| Edited by Hirokazu Inoue. Kunihiro Ohta: Corresponding author. E-mail: kohta@bio.c.u-tokyo.ac.jp |
In eukaryotic cells, chromatin structure contributes to the efficient packaging of large genomic DNA molecules within a nucleus; hence it limits spatiotemporally various DNA-related processes. For instance, chromatin remodeling has been shown to play pivotal roles in regulation of gene expression of Saccharomyces cerevisiae as well as other organisms (Narlikar et al., 2002). Alteration of local chromatin structure is also crucial for the regulation of DNA replication (Falbo and Shen, 2006; Tabancay and Forsburg, 2006) and repair (Bao and Shen, 2007; Osley et al., 2007). We previously reported that chromatin configuration is altered around Saccharomyces cerevisiae recombination hot spots in meiotic cells (Ohta et al., 1994), and around the Schizosaccharomyces pombe meiotic recombination hot spot ade6-M26 before the Spo11/Rec12-mediated formation of DNA double strand breaks (DSBs) that initiates meiotic recombination (Mizuno et al., 1997). The chromatin remodeling similar to that around the ade6-M26 hotspot can also be found in some stress-induced genes, and plays essential roles in stress-induced gene activation (Hirota et al., 2003, 2004). Possibly, chromatin remodeling is necessary for the increased local accessibility of trans-acting proteins to cis-acting DNA sequences (Wolffe, 1997).
Two classes of enzymes are known to be involved in chromatin remodeling. The first class catalyzes covalent modifications on histone N-terminal tails, such as acetylation, phosphorylation and methylation (Jenuwein and Allis, 2001). The second class is comprised of large multi-subunit complexes known as ATP-dependent chromatin remodeling complexes that use the energy from ATP hydrolysis to alter chromatin configuration (Fyodorov and Kadonaga, 2001; Vignali et al., 2000). These complexes can be further categorized into four classes: 1) SWI/SNF, 2) ISWI, 3) Mi-2 and 4) Ino80 families on the basis of the similarities of their ATPase subunits to the Swi2/Snf2, Isw1, Mi-2 and Ino80 (Mohrmann and Verrijzer, 2005). The SWI/SNF complex family consists of two subfamilies, SWI/SNF and RSC. RSC is essential in cell viability. Genetic analysis in budding yeast had revealed multiple cellular functions of RSC, which harbors Sth1/Nps1 as the Swi2/Snf2-type ATPase core subunit. Sth1 is required for the transcriptional regulation of a large number of genes (Damelin et al., 2002; Ng et al., 2002). On the other hand, Swi2/Snf2, the other Swi2/Snf2-type ADCR, is nonessential and required only for the transcription of a subset of genes (Holstege et al., 1998; Sudarsanam et al., 2000). Such distinct requirements of different ADCRs are seemingly consistent with the observation that RSC exists at ten-fold higher concentration than SWI/SNF (Cairns et al., 1996). In addition, previous studies have revealed that RSC is involved in multiple cellular processes, such as DNA repair (Shim et al., 2005; Wong et al., 2006), kinetochore function (Hsu et al., 2003; Tsuchiya et al., 1998), association of cohesins with chromosome arms (Huang et al., 2004), and maintenance of ploidy (Campsteijn et al., 2007). The in vitro chromatin remodeling activity of human and fly RSC was previously well examined (Bouazoune and Brehm, 2006; Mohrmann and Verrijzer, 2005; Xue et al., 2000), while its in vivo functions still need further investigations.
In the present study, we describe the function of the fission yeast Snf21 (GenBank Accession Number: AB162438), which we identified along with Snf22, the other Swi2/Snf2-type ADCR involved in the chromatin regulation at the ade6-M26 meiotic recombination hot spot (Yamada et al., 2004). Snf21 exhibits a close sequence similarity to S. cerevisiae Sth1. Thus, Snf21 is likely an orthologue of Sth1 rather than Swi2/Snf2 in fission yeast. As predicted, we found that Snf21 is essential for cell viability. We isolated snf21 temperature sensitive (ts) mutants and demonstrated that one of them arrests at G2-M phase under restrictive conditions. The phenotypes of this snf21 ts mutant suggest that Snf21 plays a role in chromosome segregation and centromere functions.
