Logo of molcellbPermissionsJournals.ASM.orgJournalMCB ArticleJournal InfoAuthorsReviewers
Mol Cell Biol. May 1999; 19(5): 3312–3327.

NDD1, a High-Dosage Suppressor of cdc28-1N, Is Essential for Expression of a Subset of Late-S-Phase-Specific Genes in Saccharomyces cerevisiae


cdc28-1N mutants progress through the G1 and S phases normally at the restrictive temperature but fail to undergo nuclear division. We have isolated a gene, NDD1, which at a high dosage suppresses the nuclear-division defect of cdc28-1N. NDD1 (nuclear division defective) is an essential gene. Its expression during the cell cycle is tightly regulated such that NDD1 RNA is most abundant during the S phase. Cells lacking the NDD1 gene arrest with an elongated bud, a short mitotic spindle, 2N DNA content, and an undivided nucleus, suggesting that its function is required for some aspect of nuclear division. We show that overexpression of Ndd1 results in the upregulation of both CLB1 and CLB2 transcription, suggesting that the suppression of cdc28-1N by NDD1 may be due to an accumulation of these cyclins. Overproduction of Ndd1 also enhances the expression of SWI5, whose transcription, like that of CLB1 and CLB2, is activated in the late S phase. Ndd1 is essential for the expression of CLB1, CLB2, and SWI5, since none of these genes are transcribed in its absence. Both CLB2 expression and its upregulation by NDD1 are mediated by a 240-bp promoter sequence that contains four MCM1-binding sites. However, Ndd1 does not appear to be a component of any of the protein complexes assembled on this DNA fragment, as indicated by gel mobility shift assays. Instead, overexpression of NDD1 prevents the formation of one of the complexes whose appearance correlates with the termination of CLB2 expression in G1. The inability of GAL1 promoter-driven CLB2 to suppress the lethality of NDD1 null mutant suggests that, in addition to CLB1 and CLB2, NDD1 may also be required for the transcription of other genes whose functions are necessary for G2/M transition.

An orderly progression through the cell division cycle requires that the relevant cellular events occur in a strict temporal sequence. In the budding yeast Saccharomyces cerevisiae, as in many other organisms, this is achieved by a combination of transcriptional and posttranslational regulations of various effectors. One of the best-studied examples of such posttranslational controls is the way in which the activity of the key protein kinase Cdc28 is regulated. Phosphorylation of specific tyrosine and threonine residues (19, 34, 35, 49), the association with a variety of cyclins (14, 43, 48, 53), and their proteolytic destruction at various stages of the cell cycle (2, 3, 10, 18) all contribute to sharply define not only the timing of Cdc28 activation-inactivation but also, presumably, its substrate specificity (35). Similarly, an anaphase-promoting complex whose activity is crucial for chromosome segregation appears to be regulated at the posttranslational level (24, 54, 55).

Although posttranslational modifications of the mitotic regulators are perhaps the most effective way for the cell to quickly respond to the changing cellular context, these responses may be further sharpened by controls at the transcriptional level. Indeed, it has been estimated by DNA microarray hybridization that about 800 genes of the budding yeast S. cerevisiae are transcribed only at specific stages of the cell cycle (50). Many of these periodically transcribed genes play crucial roles in cell cycle progression. For instance, the transcripts for the G1 cyclins CLN1 and CLN2 appear in late G1 phase approximately when the cells traverse START. The transcription of these genes is regulated by a sequence motif called SCB (Swi4/Swi6 cell cycle box) within their promoter (22). This motif is also found in the promoters of the genes encoding HO endonuclease and the cyclin-like protein Hcs26 (6, 11), which are also expressed in late G1. A specific heteromeric transcription factor complex (SCF) containing Swi4 and Swi6 proteins drives the cell cycle stage-specific expression by binding to the SCB element (37, 38). The S-phase cyclin genes CLB5 and CLB6 and DNA synthesis genes such as POL1, TMP1, and CDC9 are also expressed in late G1 but their expression is driven by another promoter element called MCB (Mlu1 cell cycle box) which binds a transcription factor complex (MBF) containing Mbp1 and Swi6 (23, 28). At the onset of the S phase, histones and, somewhat later, CLB3 and CLB4 cyclin genes are transcribed (15).

The transcription of another set of genes is activated near the completion of the S phase. This includes FAR1, a protein required for pheromone-induced G1 arrest, the transcription factors SWI5 and ACE2, mitotic cyclin genes CLB1 and CLB2, and CDC5, a gene encoding a Polo-like protein kinase. It is not clear whether the periodic expression of these genes is regulated by a common promoter element and transcription factors. An analysis of SWI5 promoter revealed a 55-bp element that is sufficient for the periodic expression of SWI5 (29). It has been proposed that Mcm1, a DNA-binding protein important for the determination of mating-type specificity, forms a complex with Sff (SWI5 transcription factor) and binds to this 55-bp promoter element (29). The gene encoding Sff has not yet been identified. In subsequent studies, it was shown that Mcm1 is required for the expression of CLB1, CLB2, SWI5, CDC5, and ACE2 (1, 30). This is consistent with the fact that the promoter regions of CLB1, CLB2, and SWI5 contain consensus Mcm1 binding sites. Interestingly, Mcm1 occupies these sites throughout the cell cycle, implying that the timing of expression is perhaps determined either by specific modifications of the bound factors or by other associated factors (1). It has been recently reported that a novel promoter element ECB (early cell cycle box), which renders the expression of SWI4, CDC6, CDC47, and CDC46 M/G1 specific, also binds Mcm1 (32). Since Mcm1 is also involved in the regulation of other cellular processes that are not directly linked to the cell cycle progression (26, 31, 33, 51), it seems unlikely that Mcm1 is the critical factor in determining the periodicity of gene expression.

As Cdc28 is the key protein kinase required at various stages for cell cycle progression, most CDC28 mutants exhibit defects in progression through both START and G2/M transition (39, 41). However, cdc28-1N is a unique mutant in that, although normal with respect to its START functions, it fails to undergo nuclear division at the restrictive temperature and, consequently, arrests with a large elongated bud, a short mitotic spindle, 2N DNA content, high histone H1 kinase activity, and an undivided nucleus (39, 53). Although the molecular nature of its mitotic defect is still unknown, the cdc28-1N mutant has served well for the identification of a number of novel genes. CLB1, CLB2, and CLB4 were first isolated as allele-specific, high-dosage suppressors of the cdc28-1N mutation (53). Similarly, CIM3 and CIM5 genes encoding the 26S proteosome subunits were identified as mutations that exhibit synthetic lethal behavior in combination with cdc28-1N (17). We report here the isolation of a novel allele-specific, high-dosage suppressor of cdc28-1N, termed NDD1, and examine its role in cell cycle progression. We describe its basic characterization and show that NDD1 is an essential component, as is MCM1, of the mechanism that activates the expression of a set of late-S-phase-specific genes.


Yeast strains and growth media.

Yeast strains (Table (Table1)1) used in this study were derivatives of the standard wild-type strain W303. Strain BJ2168 (US1005) was used in the mobility shift DNA-binding assay. Yeast cells were routinely grown in yeast extract-peptone medium containing adenine (50 mg/liter) supplemented with either glucose or galactose. Raffinose was also added to the galactose medium. All media were prepared as described by Rose et al. (44).

Yeast strains

To obtain cultures synchronized in G1, bar1Δ cells (US356) were treated with α-factor (0.8 μg/ml) for 3 h. In experiments where a synchronous release from late telophase was required, cdc15 cells were grown to log phase at 24°C before they were filtered and resuspended in prewarmed growth medium at 37°C. After 3 h, cells were allowed to resume cell cycle progression at 24°C.

Isolation of NDD1.

cdc28-1N cells were transformed with a library of yeast genomic fragments cloned into the BamHI site of the YEp13 (2μm) vector (36), allowed to recover at 24°C for 24 h, and then shifted to 37°C for 3 days. The plasmids were retrieved from various transformants and were categorized into seven distinct groups based on the patterns of fragments generated by restriction enzyme digestion. One group of plasmids harbored a 2.7-kb PvuII genomic fragment containing the NDD1 gene.

DNA manipulations.

