|Home | About | Journals | Submit | Contact Us | Français|
Schizosaccharomyces pombe cells divide by medial fission through the use of an actomyosin-based contractile ring. A mulitlayered division septum is assembled in concert with ring constriction. Finally, cleavage of the inner layer of the division septum results in the liberation of daughter cells. Although numerous studies have focused on actomyosin ring and division septum assembly, little information is available on the mechanism of cell separation. Here we describe a mutant, sec8-1, that is defective in cell separation but not in other aspects of cytokinesis. sec8-1 mutants accumulate ~100-nm vesicles and have reduced secretion of acid phosphatase, suggesting that they are defective in exocytosis. Sec8p is a component of the exocyst complex. Using biochemical methods, we show that Sec8p physically interacts with other members of the exocyst complex, including Sec6p, Sec10p, and Exo70p. These exocyst proteins localize to regions of active exocytosis—at the growing ends of interphase cells and in the medial region of cells undergoing cytokinesis—in an F-actin–dependent and exocytosis-independent manner. Analysis of a number of mutations in various exocyst components has established that these components are essential for cell viability. Interestingly, all exocyst mutants analyzed appear to be able to elongate and to assemble division septa but are defective for cell separation. We therefore propose that the fission yeast exocyst is involved in targeting of enzymes responsible for septum cleavage. We further propose that cell elongation and division septum assembly can continue with minimal levels of exocyst function.
Cytokinesis is the stage in the cell division cycle during which the boundaries between the two daughter cells are assembled and individual daughter cells are liberated. Given the complex nature of this event, spatial and temporal regulations are key issues underlying this process. Cytokinesis in a variety of eukaryotes is achieved through the use of an actomyosin-based contractile ring. The constriction of the actomyosin ring generates the force necessary for cell cleavage. Newly synthesized membrane is inserted at the site of division concomitant with the constriction of the actomyosin ring. Although this process has been studied at a descriptive level for decades, it is only recently that we are beginning to gain a molecular framework for understanding the mechanism and regulation of cytokinesis.
The fission yeast Schizosaccharomyces pombe is an attractive model organism for the study of cytokinesis. Like animal cells, S. pombe cells divide through the use of an actomyosin ring. This ring is assembled at the onset of mitosis. At the end of mitosis, the actomyosin ring constricts concomitant with the formation of the division septum. Genetic studies in S. pombe have identified many genes important for various steps in cytokinesis (Simanis, 1995 ). The genes mid1, plo1, and pom1 are required to position the actomyosin ring, and the division septum. Mid1p and Plo1p act possibly in a signaling pathway that integrates nuclear positioning with the position of the actomyosin ring (Bähler et al., 1998a ). The genes cdc3, cdc4, cdc8, cdc12, rng2, rng3, rng4, rng5/myo2, rlc1, and act1 are required for the assembly of the actomyosin ring. The identity of their gene products as actin cytoskeleton elements is consistent with the idea that they interact to effect actomyosin ring assembly (Balasubramanian et al., 1992 , 1994 ; Chang et al., 1997 ; Eng et al., 1998 ; McCollum et al., 1999 ; Naqvi et al., 1999 , 2000 ; Wong et al., 2000 ). After actomyosin ring assembly, the function of the ring component Cdc15p, a SH3 domain-containing protein, is required for the assembly of the F-actin patches adjacent to the actomyosin ring (Fankhauser et al., 1995 ; Balasubramanian et al., 1998 ). The genes cdc7, cdc11, cdc14, sid1, sid2, spg1/sid3, and sid4, which encode signaling molecules (collectively referred to as the Septation Initiation Network [SIN]), regulate division septum assembly during actomyosin ring constriction (Fankhauser and Simanis, 1994 ; Gould and Simanis, 1997 ; Balasubramanian et al., 1998 ; Sparks et al., 1999 ; Guertin et al., 2000 ). Genetic studies indicate that the activation of the SIN pathway might regulate Cps1p, a 1,3-β-glucan synthase essential for the assembly of the division septum (Le Goff et al., 1999 ; Liu et al., 1999 ). After assembly of the primary septum (composed primarily of unbranched 1,3-β-glucan and 1,3-α-glucan) and the secondary septa (composed of branched 1,3-β-glucan, α-galactomannan, and 1,3-α-glucan), the primary septum is cleaved to liberate two daughter cells (Humbel et al., 2001 ). Although the mechanisms of actomyosin ring assembly, constriction, and division septum assembly have received considerable attention, very little is known about how cleavage of the primary septum is achieved to effect the liberation of the two daughter cells.
In this study, we describe the characterization of sec8-1, a mutant defective in cell separation. Sec8p is a component of the exocyst complex that plays a key role in delivery of secretory vesicles in a number of organisms (Ting et al., 1995 ; TerBush et al., 1996 ; Grindstaff et al., 1998 ). The exocyst localizes to regions of active secretion in fission yeast. We come to the interesting conclusion, based on analysis of a series of mutations in members of the S. pombe exocyst complex, that the exocyst complex is rate-limiting for cell separation but that only low levels of exocyst function are required for cell elongation and division septum assembly.
