Fission yeast meu14+ is required for proper nuclear ... - CiteSeerX

JCS ePress online publication date 20 May 2003

Research Article

2721

Fission yeast meu14+ is required for proper nuclear division and accurate forespore membrane formation during meiosis II Daisuke Okuzaki1,*, Wataru Satake1,*, Aiko Hirata2 and Hiroshi Nojima1,‡ 1Department 2Department

of Molecular Genetics, Research Institute for Microbial Diseases, Osaka University, 3-1 Yamadaoka, Suita, Osaka 565-0871, Japan of Integrated Biosciences, Graduate School of Frontier Sciences Tokyo University, Bldg. FSB-101/601, 5-1-5 Kashiwanoha, Kashiwa, Chiba 277-8562, Japan

*These authors contributed equally to this work ‡Author for correspondence (e-mail: [email protected])

Accepted 18 March 2003 Journal of Cell Science 116, 2721-2735 © 2003 The Company of Biologists Ltd doi:10.1242/jcs.00496

Summary Using a meiosis-specific subtracted cDNA library of Schizosaccharomyces pombe, we identified meu14+ as a gene whose expression is upregulated during meiosis. Transcription of meu14+ is induced abruptly after the cell enters meiosis. Its transcription is dependent on the meiosis-specific transcription factor Mei4. In meu14∆ cells, the segregation and modification of the SPBs (spindle pole bodies) and microtubule elongation during meiosis II were aberrant. Meiotic meu14∆ cells consequently produced a high frequency of abnormal tetranucleate cells harboring aberrant forespore membranes and failed to produce asci. In wild-type cells harboring the integrated meu14+-gfp fusion gene, Meu14-GFP first appeared inside the nuclear region at prophase II, after which it accumulated beside the two SPBs at metaphase II. Thereafter, it formed two ring-

Introduction Meiosis produces haploid gametes in eukaryotic organisms through two successive meiotic divisions. The fission yeast Schizosaccharomyces pombe is a useful model organism to study the mechanisms that regulate processes that occur during meiosis (e.g. recombination, chromosome behavior, and sporulation) because it synchronously enters meiosis and can be easily manipulated by a variety of genetical and cell biological techniques. Meiosis in fission yeast is initiated when haploid cells of opposite mating types (h–/h+) experience nitrogen starvation and conjugate (Moreno et al., 1991). Following karyogamy (nuclear fusion), premeiotic DNA synthesis occurs. The nucleus then develops a ‘horse-tail’ shape and moves backwards and forwards about the axis of the cell. This permits homologous pairing and recombination to take place (Chikashige et al., 1994). Successive nuclear divisions, called meiosis I and meiosis II, then ensue. After meiosis II, the capsular plasma membrane forms round each haploid nucleus in the cytoplasm of the mother cell, after which spore formation occurs to generate four mature ascospores. Spore formation is initiated during meiosis II in that during this division, the dividing nuclei start to be surrounded by a double unit membrane, called the forespore membrane. A

shaped structures that surrounded the nucleus at early anaphase II. At post-anaphase II, it disappeared. Meu14GFP appears to localize at the border of the forespore membrane that later develops into spore walls at the end of sporulation. This was confirmed by coexpressing Spo3HA, a component of the forespore membrane, with Meu14GFP. Taken together, we conclude that meu14+ is crucial in meiosis in that it participates in both the nuclear division during meiosis II and the accurate formation of the forespore membrane. Movies available online Key words: Meiosi, Spore formation, Forespore membrane, Microtubule, Spindle pole body, Schizosaccharomyces pombe

previous study has shown that the forespore membrane develops near the outer plaques of the spindle pole body (SPB) during meiosis II and plays an important role in the packaging of the haploid nucleus (Hirata and Shimoda, 1994) However, the exact molecular mechanism driving forespore membrane development is unclear. For efficient spore production, the formation of the forespore membrane must be coordinated with the behavior of the dividing nuclei during meiosis II. The SPB is thought to be a key structure that links these two events. The SPB may act similarly to the mammalian centrosome, which functions as a microtubule-organizing center (MTOC). The SPB enters and leaves the nuclear envelope during the cell cycle (Ding et al., 1997), and has been observed by immunostaining to change from a dot-like image into a crescent during meiosis II (Hirata and Shimoda, 1994; Hagan and Yanagida, 1995). The correct morphological alteration of the SPB and the resulting normal sporulation depends on Spo15, an SPB component that is essential for forespore membrane formation (Ikemoto et al., 2000). Electron microscopic observation has revealed that the forespore membrane is initially assembled next to the SPB, which changes morphologically into a multilayered structure during meiotic division II (Tanaka and Hirata, 1982; Hirata and

2722

Journal of Cell Science 116 (13)

Shimoda, 1994; Hagan and Yanagida, 1995). This suggests that SPB modification is required to initiate forespore membrane formation. Thus, SPBs play an additional role in sporulation that is distinct from their known roles as MTOC scaffolds in mitosis and meiotic division I. Other molecules that appear to be required for normal sporulation are Spo20, an S. pombe phosphatidylinositoltransfer protein (Nakase et al., 2001), Spo3, a forespore membrane component (Nakamura et al., 2001), and Meu10, a molecule with no homology to well-characterized proteins (Tougan et al., 2001). Thus, sporulation appears to consist of several cytologically distinct steps that include the modification of SPBs, the assembly of forespore membranes around the SPBs, the enclosure of nuclei by the forespore membranes, the maturation of the spore wall, and the autolysis of the ascus wall. With regard to the forespore membrane, it is assembled by the fusion of vesicles perhaps derived from the endoplasmic reticulum (ER) and/or the Golgi apparatus (Tanaka and Hirata, 1982; Hirata and Shimoda, 1994). However, little is known about how the forespore membrane assembles in fission yeast, which is in contrast to the prolific studies examining the same phenomenon in budding yeast (for a review, see Pringle et al., 1997). To understand how meiosis and sporulation are regulated in S. pombe in more detail, we have established a subtracted cDNA library to comprehensively isolate meiosis/sporulationspecific genes that we have denoted as meu (meiotic expression upregulated) (Watanabe et al., 2001). Previously, we reported that the gene product generated from one of these genes, meu13+, promotes homologous pairing independently of homologous recombination and regulates the meiotic recombination checkpoint (Nabeshima et al., 2001; Shimada et al., 2002). Furthermore, we found that the meu10+ product is required for ascospore maturation (Tougan et al., 2001). Here, we characterize another meu gene, meu14+. We show that the S. pombe Meu14 protein is expressed specifically during sporulation and appears to play a role in the extension of the forespore membrane and perhaps also in the function of the SPB and/or the meiotic spindles.

Materials and Methods Cells and media The S. pombe strains used in this study are listed in Table 1. Cells were grown as described previously (Nabeshima et al., 2001). Malt extract medium (MEA) and synthetic medium (EMM2) with the appropriate auxotrophic nutrients were used for mating between the h– and h+ strains and maintenance of diploidy. To observe meiosis, h–/h+ diploid and h90 cells were cultured in EMM2 with the appropriate auxotrophic nutrients, washed in EMM-N medium (EMM2 without nitrogen) and then incubated at 28°C in EMM-N (Moreno et al., 1991). Isolation and deletion of meu14+ The genomic library (Fukushima et al., 2000) was made using partial Sau3AI DNA fragments inserted into the BamHI site of the Bluescript KS(+) vector (Stratagene, La Jolla, CA). Colony hybridization with the meu14+ cDNA fragment as a probe allowed us to clone the genomic fragment containing the meu14+ gene from this library. To disrupt meu14+, a 1.1 kb NspI fragment carrying the meu14+ genomic DNA was replaced with the 1.8 kb HindIII fragment containing ura4+ using a synthetic linker. The meu14+::ura4+ DNA fragment digested with KpnI (–1104 bp site from start codon) and PstI (1050 bp site from stop codon) was used to transform wild-type cells (TP4D5A/TP4D-1D). This procedure removed all of the coding region of Meu14, including the initiation codon. The Ura+ transformants were then screened for the disruption of one of the meu14+ gene copies by genomic Southern blot analysis. Tetrads from these cells were then dissected. Northern and Southern blot analyses were performed as described previously (Watanabe et al., 2001).