All DNA manipulations were performed as described by Sambrook et al. (46). The 2.7-kb PvuII fragment containing the entire NDD1 gene was blunt-end cloned into the SmaI-BglII site of pIC19H vector to yield pUS529. To construct the NDD1 gene under the control of GAL1 promoter, an NcoI site was created immediately upstream of the ATG start codon by PCR. The BamHI-BglII fragment from pUS529 was then replaced by the BamHI-BglII-digested 0.97-kb PCR product. The complete NDD1 gene was then excised with NcoI-HindIII and blunt-end ligated to the GAL1 promoter in a URA3- or TRP1-selectable CEN vector.

Gene disruptions were performed by using the one-step gene replacement method of Rothstein (45) with a 2.7-kb PvuII fragment in which the 450-bp EcoRV fragment within the open reading frame (ORF) was replaced by a 1.6-kb XhoI-SalI fragment containing the LEU2 gene. The disruption of the NDD1 gene was confirmed by Southern blot analysis. To tag NDD1 with three tandem copies of the hemagglutinin (HA) epitope (GRIFYPYDVPDYAGYPYDVPDYAGSYPYDVPDYAAQC), a NotI site was introduced just after the ATG start codon by PCR. The triple HA tag (as a 111-bp NotI fragment) was cloned into the newly created NotI site. The resulting HA3-tagged NDD1 was excised as an NcoI-SmaI fragment and blunt-end ligated to the GAL1 promoter in a TRP1-selectable CEN vector to yield pGAL1-HA3-NDD1.

To test the upstream activation sequence (UAS) activity, CLB2 promoter sequences were cloned upstream of a CYC1 TATA box fused to a ubiY lacZ reporter, which expresses β-galactosidase activity with a half-life of 10 min instead of more than 20 h (8). The 340-bp CLB2 UAS (extending from −863 to −523 bp from ATG) was generated by PCR and cloned into the XhoI-BglII site of pDL1498 plasmid containing the ubiY lacZ reporter (29) as a SalI-BglII fragment (sites introduced by PCR). The 240-bp CLB2 UAS (−863 to −627 bp from ATG) was made by dropping out a 110-bp XhoI-BglII fragment from the 340-bp CLB2 UAS reporter construct and religating the plasmid. The 55-bp CLB2 UAS (−698 to −643 bp from ATG) was synthesized by mutually priming two oligonucleotides (5′-TACAGAATTCTCGAGAATATAGCGACCGAATCAGGAAAAGGTCAACAACGA-3′ and 5′-ACTGAATTCAGATCTCATCCATATCGCGAACTTCGTTGTTGACCTTTTCC-3′) as described in Ausubel et al. (7) and cloned into pDL1498 as a XhoI-BglII fragment (sites introduced by PCR). The recombinant reporter constructs were excised by SmaI-NcoI digestion and blunt-end cloned into the EcoRI site of a URA3-based integrative plasmid and subsequently integrated at the URA3 locus by homologous recombination. All integration events were checked by Southern blot analysis, and β-galactosidase activities were measured in at least three independent isolates from each integrative transformation. Activities were expressed as Miller units.

To clone the full-length NDD1 in frame with the GAL4 DNA-binding domain of pGBT9 (9), NDD1 was excised from pUS529 as a NcoI-XhoI fragment and blunt-end ligated into the SmaI site of pGBT9. For GAL4-NDD1 deletion mutant analysis, all constructs were made similarly except that different restriction enzymes were used to excise the various NDD1 fragments from pUS529. The isolated fragments were subsequently cloned in frame into pGBT9 to yield the various GAL4-NDD1 deletion fusion constructs (see Table Table2).2). NDD1 lacking the polyglutamine domain was constructed by PCR with the following sets of primers: 5′-TTGATTGGATCCATGGACAGAGATATAAGC-3′ and 5′-TCTGCTGATGCTGCAGTAATATAC-3′ to obtain the sequence 5′ of the polyglutamine domain and 5′-TTCTTCTGTTCTGCAGTTCGGCAAC-3′ and 5′-TATTGTTAGATCTTAGCGGCGTTCT-3′ to obtain the sequence 3′ of the polyglutamine domain. The two PCR products were digested with NcoI-PstI and PstI-BglII and triple ligated into an NcoI-BglII-cut pUS529. This resulted in a NDD1−ΔQ construct which lacks amino acids 93 to 143. This construct was subsequently excised as an NcoI-XhoI fragment and blunt-end ligated in frame with the GAL4 DNA-binding domain in pGBT9.

Deletion analysis of Ndd1a

Random spore analysis.

To determine the terminal phenotype of cells deficient in NDD1, a diploid strain heterozygous for ndd1Δ::LEU2 (US241) was sporulated in liquid medium. Spores purified by centrifugation (7) were inoculated into medium lacking leucine. Samples were withdrawn 6 h after germination and were used for in situ immunofluorescence and photomicroscopy.

Gel mobility shift assay.

Yeast protein extracts were prepared from protease-deficient (pep4) strains essentially as described by Company et al. (12). Double-stranded DNA probes were prepared by using the Klenow enzyme to fill in the protruding 5′ end of a restriction site in the presence of radioactive [α-32P]dATP. The labelled probes were purified from the unincorporated nucleotides by using NucTrap probe purification columns (Stratagene). DNA binding reactions were carried out in 1.5-ml Eppendorf tubes. Generally, 10,000 to 20,000 cpm (Cerenkov) of probe was used for each reaction. Each reaction contained 4 μl of 5× BS buffer (100 mM Tris-HCl [pH 7.5], 250 mM NaCl, 15 mM MgCl2, 5 mM dithiothreitol, 25 mM spermidine, 250 μg of bovine serum albumin per ml, 100 mM EDTA), 2 μl of 50% glycerol, 2.5 μl of a 1-mg/ml concentration of poly(dI-dC), radiolabelled probe, water, and 20 μg of protein extract to give a final volume of 20 μl. Protein extracts were always added last, and the reactions were incubated at 24°C for 15 min followed by a further 15 min of incubation at 4°C. Conditions for competition experiments were exactly the same except that cold competitor DNAs were added to the reaction mixture before the addition of radiolabelled probe. For antibody supershift assays, the appropriate antibodies were added 6 min after the addition of the protein extracts and processed as described above. The gel used for resolving the protein-DNA complexes was 4% (20:1) acrylamide-bisacrylamide gel in 0.5× Tris-borate-EDTA (TBE). The gels were prerun at 4°C for 2 h at 200 V (~20 mA) in 0.5× TBE. After incubation, the reaction mixtures were loaded onto the gel and run at 4°C for 3 h at 200 V (~8 mA). The gels were dried and autoradiographed with an intensifying screen at −70°C.

Other techniques.

The lithium acetate method was used for all yeast transformations. Total RNA was isolated as described by Cross and Tinkelenberg (13), and Northern (RNA) blot analyses were performed as described by Price et al. (40). The method of Kilmartin and Adams (21) was used for immunofluorescence and photomicroscopy. DNA distribution analysis by flow cytometry was performed as described by Lim et al. (27).


NDD1 is an allele-specific suppressor of the cdc28-1N mutation.

Most CDC28 mutants (i.e., cdc28-1, cdc28-4, and cdc28-13) are predominantly defective in traversing START and therefore arrest in late G1 phase as unbudded cells with 1N DNA content (20, 42). Some of them also exhibit a defect in progression through mitosis (39, 41). The cdc28-1N mutant is unique in that it progresses through START and S phase normally but fails to undergo nuclear division at the nonpermissive temperature, despite the high Clb2-associated histone H1 kinase activity (39, 53). We isolated NDD1 in a genetic screen designed to identify genes whose overexpression can suppress the mitotic defect of cdc28-1N mutant. To determine whether NDD1 overexpression causes suppression of other cdc28 mutations, a high-copy-number (2μm) vector carrying NDD1 was introduced into both cdc28-4 and cdc28-1N mutants, and the transformants were tested for growth at 37°C. While NDD1 allowed growth of cdc28-1N cells at 37°C, it failed to suppress the cdc28-4 mutation (Fig. (Fig.1A).1A). Overexpression of NDD1 also failed to suppress the cdc28-4 mutation when tested at 31 and 35°C (data not included). Hence, NDD1 is an allele-specific suppressor of the cdc28-1N mutation. This behavior is identical to that of the mitotic cyclin genes CLB1 and CLB2, which were isolated in the same screening (53). This suggests that the role of NDD1 is related to the mitotic function of CDC28.