S. pombe strains used in this study are listed in Table Table1.1. Yeast cells were grown on YES medium or minimal media with appropriate supplements (Moreno et al., 1991 ). Crosses were performed by mixing appropriate strains directly on YPD plates (Moreno et al., 1991 ), except that sec8-1 was transformed with plasmid pREP3–1-sec8 before crosses. Recombinant strains were obtained by tetrad analysis. Yeast transformations were performed by the lithium acetate method (Okazaki et al., 1990 ). Kanamycin was used at 100 μg/ml. To eliminate F-actin, yeast cells were treated with latrunculin A (L-12370; Molecular Probes Inc., Eugene, OR) at a concentration of 100 μM for 3.5 h. To block ER-to-Golgi transport, cells were treated with 100 μg/ml Brefeldin A (B-7450; Molecular Probes) for 3 h. Cells treated with DMSO and ethanol, respectively, were used as controls. Thiamine was used at a final concentration of 5 μM to repress transcription from the nmt1 promoter (Basi et al., 1993 ).
Fluorescence microscopy was performed essentially as described (Balasubramanian et al., 1997 ). Cells were viewed using a Leica DMLB microscope with appropriate filters. To visualize DNA, F-actin, and septum material, cells were fixed in 3.7% formaldehyde for 1 min and stained with 4′,6-diamidino-2-phenylindole (DAPI), rhodamine-conjugated phalloidin, and Calcofluor (Sigma, St. Louis, MO), respectively, as described (Balasubramanian et al., 1997 ). For all indirect immunofluorescence, cells were fixed in formaldehyde and probed with anti-GFP antibodies (1:500 dilution; Molecular Probes), anti-Myc antibodies (1:200 dilution; Sigma), anti-Myo2p antibodies (1:400 dilution; Naqvi et al., 1999 ), anti-Mok1 antibodies (1:200 dilution; Katayama et al., 1999 ), or antitubulin antibodies (1:200 dilution; a kind gift from Dr. Keith Gull). Secondary antibodies of anti-rabbit and anti-mouse IgG conjugates (Molecular Probes) were used at 1:200.
Electron microscopy was performed on permanganate-fixed S. pombe as described (Armstrong et al., 1993 ). Briefly, S. pombe cells were grown at appropriate temperatures, washed three times in sterile water, and fixed for 1 h in 2% potassium permanganate at room temperature. Fixed cells were harvested by centrifugation and washed three times in sterile water, resuspended in 70% ethanol, and incubated overnight at 4°C. The samples were dehydrated and treated with propylene oxide before infiltration with Spurr's medium, followed by another change of medium and incubation at 65°C for 1 h. Finally, they were embedded in Spurr's resin, and the resin was allowed to polymerize at 60°C overnight. Ultrathin sections were cut on a Jung Reichert microtome (Leica Mikroskopie and Systeme GmbH, Wetzler, Germany) and examined using a JEM1010 transmission electron microscope (Jeol, Tokyo, Japan) at 100 kV.
sec8+ was cloned by complementation of the temperature-sensitive mutant mut2-1 (sec8-1). An S. pombe genomic library (Balasundaram et al., 1999 ) was introduced into mut2-1 mutant cells, and transformants were selected at 36°C. One plasmid was found to be able to reverse the temperature sensitivity. Nucleotide sequence determination and BLAST searches suggested that the rescuing DNA was located on cosmid SPCC970 (SPCC970.09, Accession no. O74562), and the only gene on this plasmid encoded a protein homologous to Sec8p in Saccharomyces cerevisiae. Three experiments were done to show that mut2-1 is defective in sec8. 1) A genetic cross between mut2-1 and sec8-GFP (marked with ura4+) showed that mut2-1 is tightly linked to the sec8 locus (no recombinants in 20 tetrads). 2) mut2-1 is fully rescued by recombination and gene conversion upon introduction of PCR fragments of sec8 that lack promoter and 5′ coding sequences. 3) Sequence determination of the sec8 locus in a mut2-1 strain revealed that it carried a mutation in the sec8 gene that changed codon 992 from CAG to TAG, thereby introducing a premature stop codon.
A search through the Sanger Center Fission Yeast Genome Sequencing Project Database for proteins homologous to S. cerevisiae Sec6p, Sec10p, Sec15p, and Exo70p identified S. pombe Sec6p, Sec10p, Sec15p, and Exo70p. sec6+, sec10+, sec15+, and exo70+ were found to reside on cosmids SPCC1235.10c (Accession no. O74846), SPAC13F5.06c (Accession no. O13705), SPCC1183.01 (Accession no. O75006), and SPBC582.02 (Accession no. Q10339), respectively.