Preparation of Meu14-GFP The Meu14 protein fused C-terminally with the green fluorescence protein (GFP) was made by first synthesizing oligonucleotides FN (5′-GGCGCGCCGCATATGGGCACTCAACCATCTTAC-3′) and FC (5′-GCGGCCGCGGCAAGAAAACAGTGGATTTTGC-3′), which correspond to the initiation and termination sites of Meu14, respectively. The NdeI and NotI sites introduced into the oligonucleotides are underlined, respectively. Using these oligonucleotides as primers and meu14+ genomic DNA as a template, we performed a polymerase chain reaction (PCR) and generated a

Table 1. S. pombe diploid strains used in this study Name CD16-1 CD16-5 JZ670 AB4 TN187 TP4D-5A/ TP4D-1D YDO100 YDO110 YDO111

Genotype h+/h– ade6-M210/ade6-M216 cyh1/+ +/lys5-391 h–/h– ade6-M210/ade6-M216 cyh1/+ +/lys5-391 h–/h– pat1-114/pat1-114 ade6-M210/ade6-M216 leu1-32/leu1-32 h–/h– pat1-114/pat1-114 ade6-M210/ade6-M216 leu1-32/leu1-32 ura4-D18/ura4-D18 h90 spo3-HA:LEU2 leu1-32 ura4-D18 h–/h+ ade6-M210/ade6-M216 leu1-32/leu1-32 ura4-D18/ura4-D18 +/his2

Source

mei4::ura4+/mei4::ura4+

MKD3 YDO50 YDO130 SI51 YDO10 YN68

h–/h+ ade6-M210/ade6-M216 leu1-32/leu1-32 ura4-D18/ura4-D18 +/his2 meu14+::ura4+/ meu14+::ura4+ h–/h+ ade6-M210/ade6-M216 leu1-32/leu1-32 ura4-D18/ura4-D18 +/his2 spo15-GFP:LEU2/spo15-GFP:LEU2 h–/h+ ade6-M210/ade6-M216 leu1-32/leu1-32 ura4-D18/ura4-D18 +/his2 spo15-GFP:LEU2/spo15-GFP:LEU2 meu14::ura4+/ meu14::ura4+ h–/h+ ade6-M210/ade6-M216 leu1-32/leu1-32 ura4-D18/ura4-D18 +/his2 meu14+:GFP/meu14+:GFP h–/h+ ade6-M210/ade6-M216 leu1-32/leu1-32 ura4-D18/ura4-D18 +/his2 meu14+:GFP/meu14+:GFP spo3-HA:LEU2/spo3-HA:LEU2 h90 spo3::ura4+ leu1-32 ura4-D18 h90 leu1-32 ura4-D18 meu14+:GFP h90 spo3::ura4+ leu1-32 ura4-D18 meu14+:GFP h90 spo15::ura4+ leu1-32 ura4-D18 h90 spo15+::ura4+ leu1-32 ura4-D18 meu14+:GFP h90 leu1-32:GFP-psy1+

YDO150

h90 leu1-32:GFP-psy1+ ura4-D18 meu14::ura4+

YDO120 YDO121

Nabeshima et al., 2001 Nabeshima et al., 2001 Nabeshima et al., 2001 Abe and Shimoda, 2000 Nakamura et al., 2001 Watanabe et al., 2001 This study This study This study This study This study Nakamura et al., 2001 This study This study Ikemoto et al., 2000 This study C. Shimoda (Osaka City University, Japan) This study

Meu14 of S. pombe meu14+ fragment harboring the NdeI and NotI sites in the N- and Ctermini, respectively. After performing a TA-cloning procedure, this DNA fragment was digested by NdeI and NotI and inserted into the pRGT1 vector (a gift from M. Yamamoto, University of Tokyo), which was designed to fuse GFP to the insert if the DNA was inserted in-frame via NdeI/NotI sites using the pREP1 vector (Maundrell, 1993). For Meu14-GFP protein expression during vegetative growth, the meu14+ DNA fragment generated by PCR to contain NdeI and NotI sites at the ends was inserted into pRGT81, a modified pRGT1 vector designed to carry a weak nmt promoter. This construct is termed pRGT81-meu14+. To obtain the 5′ upstream promoter region of meu14+, we performed PCR using the primers 5N (5′-CTGCAGATCCGAGCAAGAAGAGGCT-3′) and 5C (5′-CATATGGATTGTTTACG TTTCAGA-3′). The PstI and NdeI sites introduced into the oligonucleotides are underlined, respectively. The nmt1 promoter between the PstI and NdeI sites in the pRGT1-meu14+ construct was replaced by this amplified DNA fragment. We picked four independent clones and determined the DNA sequences of the amplified regions in order to select the clone whose sequence does not contain point mutations introduced by the PCR amplification. This construct is termed pNP-meu14+. Next, the NspI genomic DNA fragment corresponding to the 3′ downstream region of meu14+ (+61 to 1138 bp from stop codon) was inserted into the SmaI site between the GFP gene and the nmt1 terminator in the pNP-meu14+ construct. This meu14+-gfp DNA fragment carrying about 1 kb of both upstream and downstream regions was digested with PstI and KpnI, and then used to transform meu14+::ura4+ cells (YDO100). The Ura– transformants were obtained by screening for the clone that survived in the medium containing 5-Fruoroorotic Acid (5-FOA), which was confirmed by Southern blot analysis (data not shown). Immunofluorescence Meiotic cells were fixed following the procedure of Hagan and Hyams using glutaraldehyde and paraformaldehyde (Hagan and Hyams, 1988). In indirect immunofluorescence microscopy (Hagan and Yanagida, 1995), the SPB was stained with the anti-Sad1 antibody (a gift from M. Yanagida, University of Kyoto), microtubules were stained with the TAT-1 antibody [a gift from K. Gull, University of Manchester (Woods et al., 1989)], and the nuclear pore complex was stained with the MAb414 antibody (CRP, Denver, PA). Spo3-HA (a gift from C. Shimoda, City University of Osaka) was stained with the anti-HA antibody 3F10 (Boehringer Mannheim, Mannheim, Germany). Subsequently, a Texas-Red-conjugated sheep anti-mouse antibody (Amersham Biosciences, Piscataway, NJ) was used to visualize the microtubules, Spo3-HA, and the nuclear pore complex. An Alexa-488 or 594-conjugated goat anti-rabbit antibody (Molecular Probes, Eugene, OR), a Cy5-conjugated donkey anti-rabbit antibody (Jackson ImmunoResearch Laboratories, West Grove, PA), and a Texas-Redconjugated donkey anti-rabbit antibody (Amersham Biosciences, Piscataway, NJ) were used to stain Sad1. The samples were then stained with 0.2 mg/ml DAPI or Hoechst 33342 in PBS for 5 minutes and mounted with antifade-containing Vectashield mounting medium (Vector Laboratories, Burlingame, CA). Fluorescence images of these cells were observed using a fluorescence microscope (Axiophot, Zeiss, Germany or BX51, Olympus, Tokyo, Japan) with a charge-coupled device (CCD) cameras (Photometrics PXL1400) or Cool SNAP CCD camera (Roper Scientific, Sad Diego, CA). Immunofluorescence images were acquired using Adobe PhotoShop 6.0. Fluorescence in vivo imaging of Meu14 and chromosomes To stain chromosomes and detect Meu14-GFP in living meiotic cells, meu14+-gfp cells (YDO120) were first transferred to EMM2-N to induce meiosis. The meiotic cells were then stained with Hoechst 33342 (1.0 µg/ml) for a few minutes. Stained live cells were mounted on a coverslip by spotting, and microscopic observations were carried