The S. pombe strains used in this study are listed in Table 1. S. pombe cells were grown in complete medium (YE) or minimal medium (SD) (Sherman et al., 1986). For transformation of S. pombe cells, a high-efficiency lithium acetate method was employed (Davidson et al., 2004; Okazaki et al., 1990). Mating and sporulation were induced on sporulation agar plate (SPA). To synchronously arrest the cells at G1 phase, haploid cells were grown in YE to ~2 × 107 cells/ml. These cells were harvested and washed four times with minimal medium (MM) (Isshiki et al., 1992) lacking nitrogen (MM-N) and then transferred to MM-N. All medium were supplemented with appropriate nutrients. Standard genetic methods were performed as previously described (Gutz et al., 1974).
 View Details | Table 1 Strains used in this study |
Identification of snf21+ by sequence similarity was conducted as described previously (Yamada et al., 2004). Disruption of the snf21+ gene was performed by the standard one-step gene replacement method (Rothstein, 1983). About 500 base pairs (bp) of flanking sequences of both side of snf21+ open reading frame (ORF) were cloned and ligated with the ura4+ cassette (Grimm et al., 1988) to construct snf21::ura4+ in pBlueScript KS(+). A PstI-XhoI fragment encompassing snf21::ura4+ was transformed into wild-type strain D56. Then, the cells were spread on SD plates lacking uracil. The Ura+ heterozygous transformants were obtained and the correct gene targeting was confirmed by polymerase chain reaction (PCR) by using an appropriate set of primers.
C-terminal tagging of Snf21 with FLAG or green fluorescent protein (GFP) was performed by the standard integration method, using the integration vectors int2 (GFP) and int6 (FLAG), respectively (Hirota et al., 2001). The int6 vector is a derivative of int2, which only replaced GFP with FLAG. The coding region of Snf21 was amplified by PCR using primers 5’-AAC ATT CTT GCT AGA GCC CAA T-3’ and 5’-CGG ATC CGC CTC ATT TTC AAT TCT-3’ and cloned into pUC119. This resulted in the omission of the stop codon and introduction of a BamHI site (underlined). An NdeI-BamHI fragment containing the Snf21 C-terminal region was inserted in the multicloning site of int2 and int6 to make an in-frame fusion with GFP and FLAG. The resultant plasmid was digested with HincII and transformed into S. pombe cells. Stable G418-resistant transformants were selected and analyzed by PCR to confirm their correct integration. The obtained snf21+-FLAG and snf21+-GFP cells grew normally, indicating that Snf21-FLAG and Snf21-GFP are functional.
To isolate the snf21 ts alleles, error-prone PCR was performed to cover the highly conserved three motifs: SNF2_N, Helicase_C and Bromodomain. The amplified fragments were digested with EcoT22I and inserted to the PstI site in pBlueScript SK(–) carrying the ura4+ marker gene at the HindIII site. The resultant plasmid library was linearized by cutting with NruI and transformed into the wild-type haploid cells K175 and K176. The transformants carrying the ura4+ marker gene were replica-plated on SD plates lacking uracil and incubated at 34°C. Integration of the mutated genes into the snf21 locus was confirmed by PCR. Linkage of the ts phenotype to uracil prototrophy and complementation of the ts growth defect by the snf21+ gene were also confirmed. The KYP86 cells carrying the snf21-14 mutation with ura4+ were streaked on YE plate containing 5-FOA to pop out the ura4+ marker gene and the mutation site corresponding to L1080P in the Bromodomain was removed out together with the marker gene, thereby generating the snf21-36 allele.