The entire coding sequences of sec6+, sec8+, and sec10+ were deleted to create the null mutants by replacing the respective coding regions with the ura4+ gene. The following primer pairs were used to amplify the constructs containing ura4+ and the flanking 5′ and 3′ sequences of the respective gene by PCR. MOH595 (C C A G T C C G T A A A T A T A T T A A T C A A T C T G T C A G T A A A T A G A A A C G T T T G T A A G C A C T A G G T C T G C T T A T A A C T T T A A G A A A G C T A C A A A T C C C A C T G G C T A T A T G T A), containing 80 base pairs of sec6+ upstream sequence and 20 base pairs of 5′ sequence of ura4+, and MOH596 (G T A G A T C A T T A A A A T T C A G C A A C G A C T A C T T T G G A T C G A T A T T G A C G A A A C T T T T T G A C A T C A T A A T C A A A A G G A A C A T T A C T A T A G G T A A A G A T A A A C C G T A C), containing 80 base pairs of downstream sequence of sec6+ and 20 base pairs of 3′ sequence of ura4+, were used to amplify the construct for knocking out sec6. Similarly, TX-1 (T A G T G A T T T C T T A G C T C T C C T T T C A A A G A T A A T A C A G T C A A T A G A C A T A T C A A G C T A A A A C T A C T G A C T A T T T G A C T T C C G C T A C A A A T C C C A C T G G C T A T A T G T A) and TX-2 (T G C A T C T A T G T T T T T A G T T A A T A A A T T T A T T A T T A T A A A A T C A T T A C T C G T C A T T A T A A T T A A A A T T C T A T A T T A T A C G G A G A A A G C T A C A A A T C C C A C T G G C T A T A T G T A) were used for construction of the sec8-null, and MOH597 (C A C C T A C A A A C C A A A G G A A A C T T T G A T C A T T A C T T T T C T A T T C G A G A A T T G T A G A T T T A A A A T T T C T T G T C T A T T A A G A C G C T A C A A A T C C C A C T G G C T A T A T G T A) and MOH598 (T A T A A T A C A C T A T A A A A G A T A T T A T G T T T A T C T A T A G A C A A A T T A C T T C A T A A T T A A G A C A T T A A C A A A A A T G A G C G A T T G A T A T T G A C G A A A C T T T T T G A C A T C) were used for construction of the sec10-null. The purified fragments were introduced into a wild-type diploid of genotype leu1-32/leu1-32, ura4-D18/ura4-D18, ade6-210/ade6-216, h+/h−. Transformants that had undergone homologous recombination were selected by growth in supplemented minimal medium lacking uracil. Correct integration of the deletion construct was confirmed by PCR assay and nucleotide sequence determination of the genomic DNA from transformants.
Primers gma12F (CCTCCTGGTACCTAGAACACACGAGTACTTG-GACC) and gma12R (CCTCTCCCGGGGGATGATGGTTTCAAAAGATTTTG) were used to amplify the gma12 ORF (SPCC736.04c, Accession no. Q09174) by PCR, using wild-type gemonic DNA as template. Primers MOH768 (CGGCTGGTACCAAGGGGTTTTCCGTTGAC) and MOH769 (GGCGGCCCGGGTGAGCGTTCGCCACGGAG) were used to amplify the hht2 ORF (SPAC1834.04, Accession no. P09988), using wild-type genomic DNA as template. These fragments were cloned into pJK210-GFP (Naqvi, N. and Balasubramanian, M. K., unpublished results) as KpnI-SmaI fragments to generate pJK210-gma12-GFP and pJK210-hht2-GFP. These plasmids were linearized and transformed into a wild-type strain, and colonies were selected for growth on medium lacking uracil. Correct integrations were confirmed by PCR assay.
Chromosomal copies of sec6+, sec8+, sec10+, and exo70+ were tagged by the carboxy-terminal addition of GFP and/or the Myc epitope. To tag Sec6p with GFP, a 0.8-kb KpnI/SmaI fragment of the sec6+ C-terminal sequence was obtained by PCR using the primers MOH584 (GATGGTACCGAACTTTCACAGCAATTATCTG) and MOH585 (CGATCCC-GGGTAAAATTGAACTTCCAGAAAGAG) and cloned into pJK210-GFP. The resulting plasmid pJK210-sec6CT-GFP, containing sec6 fused in frame with GFP sequences, was linearized with NdeI and transformed into a wild-type strain of genotype leu1-32 ura4-D18 ade6-210. To tag Sec8p with GFP, primers MOH714 (CACCGGTACCAAGCTAATTTCGGTGGTGACTTT) and MOH715 (CTACCCCGGGATTTTTTCTCGCACCACCCACAG) were used to generate a 700-bp KpnI-SmaI fragment that was cloned into pJK210-GFP to yield pJK210-sec8CT-GFP. This plasmid was linearized with EcoRI and transformed into wild-type cells. Similarly, primers MOH586 (GATGGTACCTAGTGGA CATTAGGGAATG) and MOH587 (CGATCCCGGGACTGCTCTTTGGGGGCAATAAAG CTTC) were used to generate a 0.9-kb KpnI-SmaI fragment of sec10 that was cloned into pJK210-GFP to generate pJK210-sec10CT-GFP. This plasmid was linearized with SpeI and introduced into wild-type cells. In each case, transformants were selected on supplemented minimal medium lacking uracil, and putative integrants were subjected to PCR and Western blot analyses to confirm the desired integration.
A similar strategy was used for Myc tagging. A 1.2-kb BamHI/BglII digested fragment containing the 13myc sequence and terminator from pFA6a-13myc (Bähler et al., 1998b ) was cloned into the BamHI site of pJK210 (Keeney and Boeke, 1994 ) to generate plasmid pJK210-13myc. A 0.7-kb NotI/BamHI-digested fragment containing carboxyl-terminal sequence of sec6+ was obtained using primers MOH638 (GCTAGCGGCCGCCCGAACTTTCACAGCAATTATCTG) and MOH639 (GCTAGGATCCGTAAAATTGAACTTCCAGAAAGAG) and was cloned into the NotI/BamHI sites of pJK210-13myc. Similarly, a 0.9-kb NotI/BamHI fragment of sec10+ carboxy-terminal sequence was obtained using primers MOH623 (GCGAGCGGCCGCCCTAGTGGACATTAGGGAATGT-AAG) and MOH624 (GCTAGGATCCGACTGCTCTTTGGGGGCAAT-AAAGC) and cloned into the NotI/BamHI sites of pJK210-13myc. These resulting plasmids were linearized with NdeI and SpeI, respectively, and transformed into wild-type cells.