2723

out with the Delta Vision microscope system (Applied Precision, Issaquah, WA). Images were taken with a 0.2 second exposure at 2 minute intervals. This system allows multi-color and threedimensional acquisition of digitized images with a cooled CCD camera. The images of the live meiotic cells stained with Hoechst 33342 were obtained as described (Chikashige et al., 1994). Preparation of cell extracts and immunoblotting S. pombe cells (3.3×108 cells) were suspended in 0.3 ml of HB buffer (25 mM MOPS, pH 7.2, 15 mM MgCl2, 15 mM EGTA, 60 mM βglycerophosphate, 15 mM p-nitrophe-nylphosphate, 0.1 mM Na3VO4, 1 mM dithiothreitol, 1 mM phenylmethylsulf o-nylfluoride, and 20 mg each of leupeptin and aprotinin per ml) and disrupted with acidwashed glass beads using a bead beater. Proteins in the extracts were separated by SDS-PAGE and transferred onto polyvinylidene difluoride membranes (Immobilon; Millipore, Bedford, MA). Blots were probed with the rat anti-GFP antibody (a gift from S. Fujita of Mitsubishi Kagaku Institute of Life Sciences, Tokyo, Japan) and the rabbit anti-Cdc2 antibody (Santa Cruz Biotechnology, Santa Cruz, CA). The bands bound by the horseradish peroxidase-conjugated goat anti-rat IgG (Santa Cruz Biotechnology, Santa Cruz, CA) for the antiGFP antibody or goat anti-mouse IgG (ICN Pharmaceuticals, Costa Mesa, CA) for the anti-Cdc2 antibody were visualized using the Renaissance system (NEN Life Sciences, Boston, MA). Electron microscopic observations Wild-type diploid (TP4D-5A/TP4D-1D) and meu14∆ diploid (YDO110) cells were grown in EMM2 medium at 33°C, transferred to EMM-N medium at 30°C, and harvested by centrifugation 6 or 8 hours later. The pelleted cells were cryofixed by high-pressurefreezing using a HPM 010 (Bal-Tec, Balzers, Liechtenstein) as previously described (Humbel et al., 2001) before being subjected to electron microscopy (Hitachi H-7600, Hitachi High-Technologies, Tokyo, Japan).

Results The meu14+ gene is expressed predominantly during meiosis We identified meu14+ as one of several S. pombe genes whose transcription is only upregulated in meiosis induced by nitrogen starvation using a meiosis-specific subtracted cDNA library (Watanabe et al., 2001). To test this, we took advantage of the fact that, upon nitrogen starvation, the heterozygous CD16-1 (h+/h–) strain initiates meiosis, while the homozygous CD16-5 (h–/h–) strain does not. Northern blot analysis using RNA from the CD16-1 and CD16-5 strains and a radiolabeled meu14+ probe indeed showed that the meu14+ transcript of 1.3 kb is detected only in the CD16-1 strain and that it emerges approximately 6-12 hours after nitrogen starvation (Fig. 1A). In this experiment, the progression through the various meiotic steps was monitored by counting the frequency of 4 nuclei at 2 hours intervals by fluorescence microscopy (Fig. 1A, bottom panel). Notably, no meu14+ transcription was observed during the mitotic cell cycle (Fig. 1A,B, 0 hour time point). When we used pat1-114 cells, which synchronously initiate meiosis at the restrictive temperature, we detected the hybridized bands only at 6, 8 and 10 hours. The peak was at 8 hours, at which time many of the cells were between the two meiotic nuclear divisions, meiosis I and meiosis II (Fig. 1B). Thus, we conclude that meu14+ begins to be expressed at some point between the two meiotic nuclear divisions.

2724


Furthermore, we investigated whether meu14+ gene expression occurs under the control of the meiosis-specific transcription regulator Mei4 because the Mei4 target sequence termed FLEX (GTAAACAAACAGA) occurs in the 5′ upstream region of

meu14+ (Horie et al., 1998; Abe and Shimoda, 2000). Northern blot analysis showed that meu14+ gene expression was not detectable in the mei4∆ strain (Fig. 1B, right panel). This suggests that meu14+ expression is under the direct control of Mei4.

Fig. 1. Expression of meu14+ and its paralogs during meiosis and structural comparison of their gene products. (A) The meu14+ gene is specifically transcribed during meiosis. Diploid h+/h– (CD16-1) and h–/h– (CD16-5) cells were exposed to nitrogen starvation, which triggers meiosis in the former but not the latter strain, and cells were collected at 2 hour intervals for RNA extraction. Upper panels: Total RNA was blotted and probed by radiolabeled meu14+, mfp1+, mfp2+ and aro3+ DNA fragments. The intensities of the bands obtained by using the aro3+ probe and ethidium bromide staining were used as loading controls. Numbers at the bottom represent the cell populations harvested at each 2 hour interval and assessed for the percentages of cells that had four nuclei (counted after staining the cells with DAPI). (B) meu14+ gene expression is under the control of the sporulation-specific Mei4 transcription factor. Diploid h–/h– pat1-114/ pat1-114 (JZ670) and h–/h– pat1-114 mei4∆ pat1-114 mei4∆ (AB4) cells were synchronized to enter the G1 phase by nitrogen starvation, and then induced to meiosis by temperature shift. Nitrogen was then added, and the cells were incubated at 34°C and collected at 2 hour intervals for RNA analysis as in panel A. (C) Structure of Meu14. (i) Conserved structural motifs. The coiled-coil region (C-C) is predicted using the COILS program with a 21-residue window setting (Lupas et al., 1991). (ii) Alignment of the predicted amino acid sequence of Meu14 with S. pombe and S. cerevisiae homologues. Homologous proteins were identified in a BLAST search using the protein sequence of Meu14. Hyphens represent gaps inserted to attain maximal homology. Amino acids identical to those in Meu14 are highlighted by a gray background. Amino acid numbers are denoted at the right-hand side of each sequence. The extended coiled-coil motif found in Meu14 is indicated by a solid line. (D) The phylogenetic tree of the Meu14 homologues. The tree was constructed based on the whole amino acid sequence of each protein by Genetyx-Mac software (Software Development, Tokyo, Japan) using the neighbor joining method (Saitou and Nei, 1987). The genomic DNA sequences used are recorded in the DDBJ/EMBL/GenBank database (accession nos. CAA97088, AAB68101, CAB16733 and CAA19279, respectively).

Meu14 of S. pombe Structural characterization of the meu14+ gene and its product We used the isolated meu14+ cDNA fragment as a probe to screen the genomic library of h– L972 cells (Fukushima et al., 2000). This allowed us to clone a DNA fragment that included the surrounding regions of the meu14+ gene, and the DNA sequence was determined (DDBJ accession No. AB016983). Comparison of the genomic DNA and cDNA sequences indicated that the meu14+ gene contains one intron-bearing consensus intron splice and branch sequences (data not shown). meu14+ encodes a predicted translation product consisting of 335 amino acids with a molecular weight of ~37 kDa. The PSORT II server program (http://psort.ims.utokyo.ac.jp/) was used to search for motifs and revealed that the Meu14 protein has a predicted coiled-coil (C-C) motif (Burkhard et al., 2001) in the central region of the molecule (Fig. 1Ci). The BLAST algorism (http://www.genome.ad.jp/) was used to search for homologous genes and revealed that there are two genes in S. pombe that are paralogous to meu14+. We have denoted them as mfp1+ (SPAC3C7.02C) and mfp2+ (SPCC736.15) after meu fourteen paralog. Notably, an intimate homolog of meu14+ does not exist in the genome of S. cerevisiae. Maximum matching alignment (Fig. 1Cii) and a phylogenetic tree (Fig. 1D) based on the overall amino acid sequences of the proteins clearly show that the Mfp1 and Mfp2 proteins are more similar to the budding yeast homologues Ygr086c and Yp1004c than to Meu14. Genes with obvious homology to meu14+ have not been identified in any other organism. Northern blot analysis indicated that, like meu14+, mfp1+ transcription is only induced during the meiosis/sporulation process, although it was expressed later (8-12 hours; Fig. 1A). It is strange, however, that when the pat1-114 strain is used, mfp1+ displays two peaks (Fig. 1B). We obtained this result repeatedly using different northern blots. The reason for this is currently under investigation. Nevertheless, when we constructed a null mutation of mfp1+, abnormal phenotypes in vegetative growth or the meiosis/sporulation of diploid cells were not observed (data not shown). With regard to mfp2+, northern blot analysis revealed that it is transcribed both during vegetative growth (0 hours) and meiosis (Fig. 1A). Moreover, neither mfp1+ nor mfp2+ seems to be under the control of Mie4 transcription factor (Fig. 1B, right panel). Thus, mfp1+ and mfp2+ genes were not studied further.

