization of mutants defective in repression by a1-alpha 2

The alpha 2 protein, the product of the MAT alpha 2 cistron, represses various genes specific to the a mating type (alpha 2 repression), and when combined with the MATa1 gene product, it represses MAT alpha 1 and various haploid-specific genes (a1-alpha 2 repression). One target of a1-alpha 2 repression is RME1, which is a negative regulator of a/alpha-specific genes. We have isolated 13 recessive mutants whose a1-alpha 2 repression is defective but which retain alpha 2 repression in a genetic background of ho MATa HML alpha HMRa sir3 or ho MAT alpha HMRa HMRa sir3. These mutations can be divided into three different classes. One class contains a missense mutation, designated hml alpha 2-102, in the alpha 2 cistron of HML, and another class contains two mat alpha 2-202, in the MAT alpha locus. These three mutants each have an amino acid substitution of tyrosine or acid substitution of tyrosine or phenylalanine for cysteine at the 33rd codon from the translation initiation codon in the alpha 2 cistron of HML alpha or MAT alpha. The remaining 10 mutants make up the third class and form a single complementation group, having mutations designated aar1 (a1-alpha 2 repression), at a gene other than MAT, HML, HMR, RME1, or the four SIR genes. Although a diploid cell homozygous for the aarl and sir3 mutations and for the MATa, HML alpha, and HMRa alleles showed alpha mating type, it could sporulate and gave rise to asci containing four alpha mating-type spores. These facts indicate that the domain for alpha2 repression is separable from that for a1-alpha2 protein interaction or complex formation in the alpha2 protein and that an additional regulation gene, AAR1, is associated with the a1-alpha2 repression of the alpha1 cistron and haploid-specific genes.

Zinc-finger motifs expressed in E. coli and folded in vitro direct specific binding to DNA

The short sequence motif named 'zinc finger', first recognized repeated in tandem in the Xenopus transcription factor IIIA (TFIIIA), is also found in the yeast transcriptional activator SWI5 (ref. 3) and many other regulator proteins. Embedded in the 709-amino-acid polypeptide chain of SWI5 are three tandemly repeated zinc-finger motifs. Because the zinc fingers of TFIIIA are known to bind to DNA, it is probable that in the case of SWI5 these finger motifs also play an important, but not necessarily exclusive, role in the sequence-specific binding of the protein to DNA. To test this prediction we have expressed the 89-amino-acid sequence of the domain containing the three zinc fingers of SWI5 in Escherichia coli as a cleavable fusion protein, purified under denaturing conditions and folded in vitro. This experimental approach allows us to study directly both the metal requirement and DNA-binding properties of the isolated polypeptide. We find that zinc is required for specific DNA recognition and, most significantly, DNaseI protection studies show that the isolated three-fingered domain is sufficient for sequence-specific binding to DNA.

Purification and cloning of a DNA binding protein from yeast that binds to both silencer and activator elements

A DNA binding protein (RAP1, previously called SBF-E) has been shown to bind to putative regulatory sites at both yeast mating-type silencers, yet is not the product of genetically identified regulators of the silent loci. Here, we report the purification of RAP1 by DNA affinity chromatography, and the isolation of its gene from a lambda gt11 genomic library using antibodies raised against the protein. Disruption of the chromosomal copy of this gene is lethal. We show that RAP1 protein also binds in vitro to the upstream activation site (UAS) of MAT alpha and ribosomal protein genes. In addition, we show that two different UAS-associated RAP1 binding sites can substitute in vivo for a silencer binding site. Our results suggest that RAP1 may be a transcriptional regulator that can play a role in either repression or activation of transcription, depending upon the context of its binding site.

A yeast silencer contains sequences that can promote autonomous plasmid replication and transcriptional activation

Repression of the yeast silent mating type loci requires cis-acting sequences located over 1 kb from the regulated promoters. One of these sites, a "silencer," exhibits enhancer-like distance- and orientation-independence. The silencer demonstrates both autonomous replication sequence (ARS) activity and a centromere-like segregation function, suggesting roles for DNA replication and segregation in transcriptional repression. Here we identify three sequences (A, E, and B) involved both in repression and in either ARS or segregation activity. The sequences are functionally redundant: no one is essential for complete transcriptional control, but mutations in any two inactivate the silencer. Surprisingly, elements E and B can each activate transcription from heterologous promoters, and E shows striking homology to several yeast upstream activation sequences. Therefore, sequences individually involved in replication, segregation, and transcriptional activation can, at the silencer, efficiently repress transcription.