The tagging of exo70 with 13myc was done according to the methods described by Bähler et al. (1998b) . A PCR fragment was generated using plasmid pFA6a-kanMX6 as template and primers exo70–5′ (C T C A T T A C G T A G T A T A T C A A A T T T A C A A A G G C T G A T T T A G A T T C T T T T A T T A C A A G C G C G T T T G C T C C T T C C C T A C G G A T C C C C G G G T T A A T T A A; contains 75 base pairs of exo70 C-terminal sequence and 20 base pairs of 13myc sequence) and exo70–3′ (T T C A A A G A A A A G T G A G A A T G C C A G T A C A C C C A C T T T A G T A C T A T A T T A T G G A A T T T C A A A G G A C C C A A A T T C A T C G A A T T C G A G C T C G T T T A A A C; contains 75 base pairs of exo70 3′UTR sequence and 20 base pairs of vector sequence). The PCR fragment was purified and transformed into a wild-type strain. The desired integrations were confirmed by PCR assay and Western blot analysis.
To assess whether the maternal Sec8p was present in sec8-null cells, a diploid strain was constructed in which one sec8 locus was replaced with ura4+, and the other sec8 locus was tagged with the Myc epitope marked with leu1+ and the kanamycin-resistance gene. A 2.4-kb fragment containing Myc and kanR sequences was obtained from pFA6a-13myc-kanMX6 (Bähler et al., 1998b ) by digestion with SmaI and SacI and then cloned into pJK148 (Keeney and Boeke, 1994 ) digested with SmaI and SacI. The resulting plasmid (pJK148-Myc-Kan) was then digested with KpnI and SmaI and ligated to a 0.7-kb KpnI-SmaI fragment containing the carboxy-terminal sequence of sec8 as described above. This plasmid (pJK148-Myc-Kan-sec8CT) was linearized with XbaI, transformed into sec8::ura4+/sec8+ diploid cells (see above), and selected for kanamycin-resistance. Correct integration was confirmed by PCR assay and immunoblotting.
To construct a sec8 shut-off strain, we made a tagging cassette containing 5′ upstream regulatory sequence of sec8, ura4+, the 81nmt1 promoter, and the coding sequences of sec8 sequentially to replace the sec8 gene with the 81nmt1 promoter-controlled sec8. The 81nmt1 promoter region from pREP81 was cloned into pSK-ura4 (Naqvi, N., and Balasubramanian, M. K., unpublished results) to generate pSK-ura4-81nmt1 (Rajagopalan, S., and Balasubramanian, M. K., unpublished results). A 0.5-kb fragment containing 5′ sequence of sec8 was obtained by PCR using primers MOH724 (CATGGTACCGTATGATCGAGGATACGTACGAGG) and MOH790 (CCATCGATAAGGGTGTGTAACTAAGC), and the N-terminal sequence of sec8 was amplified by PCR using primer MOH746 (CGGGATCCCATATGGATACCAGAGGCTATTCGGAAACG) and MOH727 (CGCTCGAGCTCAGCGGATTTGGAGAGCAGTAC), respectively. The first fragment was digested with KpnI and ClaI, and the latter fragment was digested with NdeI and SacI. These two fragments were cloned into pSK-ura4-81nmt1 sequentially to yield a plasmid that was then linearized with SacI and XhoI. The linearized DNA was used to transform wild-type cells of genotype leu1-32 ura4-D18 ade6-210. Transformants were selected on medium lacking uracil, and correct integration was confirmed by PCR assay. The sec8 shut-off strain was maintained in minimal medium lacking thiamine. To shut off Sec8p expression, cells were grown in minimal medium containing thiamine at 30°C for 14 h.
MBY888 (sec8-1) was generated by crossing MBY887 (sec8-1, ura4-D18, leu1-32, h+) to wild-type strain 972 (Leupold, 1970 ). MBY888 and wild-type cells were grown in minimal medium overnight at 24°C to early log phase (optical density595 < 0.4). Cells were washed three times with minimal medium lacking nitrogen, resuspended in the same medium, and grown for 18 h at 24°C to arrest in the G1 phase. Cells were shifted to 36°C for 1 h to inactivate Sec8-1p and then transferred into YES (rich medium) to release cells from G1 and allow mitotic cell cycle progression at 36°C. Cell samples were taken just before the release (0) and every hour after the release (1–8).
Immunoprecipitation and immunoblotting were performed essentially as described (Naqvi et al., 1999 ). Cells were grown to exponential phase in YES medium at 24°C and harvested. Cell lysis was achieved by the addition of 500 μl of acid-washed glass beads to the cell pellet and subsequent disruption using a mini-bead beater (three cycles of 30-s duration and 2-min cooling intervals). For immunoprecipitation, 500 μl NP-40 buffer (1% Triton X-100, 150 mM NaCl, 2 mM EDTA, 6 mM Na2HPO4, 4 mM NaH2PO4, 1 mM PMSF, 2 mM Benzamidine) was used to extract soluble proteins. Cell extracts were clarified by centrifugation at 14,000 rpm for 10 min at 4°C. For each immunoprecipitation, 500 μl of soluble protein was incubated with 2 μl of anti-GFP antibodies for 1 h at 4°C. Sepharose-Protein G beads were added to the antigen-antibody immunocomplex and incubated for 45 min at 4°C. Beads were washed six times with 1 ml NP-40 buffer, resuspended in gel loading buffer, and heated at 95°C for 3 min.