meu14∆ cells fail to sporulate and produce a high frequency of abnormal tetranucleated cells To assess the physiological role of meu14+, a deletion mutant that expresses no Meu14 protein was constructed by one-step gene replacement (see Materials and Methods). We confirmed the successful disruption of meu14+ by examining the transformants for the presence or absence of the uracil auxotrophic marker and by Southern analysis. Diploid cells in which one of the meu14+ genes had been replaced by ura4+ were sporulated and germinated. The segregation ratio compared to the wild-type was 1:1. All of the resulting spores were viable, indicating that the meu14+ gene is not essential for vegetative growth. The growth properties and the cell size and morphology of meu14∆ cells were also

2725

indistinguishable from those of the wild-type cells. Thus, we conclude that the meu14+ gene has no obvious function in the vegetative growth phase. FACS analysis indicated that premeiotic DNA synthesis was normal in meu14∆ cells (data not shown). Spore walls were not formed by diploid cells homozygous for the meu14∆ gene (Fig. 2A,B), indicating that meu14∆ cells are defective in sporulation. Progression through the various meiotic steps was monitored by counting the frequency of various numbers of nuclei at 2 hour intervals by fluorescence microscopy and showed that meu14∆ cells proceed normally through the onset of the two meiotic divisions, as its kinetics are comparable to that of the parental wild-type control (Fig. 2C). Thus, the onset of the nuclear divisions occurring during meiosis I and II is normal in meu14∆ cells. However, at 12 hours, at which point 80% of the wild-type cells had undergone sporulation, no asci were observed in meu14∆ cells, confirming sporulation in meu14∆ cells is severely impaired. Notably, horse-tail phase seems longer in meu14∆ cells than wild-type, which suggests that Meu14 may also play a role at this phase. The meu14∆ cells produced abnormal tetranucleate cells at a high frequency (Fig. 2D). We could classify these abnormal tetranucleate cells into five classes (see legend to Fig. 2D). Taken together, we conclude that meu14∆ cells become abnormal after meiosis II has commenced. SPB assembly is aberrant in meu14∆ cells during meiosis II To determine if the abnormal tetranucleate cells of meu14∆ strain are due to the aberrant function of the microtubules or the SPBs, meu14∆ cells were induced to enter meiosis. The cell populations enriched in anaphase II cells were then stained by Hoechst 33342 to reveal the nuclei and immunostained with the anti--tubulin antibody (TAT-1) and the anti-Sad1 antibody to delineate the microtubules and the SPBs (Hagan and Yanagida, 1995), respectively. We found that meu14∆ cells are normal at meiosis I (Fig. 3Av,vi). However, many of the meu14∆ cells progressing though meiosis II contained fragmented or unequally-shaped microtubule bundles (Fig. 3Ai-iv). Moreover, multiple Sad1stained objects were observed more frequently in meu14∆ cells (Fig. 3A) To confirm that the anti-Sad1 antibody specifically recognize the SPBs in meu14∆ cells, we induced cells that express the Spo15 protein fused to GFP to enter meiosis II and then stained them with Hoechst 33342 and the anti-Sad1 antibody. Spo15 is a SPB component that is required for SPB alteration in Meiosis II and sporulation (Ikemoto et al., 2000). The abnormal Sad1 signals colocalized with Spo15-GFP fluorescence signals (Fig. 3B). Thus, the organization or segregation of the SPBs in meu14∆ cells is abnormal. This suggests that the SPBs may not be able to act properly as scaffolds for forespore membrane formation, which may explain the lack of spore formation in meu14∆ cells. Supporting this is that the meu14∆ cells lack the crescent form of SPBs, which resembles the abnormality observed in spo15∆ cells (Ikemoto et al., 2000). Thus, the abnormal spore formation in meu14∆ cells may be due to a failure in the formation or segregation of SPBs during meiosis II.

2726


Subcellular localization of Meu14-GFP during meiosis To assess the behavior of the endogenous Meu14 protein during meiosis, we constructed a strain in which meu14+ was replaced by a meu14+-gfp gene that is designed to express the Meu14 protein fused with GFP at its C-terminal end. Since expression of this meu14+-gfp gene is controlled by the native meu14+ promoter, its expression is expected to be identical to that of the intact meu14+ gene. That the meu14+-gfp gene could fully complement the meiotic defects in meu14∆ diploid cells, and that these cells generated normal ascospores implies that the Meu14-GFP protein retains its function (data not shown). Western blot analysis using cell lysates reveals that the expressed Meu14-GFP protein migrates at the expected size, indicating it is not degraded in vivo (Fig. 4A). Western blotting also showed it is expressed only during meiosis II (4 to 8 hours) (Fig. 4A). Cells carrying meu14+-gfp were induced to enter meiosis and were fixed before fluorescence analysis of DNA, Meu14GFP, and tubulin. The timing of meiotic progression in the meu14+-gfp cells was confirmed to be quite similar to that of wild-type cells (data not shown). Based on the number of nuclei and the morphology of the microtubules, we judged the stage of meiosis of the individual cells and collected typical images to represent each stage of meiosis. Fluorescence signals from Meu14-GFP were not detected in mitotically growing cells or in early meiotic cells (at interphase, horse-tail, or meiosis I; Fig. 4B). The Meu14-GFP signal first appeared at prometaphase II as a blur in both the nucleus and the

cytoplasm. After this, four strong rings appeared in the cells at metaphase II, and these four rings became bigger at anaphase II. The rings shrank at post-anaphase II and were again visualized as four small rings. When we measured the distances between the two Meu14GFP rings and plotted these versus the nuclear distance (Fig. 4C; upper right panel) or the length of microtubules (lower right panel), we found that the curves are similar, indicating that the Meu14-GFP rings behave in good coordination with both nuclear division and microtubule elongation. By counting the number of cells displaying Meu14-GFP signals, we found that Meu14-GFP was detectable in 78% of short-spindle cells at prometaphase II, in 100% of elongated-spindle cells at anaphase II, and in 48% of spindle-negative cells at post-anaphase II. Immunostaining at various stages of meiosis also revealed that Meu14-GFP localizes next to the Sad1 signal on the SPB only at metaphase II (Fig. 4D; white arrow). This suggests that Meu14-GFP is located on the cytoplasmic side of the SPB. When the nuclear pore complex (NPC) was immunostained with the mAB414 antibody (Wilkinson et al., 2000), Meu14GFP was detected outside of the NPC at late meiosis II (Fig. 4E). Localization of Meu14-GFP in live cells during meiosis The behavior of Meu14-GFP during meiosis suggests that it is transiently expressed and then degraded (Fig. 1B, Fig. 4A), and that before its degradation it moves around in the cell. To further investigate this dynamic behavior of Meu14-GFP, we observed the protein in live meu14+-gfp cells that were induced to enter meiosis and then stained with Hoechst 33342 to visualize the nucleus. The images have been stored as a file that can be run as an animation Fig. 2. Phenotypes of meu14∆ cells. (A) The meu14 disruptant is defective in spore formation. Shown are DIC (differential interference contrast) and fluorescence photographs of Hoechst 33342-stained cells of the wild-type (TP4D-5A/TP4D-1D) and the meu14∆ mutant (YDO100) strains after 16 hours of nitrogen starvation. Bar, 10 µm. (B) Absence of spore walls in meu14∆ cells induced to sporulate. Wt and meu14∆ cells were streaked on a sporulation plate (EMM-N) and incubated at 28°C for 4 days. They were then exposed for 5 minutes to iodine vapor, which stains cells that have sporulated dark brown. (C) Progression of meiosis in wild-type and meu14∆ cells. Up to 1×107 cells/ml cultured overnight in liquid growth medium (EMM2) were incubated with shaking at 30°C in liquid sporulation medium (EMM-N). A portion of the culture was taken every 2 hours and stained with DAPI. Cells were classified based on the numbers of nuclei. 䊐, interphase (mononucleate); 䉫, Horse-tail; 䊊, binucleate; 䉭, tetranucleate. For each sample, approximately 200 cells were counted. Values depict one representative result of four independent experiments. (D) The meu14∆ cells frequently produce abnormal tetranucleate cells. Tetranucleate cells were classified according to the number and position of the four nuclei per cell. 1, normal pattern; 2, unequally segregated nuclei; 3, missegregated nuclei; 4, unsegregated nuclei; 5, abnormally distributed nuclei; 6, distorted nuclei. The percentages of Wt and meu14∆ cells in each category after 12 hours of nitrogen starvation were compared. At least 200 cells were counted.