Similarity between cell-cycle genes of budding yeast and fission yeast and the Notch gene of Drosophila

The HO gene of Saccharomyces cerevisiae encodes the endonuclease that initiates mating-type switching. To prevent inopportune switching, HO transcription is restricted to a specific period in the haploid cell cycle, which is just after, and dependent on, the start of the mitotic cell cycle. A repeated promoter element (CACGA4) (refs 7-9) and two trans-acting activators (SWI4 and SWI6) have been identified, which are responsible for the periodic and start-dependent transcription of HO. To understand further the link between start and HO transcription, the SWI6 gene has been cloned and sequenced. The SWI6 protein is similar to the protein in Schizosaccharomyces pombe that is encoded by cdc10 an essential gene specifically required at the start of the cell cycle. The similarity between the SWI6 and cdc10 products, and their common involvement with 'start', suggest that they may share a common mechanism for sensing or executing this critical control step in the cell cycle. The SWI6 and cdc10 proteins also contain two copies of a repeated motif that occurs at least five times in the cytoplasmic domain of the Notch protein of Drosophila melanogaster.

Transcriptional regulation in the yeast life cycle

The transition from haploid to diploid in homothallic yeast involves a defined sequence of events which are regulated at the level of transcription. Transcription factors encoded by SWI genes activate the HO endonuclease gene at a precise stage in the cell cycle of mother cells. The HO endonuclease initiates a transposition event which activates genes of the opposite mating type by causing them to move away from a silencer element. The activated mating type genes then regulate genes involved in cell signaling such as the mating type-specific pheromones and their receptors. Since HO is only activated in one of the sister cells after division (the mother), adjacent cells of opposite mating type are generated which respond to each others' secreted pheromones by inducing genes involved in conjugation. This leads to the formation of a diploid in which many of the genes involved in mating and mating-type switching become repressed due to the heterozygosity of the mating-type locus. This article summarizes what is known about these transcriptional controls and discusses possible parallels in higher eukaryotes.

Cell cycle regulation of SW15 is required for mother-cell-specific HO transcription in yeast

In haploid homothallic yeast, cell division gives rise to a mother cell that transiently transcribes the HO gene (as it undergoes START) and a daughter cell that does not. Consequently, only mother cells switch their mating types. Here, we test the proposition that a transcription factor called SWI5 is the "determinant" of mother-cell-specific HO transcription; that is, that SWI5 is the only factor missing in daughter cells. We show that SWI5 RNAs are cell-cycle regulated so that they are only produced after the post-START window of HO transcription. This regulation is vital for mother-cell specificity since constitutive transcription of SWI5 causes daughter cells to switch their mating types. We propose that SWI5 gene products are partitioned asymmetrically at cell division.

Both positive and negative regulators of HO transcription are required for mother-cell-specific mating-type switching in yeast

The HO gene, which encodes an endonuclease responsible for initiating mating type switching in yeast, is transcribed at START during the cell cycle of mother cells but not at all during the cell cycle of daughter cells. At least six genes, called SWI1-6, are necessary for HO transcription. We describe the isolation and characterization of mutations in two new genes called SDI1 and SDI2, which partially suppress the requirement for SWI5 and which cause daughter cells to express HO. The analysis of mating type switching in swi5- sdi1- and SWI5+ sdi1- strains suggests that the mother cell specificity of HO transcription is due exclusively to the selective action of SWI5 in mother cells. SDI1 encodes (or regulates) a repressor protein that binds to the HO promoter and prevents HO transcription in daughter cells by causing HO to be fully SWI5 dependent.