For sporulation, heterozygous diploid cells were suspended in distilled H2O (dH2O) and spotted on SPA plates. After incubation at 30°C for 3 days, cells were collected and resuspended in 0.5% glusulase solution. After agitating at 30°C for 30 min, ethanol was added at a final concentration of 30% and incubated at room temperature for 5 min. The suspension was centrifuged and the precipitates were washed three times with dH2O. To induce germination of snf21::ura4+ spores, an appropriate number of spores were grown in SD lacking uracil at 30°C.
Synchronization of temperature-sensitive S. pombe cells was carried out as described with slight modifications (Alfa et al., 1993). Exponentially growing cdc25-22 mutant cells carrying the tagged snf21+ gene (Snf21-FLAG) at 25°C were collected, transferred to YE pre-warmed at 36°C and incubated at 36°C for 4 hours. Cells were harvested and transferred to YE at 25°C. Aliquots were taken at 20 min intervals until 240 min. Synchronization was monitored by counting the percentage of septated cells.
A total of 1–2 × 107 cells were fixed with cold 70% ethanol. After washing with 0.5 ml of buffer A (0.2 M Tris-Cl pH 6.8, 0.05 M EDTA), cells were suspended in 0.5 ml of buffer A containing RNaseA (final conc. 0.2 mg/ml), sonicated gently and incubated at 37°C for 2 hours. 0.5 ml of buffer A containing propidium iodide (final conc. 2.5 μg/ml) was added and cells were stained. The fluorescence intensities of stained cells were measured and analyzed on a flow cytometer (Becton Dickinson) using the CellQuest software. Ten thousand events were analyzed per sample.
For the observation of Snf21-GFP in living cells, wild-type KYP25 cells were logarithmically grown in YE medium, harvested and suspended in YE on a slide glass. Hoechst 33342 (final conc. 1 μg/ml) was added for microscopic observation. Fluorescence images of living cells were taken with a cooled charge-coupled device camera and stored digitally using MetaMorph software (Universal Imaging, Downingtown, PA). For the microscopic analysis to monitor the nuclear and cell morphology, cells were fixed with cold 70% ethanol, washed with phosphate-buffered saline (PBS) and suspended in PBS containing 1 μg/ml Hoechst 33342.
Total RNA was prepared from S. pombe cells according to the standard method (Elder et al., 1983). For Northern analysis, 5 μg of total RNA was denatured with formamide, separated on 1.2% agarose gel containing formaldehyde (Sambrook et al., 1989) and blotted on a charged nylon membrane (Hybond-XL, GE Healthcare). The probes to detect transcripts of snf21+ and cam1+ were prepared from PCR products and labeled with 32P using a random priming kit (Amersham Ready-To-Go DNA Labelling Beads, GE Healthcare) according to the manufacturer’s instructions. The following primers were used for snf21+: 5’-CAT ATG CGT GCA GAG AAG CAA T-3’ and 5’-GGA TCC TCA AGC CTC ATT TTC AAT TC-3’. The nucleotide sequences of the primers used for cam1+ were as described (Hirota et al., 2003).
Cell extracts were prepared as follows. Cells were collected and washed with STOP buffer (150 mM NaCl, 50 mM NaF, 10 mM EDTA pH 8.0, 1 M NaN3). Cell pellets were resuspended in 1 x sodium dodecyl sulfate (SDS) sample buffer and boiled at 100°C for 5 min. After vortexing with glass beads, supernatants from cell suspensions were prepared. Protein samples were loaded on an SDS-polyacrylamide gel, electrophoretically separated, transferred to a PVDF membrane (Immobilon-P, Millipore) and detected with the ECL detection kit (GE Healthcare). The following antibodies were used as primary antibodies: mouse monoclonal anti-FLAG M2 antibody (1:2500) (SIGMA) and mouse monoclonal anti-α-tubulin B-5-1-2 antibody (1:5000) (SIGMA). Horseradish peroxidase (HRP)-conjugated anti-mouse IgG F(ab’)2 fragment (1:10000) (GE Healthcare) was used for secondary antibody.