Meu14 of S. pombe available on the internet. We show in Fig. 5 several time lapse images taken at 6 minute intervals of a cell passing from prophase II to metaphase II (Fig. 5A; see Movies 1 and 2, http://jcs.biologists.org/supplemental) and of a cell progressing from metaphase II to anaphase II (Fig. 5B; see Movies 3 and 4). Meu14-GFP first appears at early meiosis II as a dot at the periphery of each of the two nuclei. The dots then duplicate and move around the nuclear membrane (as indicated by white arrowheads in Fig. 5Ai,ii). This behavior of Meu14-GFP resembles that of the SPB. Thereafter, the Meu14-GFP signals start to increase (time 0 in Fig. 5A) from the opposite ends of the two nuclei. Each nucleus develops small ring-shaped structures at either end. These structures gradually increase their diameters in a synchronized manner. The two rings approach each other until their diameters are maximized, after

Fig. 3. Meu14 is required for the proper segregation of SPBs. (A) Chromosomal DNA, SPBs, microtubules in meu14∆ (YDO100) cells were stained by Hoechst 33342 (blue), anti-Sad1 antibody (green) and anti-αtubulin antibody TAT-1 (red), respectively. Merged images are also shown. The following typical images are displayed. (i) Microtubule extension between the two pairs of dividing nuclei at meiosis II is not synchronized. (ii) Multiple foci of Sad1 staining are observed. (iii) Microtubules between the two nuclei are not detected. (iv) Localization of microtubules is abnormal and there are more than four Sad1 foci. (v,vi) meu14∆ (YDO100) cells at metaphase I (v) or anaphase I (vi). (B) Sad1 colocalizes with the SPB protein Spo15 during meiosis II. The following typical images are displayed for meu14∆ (YDO111) (i-iii) or wild-type (iv,v) cells. (i) One of the four SPBs is abnormal. (ii) More than four SPBs are observed. (iii) Nuclei are not properly divided at meiosis II. (iv,v) Wild-type cells at metaphase II (iv) and anaphase II (v). Bar, 10 µm.

2727

which they start to separate because of the nuclear division of meiosis II. Subsequently, as the nuclei separate from each other, the diameters of the two rings gradually shrink until they become two dots in the center of the two dividing nuclei (Fig. 5B). In post-anaphase II, the strengths of the Meu14-GFP ring/dot signals gradually diminish and the Meu14-GFP signal is observed to be diffusely distributed among the four spores. The diminished Meu14-GFP signal just before the maturation of the ascospores coincides with the disappearance of the Meu14-GFP band in immunoblotting (Fig. 4A). At the late stage of meiosis II, the dim stain throughout the nuclear region almost completely disappears and only rings are observed, which is shown in Fig. 5Aii and Fig. 5Bii. The pictures are colorless so as to highlight the GFP signals. These pictures suggest that Meu14 localizes in the nuclear region and then assembles into rings during meiosis II.

2728


Meu14 of S. pombe Fig. 4. Subcellular localization of Meu14-GFP in fixed cells. (A) Detection of Meu14-GFP protein by western blotting, which is expressed in a meiosis-specific manner. A band corresponding to Meu14-GFP with the expected size (shown by an arrowhead) is detected predominantly 4-8 hours after nitrogen starvation, after which it suddenly disappears and cannot be detected at all at 10 hours and 12 hours (uppermost panels). In contrast, a band for GFP alone is detected primarily at 6-12 hours (middle panels), probably because the Meu14 portion of Meu14-GFP is easily degraded, releasing the GFP portion. Western blot probed by an anti-Cdc2 antibody was used as a loading control. Asterisks indicate nonspecific bands. The extracts of wild-type cells (TP4D-5A/TP4D1D) transformed with pRGT81-meu14+ (meu14+-GFP) or pRGT81 (GFP vector) are used to identify the expressed Meu14-GFP protein. These extracts are denoted by 14 or G, respectively (rightmost lanes).

Fig. 5. Dynamic movement of Meu14-GFP in live meiotic cells as revealed by time lapse observations. (A) Cells (YDO120) passing from prometaphase II to metaphase II. (i) Merged images of Meu14GFP (green) and nuclei stained by Hoechst 33342 (blue) recorded at 2 minute intervals and shown at 6 minute intervals. (ii) An enlarged Meu14-GFP image at 0 minutes is shown to highlight the appearance of the Meu14-GFP dot/ring structure (arrowhead) at the end of the nucleus. (B) Cells passing from metaphase II to anaphase II. (C) Three-dimensional images of the Meu14-GFP structure during anaphase II. (D) Enlarged views of Meu14-GFP localization at the late stage of meiosis II. A thin pouch-like image containing trace amounts of Meu14-GFP is observed, at the end of which a strong band of Meu14-GFP is detected. Bar, 10 µm. Please refer to Movies 1-4 for the dynamic Meu14-GFP profile and Movie 5 for three-dimensional images of Meu14-GFP rings.

2729

(B) Fluorescence from Meu14-GFP observed at various stages of meiosis. Cells were immunostained with the TAT1 antibody to mark the microtubules (red). Cells bearing meu14+-gfp (YDO120) were induced to enter meiosis and then chemically fixed and analyzed by fluorescence microscopy for DNA (blue), Meu14-GFP (green), and microtubules (red). The three images are merged and shown in the rightmost panels. (C) Correlation of the distance between the two Meu14-GFP rings (c) with either (a) the distance between two nuclei or (b) the length of the microtubules. (D) meu14+-gfp (YDO120) cells immunostained by the anti-Sad1 antibody to visualize the SPB at various stages of meiosis. It appears that Meu14-GFP is located on the cytoplasmic side of the SPB (white arrow). (E) meu14+-gfp (YDO120) cells immunostained by the anti-NPC antibody mAB414. Bar, 10 µm.