FIG. 1FIG. 1
(A) NDD1 is an allele-specific suppressor of the cdc28-1N mutation. cdc28-4 and cdc28-1N cells were transformed with YEp13 (2μm) or the vector containing NDD1 as the 2.7-kb PvuII fragment. The transformants were plated on leucine-deficient plates ...

NDD1 gene, located on chromosome XV, contains a 1.66-kb ORF (Saccharomyces Genome Database ORF sequence YOR372c) that encodes a protein of 554 amino acids (Fig. (Fig.1B)1B) with no significant overall homology to any known protein in the databases. The protein contains an unbroken stretch of 15 glutamine residues in its N terminus. The glutamine stretches are found in a number of proteins with diverse functions, including some transcription activators, but the precise role this motif serves in these proteins is not known. At least in the case of Ndd1, the glutamine tract does not seem to play a crucial role for Ndd1 function, since Ndd1 lacking this stretch can complement NDD1-null mutant and is also able to suppress cdc28-1N mutation (see below). Ndd1 also contains three putative Cdc28 phosphorylation sites in the middle region, but their relevance to Ndd1 function is not clear. As there are no other features that match any of the known motifs, the amino acid sequence of Ndd1 currently provides no clues to its possible function.

NDD1 is an essential gene whose function is required during mitosis.

To determine whether NDD1 is necessary for cell viability, the ORF in one of the two copies of NDD1 in a wild-type diploid was replaced by the LEU2 gene (Fig. (Fig.2A).2A). The resulting heterozygous mutant (US241) was allowed to sporulate, and the tetrads were dissected on rich medium. Only two spores in each tetrad survived and gave rise to normal colonies that failed to grow when replica plated on leucine-deficient medium. By inference, the absence of viable Leu+ segregants in these dissections suggests that all segregants that were Leu+ and therefore lacked intact NDD1 ORF were nonviable. Microscopic examination of these segregants revealed that the spores had germinated but that most had arrested as large-budded cells, while others had undergone one or two divisions before cessation of growth. This suggests that NDD1 is essential for vegetative growth.

FIG. 2
NDD1 is an essential gene. (A) Partial restriction map of the 2.7-kb PvuII fragment containing NDD1. The 0.45-kb EcoRV fragment in the ORF was replaced by the LEU2 gene to generate the disruption mutant ndd1Δ::LEU2. The scale bar is in base pairs. ...

To characterize the cell cycle arrest phenotype of NDD1-deficient cells, diploid cells heterozygous for NDD1 disruption (US241) were grown in YPD medium and then transferred to sporulation medium. After 20 h, spores were purified and inoculated into medium lacking leucine to allow germination and growth of only those spores in which the NDD1 ORF had been replaced by the LEU2 gene. Immunofluorescence microscopy revealed that NDD1-deficient cells arrested with a large, elongated bud, an undivided nucleus, and a short mitotic spindle (Fig. (Fig.2B).2B). In a parallel experiment, NDD1-deficient haploid cells kept alive by a CEN vector carrying GAL-NDD1 (US262) was first grown in galactose and then transferred to glucose medium. At the end of 6 h, the cells had arrested, with a phenotype identical to that observed in the spore outgrowth experiment. Fluorescence-activated cell sorter analysis showed that these cells had arrested with 2N DNA content (Fig. (Fig.2B).2B). Thus, NDD1-deficient cells are capable of bud emergence, DNA replication, and spindle formation, but they fail to proceed to anaphase. This suggests that NDD1 is essential for some aspect of nuclear division.

Expression and localization of Ndd1 during the cell cycle.

Since NDD1 appeared to be necessary for the nuclear division process, we asked whether its expression is restricted to this stage of the cell cycle. Exponentially growing wild-type cells were synchronized in G1 by α-factor treatment for 3 h at 25°C and then allowed to resume cell cycle progression in α-factor-free medium. Total RNA was isolated, and NDD1-specific transcript was detected by Northern blot analysis. The NDD1 RNA, first detectable above the basal level at 30 min after the release from G1 arrest, peaks at 50 to 60 min and then declines ca. 30 min before the majority of cells reach anaphase (Fig. (Fig.3A).3A). The transcript appears again in the next cycle when the proportion of cells with anaphase spindle has reached its minimum. The window of NDD1 expression lies between the peaks of CLN1 and CLB2 transcription (Fig. (Fig.3A),3A), a period that corresponds to the S phase. It is not yet clear as to what restricts NDD1 transcription to the S phase. So far we have not found any known cis-regulatory elements within the 2-kb region upstream of the ORF except one loosely conserved SCB (Swi4/Swi6 cell cycle box) in reverse orientation.

FIG. 3FIG. 3
(A) NDD1 gene expression during the cell cycle. Wild-type cells lacking BAR1 gene were arrested in G1 by α-factor treatment and then released into growth medium without α-factor. Samples were withdrawn at 10-min intervals, and total RNA ...

To determine the cellular location of Ndd1 protein, NDD1-ORF was fused in frame with the gene encoding β-galactosidase and put under the control of GAL1 promoter on a CEN plasmid. This fusion construct was fully functional, since a NDD1 null mutant harboring this plasmid grew normally on galactose but remained nonviable in glucose medium. Wild-type cells carrying the fusion construct were grown in galactose medium for 3 h and then fixed with formaldehyde. Immunofluorescence staining with anti-β-galactosidase antibodies showed brightly stained nuclei (Fig. (Fig.3B).3B). No nuclear staining was observed in cells expressing only β-galactosidase from GAL1 promoter (not shown). This implies that Ndd1 is a nuclear protein (Fig. (Fig.3B).3B). The protein was always seen in the nucleus, irrespective of both the bud size and the state of nuclear division, suggesting that nuclear localization of Ndd1 is perhaps not cell cycle stage dependent.

Overexpression of Ndd1 enhances the expression of CLB1, CLB2, and SWI5 genes.

It is known that elevated levels of the B-type cyclins Clb1, Clb2, and Clb4 can suppress the nuclear division defect of the cdc28-1N mutant (53). We reasoned that any gene whose overexpression can cause the accumulation of these cyclins, either by upregulation of transcription or by increasing their stabilization, would also be identified by our genetic screen. Therefore, we tested NDD1 for its ability to enhance the transcription of CLB1 and CLB2. Wild-type cells carrying GAL-NDD1 on a CEN plasmid (US737) were grown in raffinose medium (in which GAL1 promoter is not active) for 3 h and then one-half of the culture was induced to express Ndd1 by the addition of galactose. Total RNA was prepared at various time points and analyzed for the presence of CLB1, CLB2, CLB3, SWI5, CDC20, and NDD1 transcripts. Of these, the CLB1, CLB2, SWI5, and CDC20 genes are transcribed in late S/G2 phase, whereas CLB3 is expressed slightly earlier (22). The transcription of CLB1, CLB2, and SWI5 continues throughout mitosis and is switched off as the cells exit from mitosis (29, 53). URA3 served as an internal control for equal RNA loading since it is transcribed constitutively during the cell cycle. The overexpression of NDD1 did not cause any change in the level of the CLB3, CDC20, and URA3 transcripts. However, there was a three- to fivefold increase in the steady-state levels of CLB1 and CLB2 RNA (Fig. (Fig.4A).4A). The SWI5 expression was also noticeably enhanced (Fig. (Fig.4A).4A). CDC5 and ACE2 are two other genes that are expressed in late-S/G2 phase, but their expression is not affected by the overexpression of NDD1 (data not shown). Thus, NDD1 modulates expression of only a subset of the late-S/G2 genes. These data also suggest that the ability of NDD1 to suppress the cdc28-1N mutation is perhaps linked to its capacity to augment the expression of the mitotic cyclins CLB1 and CLB2.

FIG. 4
(A) Overexpression of NDD1 results in enhanced expression of CLB1, CLB2, and SWI5. A wild-type strain containing pGAL1-NDD1 was grown to mid-log phase in raffinose medium. While one-half of the culture continued to grow in raffinose medium (left lanes), ...