For detection of Myc- or GFP-tagged proteins from total protein extracts or immunoprecipitates, proteins were separated on 6% SDS-PAGE (Mini-protein II system; Bio-Rad Laboratories, Hercules CA) at 120 V for 1 h and transferred (Trans-Blot system; Bio-Rad Laboratories) at 85 V for 2 h to a nitrocellulose membrane (Pierce Chemical Co., Rockford, IL). The membranes were blocked with 10% nonfat milk in TBS-Tween 20 (50 mM Tris-HCl, 150 mM NaCl, 0.05% Tween 20, pH 7.6) for 1 h at room temperature. Primary anti-GFP and anti-Myc antibodies were used at 1:700 and 1:1000 dilutions, respectively. Peroxidase-conjugated anti-rabbit and anti-mouse IgG (Sigma) were used at 1:4000 dilutions, and the enhanced chemiluminescent signal was detected using a 1:1 mixture of ECL1 (2.5 mM 3-aminophytaldrazide dissolved in DMSO, 0.4 mM p-coumaric acid, 100 mM Tris-HCl, pH 8.5) and ECL2 (0.02% H2O2, 100 mM Tris-HCl, pH 8.5; Schneppenheim et al., 1991 ).
Acid phosphatase secretion was assayed as follows (modified from Craighead et al., 1993 ; Tanaka and Okayama, 2000 ). Because up to 40% of acid phosphatase is secreted into the medium in fission yeast, enzyme activity was assayed in the culture supernatant. Cells were grown to log phase in minimal medium (MM) at 24°C, pelleted, washed twice with MM, and resuspended in fresh MM at 24 or 36°C. Samples were taken at 0 h (time of resuspension) and at hourly intervals thereafter. For each sample, 1 ml of culture was centrifuged, and 500 μl of the supernatant was added to 500 μl of substrate solution (2 mM p-nitrophenyl phosphate, 0.1 M sodium acetate, pH 4.0; prewarmed to 30°C) and incubated at 30°C for 5 min. Reactions were stopped by the addition of 500 μl of 1 M sodium hydroxide. The absorbance at 405 nm was measured, using the 0-h sample as a blank control.
We performed a screen to identify mutants defective in cytokinesis in S. pombe. To visualize nuclei easily, a strain was constructed in which the coding region of the histone H3 gene (hht2) was fused to green fluorescent protein sequences (hht2-GFP). As expected, Hht2-GFP localized to the nucleus throughout the cell cycle (our unpublished results). This starting strain (MBY816) was mutagenized by UV irradiation, and the resulting ts− mutants were subjected to microscopic analysis to detect mutants that accumulated multiple nuclei (Tang, X., and Balasubramanian, M.K., unpublished results). The characterization of one such mutant, mut2-1, is described in this study. mut2-1 cells grew and formed colonies at 24°C (permissive temperature) but were unable to do so at 36°C (restrictive temperature). Although wild-type cells continued to grow and divide upon temperature shift from 24 to 36°C, the cell number of a mut2-1 strain did not increase after an identical temperature shift (Figure (Figure1A),1A), whereas the number of attached cell bodies increased, indicating failed cell separation. To better characterize the phenotype of mut2-1, we monitored changes in the subcellular distribution of F-actin and cell wall material after a shift from 24 to 36°C (Figure (Figure1B).1B). Under permissive conditions, F-actin rings and septa in the majority of mut2-1 cells resembled those found in wild-type cells (Figure (Figure1B,1B, 0 h). After 4 h at 36°C, >50% of mut2-1 cells contained four nuclei, indicative of the successful completion of two rounds of mitosis despite the aberrant cytokinesis (Figure (Figure1B,1B, 4 h). Interestingly, under these conditions, assembly and constriction of the actomyosin ring were not impaired in mut2-1 cells (Figure (Figure1B,1B, arrow). In addition, mut2-1 cells were also capable of assembling medial division septa (Figure (Figure1B).1B). However, the septa apparently could not be disassembled in mut2-1 cells, leading to the accumulation of elongated cells with one or three septa. Thus, mut2-1 identifies a protein important for cell separation after assembly of the division septum.
To identify the gene responsible for the mut2-1 phenotype, a plasmid rescuing the temperature-sensitive lethality of mut2-1 was identified (see MATERIALS AND METHODS). The rescuing DNA encoded a 1088-amino acid polypeptide. Database searches using the predicted protein sequence showed that it was related to S. cerevisiae Sec8p (16% amino acid identity), a component of the exocyst, as well as to Sec8p-like proteins from humans (13% identity, Figure Figure2).2). Several lines of evidence established that mut2-1 is an allele of sec8+ (see MATERIALS AND METHODS).
The exocyst is required for polarized cell growth and cell surface expansion in S. cerevisiae (TerBush et al., 1996 ; Roth et al., 1998 ). In contrast, the sec8-1 mutant described in this study appeared to be defective only in septum disassembly and cell separation. To test the role of Sec8p in cell elongation, wild-type and sec8-1 cells were synchronized by nitrogen starvation and monitored for the ability to undergo polarized cell growth and septum assembly. This protocol provides a convenient means to assess the function of a protein in cell elongation, because wild-type cells start at the length of 4 μm and elongate to 12–14 μm before division (Figure (Figure3A).3A). After release into rich medium, sec8-1 cells were able to elongate, enter mitosis, and assemble division septa with kinetics similar to that of wild-type cells (Figure (Figure3B).3B). However, unlike wild-type cells, sec8-1 cells failed to disassemble the division septa, leading to the accumulation of binucleate cells with a medial division septum. Even although septum cleavage and cell separation failed, sec8-1 cells reinitiated polarized growth and underwent a second round of mitosis and division septum assembly 7 h after release into the rich medium. Septum cleavage again failed in these cells resulting in the accumulation of tetranucleate cells with three septa. On prolonged incubation (12 h) sec8-1 cells lysed. Cell length and the percentage of septated cells at each time point were also quantified (Table (Table2),2), which indicated that sec8-1 is not defective in cell elongation and is not delayed for septum assembly. These data suggest that sec8-1 cells are specifically defective in septum cleavage and cell separation.