2730


When the Meu14-GFP rings are closely examined by an enlarged view rotated on the Y-axis by an angle of seventy-two degrees to obtain three-dimensional images, they are found to form a structure like ‘assembled beads’ (Fig. 5C, Movie 5). We counted more than ten beads that composed each Meu14-GFP ring in these images. Notably, at the end of meiosis II, when the strong Meu14-GFP signal is observed as an intense dot at the edge of the nucleus (Fig. 5B), the weak Meu14-GFP signal was also observed to have a balloon-like shape (Fig. 5D). This is probably the forespore membrane, indicating that some portion of Meu14-GFP is distributed at the forespore membrane. Meu14 is required for the extension of the forespore membrane from the SPBs To investigate whether Meu14 is located on the forespore membrane, we compared the localization of Meu14-GFP to that of the Spo3-HA molecule, which is a forespore membrane component (Nakamura et al., 2001). Spo3-HA fluorescence signals appeared as two semicircular structures that enclose the nucleus at metaphase II, and as four capsular structures enclosing the nucleus at anaphase II (Fig. 6A). The SPBs were situated at the center of each semicircle displayed by Spo3-HA. At the edges of these semicircles and capsules, Meu14-GFP could be detected. Thus, Meu14-GFP occurs at the leading edge of the forespore membrane during meiosis II. The fluorescence of Meu14-GFP that localized at the edge of the forespore membrane became reduced as the edge of the forespore membrane encloses itself, while Spo3, the component of the forespore membrane, remains until the forespore membrane growth is completed (Fig. 6Ai-v) (Nakamura et al., 2001). The edge of the extended forespore membrane appears to play an important step for the mature spores. Next, we examined the localization of Psyl-GFP in meu14∆ cells by fluorescence microscopy. Psyl, like Spo3, also serves as a marker for the formation of the forespore membrane (Nakamura et al., 2001). The reason for using the Psy1-GFP is that fluorescence of Psy1-GFP can be detected much more easily during fluorescence microscopy than the fluorescence of Spo3-HA (data not shown). Psy1-GFP was integrated into the genome of meu14∆ cells and wild-type cells. The cells were then induced to enter meiosis II and stained with the anti-Sad1 antibody to mark the SPBs (Fig. 6B). In wild-type cells, the forespore membranes visualized by Psy1-GFP assembled normally as a circle and at meiosis II enclosed the haploid nucleus stained by Hoechst 33342 (Fig. 6Bi). In wild-type cells, the forespore membrane extends symmetrically from the SPB and results in this circular structure [data not shown (Nakamura et al., 2001). Here, the position of SPBs indicates the orientation of the nuclear division at meiosis II. However, in most of the meu14∆ cells, the semicircular forespore membrane is formed in an inappropriate place, thus failing to properly encircle the nucleus (Fig. 6Bii-iv). In meu14∆ cells, only few of the forespore membrane leading edges were normally assembled as in wild-type cells (Fig. 6C). The majority were abnormally assembled. These observations indicate that Meu14 is required for the proper extension of forespore membrane from the SPBs. As shown in Fig. 6C, the meu14∆ ascospores were abnormal in number and shape.

In contrast, in wild-type cells, the forespore membrane encapsulated each haploid nucleus (>90%), while the initiation of the forespore membrane in the meu14∆ cells is normal, most later fail to develop a normal morphology. Meu14-GFP can form ring shapes in the absence of Spo3 or Spo15 Previous reports show that spo15+ is required for the alternation of the SPBs in meiosis II and spo3+ is required for the accurate formation of the forespore (Ikemoto et al., 2000; Nakamura et al., 2001). To examine if Spo3 and Spo15 are required for the localization of Meu14-GFP at the edge of forespore membrane, we constructed the spo3∆ meu14+-gfp and spo15∆ meu14+-gfp strains (Fig. 6D). The h90spo3∆ meu14+-gfp or h90 spo15∆ meu14+-gfp strains were induced to proceed with zygotic meiosis and sporulation by nitrogen starvation, after which the localization of Meu14-GFP was visualized by fluorescence microscopy. As shown in Fig. 6Di, Meu14-GFP was detected at the edge of the forespore membrane in zygotic meiosis as also observed in azygotic meiosis (Fig. 4D). The fluorescence of Meu14-GFP was found to form ring shapes in both spo3∆ and spo15∆. The four ring shapes of Meu14-GFP showed a synchronized action in spo3∆ but not in spo15∆. In spo3∆ cells, we found that Meu14 is situated at the edge of a spore-like body (Fig. 6Diii,iv). In many spo15∆ cells, Meu14 was undetectable (Fig. 6Dv) while in others, the number, size and position of Meu14 signals were abnormal (Fig. 6Dvi,vii). Thus, Meu14-GFP is able to localize at the edge of the abnormal forespore membrane in the absence of Spo3 and Spo15. Morphology observed by electron microscopy To investigate the structure of the forespore membrane and the spore wall of meu14∆ cells in more detail, we examined their morphology by thin-section electron microscopy (EM). We treated the cells by high-pressure freezing (Humbel et al., 2001) to minimize the artifacts that can arise during the prefixation process. The substituted cells were then fixed and observed by electron microscopy. From the meu14∆ cells, we obtained images showing that the forespore membranes and spore wall failed to form accurately and displayed abnormal morphology (Fig. 7Ai) as compared to those of wild-type cells (Fig. 7Aii). Enlarged pictures show some of the various abnormalities of the meu14∆ spore wall observed, such as a small spore-like structure without a nucleus (iii), an incompletely encapsulated nucleus (iv), and a nucleus without a spore wall (v). Furthermore, in meu14∆ cells, the thickness of the spore wall is not homogeneous compared to that of the wild-type cells (vi). EM images also show that the interaction of the forespore membrane (black arrows) with the SPB (white arrows) becomes abnormal in meu14∆ cells (Fig. 7Bi). That is, the forespore membrane segregates from the SPB, thus extending abnormally to the opposite direction to the nucleus (ii). The relative orientation of the forespore membrane with the SPB (iii) is distinct from that of wild-type cells (iv). This abnormal segregation of the forespore membrane from SPB is consistent with our indirect immunofluorescence data (Fig. 6B). Some meu14∆ cells showed abnormally shaped SPBs as indicated by

Meu14 of S. pombe

2731

Fig. 6. Meu14 is localized at the leading edge of the forespore membrane and is required for its extension. (A) Localization of Meu14-GFP, Spo3-HA, and Sad1 during meiosis II. meu14+-gfp spo3::HA diploid cells (YDO121) at metaphase II (panel i and ii) and anaphase II (panels iii-v) were analyzed by immunofluorescence microscopy. Hoechst 33342 staining is blue, Meu14-GFP is green, anti-HA staining is red, and anti-Sad1 staining is yellow. (B) Psy1, a forespore membrane component, localizes abnormally in h90meu14∆psy1+-gfp cells (YDO150). h90psy1+-gfp cells (YN68) (panel (i) and h90meu14∆psy1+-gfp cells (panels iiiv) harboring the integrated psy1+-gfp gene were induced to enter meiosis II and were analyzed by immunofluorescence microscopy. Hoechst 33342 staining is blue, anti-Sad1 staining is red, and Psy1-GFP is green. (C) Frequency of the abnormal forespore formation in h90meu14∆ cells. Stained cells were classified into the following 7 classes: class I, all four forespore membranes were formed next to nuclei; class II, all four membranes were formed next to nuclei but were crushed; class III, one or two forespore membranes alone were formed normally; class IV, membranes were formed but they do not properly enclose the nucleus; class V, one or two membranes were formed but others formed next to the nuclei; class VI, all four membranes were crushed and formed aggregates near the nuclei; class VII, all four membranes were formed normally. At least 250 cells were counted. (D) Localization of Meu14-GFP in the absence of Spo3 or Spo15. h90meu14+-gfp cells (YDO50; panels i and ii), h90 meu14+-gfp spo3::ura4+ cells (YDO130; panels iii and iv), and h90 meu14+-gfp spo15::ura4+ cells (YDO10; panels v-vii) undergoing meiosis II were analyzed by immunofluorescence microscopy. Hoechst 33342 staining is blue, Meu14-GFP is green, and anti-Sad1 staining is red. Bar, 10 µm.

white arrows in Fig. 7C. In such cases, encapsulation of the nucleus by the forespore membrane is mostly unsuccessful (i and iv). Enlarged pictures display the abnormally enclosed forespore membrane without a nucleus (ii) next to the naked

nucleus (iii), both of which carry improperly formed SPBs. Moreover, some of the SPBs do not localize between the nucleus and forespore membrane but rather, are situated outside of these structures (v and vi).

2732


The organization of the spindles are also abnormal in meu14∆ cells (Fig. 7D), as some cells have abnormally bundled spindle-like structures (i) that are improperly enclosed by the abnormally formed forespore membranes. Some meu14∆ cells showed seemingly normal spindle formation (ii-iv), but enclosure of the bundle is abnormal when no SPB is found (iii). It is also unusual that a spindle-like structure is observed in the spores, indicating that the handling of the spindles is also abnormal in meu14∆ cells (Fig. 7Av).