The increase in the steady-state level of CLB1 and CLB2 RNAs may be because NDD1 overexpression abolishes the cell cycle regulation of these cyclins so that they are constitutively expressed in response to excess Ndd1. To test this, cdc15 cells carrying GAL-NDD1 on a CEN vector (US1354) were grown in raffinose medium at 25°C and then synchronized in telophase by incubation at 37°C for 2 h. Cells were induced to produce NDD1 for the next 2 h by the addition of galactose before they were allowed to resume cell cycle progression at 25°C in galactose medium. In a control experiment, cdc15 cells carrying a vector without GAL-NDD1 were subjected to an identical experimental regimen. When released from telophase arrest, cdc15 cells traverse the cell cycle in a highly synchronous manner (52). Total RNA was prepared at various time points and was analyzed for the presence of CLB2 transcript by Northern blotting. As expected, CLB2 expression in cells without GAL-NDD1 showed characteristic undulation such that it is switched off 30 min after the release from telophase and is turned on again when the cells are well into the next cycle (Fig. (Fig.4B).4B). The overexpression of NDD1 from GAL1 promoter does not change this pattern; instead it causes a three- to fivefold increase in the overall expression of CLB2 (Fig. (Fig.4B).4B). Thus, excess NDD1 amplifies the CLB2 expression but does not abrogate its cell cycle regulation.

NDD1 is essential for CLB1, CLB2, and SWI5 expression.

Since Ndd1 can modulate the expression of CLB1, CLB2, and SWI5, we asked whether NDD1 is essential for the transcription of these genes. Since our efforts to isolate a “tight” temperature-sensitive allele of NDD1 were not successful, we used for these experiments an ndd1Δ strain kept alive by a CEN plasmid carrying GAL-NDD1 (US262). When transferred to glucose medium, this strain exhibits an arrest phenotype within 4 to 5 h. The cells were first grown in galactose until they reached log phase and were then shifted to glucose medium to switch off the NDD1 expression. Total RNA was isolated from samples withdrawn at various time intervals and analyzed for the presence of CLB1, CLB2, SWI5, CLB3, CDC20, NDD1, and URA3 RNAs by Northern blotting. The cells began to exhibit their characteristic terminal phenotype after 5 h and thereafter remained arrested with 2N content DNA, a short mitotic spindle, and an undivided nucleus. While the CLB3, CDC20, and URA3 transcripts were present throughout the course of the experiment, the CLB1, CLB2, and SWI5 RNAs became undetectable within 4 h of shifting to glucose medium and remained so for the remainder of the time course (Fig. (Fig.5A).5A). As expected, NDD1 was not transcribed during this period. These results suggest that NDD1 is essential for the transcriptional activation of CLB1, CLB2, and SWI5 genes but not for the expression of CLB3 or CDC20.

FIG. 5
(A) NDD1 is required for the normal expression of CLB1, CLB2, and SWI5. ndd1Δ::LEU2 cells carrying pGAL1-NDD1 were grown to log phase in galactose medium. After 3 h, NDD1 expression from the GAL1 promoter was terminated by transferring the cells ...

Clb1 and Clb2 are B-type cyclins that associate with the protein kinase Cdc28 and govern its mitotic activity (16, 53). Cells lacking either CLB1 or CLB2 are viable but the clb1Δ clb2Δ double mutant is unable to proceed through mitosis and arrest with a short spindle, an undivided nucleus, and 2N DNA content (53). This phenotype is very similar to that of the ndd1Δ mutant. Since NDD1 is required for CLB1 and CLB2 transcription, it is possible that the nonviability of ndd1Δ mutant may be entirely due to a lack of these mitotic cyclins. To test this, we performed a plasmid shuffle experiment in which a TRP1-selectable CEN plasmid carrying either GAL-CLB2 or GAL-NDD1 was introduced into an ndd1Δ mutant kept alive by the native-promoter-driven NDD1 on a URA3-selectable vector. The transformants were plated on galactose medium containing 5-fluoro-orotic acid (5-FOA) to select for clones that had lost the URA3 vector bearing NDD1 but that expressed either GAL-NDD1 or GAL-CLB2. While the cells containing GAL-NDD1 grew readily on 5-FOA plates, the strain harboring GAL-CLB2 did not give rise to any colonies (Fig. (Fig.5B),5B), suggesting that Clb2 overexpression alone cannot compensate for the absence of Ndd1. Hence, the lethality caused by NDD1 deficiency is not solely due to the lack of mitotic cyclins. This implies that, in addition to CLB1 and CLB2, NDD1 may also be required for the expression of other genes whose function is necessary for the G2/M transition.

Transcriptional activation by NDD1.

Since our results raise the possibility that Ndd1 may be a transcription activator, we tested its ability to activate transcription in a heterologous context by using a yeast one-hybrid assay. The NDD1 coding region, fused in frame with the DNA-binding domain of the transcription factor GAL4, was put under the control of the ADH promoter and was transformed into a tester strain that carried the lacZ and HIS3 genes as reporters, both driven by three copies of a 17-mer Gal4-binding sequence. The Gal4-Ndd1 fusion was able to activate transcription of the reporter constructs, resulting in both the appearance of His+ colonies and the expression of β-galactosidase (data not shown).

It has been shown previously that MCM1 participates in the mechanism that coordinates the expression of a group of genes (including CLB1, CLB2, and SWI5) whose transcription is activated during the late-S/G2 phase of the cell cycle (1). In light of our observation that NDD1 is required for the transcription of a subset of these genes and that it can activate transcription in the yeast one-hybrid assay, it is possible that NDD1 interacts functionally with MCM1. Indeed, a multicopy vector (2μm) carrying native-promoter-driven NDD1, although not able to support growth of the mcm1 mutant cells at 37°C, does allow them to grow at 35°C (1a). Thus, excess NDD1 is able to suppress, to some extent, the lethality caused by the lack of Mcm1 function.

Normal periodic transcription of CLB2 requires a 240-bp promoter sequence.

Maher et al. (30) have described a 55-bp UAS within the CLB2 promoter which contains one consensus Mcm1 binding site [DCCY(A/T)(A/T)(T/A)NN(G/A)G; D ≠ C, Y = T or C] (25) and is both necessary and sufficient for the cell-cycle-regulated expression of CLB2. This 55-bp UAS is similar to a 55-bp element in the SWI5 promoter in both its sequence and its ability to bind a Mcm1-containing ternary complex (30).

In addition to the one Mcm1 consensus element (MCE) and an Sff response element (SFRE) in the 55-bp UAS, there are four other MCEs in the CLB2 promoter. The relative positions of these elements are shown in Fig. Fig.6A.6A. Some of our initial experiments had led us to suspect that the previously documented 55-bp UAS (30) may not be sufficient for the normal cell-cycle-regulated expression of CLB2. We reasoned that MCEs contained within the sequences flanking the 55-bp UAS might also be required. We therefore compared the 55-, 240-, and 340-bp promoter fragments, which contained one, four, and five MCEs, respectively, for their ability to regulate transcription of a reporter gene. The promoter fragments were placed immediately upstream of the CYC1 TATA box linked to a reporter gene ubiY lacZ (29), and these constructs were subsequently integrated in chromosome I at the URA3 locus in wild-type cells. Constructs without any UAS (“dead”) or with the 55-bp UAS of SWI5 gene were used as negative and positive controls, respectively. To determine the pattern of lacZ expression at various stages of the cell cycle, the growth of wild-type cells carrying these constructs was arrested by treatment with α-factor (late G1), hydroxyurea (early S phase), or nocodazole (pre-nuclear division). Total RNA was prepared and analyzed by Northern blotting for the presence of both lacZ and the endogenous CLB2 transcripts. The 240-bp, 340-bp, and SWI5 UAS fragments were all able to drive the lacZ expression in cycling, hydroxyurea (HU)- and nocodazole (NOC)-arrested cells but were transcriptionally silent in α-factor-treated cells (Fig. (Fig.6B).6B). This expression pattern closely matches with that of the endogenous CLB2 RNA, except that the level of lacZ transcription from the recombinant constructs is somewhat lower. The 55-bp fragment of CLB2 promoter, however, failed to activate lacZ transcription in all instances (Fig. (Fig.6B).6B). Identical results were obtained when the reporter constructs were tested on 2μm vectors (results not shown).

FIG. 6FIG. 6FIG. 6
(A) The 340-bp region of the CLB2 promoter. The 5′ flanking region of CLB2 gene from −870 to −520 bp is shown. The boxed region indicates the presumptive 55-bp core promoter sequence. Mcm1 binding sites (MCE I, II, III, IV, and ...