The exocyst in S. cerevisiae is a mulitprotein complex comprised of Sec3p, Sec5p, Sec6p, Sec8p, Sec10p, Sec15p, Exo70p, and Exo84p (TerBush et al., 1996 ). We have identified Sec8p in S. pombe as a homologue of one component of the exocyst. We therefore searched the S. pombe databases to determine whether other exocyst components were also present in fission yeast. Interestingly, homologues of S. cerevisiae Sec6p, Sec10p, Sec15p, and Exo70p were also present in S. pombe (see MATERIALS AND METHODS). We were unable to identify proteins related to Sec3p, Sec5p, or Exo84p. The alignments of the S. pombe exocyst proteins (named like their S. cerevisiae homologues) with their counterparts in other organisms are shown in Figure Figure4.4. These four exocyst proteins in S. pombe share ~20% identities in sequences and align through their entire lengths with their homologues. Thus, several components of the exocyst complex are conserved in S. pombe.
Immunoprecipitation experiments were performed in order to determine whether S. pombe Sec6p, Sec8p, Sec10p, and Exo70p form a complex in vivo, as has been demonstrated with their counterparts in other organisms. A number of strains expressing either c-Myc– or GFP-tagged versions of Sec6p, Sec8p, Sec10p, and Exo70p were constructed. To test the interaction between Sec8p and Sec6p, protein extracts from strains expressing sec8-GFP alone, sec8-GFP and sec6-Myc, or sec6-Myc alone were immunoprecipitated using anti-GFP antibodies and analyzed using a Myc mAb. Sec6p-Myc was only detected in the immunoprecipitates from the sec8-GFP sec6-Myc strain (Figure (Figure5A),5A), suggesting that these two proteins associate in vivo. To test the interaction of Sec8p with Sec10p, similar immunoprecipitations were performed using extracts of strains expressing sec8-GFP alone, sec8-GFP and sec10-Myc, or sec10-Myc alone. Sec10-Myc was detected only in the immunoprecipitates from sec8-GFP sec10-Myc cells (Figure (Figure5B).5B). Thus, Sec8p also interacts with Sec10p. Finally, the interaction of Sec8p with Exo70p was demonstrated in similar experiments (Figure (Figure5C).5C). In addition, we observed interactions in the other pairwise combinations (Sec6p-Sec10p, Sec6p-Exo70p, and Sec10p-Exo70p; our unpublished results). Thus, the exocyst components Sec6p, Sec8p, Sec10p, and Exo70p physically interact with each other in S. pombe.
The subcellular localization of Sec8p was determined by tagging the chromosomal copy of sec8+ with GFP sequences. In this strain, the expression of Sec8p-GFP was under the control of the sec8+ promoter. The sec8-GFP cells resembled wild-type cells in morphology and growth rates, establishing that the addition of GFP did not compromise the function of Sec8p. However, the Sec8p-GFP signal was prone to rapid photobleaching. Therefore, indirect immunofluorescence was performed to visualize Sec8p-GFP. In interphase cells, identified as uninucleate cells with uncondensed chromosomes, tip localization was observed in 55% (Figure (Figure6A,6A, marked with arrowheads). In early mitotic cells, tip localization was absent and Sec8-GFP was seen as a ring in the medial region of the cell that resembled the actomyosin ring (Figure (Figure6A,6A, cells marked with 1 and 4). However, in late mitotic cells, unlike the actomyosin ring, which undergoes constriction, medial staining of Sec8p-GFP was detected as double rings (Figure (Figure6A,6A, cells marked with 2 and 3). To examine whether these structures were real ring structures, confocal microscopy and 3D-projection software were used to determine the localization of Sec8-GFP. When Sec8-GFP double ring images were rotated, they appeared clearly as rings (Figure (Figure6B;6B; arrow marks the entire ring visualized upon rotation by 139°). Essentially identical localization patterns were observed for Sec6p-GFP, Sec10p-GFP, Sec6p-Myc, Sec10p-Myc, and Exo70p-Myc (Figure (Figure6,6, A, C, and D). Thus, consistent with their coimmunoprecipitation, components of the exocyst also colocalized in S. pombe cells, supporting the hypothesis that the exocyst components interact in vivo.
To investigate the localization of exocyst components in relation to the actomyosin ring, we examined the localization of Sec10p-GFP and Myo2p (an actomyosin ring component) in the same cells. Both proteins assembled into ring structures at early mitosis and approximately colocalized (Figure (Figure6E,6E, top panel). However, in cells undergoing actomyosin ring constriction (Figure (Figure6E,6E, bottom panel), constriction of the Sec10-GFP rings was not observed. Instead, the Sec10-GFP rings split into a pair of rings on either side of the constricting actomyosin ring.