Discussion Meu14 participates in meiosis II In the present study, we have shown that meu14+ is a meiosisspecific gene, and the results are consistent with the possibility that meu14+ expression is under the direct control of Mei4. Meu14 was consistently observed to be present only at meiosis II as judged by northern blotting (Fig. 1A,B), western blotting (Fig. 4A) and fluorescence microscopy of chemically fixed (Fig. 4B) and live cells (Fig. 5A,B) expressing the Meu14-GPF

Meu14 of S. pombe

2733

Fig. 7. Thin section electron microscopy of meu14∆ cells. (A) EM images of the meu14∆ (YDO100) (i) and the wild-type (TP4D-5A/TP4D1D) (ii) strains, and enlarged views (iii-vi) after spore wall formation had commenced. (B) EM images of the meu14∆ (i) and the wild-type (iv) strains, and enlarged views (ii and iii) to show the abnormal encapsulation of the nuclei in the meu14∆ cells. (C) EM images of the meu14∆ cells (i and iv), and enlarged views (ii, iii, v and vi) to show the abnormal localization of the SPBs in the meu14∆ cells. (D) An EM image of the meu14∆ cells shows that the forespore membrane abnormally encloses the spindle (i). An example (ii) demonstrates that the absence (iii) or presence (iv) of SPB affects the proper formation of the forespore membrane in the meu14∆ cells. Pictures of wild-type cells are encircled by black lines. White arrowheads (spindle), white arrows (SPB) and black arrows (forespore membrane). White scale bar, 1 µm. Black scale bar, 0.5 µm.

fusion protein. We found that meu14∆ cells were defective in microtubule elongation and organization as well as in the segregation and duplication of SPBs, and that this caused abnormal nuclear division at meiosis II (Fig. 3). Furthermore, Meu14 seems to act not only to stabilize the proper segregation of SPBs during meiosis II but also to accurately encapsulate the haploid nuclei by guiding forespore membrane formation (Fig. 5).

During meiosis II, Meu14-GFP is found first at the cytoplasmic side of the SPB (Fig. 4D) and then at the leading edge of the forespore membrane (Fig. 6A). Time-lapse microscopic analysis using a meu14+-gfp strain (Fig. 5A,B) allowed us to observe the dynamic behavior of Meu14-GFP in live cells during meiosis II. Meu14-GFP first appears as a dot at the periphery of the nucleus, where it is duplicated and then moves to either end of the nucleus to form a pair of rings. The

2734


rings gradually increase in diameter. At the same time, the forespore membrane, which grows from the SPB, also increases. After the forespore membrane completely encloses the haploid nucleus at the end of meiosis II, the diameter of the rings gradually shrink until they appear as dots, after which they disappear. Thus, Meu14 is required specifically at meiosis II, where it seems important for several processes: (i) it guides the formation of the forespore membrane as it develops during meiosis II and sporulation, and (ii) it assists the equal partitioning of the nucleus and the temporal coordination of this event with chromosome segregation during meiosis II. Many studies have been conducted to comprehensively identify and characterize the subcellular localization of yeast proteins involved in meiosis and mitosis (Sawin and Nurse, 1996; Ding et al., 2000). Of note is that none of these studies identified Meu14. Thus, the subcellular localization of Meu14GFP during S. pombe meiosis is unique. Loss of Meu14 may destabilize the SPB The dynamic movement of Meu14 at meiosis II indicates that Meu14 is a novel type of molecule that regulates the architecture of cellular structures by using the SPB as one of its scaffolds. The SPB is a multilayered proteinaceous structure that is inserted into the nuclear envelope in S. pombe (Hirata and Tanaka, 1982; Ding et al., 1997). It is thought that the SPB buried in the nuclear membrane is stable due to astral microtubule interaction with the cell cortex. In the second meiotic nuclear division, the astral microtubules on the cytoplasmic side are exchanged with a multilayered SPB structure from which the forespore membrane, whose analogue in S. cerevisiae is the prospore membrane (Neiman, 1998), extends to enclose the spore nucleus. Thus, during meiosis II, astral microtubules are absent. We believe that the lack of Meu14 causes the SPB buried in the nuclear membrane to become unstable, which would explain the fact that meu14∆ cells display deformed and multiple SPBs that are not embedded in the nuclear envelope during meiosis II (Fig. 7C). This also explains our observation that the SPBs in meu14∆ cells, when observed by Spo15-GFP, a protein that regulates the onset of forespore membrane extension (Ikemoto et al., 2000), are abnormally organized or segregated. It is likely that Meu14 also plays a role in the structure of meiosis I SPB, possibly during anaphase I in preparation for meiosis II. Thus, Meu14 may already associate with the meiosis I SPBs, but it cannot be detected because Meu14-GFP at this stage is too faint to be recognized as GFP signals (Fig. 4B). Meu14 guides the formation of the forespore membrane Recently, Spo3 and Psy1 of S. pombe and Don1, Ady3, and Ssp1 of S. cerevisiae (Knop and Strasser, 2000; MorenoBorchart et al., 2001; Nickas and Neiman, 2002) were identified to be protein components of the forespore membrane. All have been shown to be localized at the cytoplasmic side of SPB. We used a Psy1-GFP integrated cell to trace the assembly and extension of the forespore membrane in meu14∆ cells that were also stained with markers for the nucleus and SPBs. We observed that the forespore membrane had an abnormally assembled leading edge that could not

extend out from the SPB and thus could not properly encircle the nucleus in most meu14∆ cells (Fig. 6B). We also observed that the loss of meu14+, as in the spo3 mutant, caused sporulating cells to generate spore-like structures lacking nuclei (Fig. 6Bii-iv). This indicates that Spo3 and Meu14 are both required for the accurate enclosure of the spore nucleus by the forespore membrane. However, unlike Spo3, whose localization is distinct from that of Meu14 and stays in the membrane until forespore membrane development is completed, the intensity of the Meu14-GFP signal at the leading edge of the forespore membrane diminishes as the forespore membrane grows, indicating that Meu14 serves a different function to Spo3. With regard to Don1, its subcellular localization is almost identical to that of Meu14 except that Don1 appears in the cytoplasm during meiosis I (Moreno-Borchart et al., 2001). In contrast, Meu14 appears in the cytoplasm for the first time at meiosis II (Fig. 5A). Furthermore, there is no homology in the amino acid sequences of Don1 and Meu14 and the DON1 gene is not essential for forespore membrane formation (MorenoBorchart et al., 2001). Thus, Meu14 is distinct from Don1 in both its structure and function. Taken together, it appears that one function of Meu14 is to guide the formation of the forespore membrane at its front edge, acting to all intents and purposes as a size-adjustable steel ring that is used to make a large soap bubble. We thank C. Shimoda and M. Yamamoto for S. pombe cells and vectors, M. Yanagida for the anti-Sad1 antibody, S. Fujita for the antiGFP antibody and K. Gull for the anti-tubulin antibody (TAT-1). In particular, we thank C. Shimoda for the gift of the psy1+-gfp strain. We also thank P. Hughes for critically reading the manuscript and T. Nakamura for technical suggestions and critical discussions. This work was supported by a Grant-in-aid for Scientific Research on Priority Areas from the Ministry of Education, Science, Sports and Culture of Japan, and a grant from The Uehara Memorial Foundation to H.N.