To confirm that the 240-bp fragment can regulate lacZ expression during the progression through the cell cycle, cells containing an integrated copy of the 240-bp-driven ubiY lacZ were first synchronized in G1 by α-factor treatment and then allowed to traverse the cell cycle in α-factor-free medium. Total RNA was isolated from samples collected at various times and analyzed for the presence of the lacZ and endogenous CLB2 RNAs by Northern blotting. As in the case of cells arrested at various stages of the cell cycle, the pattern of lacZ expression in synchronously cycling cells is indistinguishable from that of endogenous CLB2 gene (Fig. (Fig.6C),6C), except that the level of lacZ expression is lower in the second cycle. These results suggest that the 240-bp promoter fragment, but not the 55-bp element, is sufficient for the cell-cycle-regulated expression of CLB2.

The 240-bp fragment mediates potentiation of CLB2 expression by NDD1 overexpression.

Since the 240-bp sequence is sufficient to impose cell cycle regulation on CLB2 transcription, we asked whether the enhancement of CLB2 expression by excess Ndd1 is effected through this promoter fragment. Therefore, we compared the efficacy of the 55-, 240-, and 340-bp fragments to elicit the upregulation of CLB2 transcription. A CEN vector carrying GAL-NDD1 was introduced into wild-type strains containing ubiY lacZ reporters driven by the various promoter fragments (described in the preceding section) at the URA3 locus. The cells were first grown in raffinose medium and then induced to express NDD1 for 2 h by the addition of galactose. The total RNA was analyzed for the presence of lacZ and the endogenous CLB2 transcripts. Whereas no lacZ RNA was detected when the reporter construct was under control of the 55-bp fragment, both the 240- and the 340-bp-fragment-driven reporters were transcribed to a readily detectable level (Fig. (Fig.7).7). The transcription from both the 240- and the 340-bp constructs was upregulated in response to the overexpression of NDD1 (Fig. (Fig.7).7). All three strains transcribed the endogenous CLB2 gene and, as expected, the expression was enhanced when Ndd1 was overexpressed. Hence, Ndd1 exerts its effect on CLB2 transcription via the 240-bp promoter sequence.

FIG. 7
Enhancement of CLB2 transcription by excess Ndd1 requires the 240-bp CLB2 UAS. Strains containing the 55-, 240-, or 340-bp CLB2 UAS-ubiY lacZ reporter were transformed with pGAL1-NDD1 and grown in raffinose medium to mid-log phase. Galactose was then ...

Ndd1 prevents formation of protein complexes on the 240-bp sequence, whose absence correlated with the expression of Clb2.

Our results thus far strongly suggest that the 240-bp promoter fragment mediates both the cell-cycle-regulated expression of CLB2 and its potentiation by excess Ndd1. To determine whether Ndd1 is recruited to the protein complexes assembled on this cis-regulatory sequence, a radioactively labelled 240-bp fragment and the protein extract from cells expressing native-promoter-driven, fully functional HA3-NDD1 were analyzed in a gel mobility shift assay. The assay yielded four specific complexes (designated I, II, III, and IV) (Fig. (Fig.8A,8A, left panel, lane 2), all of which disappeared when challenged with a molar excess of nonradioactive probe (Fig. (Fig.8A,8A, left panel, lane 6). However, when a molar excess of nonradioactive 55-bp fragment was used to challenge the binding, complex IV (and to some extent complexes I and II) was abrogated. This suggests that this particular complex is assembled on the core 55-bp region of the 240-bp fragment (Fig. (Fig.8A,8A, right panel, lane 4). Complexes I, II, and IV contained Mcm1 because inclusion of anti-Mcm1 antibodies in the binding reactions abolished their formation (Fig. (Fig.8A,8A, left panel, lane 5). Mcm1 is perhaps also present in complex III since anti-Mcm1 antibodies diminish, though do not abolish, its formation. However, anti-HA antibodies failed to supershift any of the complexes, indicating that Ndd1 is not present in these complexes (Fig. (Fig.8A,8A, left panel, lanes 3 and 4). The failure of anti-HA antibodies to elicit any supershift in our experiments is not due to their general ineffectiveness in gel retardation assays, since the same antibodies (12CA5) have been shown to cause supershifts in a variety of different contexts (37a).

FIG. 8
(A) DNA-protein complexes formed on 240-bp CLB2 UAS do not contain Ndd1. Radioactively labelled double-stranded 240-bp CLB2 UAS probe was used for gel retardation assays. The probe was mixed with 20 μg of protein extracts prepared from pep4 ...

If Ndd1 does not appear to physically associate with the protein assemblage on the 240-bp promoter sequence, then how does it modulate the expression of CLB2 and the other target genes? To address this question, we first compared the pattern of protein complexes formed on the 240-bp fragment during the stages when CLB2 promoter is either active or inactive. pep4 cells (US1005) were arrested in late G1 phase, early S phase, or in mitosis by treatment with α-factor, HU, or NOC, respectively. Cell extracts from these cultures were used in a gel mobility shift assay with the 240-bp sequence as a probe, and total RNA was analyzed for the presence of the CLB2 transcript. As expected, the cycling extract yielded four specific protein complexes (I, II, III, and IV), identical to the ones described in the preceding paragraph (Fig. (Fig.8B,8B, lane 2). While the same four bands were detected in extracts from HU-arrested cells, complexes I, II, and III were either absent or diminished in NOC extracts (Fig. (Fig.8B,8B, lanes 4 and 5). Interestingly, a slow-migrating complex (N) was detected in α-factor extract, which was conspicuously absent from early-S-phase or mitotic extracts (Fig. (Fig.8B,8B, lane 3). Anti-Mcm1 antibodies abolish its formation, suggesting that this complex also contains Mcm1 (Fig. (Fig.8B,8B, lane 7). As expected, CLB2 was not expressed in G1 (α-factor arrest), but it was efficiently transcribed in both cycling and HU- or NOC-arrested cells (Fig. (Fig.8B,8B, bottom panel). Hence, the appearance of the slow-migrating N complex in α-factor extracts correlates with the absence of CLB2 RNA. Similarly, the high-level expression of CLB2 in NOC-arrested extracts is concomitant with a dramatic reduction in the abundance of complex III.

The CLB2 transcription is activated in late S/G2 phase; it continues through most of the M phase and is then switched off when the cells exit mitosis to enter the G1 phase of a new cycle (15, 53). Since Ndd1 is required for the expression of CLB2, we wondered whether the turning off of CLB2 transcription in G1 is, at least in part, caused by the absence of the Ndd1 protein. This inquiry seemed pertinent in light of our observation that CLB2 transcription shuts off less efficiently in cells expressing GAL-NDD1 (Fig. (Fig.4B4B and our unpublished observations). We therefore determined the relative stability of Ndd1 protein during the cell cycle. cdc15 mutant cells carrying GAL-HA3-NDD1 on a CEN plasmid (US1354) were first synchronized in late telophase by incubation at 37°C for 2 h in raffinose medium. They were induced to express HA3-Ndd1 for the next 2 h by the addition of galactose and then allowed to resume cell cycle progression at 25°C in galactose medium. Constitutive expression of Ndd1 from GAL1 promoter in this experiment allowed us to determine not only the timing of its degradation but also the timing of its stabilization. Cell extracts were prepared from samples withdrawn at various time points after the release and were analyzed for the presence of both HA3-Ndd1 and Clb2 proteins by Western blotting. The abundance of Ndd1 protein begins to decline 30 min after the release from telophase arrest, as the cells begin to disassemble the mitotic spindles (Fig. (Fig.8C).8C). Ndd1 remains unstable until cells enter a new cycle at 75 min, as indicated by the absence of anaphase spindles; thereafter, the protein continues to accumulate until it becomes unstable again at 180 min, when the second cycle of spindle disassembly begins. The timing of Ndd1 degradation after the release from telophase arrest matched closely with that of endogenous Clb2, but Clb2 accumulation showed an apparent delay of 40 min (Fig. (Fig.8C).8C). This delay is expected in this experiment since the endogenous CLB2 gene, unlike GAL1-driven NDD1, is under the control of its native promoter and is transcribed significantly later than is Ndd1. Thus, the disappearance of Ndd1 protein in G1 correlates well with the silencing of CLB2 transcription.