Saccharomyces cerevisiae Sth1 was isolated on the basis of its similarity to Swi2/Snf2 and shown to be essential for viability (Laurent et al., 1992; Tsuchiya et al., 1992). In contrast, Swi2/Snf2 is nonessential. We previously identified snf22+ as a homologue of S. cerevisiae SWI2/SNF2 using the sequences of conserved domains (Fig. 1A) and reported that the deletion of the nonessential snf22+ gene affects chromatin remodeling, histone acetylation, meiotic recombination at the ade6-M26 hot spot, and the regulation of transcription response to osmotic stress (Hirota et al., 2008; Yamada et al., 2004). In addition, we also identified a gene referred as snf21+, which shows close similarity to STH1. Therefore, here we focus on the function of the snf21+ gene.
 View Details | Fig. 1 Fission yeast snf21+ is essential for viability. (A) Schematic representation of the domain structures of Swi2/Snf2 family ATP-dependent chromatin remodelers in yeast: Saccharomyces cerevisiae Sth1 (ScSth1), Swi2/Snf2 (ScSwi2/Snf2) and Schizosaccharomyces pombe Snf21 (SpSnf21), Snf22 (SpSnf22). Conserved domains are indicated as follows: SNF2_N, SNF2 family N-terminal domain; Helicase_C, Helicase conserved C-terminal domain; Bromo, Bromodomain. The entire length of each protein is shown on the right. Similarity (%) of overall amino acid sequences with ScSth1 or ScSwi2/Snf2 are also shown on the far right. (B) The heterozygous (snf21::ura4+/snf21+) diploid strain (KYP3) was sporulated on SPA plate and dissected on YE plate. Six tetrads are shown, with the four spores (a–d) from each tetrad in a horizontal row. (C) Microscopic observation of lethal spores in B. The image shows that spores have germinated, divided a few times and arrested with bent and elongated morphology. (D) Chromosome condensation in snf21Δ cells. The diploid strain KYP73 (wild type) and KYP3 (snf21Δ) was sporulated, selectively germinated in SD lacking uracil (30°C, 24 h for KYP73 and 48 h for KYP3) and stained with Hoechst 33342. Bar, 10 μm. |
First we tested whether Snf21 is essential for viability, as observed with the Sth1 deletion in S. cerevisiae. We made a heterozygous diploid with the deletion of snf21+ flanked with the ura4+ insertion, and observed the phenotype for formation and viability of ascospores. For all the tetrads tested, two spores formed regular-sized colonies but the other two spores did not grow to visible colonies (Fig. 1B). All well-grown colonies exhibited a uracil auxotrophic phenotype, indicating that snf21+ is essential for cell viability. Microscopic observation showed that the inviable snf21Δ spores stopped growing after several rounds of cell divisions (Fig. 1C). To observe the cell shape and nuclear morphology, the snf21Δ spores were selectively germinated and analyzed microscopically (Fig. 1D). We observed that cells germinated from the snf21Δ spores showed larger, bent and elongated cell shape, which is a typical phenotype in mutants defective in cell cycle progression. In addition, we found an accumulation of condensed nuclei, instead of nuclei of hemispherical appearance like in wild type cells. Since condensed nuclei are characteristic of the mitotic phase, these suggest that Snf21 is required for mitotic progression.
We next examined the snf21+ expression and subcellular localization throughout the cell cycle. Using the temperature-sensitive (ts) cdc25-22 mutant, cells were arrested in G2/M phase at a restrictive temperature followed by a release at a permissive temperature. Cell cycle progression was monitored by septation index and ~50% of the cells were synchronized in S phase at 100 min after the release (Fig. 2A). The expression of snf21+ was then analyzed by northern and western blotting (Fig. 2B, C). We observed that the snf21+ mRNA and Snf21 protein exhibit a substantial decrease around 100 min after the release, when the most of the cells were in S phase as shown by the high septation index. In contrast, they were relatively abundant at around the G2/M phase. We next examined the localization of Snf21-GFP, which had been demonstrated to be functional, and found that it was localized to nuclei throughout the cell cycle (Fig. 2D). These data suggest that Snf21 is a nuclear protein fluctuated in a cell cycle-dependent manner.