Given that the S. pombe exocyst components assembled as a medial ring that colocalized approximately with the actomyosin ring at early mitosis, we addressed the roles of the F-actin cytoskeleton and of proteins important for actomyosin ring formation in the assembly of the exocyst complex at the division site. First, we monitored the localization of Sec10-GFP after treatment of G2-synchronized cells with latrunculin A (LatA), a drug that prevents actin polymerization. Although DMSO alone did not affect assembly of medial Sec10-GFP rings, cells treated with LatA in DMSO were unable to assemble medial Sec10p-GFP rings (Figure (Figure6F).6F). Thus, the proper assembly of Sec10p and, by inference, the other exocyst components as a medial ring at the division site is F-actin dependent. We then examined Sec10p-GFP localization in cdc8-110 (Balasubramanian et al., 1992 ), cdc12-112 (Chang et al., 1997 ), and cdc15-140 (Fankhauser et al., 1995 ) mutants. Although Sec10p-GFP was observed as a medial ring at 24°C in all these mutants, at 36°C none of the mutants was able to assemble Sec10p-GFP into ring structures (Figure (Figure6,6, G and H, and our unpublished data). Thus, the assembly of the exocyst to the medial region appears to depend on the proteins essential for actomyosin ring assembly and actin patch mobilization.
Because the exocyst in S. pombe is potentially involved in secretion, we wanted to ascertain whether the localization of the exocyst as a medial ring is dependent on the secretory pathway. Synchronous cells expressing Sec10-GFP were treated with brefeldin A (BFA), a drug blocking membrane trafficking of newly synthesized proteins from endoplasmic reticulum (ER) to Golgi (Turi et al., 1994 ). Gma12p, a Golgi marker protein that has been reported to relocate from Golgi to ER upon BFA treatment (Brazer et al., 2000 ), was used to test the efficacy of BFA treatment. The ER in S. pombe is distributed primarily around the nuclear membrane region, whereas Golgi is seen as patches throughout the cytoplasm (Brazer et al., 2000 ). As expected, Gma12p-GFP relocated from Golgi to ER upon treatment with BFA (Figure (Figure6I).6I). In contrast, the localization of Sec10p was not affected by BFA (Figure (Figure6I),6I), indicating that the exocyst localization is independent of exocytosis. Thus, the exocyst complex in S. pombe could serve as a landmark for the targeting of the exocytic machinery.
Although exocyst mutants in S. cerevisiae are defective in polarized growth (Hsu et al., 1999 ), sec8-1 mutants in S. pombe appear to be unaffected with respect to polarized growth. Given this dramatic difference in phenotype, it seemed possible that sec8-1 was not defective in a polarized growth function of Sec8p. In this case, a sec8 null mutant would be expected to show a stronger phenotypic defect with respect to cell growth. To test this, we replaced sec8+ with ura4+ in a diploid strain. By analysis of meiotic products from the heterozygous strain, we found that spores bearing the sec8-null mutation were incapable of forming colonies. Thus, Sec8p is essential for cell viability. To characterize the terminal phenotype, the mutant spores were germinated and stained to visualize F-actin, DNA, and septa (Figure (Figure7A).7A). The mutant spores were capable of germination, cell elongation, mitosis, actomyosin ring assembly, and septum assembly. However, the septa assembled in the germinating mutant cells were not cleaved. Cell growth, cell elongation, and mitosis continued in the unseparated mutant cells, leading eventually to the accumulation of tetranucleate cells with septa placed between each pair of nuclei. Similar results were obtained with sec6 and sec10 null mutants (Figure (Figure7,7, D and E).
To ensure that the phenotype was not due to inherited maternal exocyst proteins, we tested whether the maternal Sec8p was present in sec8-null cells using a diploid strain in which one sec8 locus was replaced with ura4+ and the other was tagged with Myc and leu1+. We examined whether the maternal Sec8-Myc protein was present in the germinated null mutant cells. Spores were germinated in medium selective for ura4+ or leu1+ and stained with antibodies against Myc and Mok1p to visualize Sec8-Myc and the α-glucan synthase Mok1p (Katayama et al., 1999 ). Although Sec8-Myc localization was clearly observed in sec8-Myc cells (our unpublished results), it was not observed in the sec8 null cells (Figure (Figure7B),7B), suggesting that there was no significant carry-over of maternal Sec8p in these cells. Mok1p, used as a control, was observed in both cases as expected.
To analyze the sec8 loss-of-function phenotype using a different approach, we constructed a sec8 shut-off strain in which sec8 transcription was under the control of the low-strength and thiamine-repressible 81nmt1 promoter. On growth under repressing conditions, sec8 shut-off cells again appeared defective only in the disassembly of division septa but not in polarized cell growth (Figure (Figure7C).7C). We conclude that the exocyst is essential for septum disassembly and cell separation, whereas cell elongation and division septum assembly might require reduced levels of exocyst function or might be independent of it.
The exocyst in S. cerevisiae and mammals is involved in membrane trafficking from the Golgi apparatus to the plasma membrane (Hsu et al., 1999 ). To test whether the exocyst in S. pombe has a role in exocytosis, we used electron microscopy to ask if the targeting and fusion of secretory vesicles with the plasma membrane could occur normally in sec8 mutant cells. Presumed secretory vesicles (100 nm in diameter) were observed only rarely in wild-type cells (Figure (Figure8A).8A). In contrast, 60–100 such vesicles were detected in every section in the sec8-1 mutant (Figure (Figure8B;8B; average three vesicles/μm2). These vesicles were stained intensely after permanganate fixation and most likely represent post-Golgi secretory vesicles (Armstrong et al., 1993 ). In mutant cells undergoing septum assembly, most of the vesicles were clustered approximately in the vicinity of the septa. These observations suggested that targeting of secretory vesicles to the correct location occurs in sec8-1 cells but that the subsequent docking and/or fusion with the plasma membrane failed. During interphase, sec8-1 cells were also found to accumulate ~100 nm vesicles, indicating that Sec8p might also participate in exocytic events during interphase. Similarly, sec8 shut-off cells accumulated a large number of ~100 nm vesicles under repressing conditions (Figure (Figure8D),8D), whereas cells under nonrepressing conditions resembled wild-type cells (Figure (Figure8C).8C).