The rapid degradation of Ndd1 upon exit from mitosis also correlates with the appearance of the slow-migrating DNA-protein complex (N) (Fig. (Fig.8B).8B). Therefore, we asked if the expression of Ndd1 in G1 would abrogate the formation of this protein complex. pep4 cells carrying three copies of GAL-NDD1 integrated at the URA3 locus were first synchronized in G1 by α-factor treatment in raffinose medium for 2 h. One-half of the culture was then induced to express NDD1 by the addition of galactose, while the other half received glucose to repress the GAL1 promoter. After 2 h, cell extracts were prepared and were used in combination with the 240-bp promoter fragment in a gel mobility shift assay. The G1 arrest induced by α-factor was maintained through the entire course of the experiment. While both the slow-migrating N complex and complex III were present in the glucose-grown cells, their abundance was dramatically reduced in cells expressing Ndd1 from GAL1 promoter (Fig. (Fig.8D,8D, lanes 3 and 4), suggesting that Ndd1 may prevent the formation of these protein complexes on the CLB2 promoter. As before (Fig. (Fig.8B),8B), the N complex was competed out by both the anti-Mcm1 antibodies and a molar excess of the nonradioactive probe (Fig. (Fig.8D,8D, lanes 5 and 6). The overexpression of Ndd1 also diminishes complexes I, II, and III, suggesting that presence of Ndd1 in G1 also affects their assembly.

Together, these data suggest a possible mechanism by which NDD1 may participate in the regulation of CLB2 expression. According to this scheme, complexes such as N and III may act as repressors of CLB2 transcription, since their appearance is inversely correlated to the expression of CLB2. That Ndd1 can cause abrogation of both complexes N and III (Fig. (Fig.8D)8D) implies that Ndd1 may modulate CLB2 transcription by preventing the formation of these repressor complexes, thereby allowing the activation of transcription. Thus, the rapid degradation of Ndd1 in G1, which would allow (or enhance) the assembly of various repressor complexes, may constitute an important step in the silencing of CLB2 expression upon exit from mitosis. Although we describe this scheme with a particular focus on complexes N and III, complexes I and II could also be repressor aggregates of various protein compositions since, like N complex, they are present in protein extracts from α-factor-arrested cells and diminish in abundance in NOC extracts.

A deletion analysis of Ndd1.

Based on the requirement of Ndd1 for CLB1, CLB2 and SWI5 expression and its ability to activate transcription in a yeast one-hybrid assay, we had raised the possibility that Ndd1 may act as an activator of transcription. Since transcription activation in a heterologous context such as a one-hybrid assay can be artifactual, we asked whether the biological activity of Ndd1 (such as the suppression of cdc28-1N and mcm1 mutations) is related to its ability to activate transcription in a one-hybrid assay. Therefore, we made various constructs in which the DNA-binding domain of the GAL4 transcription factor was fused in frame with different parts of the NDD1 ORF and then tested these constructs for both their biological activity and their ability to activate transcription. Such a deletion analysis was also expected to identify the regions important for Ndd1 function.

The deletion constructs (Table (Table2)2) were introduced into a yeast strain carrying the lacZ reporter under the control of GAL4 UAS (US449), and the extent of transcriptional activation was estimated by quantitative measurement of the β-galactosidase activity. To determine the biological activity, the hybrid fusions were also introduced into the ndd1Δ, cdc28-1N, and mcm1 mutants. As evident from Table Table2,2, the full-length NDD1 fused to the GAL4 DNA-binding domain is able to both activate transcription and suppress the growth defects of ndd1Δ, cdc28-1N, and mcm1 mutants (row 2). A comparison of various deletion mutants shows that the elimination of the C-terminal portion (residues 232 to 554; Table Table2,2, rows 3, 4, and 5) results in a loss of the biological activity of Ndd1 but it does not diminish its ability to activate transcription. Hence, the N-terminal part of the protein is sufficient for transcriptional activation but not for the biological function. Although the C-terminal half is required for the Ndd1 function, by itself it is both biologically and transcriptionally inactive (row 9). Interestingly, the polyglutamine stretch in the N-terminal half, the most conspicuous feature of the protein, is dispensable since its deletion does not affect Ndd1 function, although it does elicit a higher level of transcription activation (row 12). These results imply that the ability of Ndd1 to activate transcription in a one-hybrid assay may be unrelated to its biological function. However, it is possible that Ndd1 serves as a transcription activator in other cellular contexts.


The activities of many of the cell cycle effectors are regulated by protein modification, as well as by the controls operating at the level of transcription. Such multitiered regulation is presumably necessary in order to sharpen the timing of various cellular events during the progression through the division cycle. The cis-acting DNA sequences and the factors that activate the transcription of genes at START or at the onset of the S phase have been described in some detail (23, 28, 37, 38). However, the transcriptional regulation of the late S/G2-phase-specific genes is poorly understood. So far, Mcm1 and another putative transcription factor, Sff, have been shown to participate in the control of gene expression at this stage of the cell cycle (1, 29). However, since Mcm1 is also required for the regulation of a number of other genes whose transcription is not periodic (31, 33, 51), it is unlikely that Mcm1 itself is the major target for cell cycle regulation. This notion is consistent with the finding that Mcm1 occupies both SWI5 and CLB2 promoters at all times, although their expression is cyclic (1).

We have identified and characterized a new gene, NDD1, that plays an essential role in the expression of the late-S-phase genes CLB1, CLB2, and SWI5, such that none of these genes are transcribed in its absence. Thus, it is a new player in the regulation of gene expression during late S phase. Ndd1 is not only essential for the transcription of CLB1 and CLB2, but its excess enhances the level of their expression. This observation may provide an explanation for the ability of Ndd1 to suppress the cdc28-1N mutation. The requirement of NDD1 for CLB1 and CLB2 expression may argue that the lethality of the ndd1Δ mutant is solely due to a lack of these mitotic cyclins. However, overexpression of CLB2 from the GAL1 promoter fails to suppress the nuclear-division defect of NDD1-deficient cells (Fig. (Fig.5B),5B), implying that NDD1 is necessary for the expression of additional genes that participate in the process of nuclear division. NDD1 was also independently identified as a gene whose overexpression could suppress a crippling mutation in the SWI5 UAS, which abolishes SWI5 transcription (29a).

NDD1 is expressed in a cell-cycle-stage-dependent manner such that its expression peaks in S phase just prior to the expression of its target gene CLB2 (Fig. (Fig.3A).3A). This may implicate NDD1 as a critical factor in the timing of CLB2 expression. Our observation that the constitutive expression of Ndd1 does not alter the cell-cycle-regulated pattern of CLB2 transcription (Fig. (Fig.4B)4B) suggests that NDD1, though necessary for the activation of a subset of late-S-phase genes, is not a crucial determinant in the temporal control of their transcription. However, since the constitutive expression of Ndd1 leads to inefficient silencing of CLB2 transcription in late telophase (unpublished observations), the turning off of CLB2 transcription in late telophase may involve regulation of the Ndd1 protein. It is conceivable that the inactivation of the Ndd1 protein might be one of the steps required to switch off CLB2 expression at the end of mitosis. The instability of Ndd1 due to rapid degradation during mitotic exit is consistent with this notion (Fig. (Fig.8C).8C). The mechanism, which renders Ndd1 unstable in G1, remains unclear. We found four destruction-box-like sequences in the middle region of the protein, but their removal did not affect its stability (unpublished results). Another important factor in silencing CLB2 expression may be the Cdc28-Clb kinase itself. It has been previously shown that Cdc28-Clb kinase complex stimulates CLB2 expression via a positive feedback loop (4). Consequently, inactivation of the kinase by abrupt proteolysis of Clb proteins during the M/G1 transition could lead to a rapid decline in CLB2 expression.

The genetic interaction with MCM1 and its ability to enhance CLB1, CLB2, and SWI5 transcription lead us to suspect that NDD1 may influence gene expression by modulating their promoter activity. During the course of our studies, we discovered that the previously reported 55-bp UAS (30) in the CLB2 promoter, which contains a pair of regulatory sites comprised of one MCE and one SFRE, is unable to drive the expression of a reporter gene in our assay system (see Materials and Methods). In an identical assay, the 55-bp UAS of SWI5 gene not only elicited transcription but also showed the expected pattern of expression during the cell cycle. This suggests that the inactivity of the 55-bp CLB2 UAS in our experiments is not due to a faulty assay system. By further investigations, we have identified a 240-bp fragment that is sufficient for both the expression and the cell cycle regulation of CLB2. The 240-bp sequence contains within it the 55-bp presumptive UAS and another pair of MCE and SFRE flanked by two additional MCE (Fig. (Fig.6A)6A) (reference 1 and this study). This configuration is also capable of mediating the transcriptional enhancement caused by the overexpression of Ndd1; the 55-bp UAS alone, on the other hand, remains unresponsive (Fig. (Fig.7).7). The CLB1 promoter also harbors two pairs of MCE and SFRE but their spatial arrangement is not identical to that in the CLB2 promoter (1). It is intriguing that while one set of MCE and SFRE can appropriately regulate SWI5 expression, it is not sufficient for the activation of CLB2 transcription. Perhaps this is due to the differences in both the regulatory sequences themselves and in the promoter sequence within which these elements are embedded.