 View Details | Fig. 2 Expression and nuclear localization of Snf21. (A) Percentage of septated cells. Exponentially growing cdc25-22 mutant cells harboring Snf21-FLAG (KYP61) were arrested at 36°C for 4 h and shifted down to 25°C. Cells were taken at 20 min intervals and septating cells were counted. (B–C) Expression patterns of snf21+ transcripts (B) and Snf21 protein (C) through cell cycle. cam1+ and α-Tubulin were used as loading control, respectively. The no-tagged KKT142 strain was also used as a control. (D) Localization of Snf21. Haploid strain KYP25 was logarithmically grown in YE medium, the nucleus was stained with Hoechst 33342 (nuclear DNA, blue; Snf21-GFP, Red). S phase and M phase cells were indicated by arrowheads and an arrow, respectively. Bar, 10 μm. |
To further investigate the in vivo functions of Snf21, we isolated several temperature sensitive (ts) alleles of snf21+. Among them, three ts mutants (snf21-1, snf21-4 and snf21-14) conferred a severe growth defect under non-permissive conditions over 30°C (Fig. 3A, B). Since snf21-14 showed the most severe ts phenotype, we selected it and removed the ura4+ marker gene from snf21-14, thereby generating a ts allele referred to as snf21-36, which has amino acid substitutions only in the conserved ATPase motif, SNF2_N. This mutant showed basically the same ts phenotypes as snf21-14. The growth defect at 34°C in snf21-36 was rescued by an expression of wild-type snf21+ in a pREP41-based plasmid (Basi et al., 1993; Maundrell, 1990) (Fig. 3C), indicating that the phenotypes of the snf21-36 strain are due to the recessive snf21-36 mutations. Then, we further analyzed the phenotype of snf21-36. At permissive temperature, we observed no defect in cell shape and nuclear morphology (Fig. 3D). However, we found that snf21-36 is sensitive to a microtubule-destabilizing agent, thiabendazole (TBZ) at permissive temperature, which causes instability of spindle microtubules and thereby perturbs chromosome segregation (Fig. 3E). When we combined snf21-36 with the deletion of mad2+ gene, which is involved in spindle assembly checkpoint (SAC) at kinetochores, the sensitivity to TBZ of the double mutant was higher than in the snf21-36 or mad2Δ single mutants.
 View Details | Fig. 3 Isolation and characterization of snf21 ts alleles. (A) Diagram of the structures of the snf21 ts gene products. Mutations found in the ts alleles and conserved domains (SNF2_N, SNF2 family N-terminal domain; H, Helicase conserved C-terminal domain; B, Bromodomain) are indicated. (B) Growth defect of snf21 ts mutants at 25°C, 30°C, 34°C and 37°C. Wild type and snf21 ts cells were grown in YE medium at 25°C and spotted on YE plates (1 × 107 cells/ml in the far-left spots for each plate and then diluted 5-fold in each subsequent spot rightwards). Plates were incubated at indicated temperature for 3 days. (C) Complementation of growth defect in snf21-36 by expression of snf21+. Cells were grown in SD medium lacking uracil at 25°C and spotted on MM plates lacking uracil (5 × 107 cells/ml in the far-left spots for each plate and then diluted 5-fold in each subsequent spot rightwards). Plates were incubated at 25°C and 34°C for 3 days. (D) Cell and nuclear morphology of wild type and snf21-36 cells. Cells were exponentially grown in YE medium at 25°C, and observed after staining with Hoechst 33342 and observed. Bar, 10 μm. (E) TBZ sensitivity of snf21-36. Various strains as indicated were spotted on YE plates containing 0, 5 and 10 μg/ml TBZ in a similar manner as in B (5 × 107 cells/ml in the far-left spots for each plate and then diluted 5-fold in each subsequent spot rightwards) and incubated at 25°C for 4 days. |
Defects in the spindle association with kinetochores are monitored by the SAC machinery, then checkpoint signals operate to arrest mitotic cell cycles. Mad2 is a core component of SAC machinery. Thus, we next investigated the effects of the snf21-36 and mad2Δ mutations on cell cycle progression. Fig. 4 illustrates that most of snf21-36 cells had 2C DNA content at any of the tested temperatures (25°C, 30°C and 34°C). This effect of snf21-36 was alleviated by the introduction of mad2Δ mutation at 25°C and 30°C, but not at 34°C. Therefore, kinetochores in snf21-36 at 34°C are seemingly nonfunctional even when the SAC pathway is absent, although they may be functional enough at 25°C and 30°C.