Cell cycle control of the yeast HO gene: cis- and trans-acting regulators

In this paper, we investigate the role of a short repeated sequence (CACGA4) in the cell-cycle regulation of HO. We show that this sequence activates transcription of a heterologous gene in a cell-cycle-dependent fashion indistinguishable from that of the wild-type HO promoter. We also show that, in addition to SWI1 through SWI5, at least five other genes (SWI6 through SWI10) are required for HO transcription. These genes fit into three distinct classes with respect to their targets within the HO promoter. SWI4 and SWI6 are specifically required for CACGA4-mediated activation of transcription. SWI1, SWI2, and SWI5 are required for transcription from sequences physically separate from and independent of the CACGA4 sequences. SWI3 may be required for both. Since all the SWI genes are required for HO transcription, the HO promoter must contain at least two essential upstream activation sequences, which are affected by different trans-acting factors and are subject to different types of control.

The determination of mother cell-specific mating type of switching in yeast by a specific regulator of HO transcription

In haploid homothallic budding yeast, cell division gives rise to a mother cell which proceeds to switch its mating type and a daughter cell (the bud) which does not. Switching is initiated by a specific double strand cleavage of mating type DNA by an endonuclease encoded by the HO gene. Previous data suggest that the pattern of HO transcription is responsible for the mother cell specificity of switching. HO is transcribed transiently, at START, during the cell cycle of mother cells but not at all during the cell cycle of daughter cells. The HO promoter is complex. Sequences between -1000 and -1400 (called URS1) are essential for transcription, whereas sequences between -150 and -900 (called URS2) are necessary for cell cycle control. Moreover, 10 trans-acting gene products called SWI1-10 are necessary for maximum expression. In an attempt to identify the cis-acting DNA sequences which are responsible for mother cell specificity and to identify which SW1 genes are involved, a hybrid GAL/HO promoter was constructed in which the upstream activation region putatively involved in mother cell-specific activation (URS1) is replaced by the upstream activation region of the GAL1-10 promoter. The properties of this hybrid promoter show, for the first time, that: (i) the HO promoter is modular since mother cell specificity can be replaced by galactose dependence without compromising cell cycle control or a/alpha repression; (ii) transcription of HO is indeed the major rate-limiting event for switching which is absent in daughter cells; (iii) SWI1,2,3, 4,6,7,8.9 and 10 are unlikely to be involved in mother cell specificity but SW15 probably is.

At least 1400 base pairs of 5'-flanking DNA is required for the correct expression of the HO gene in yeast

Homothallic yeast cells undergo a specific pattern of mating-type switching initiated by an endonuclease encoded by the HO gene. HO transcription is affected by cell type (a, alpha, and a/alpha), by cell age (mother or daughter), and by the cell cycle. This paper investigates the sequences involved in HO transcription by replacing genomic DNA with copies mutated in vitro. A region between -1000 and 1400 (called URS1) is necessary for transcription in addition to a "TATA"-like region at -90. The 900 bp of DNA separating URS1 from the "TATA" box is not necessary for transcription nor for a/alpha repression and some measure of mother/daughter control, but it is necessary for correct cell cycle control.

A repetitive DNA sequence that confers cell-cycle START (CDC28)-dependent transcription of the HO gene in yeast

The interconversion of mating types in yeast is initiated by an endonuclease encoded by the HO gene. HO is transcribed only transiently during the G1 phase of the cell cycle, as cells undergo START. A deletion analysis of HO 5'-flanking DNA suggests that there must be multiple copies of the sequence that confers START-specific transcription in the interval from -150 to -900. Analysis of this interval revealed 10 occurrences of sequences that closely match the consensus Pur N N Pyr C A C G A4. To test whether these sequences are the putative cell-cycle control elements, synthetically derived copies of the consensus were inserted at the break points of constitutive deletions and shown to restore START-dependent cell-cycle control.

Characterization of a "silencer" in yeast: a DNA sequence with properties opposite to those of a transcriptional enhancer

The mating type of yeast is determined by the allele, either a or alpha, at the MAT locus. Two other loci, HML and HMR, contain complete copies of the alpha and a genes, respectively, which are not expressed. The four SIR gene products are required in trans for repression of the silent loci, as are cis-acting sites on either side of HML and HMR, about 1000 bp from the mating-type promoters. We demonstrate that one of these cis-acting sequences, HMRE, is able to switch off at least two nonmating-type promoters. In common with enhancers, it is able to function in either orientation, relatively independently of its position with respect to the regulated promoter, and can act on promoters 2600 bp away. However since HMRE represses, rather than enhances, transcription we have called it a "silencer" sequence.