The 240-bp promoter fragment promotes the assembly of four prominent protein complexes. While the exact composition of these complexes remains unknown, they all appear to contain Mcm1 (Fig. (Fig.8A).8A). Extracts prepared from cells traversing synchronously through the cell cycle are capable of assembling, albeit to various extents, all four complexes on this CLB2 promoter fragment (data not shown). Interestingly, Ndd1 is not recruited to any of these protein complexes, as is suggested by the gel mobility shift assay (Fig. (Fig.8A),8A), despite its requirement for CLB1, CLB2, and SWI5 transcription and its ability to enhance gene expression. In immunoprecipitation experiments, Ndd1 neither associates with Mcm1 nor with any of the major components of the general transcription machinery such as yeast TBP, TAF145, or TAF90 (unpublished data). These observations suggest that Ndd1 cooperates indirectly with the transcriptional apparatus, possibly through intermediary proteins, to regulate the expression of late-S-phase genes. How then, does Ndd1 exert its effect on gene expression? One possibility is that Ndd1 is an effector of Cdc28-Clb kinase activity and that it affects CLB2 transcription by modifying the efficacy of the positive-feedback loop. However, this seems unlikely, because the Ndd1-depleted cells during NOC-induced arrest contain significant levels of the mitotic kinase activity but yet do not transcribe CLB2 (unpublished results).

Alternatively, Ndd1 may prevent a repressor from binding to the CLB2 promoter, thus allowing the activation of transcription. The finding that Mcm1, an activator of transcription, occupies the CLB2 promoter throughout the cell cycle (1) strengthens the possibility that a repressor may be, in part, responsible for the termination of CLB2 transcription in G1. In the context of Ndd1 function, this repressor model is consistent with some of our findings: (i) a new protein complex N is detected on the 240-bp promoter fragment during G1 when CLB2 transcription is abruptly switched off; (ii) this complex is undetectable in HU- or NOC-arrested cells where CLB2 is actively transcribed; (iii) overexpression of Ndd1 in G1 abolishes the formation of the N complex; and (iv) in wild-type cells, Ndd1 is rapidly degraded in G1, concomitant with the silencing of CLB2 expression. Complex III behaves like complex N in that its abundance is enhanced during G1, dramatically reduced during NOC-induced arrest when CLB2 is maximally transcribed, and negatively influenced by the overexpression of Ndd1. Thus, complexes N and III could be repressor assemblages capable of terminating CLB2 expression. Although the presence of Ndd1 abolishes the formation of the presumptive repressor complexes, its overexpression does not elicit CLB2 expression in G1. This implies that a lack of Ndd1 protein may not be the sole reason for the termination of CLB2 transcription; a progressive weakening of the positive-feedback loop upon proteolysis of Clb proteins may also be critical. The nature of the hypothetical repressor is so far unknown. To identify such a repressor, we have embarked upon a genetic screen to isolate mutations that will lead to inappropriate CLB2 expression in G1.

With the identification of NDD1, we have added a new element in the regulation of gene expression in late S phase. However, further investigations will be required to uncover the mechanism that ensures the correct timing of the onset and termination of the expression of these genes.


We thank Kim Nasmyth, in whose laboratory the plasmid containing NDD1 was first isolated by U.S. We are grateful to Gustav Ammerer for various strains, anti-Mcm1 antibodies, suggestions, and fruitful discussions.

This work was supported by the National Science and Technology Board, Singapore.