 View Details | Fig. 4 Accumulation of the snf21-36 mutant in G2/M phase at permissive temperature in a Mad2-dependent manner. The haploid strains (wild type, mad2Δ, snf21-36 and mad2Δ snf21-36) were cultured at 25°C in YE (+N) to mid-log phase, washed four times with MM-N, transferred to MM-N and incubated at 25°C. The DNA content of each sample was analyzed by flow cytometry. The percentage of cells with DNA contents (1C and 2C) is shown in pie graphs. |
To further elucidate the Snf21 function in kinetochores, we observed cell viability and chromosome segregation in the snf21-36 mad2Δ double mutant at 34°C. We found that the double and snf21-36 single mutants grew very slowly at 34°C, causing cell death partly. The double mutant showed a severer cell death phenotype than the snf21-36 single mutant, suggesting that the cell viability in snf21-36 is partly dependent on SAC (Fig. 5A). Both mutants exhibited a defect in chromosome segregation at 34°C, although there seemed to be no significant differences between snf21-36 and mad2Δ snf21-36 cells (Fig. 5B). We discovered that the snf21Δ and snf21-36 cells lengthened at 30°C and 34°C, respectively (Fig. 1D, Fig. 5B). The lengths of the snf21-36 and mad2Δ snf21-36 cells were measured and their distribution was reported as histograms (Fig. 5C). At 12 hours after the temperature shift to 34°C, the median of the mad2Δ snf21-36 cell lengths decreases to 14.5 μm from 17.0 μm in the snf21-36 single mutant. More importantly, the frequency of the cells showing highly elongating cell shapes (over 20 μm) was significantly reduced in the mad2Δ snf21-36 mutant, indicating that the mad2+ deletion partially suppresses the accumulation of elongated cells. Taken together, these results suggest that Snf21 may be involved in chromosome segregation during mitosis possibly in kinetochores.
 View Details | Fig. 5 Snf21 is involved in chromosome segregation process. (A) Cell number (left) and viability (right) of the wild type, mad2Δ, snf21-36 and mad2Δ snf21-36 strains. Cells were logarithmically cultured in YE medium at 25°C, harvested and transferred to YE pre-warmed at 34°C. After transfer to 34°C (0 hr), aliquots of the culture were collected, the number of cells was counted and cells were plated on YE at 25°C at each time point. Cell viability was measured by counting the number of colonies. (B) Nuclear and cell morphology of the snf21-36 and the mad2Δ snf21-36 strains. Bar, 10 μm. (C) A distribution histogram of the cell length in the wild type, mad2Δ, snf21-36 and mad2Δ snf21-36 strains. The number of counted cells (n), the cell length median (m) and standard deviation (σ) are indicated. |
We demonstrated in this study that snf21+ is an essential gene. This is consistent with the close sequence similarity of Snf21 to Sth1, which is a component of RSC complex in budding yeast and indispensable for cell growth. Fission yeast has another non-essential Swi2/Snf2-type ADCR, Snf22, which is an orthologue of Swi2/Snf2, the core subunit of the SWI/SNF chromatin remodeling complex. Thus, there is a division of labors in fission yeast between two evolutionally conserved subfamilies of ADCRs. The members of the first subfamily, Snf22 and Swi2/Snf2, are assumed to be important in the chromatin regulation of some specific genes (for example stress response genes) that are not essential at least under normal physiological conditions. On the other hand, the second subfamily members, Snf21 and Sth1, may play crucial roles in cell viability or cell growth at vegetative stage. The snf21-36 ts mutant exhibits marked slow-down of cell growth and cell death in part at restrictive temperature. Therefore, it is likely that Snf21, as Sth1 does, plays a pivotal role in the cell growth process.