References Abe, H. and Shimoda, C. (2000). Autoregulated expression of Schizosaccharomyces pombe meiosis-specific transcription factor Mei4 and a genome-wide search for its target genes. Genetics 154, 1497-1508. Alfa, C., Fantes, P., Hyams, J., McLeod, M. and Warbrick, E. (1993). Experiments With Fission Yeast: A Laboratory Course Manual. Cold Spring Harbor Press, Cold Spring Harbor, NY, USA. Burkhard, P., Stetefeld, J. and Strelkov, S. V. (2001). Coiled coils: a highly versatile protein folding motif. Trends Cell Biol. 11, 82-88. Chikashige, Y., Ding, D. Q., Funabiki, H., Haraguchi, T., Mashiko, S., Yanagida, M. and Hiraoka, M. (1994). Telomere-led premeiotic chromosome movement in fission yeast. Science 264, 270-273. Ding, R., West, R. R., Morphew, D. M., Oakley, B. R. and McIntosh, J. R. (1997). The spindle pole body of Schizosaccharomyces pombe enters and leaves the nuclear envelope as the cell cycle proceeds. Mol. Biol. Cell 8, 1461-1479. Ding, D. Q., Tomita, Y., Yamamoto, A., Chikashige, Y., Haraguchi, T. and Hiraoka, Y. (2000). Large-scale screening of intracellular protein localization in living fission yeast cells by the use of a GFP-fusion genomic DNA library. Genes Cells 5, 169-190. Fukushima, K., Tanaka, Y., Nabeshima, K., Yoneki, T., Tougan, T., Tanaka, S. and Nojima, H. (2000). Dmc1 of Schizosaccharomyces pombe plays a role in meiotic recombination. Nucleic Acids Res. 28, 27092714. Hagan, I. and Hyams, S. J. (1988). The use of cell division cycle mutants to investigate the control of microtubule distribution in the fission yeast Schizosaccharomyces pombe. J. Cell Sci. 89, 343-357. Hagan, I. and Yanagida, M. (1995). The product of the spindle formation

Meu14 of S. pombe gene sad1+ associates with the fission yeast spindle pole body and is essential for viability. J. Cell Biol. 129, 1033-1047. Hirata, A. and Shimoda, C. (1994). Structural modification of spindle pole bodies during meiosis II is essential for the normal formation of ascospores in Schizosaccharomyces pombe: ultrastructural analysis of spo mutants. Yeast 10, 173-183. Hirata, A. and Tanaka, K. (1982). Nuclear behavior during conjugation and meiosis in the fission yeast Schizosaccharomyces pombe. J. Gen. Appl. Microbiol. 28, 263-274. Horie, S., Watanabe, Y., Tanaka, K., Nishiwaki, S., Fujioka, H., Abe, H., Yamamoto, M. and Shimoda, C. (1998). The Schizosaccharomyces pombe mei4+ gene encodes a meiosis-specific transcription factor containing a forkhead DNA-binding domain. Mol. Cell. Biol. 18, 2118-2129. Humbel, B. M., Konomi, M., Takagi, T., Kamasawa, N., Ishijima, S. A. and Osumi, M. (2001). In situ localization of beta-glucans in the cell wall of Schizosaccharomyces pombe. Yeast 18, 433-444. Ikemoto, S., Nakamura, T., Kubo, M. and Shimoda, C. (2000). S. pombe sporulation-specific coiled-coil protein Spo15p is localized to the spindle pole body and essential for its modification. J. Cell Sci. 113, 545-554. Knop, M. and Strasser, K. (2000). Role of the spindle pole body of yeast in mediating assembly of the prospore membrane during meiosis. EMBO J. 19, 3657-3667. Lupas, A., van Dyke, M. and Stock, J. (1991). Predicting coiled coils from protein sequences. Science 252, 1162-1164. Maundrell, K. (1993). Thiamine-repressible expression vectors pREP and pRIP for fission yeast. Gene 123, 127-130. Moreno, S., Klar, A. and Nurse, P. (1991). Molecular genetic analysis of fission yeast Schizosaccharomyces pombe. Methods Enzymol. 194, 795823. Moreno-Borchart, A. C., Strasser, K., Finkbeiner, M. G., Shevchenko, A. and Knop, M. (2001). Prospore membrane formation linked to the leading edge protein (LEP) coat assembly. EMBO J. 20, 6946-6957. Nabeshima, K., Kakihara, Y., Hiraoka, Y. and Nojima, H. (2001). A novel meiosis-specific protein of fission yeast, Meu13p, promotes homologous pairing independently of homologous recombination. EMBO J. 20, 38713881. Nakamura, T., Nakamura-Kubo, M., Hirata, A. and Shimoda, C. (2001). The Schizosaccharomyces pombe spo3+ gene is required for assembly of the forespore membrane and genetically interacts with psy1+-encoding syntaxin-like protein. Mol. Biol. Cell 12, 3955-3972. Nakase, Y., Nakamura, T., Hirata, A., Routt, S. M., Skinner, H. B., Bankaitis, V. A. and Shimoda, C. (2001). The Schizosaccharomyces pombe spo20(+) gene encoding a homologue of Saccharomyces cerevisiae Sec14 plays an important role in forespore membrane formation. Mol. Biol. Cell 12, 901-917. Neiman, A. M. (1998). Prospore membrane formation defines a developmentally regulated branch of the secretory pathway in yeast. J. Cell Biol. 140, 29-37. Neiman, A. M., Katz, L. and Brennwald, P. J. (2000). Identification of

2735

domains required for developmentally regulated SNARE function in Saccharomyces cerevisiae. Genetics 155, 1643-1655. Nickas, M. E. and Neiman, A. M. (2002). Ady3p links spindle pole body function to spore wall synthesis in Saccharomyces cerevisiae. Genetics 160, 1439-1450. Pringle, J. R., Broach, J. R. and Jones, E. W. (1997). The Yeast Saccharomyces: Cell Cycle and Cell Biology. Cold Spring Harbor, NY: Cold Spring Harbor Laboratory Press. Saitou, N. and Nei, M. (1987). The neighbor-joining method: a new method for reconstructing phylogenetic trees. Mol. Biol. Evol. 4, 406-425. Sawin, K. E. and Nurse, P. (1996). Identification of fission yeast nuclear markers using random polypeptide fusions with green fluorescent protein. Proc. Natl. Acad. Sci. USA 93, 15146-15151. Shaw, S. L., Yeh, E., Maddox, P., Salmon, E. D. and Bloom, K. (1997). Astral microtubule dynamics in yeast: a microtubule-based searching mechanism for spindle orientation and nuclear migration into the bud. J. Cell Biol. 139, 985-994. Shimada, M., Nabeshima, K., Tougan, T. and Nojima, H. (2002). The meiotic recombination checkpoint is regulated by checkpoint rad+ genes in fission yeast. EMBO J. 21, 2807-2818. Sun, G. H., Hirata, A., Ohya, Y. and Anraku, Y. (1992). Mutations in yeast calmodulin cause defects in spindle pole body functions and nuclear integrity. J. Cell Biol. 119, 1625-1639. Tanaka, K. and Hirata, A. (1982). Ascospore development in the fission yeasts Schizosaccharomyces pombe and S. japonicus. J. Cell Sci. 56, 263279. Tougan, T., Chiba, Y., Kakihara, Y., Hirata, A. and Nojima, H. (2001). Meu10 is required for spore wall maturation in Schizosaccharomyces pombe. Genes Cells 7, 217-231. Watanabe, T., Miyashita, K., Saito, T. T., Yoneki, T., Kakihara, Y., Nabeshima, K., Kishi, Y. A., Shimoda, C. and Nojima, H. (2001). Comprehensive isolation of meiosis specific genes identifies many unusual non-coding transcripts in Schizosaccharomyces pombe. Nucleic Acids Res. 29, 2327-2337. Wilkinson, C. R., Ferrell, K., Penney, M., Wallace, M., Dubiel, W. and Gordon, C. (2000). Analysis of a gene encoding Rpn10 of the fission yeast proteasome reveals that the polyubiquitin-binding site of this subunit is essential when Rpn12/Mts3 activity is compromised. J. Biol. Chem. 275, 15182-15192. Woods, A., Sherwin, T., Sasse, R., MacRae, T. H., Baines, A. J. and Gull, K. (1989). Definition of individual components within the cytoskeleton of Trypanosoma brucei by a library of monoclonal antibodies. J. Cell Sci. 93, 491-500. Yamamoto, M., Imai, Y. and Watanabe, Y. (1997). Mating and sporulation in Schizosaccharomyces pombe (ed. J. R. Pringle, J. B. Broach and E. W. Jones), pp. 1037-1106. New York: Cold Spring Harbor Laboratory Press. Yoo, B. Y., Calleja, G. B. and Johnson, B. F. (1973). Ultrastructural changes of the fission yeast (Schizosaccharomyces pombe) during ascospore formation. Arch. Mikrobiol. 91, 1-10.