1. Althoefer H, Schleiffer A, Wassmann K, Nordheim A, Ammerer G. Mcm1 is required to coordinate G2-specific transcription in Saccharomyces cerevisiae. Mol Cell Biol. 1995;15:5917–5928. [PMC free article] [PubMed]
1a. Ammerer, G. Personal communication.
2. Amon A. Regulation of B-type cyclin proteolysis by Cdc28-associated kinase in budding yeast. EMBO J. 1997;16:2693–2702. [PMC free article] [PubMed]
3. Amon A, Irniger S, Nasmyth K. Closing the cell cycle circle in yeast: G2 cyclin proteolysis initiated at mitosis persists until the activation of G1 cyclins in the next cell cycle. Cell. 1994;77:1037–1050. [PubMed]
4. Amon A, Tyers M, Futcher B, Nasmyth K. Mechanisms that help the yeast cell cycle clock tick: G2 cyclins transcriptionally activate G2 cyclins and repress G1 cyclins. Cell. 1993;74:993–1007. [PubMed]
5. Amon A, Surana U, Muroff I, Nasmyth K. Regulation of p34CDC28 tyrosine phosphorylation is not required for entry into mitosis in S. cerevisiae. Nature (London) 1992;355:368–371. [PubMed]
6. Andrews B J, Herskowitz I. Identification of a DNA binding factor involved in cell-cycle control of the yeast HO gene. Cell. 1989;57:21–29. [PubMed]
7. Ausubel F A, Brent R, Kingston R E, Moore D D, Seidman J G, Smith J A, Struhl K, editors. Current protocols in molecular biology. New York, N.Y: John Wiley & Sons; 1998.
8. Bachmair A, Finley D, Varshavsky A. In vivo half-life of a protein is a function of its amino-terminal residue. Science. 1986;234:179–186. [PubMed]
9. Bartel P, Chien C T, Sternglanz R, Fields S. Elimination of false positives that arise in using the two-hybrid system. BioTechniques. 1993;14:920–924. [PubMed]
10. Blondel M, Mann C. G2 cyclins are required for the degradation of G1 cyclins in yeast. Nature (London) 1996;384:279–282. [PubMed]
11. Breeden L, Nasmyth K. Cell cycle control of the yeast HO gene: cis and trans-acting regulators. Cell. 1987;48:389–397. [PubMed]
12. Company M, Adler C, Errede B. Identification of a Ty1 regulatory sequence responsive to STE7 and STE12. Mol Cell Biol. 1988;8:2545–2554. [PMC free article] [PubMed]
13. Cross F, Tinkelenberg A H. A positive feedback loop controlling CLN1 and CLN2 gene expression at the START of yeast cell cycle. Cell. 1991;65:875–883. [PubMed]
14. Epstein C B, Cross F. CLB5: a novel B type cyclin from yeast with a role in S-phase. Genes Dev. 1992;6:1695–1706. [PubMed]
15. Fitch I, Dahman C, Surana U, Amon A, Goetsh L, Byers B, Futcher B. Characterization of four B-type cyclin genes of the budding yeast S. cerevisiae. Mol Biol Cell. 1992;3:805–818. [PMC free article] [PubMed]
16. Ghiara J B, Richardson H E, Sugimoto K, Henze M, Lew D J, Wittenberg C, Reed S I. A cyclin B homolog in S. cerevisiae. Chronic activation of the Cdc28 protein kinase by cyclin prevents exit from mitosis. Cell. 1991;65:163–174. [PubMed]
17. Ghislain M, Udvardy A, Mann C. S. cerevisiae 26S protease mutants arrest cell division in G2/metaphase. Nature (London) 1993;366:358–362. [PubMed]
18. Glotzer M, Murray A W, Kirshner M W. Cyclin is degraded by the ubiquitin pathway. Nature (London) 1991;349:132–138. [PubMed]
19. Gu Y, Rosenblatt J, Morgan D O. Cell cycle regulation of CDK2 activity by phosphorylation of Thr160 and Tyr15. EMBO J. 1992;11:3995–4005. [PMC free article] [PubMed]
20. Hartwell L H, Culotti J R, Reid B J. Genetic control of the cell cycle in yeast. Science. 1974;183:46–51. [PubMed]
21. Kilmartin J V, Adams A E M. Structural arrangements of tubulin and actin during the cell cycle of yeast Saccharomyces cerevisiae. J Cell Biol. 1984;98:922–933. [PMC free article] [PubMed]
22. Koch C, Nasmyth K. Cell cycle regulated transcription in yeast. Curr Opin Cell Biol. 1994;6:451–459. [PubMed]
23. Koch C, Moll T, Neuberg M, Ahorn H, Nasmyth K. A role for the transcription factors Mbp1 and Swi4 in progression from G1 to S phase. Science. 1993;261:1551–1557. [PubMed]
24. Kotani S, Tugendreich S, Fujii M, Jorgensen P M, Watanabe N, Hoog C, Hieter P, Todokoro K. PKA and MPF-activated polo-like kinase regulate anaphase-promoting complex activity and mitosis progression. Mol Cell. 1998;1:371–380. [PubMed]
25. Kuo M-H, Grayhack E J. A library of yeast genomic Mcm1 binding sites contains genes involved in cell cycle control, cell wall and membrane structure, and metabolism. Mol Cell Biol. 1994;14:348–359. [PMC free article] [PubMed]
26. Kuo M-H, Nadeau E T, Grayhack E J. Multiple phosphorylated forms of the Saccharomyces cerevisiae Mcm1 protein include an isoform induced in response to high salt concentrations. Mol Cell Biol. 1997;17:819–832. [PMC free article] [PubMed]
27. Lim H H, Loy C J, Zaman S, Surana U. Dephosphorylation of threonine 169 of Cdc28 is not required for the exit from mitosis but may be necessary for START in Saccharomyces cerevisiae. Mol Cell Biol. 1996;16:4573–4583. [PMC free article] [PubMed]
28. Lowndes N F, Johnson A L, Breeden L, Johnston L H. Swi6 protein is required for transcription of the periodically expressed DNA synthesis genes in budding yeast. Nature (London) 1992;357:505–508. [PubMed]
29. Lydall D, Ammerer G, Nasmyth K. A new role for MCM1 in yeast: cell cycle regulation of SWI5 transcription. Genes Dev. 1991;5:2405–2419. [PubMed]
29a. Lydall, D., and K. Nasmyth. Unpublished results.
30. Maher M, Cong F, Kindelberger D, Nasmyth K, Dalton S. Cell cycle-regulated transcription of the CLB2 gene is dependent on Mcm1 and a ternary complex factor. Mol Cell Biol. 1995;15:3129–3137. [PMC free article] [PubMed]
31. Maine G, Sinha P, Tye B-K. Mutants of S. cerevisiae defective in the maintenance of minichromosomes. Genetics. 1984;106:365–385. [PMC free article] [PubMed]
32. McInerny C J, Partridge J F, Mikesell G E, Creemer D P, Breeden L L. A novel Mcm1-dependent element in the SWI4, CLN3, CDC6, and CDC47 promoters activates M/G1-specific transcription. Genes Dev. 1997;11:1277–1288. [PubMed]
33. Messenguy F, Dubois E. Genetic evidence for a role for MCM1 in the regulation of arginine metabolism in Saccharomyces cerevisiae. Mol Cell Biol. 1993;13:2586–2592. [PMC free article] [PubMed]
34. Morgan D O. Principles of CDK regulation. Nature (London) 1995;374:131–134. [PubMed]
35. Nasmyth K. Control of the yeast cell cycle by the Cdc28 protein kinase. Curr Opin Cell Biol. 1993;5:166–179. [PubMed]
36. Nasmyth K, Tatchell K. The structure of transposable yeast mating type loci. Cell. 1980;19:753–764. [PubMed]
37. Nasmyth K, Dirick L. The role of SWI4 and SWI6 in the activity of G1 cyclins in yeast. Cell. 1991;66:995–1013. [PubMed]
37a. O’Connor, M. Personal communication.
38. Ogas J, Andrews B J, Herskowitz I. Transcriptional activation of CLN1, CLN2, and a putative new G1 cyclin (HCS26) by SWI4, a positive regulator of G1-specific transcription. Cell. 1991;66:1015–1026. [PubMed]
39. Piggot J R, Rai R, Carter B L A. A bifunctional gene product involved in two phases of the yeast cell cycle. Nature (London) 1982;298:391–393. [PubMed]
40. Price C, Nasmyth K, Schuster T. A general approach to the isolation of cell cycle-regulated genes in the budding yeast Saccharomyces cerevisiae. J Cell Biol. 1991;218:543–556. [PubMed]
41. Reed S I, Wittenberg C. Mitotic role for the Cdc28 protein kinase of S. cerevisiae. Proc Natl Acad Sci USA. 1990;87:5697–5701. [PMC free article] [PubMed]
42. Reed S I, Hadwiger J A, Lorincz A. Protein kinase activity associated with the product of the yeast cell division cycle gene CDC28. Proc Natl Acad Sci USA. 1985;82:4055–4059. [PMC free article] [PubMed]
43. Richardson H E, Wittenberg C, Cross F, Reed S. An essential G1 function for cyclin-like proteins in yeast. Cell. 1989;59:1127–1133. [PubMed]
44. Rose M D, Winston F, Hieter P. Methods in yeast genetics: a laboratory course manual. Cold Spring Harbor, N.Y: Cold Spring Harbor Laboratory Press; 1990.
45. Rothstein R. Targeting, disruption, replacement, and allele rescue: integrative DNA transformation in yeast. Methods Enzymol. 1991;194:281–301. [PubMed]
46. Sambrook J, Fritsch E F, Maniatis T. Molecular cloning: a laboratory manual. 2nd ed. Cold Spring Harbor, N.Y: Cold Spring Harbor Laboratory Press; 1989.
47. Schneider B L, Patton E E, Lanker S, Mendenhall M D, Wittenberg C, Futcher B, Tyers M. Yeast G1 cyclins are unstable in G1 phase. Nature (London) 1998;395:86–89. [PubMed]
48. Schwob E, Nasmyth K. CLB5 and CLB6, a new pair of B cyclins involved in S phase and mitotic spindle formation in S. cerevisiae. Genes Dev. 1993;7:1160–1175. [PubMed]
49. Sorger P K, Murray A W. S-phase feedback control in budding yeast is independent of tyrosine phosphorylation of p34CDC28. Nature (London) 1992;355:365–368. [PubMed]
50. Spellman P T, Sherlock G, Zhang M Q, Iyer V R, Anders K, Eisen M B, Brown P O, Botstein D, Futcher B. Comprehensive identification of cell cycle-regulated genes of the yeast Saccharomyces cerevisiae by microarray hybridization. Mol Biol Cell. 1998;9:3273–3297. [PMC free article] [PubMed]
51. Sprague G. Combinatorial associations of regulatory proteins and the control of cell type in yeast. Adv Genet. 1990;27:33–62. [PubMed]
52. Surana U, Amon A, Dowzer G, McGrew J, Byers B, Nasmyth K. Destruction of the CDC28/CLB mitotic kinase is not required for the metaphase to anaphase transition in budding yeast. EMBO J. 1993;12:1969–1978. [PMC free article] [PubMed]
53. Surana U, Robitsch H, Price C, Schuster T, Fitch I, Futcher A B, Nasmyth K. The role of CDC28 and cyclins during mitosis in the budding yeast S. cerevisiae. Cell. 1991;65:145–161. [PubMed]
54. Townsley F M, Ruderman J V. Proteolytic ratchets that control progression through mitosis. Trends Cell Biol. 1998;8:238–244. [PubMed]
55. Wolf D A, Jackson P K. Cell cycle: oiling the gears of anaphase. Curr Biol. 1998;8:636–639. [PubMed]

Articles from Molecular and Cellular Biology are provided here courtesy of American Society for Microbiology (ASM)
PubReader format: click here to try


Related citations in PubMed

See reviews...See all...

Cited by other articles in PMC

See all...


  • Gene
    Gene links
  • GEO Profiles
    GEO Profiles
    Related GEO records
  • HomoloGene
    HomoloGene links
  • MedGen
    Related information in MedGen
  • Pathways + GO
    Pathways + GO
    Pathways, annotations and biological systems (BioSystems) that cite the current article.
  • Protein
    Published protein sequences
  • PubMed
    PubMed citations for these articles
  • Substance
    PubChem Substance links
  • Taxonomy
    Related taxonomy entry
  • Taxonomy Tree
    Taxonomy Tree

Recent Activity

Your browsing activity is empty.

Activity recording is turned off.

Turn recording back on

See more...