This notion is supported by the observation that snf21-36 cells showed abnormality in chromosome segregation at restrictive temperature. The snf21-36 haploid tends to accumulate cells with 2C DNA content at both permissive and non-permissive temperatures. Interestingly, deletion of mad2+ can alleviate this phenotype only at low temperatures (25–30°C), but not at 34°C. Thus, there may be two distinct conditions of kinetochores-spindle complex in snf21-36: mad2+ deletion can compensate one, but not the other. Snf21 may have a redundant function with the Mad2 SAC pathway at kinetochores on mitotic chromosomes.
An alternative, but not mutually exclusive possibility is that Snf21 may be important for the establishment of the proper differentiation of centromeric chromatin structure in mitosis. The centromere has been demonstrated to have distinct chromatin with a histone H3 variant CENP-A, referred to Cse4 and Cnp1 in budding and fission yeast, respectively (Meluh et al., 1998; Smith, 2002; Takahashi et al., 2000). In budding yeast, RSC is required for kinetochore function, although not required for centromeric deposition of Cse4 (Hsu et al., 2003). Thus, Snf21, the central subunit of RSC in fission yeast, may play a pivotal role in kinetochore. Recently, fission yeast SWI/SNF and RSC complexes were characterized (Monahan et al., 2008). Interestingly, fission yeast SWI/SNF and RSC complexes differ from those in budding yeast in some ways. Given that chromatin structure of fission yeast is distinct from that of budding yeast and more similar to higher eukaryotes (Cleveland et al., 2003), there is potentially another regulatory mechanism of kinetochore, distinct from budding yeast. In fission yeast, deposition of Cnp1 occurs at two stages, S and G2 phases (Takayama et al., 2008). The latter process functions as a salvage pathway, and is dependent on Hip1, the fission yeast homolog of HIRA histone chaperone. Hrp1, a CHD family ADCR, was shown to be required to maintain high levels of Cnp1 at centromeres during DNA replication (Walfridsson et al., 2005). At G2 phase, Snf21 may be involved in the deposition of Cnp1 at centromeres together with Hip1 chaperone.
A third possibility is that Snf21 could function in transcriptional regulation and affects the cell cycle progression via the expression of cell cycle regulatory genes. RSC has been shown to be involved in global transcription in budding yeast (Damelin et al., 2002; Ng et al., 2002). A temperature sensitive mutant of STH1, sht1-3ts, exhibited a defect in G2/M transition (Du et al., 1998). At restrictive temperature, CLB2 and CSE4, which encodes a B-type cyclin and a homologue of CENP-A, were normally transcribed in the sth1-3ts mutant. However, the recent study had shown that the gene regulated by SWI/SNF and/or RSC complexes in fission yeast are distinct from that of budding yeast (Monahan et al., 2008). Therefore, we need to analyze the expression of a set of genes required for cell cycle progression.
The data of this study revealed essential roles of Snf21 protein in fission yeast. Since fission yeast provides a good system to investigate chromosome segregation, cell cycle control, chromatin modification and heterochromatin formation, these findings provide important clues to understand conserved essential functions of the RSC subfamily in the eukaryotic chromosome metabolism.
We thank Y. Ichikawa and R. Nakazawa for DNA sequencing, Y. Sakuma for technical assistance, and the Yeast Genetic Resource Center (YGRC) for yeast strains. We are also grateful to W. Lin for critical reading of the manuscript, and other members in our laboratory and Cellular and Molecular Biology Laboratory in RIKEN for their support and valuable comments. This work was supported by grants from the following sources: basic research from the Bio-oriented Technology Research Advancement Institution (to T. S. and K. O.); grants-in-aid for scientific research on priority areas from the Ministry of Education, Science, Culture and Sports, Japan (to K. O.).
|
Index