Molecular analysis of a cell lineage

Mating type interconversion has a precise lineage which serves to minimize the time taken for yeast cells to achieve the diploid state. The HO gene either encodes or regulates an endonuclease which initiates the interconversion process. Expression of this gene is switched on during the G1 phase of mother cells and not at all during the cell cycle of their daughters. This behaviour can explain what is known about the lineage.

Cell division cycle mutants altered in DNA replication and mitosis in the fission yeast Schizosaccharomyces pombe

A total of 59 new temperature sensitive cdc mutants are described which grow normally at 25 degrees C but become blocked at DNA replication or mitosis when incubated at 36 degrees C. Thirty-nine of the mutants are altered in cdc genes which have been identified previously. The remaining 20 mutants define 10 new cdc genes. These have been characterised physiologically, and 6 of the genes (cdc 17, 20, 21, 22, 23, 24) were found to be required for DNA replication, 2 for mitosis (cdc 27, 28), and 2 (cdc 18, 19), could not be unambiguously assigned to either DNA replication or mitosis but were definitely required for one or the other. Three genes, the previously identified cdc 10, and cdc 20, 22 are likely to be required for the initiation of DNA replication. Mutants in two genes, cdc 17, 24 undergo bulk DNA synthesis at 36 degrees C, but this DNA is defective. In the case of cdc 17 the defect is in the ligation of Okazaki fragments. cdc 23 is required for bulk DNA synthesis, whilst cdc 21 may possibly be required for the initiation of a particular sub-set of replicons. A previously isolated mutant cdc 13.117 is also further described. This mutant becomes blocked in the middle of mitosis with apparently condensed chromosomes.

The Effect of Cell Mass on the Cell Cycle Timing and Duration of S-Phase in Fission Yeast

Request for reprints to Paul Nurse.

Two isotopic methods for measuring DNA replication in the fission yeast Schizosaccharomyces pombe are described. The first is a method for measuring the total quantity of [3H]uracil incorporated into DNA after pulse labelling. The second is a means of detecting DNA replication in single cells by autoradiography. Both of these techniques have been used to investigate the timing and duration of S-phase in a series of mutant strains whose cell mass at division varies over a 3-fold range. The results support the hypothesis that in S. pombe there are 2 different controls over the timing of S-phase: an attainment of a critical cell mass and a dependency upon the completion of the previous mitosis coupled with a short minimum time in G1. Strains whose cell mass at birth is above this critical level initiate DNA replication almost immediately after septation, that is, very soon after the previous mitosis. Strains whose cell mass at birth is below the critical level do not initiate replication until the critical cell mass is attained. The duration of S-phase has been estimated from the proportion of cells whose nuclei are labelled after a pulse of given duration. S-phase is short in S. pombe, lasting only about 0.1 of a cell cycle in wild type. Cell mass at S-phase does not have any consistent effect on this length. We have also investigated the degree of synchrony of S-phase initiation in daughter cells, and have found that, in a cell cycle 240 min long, their S-phases are initiated within 1–2 min of each other. This result indicates that between sisters variability in the duration of the G1 phase is small compared with variability in the total cell cycle time, and argues against the hypothesis that the rate of cell cycle traverse is determined by a random transition in G1.

Genetic control of the cell division cycle in the fission yeast Schizosaccharomyces pombe

Twenty seven recessive temperature sensitive mutants have been isolated in Schizosaccharomyces pombe which are unable to complete the cell division cycle at the restrictive temperature. These mutants define 14 unlinked genes which are involved in DNA synthesis, nuclear division and cell plate formation. The products from most of these genes complete their function just before the cell cycle event in which they are involved. Physiological characterisation of the mutants has shown that DNA synthesis and nuclear division form a cycle of mutually dependent events which can operate in the absence of cell plate formation. Cell plate formation itself is usually dependent upon the completion of nuclear division.