To test whether sec8 mutants are defective in exocytosis using a different approach, we monitored the transport of the enzyme acid phosphatase through the S. pombe secretory pathway in sec8-1 cells (Figure (Figure8E).8E). The activity of secreted acid phosphatase was assayed using the culture supernatant (see MATERIALS AND METHODS). Wild-type cells at 36°C secreted acid phosphatase about twice as fast as at 24°C. sec8-1 cells secreted much less acid phosphatase than wild-type cells at both temperatures. After 4 h, they secreted 67% of the level of activity of wild-type cells at 24°C and 42% of the activity at 36°C. Thus, sec8-1 is indeed defective in exocytosis.
Previous studies of cytokinesis in the fission yeast S. pombe have focused on actomyosin ring assembly, actin patch movement, signaling events that control septum delivery, and on the study of enzymes responsible for septum assembly (Simanis, 1995 ). However, little information was available on the regulation of cell separation. In this study, we have described the isolation of sec8-1, a mutant that is defective in cell separation after assembly of the division septum. Molecular cloning established that Sec8p is a component of the exocyst protein complex with homologues in several other organisms including the prototypic Sec8p from the budding yeast S. cerevisiae. The exocyst is a mulitprotein complex that has been identified in a number of organisms (Hsu et al., 1999 ). In budding yeast the exocyst consists of seven core subunits: Sec5p, Sec6p, Sec8p, Sec10p, Sec15p, Exo70p, and Exo84p (Potenza et al., 1992 ; TerBush and Novick, 1995 ; TerBush et al., 1996 ; Guo et al., 1999 ). In addition, the budding yeast exocyst complex interacts with its targeting factor Sec3p and the rab-related GTPase Sec4p (Finger et al., 1998 ; Guo et al., 1999 ). Although homologues of Sec5p, Sec6p, Sec8p, Sec10p, Sec15p, and Exo70 have been identified in other organisms including in mammalian cells, Sec3p and Exo84p-related proteins have been identified only in S. cerevisiae (Kee et al., 1997 ). The exocyst proteins appear to be important for transport between the Golgi apparatus and the plasma membrane and have been implicated in targeting and fusion of Golgi derived vesicles with the plasma membrane (Bowser and Novick, 1991 ; Potenza et al., 1992 ; Roth et al., 1998 ; Hsu et al., 1999 ).
Using sequences of the budding yeast exocyst proteins, we have identified proteins related to Sec6p, Sec10p, Sec15p, and Exo70 in S. pombe. These four proteins are ~20% identical in protein sequence with the budding yeast, plant, and rat counterparts. Using biochemical methods we have shown that Sec6p, Sec8p, Sec10p, and Exo70p interact physically. We therefore conclude that an exocyst-like complex is present in S. pombe. However, the S. pombe exocyst complex appears to lack proteins related to the budding yeast Sec5p and Exo84p. It will be interesting to test if proteins structurally related to the budding yeast Sec5p and Exo84p associate with the S. pombe exocyst complex. The fact that sec8-1 mutants accumulate ~100 nm vesicles at the restrictive temperature indicates that the exocyst complex in S. pombe, as in budding yeast and in mammalian cells, is important for exocytic events. The accumulation of ~100-nm vesicles in interphase as well as mitotic cells suggests that the exocyst might participate in exocytic events in all phases of the cell cycle. Independent experiments on secretion of acid phosphatase in wild-type and sec8-1 cells also conforms a role for the S. pombe exocyst in exocytosis. Whether the exocyst is required for all exocytosis events remains to be established.
We show that the fission yeast exocyst proteins localize to both cell tips as well as the site of cell division. In early mitosis, the exocyst colocalizes with the actomyosin ring and later splits into two rings upon constriction of the actomyosin ring. We have shown that the localization of the exocyst complex to the division site is dependent on an intact F-actin cytoskeleton and also on the molecules that are important for actomyosin ring assembly. Thus, the actomyosin ring might serve as a spatial landmark for targeting of the exocyst complex. It is also possible that the exocyst complex might be transported to the division site along the F-actin cables (Marks and Hyams, 1985 ; Balasubramanian et al., 1996 ; Pelham and Chang, 2001 ) that are attached to the actomyosin ring. The function of Cdc15p, an SH3 domain containing protein is also essential for assembly of the exocyst at the division site (Fankhauser et al., 1995 ). It is interesting to note that Cdc15p is also related to proteins of the PACSIN family, which are important for membrane transport events (Lippincott and Li, 2000 ). It is likely that Cdc15p might participate in membrane transport events pertaining to cytokinesis and might allow the targeting of proteins that specify exocytic events or allow the localization of proteins that themselves utilize the exocytic pathway during cytokinesis.
The localization of the exocyst appears to be independent of secretion because disruption of the Golgi apparatus by treatment with BFA does not impair the ability of the exocyst complex to localize to the actomyosin ring. The secretion-independent localization of the S. pombe exocyst is different from the situation in S. cerevisiae, where it has been shown that the localization of all members of the exocyst complex (with the exception of the targeting subunit, Sec3p) depends on the secretory pathway (Finger et al., 1998 ). We have been unable to find a Sec3p-like protein in S. pombe. Thus, in the absence of a Sec3p-like protein the other components might have evolved additional secretion-independent mechanisms to achieve their intracellular localizations in S. pombe. Thus, the exocyst complex might localize to the division site in a secretion-independent and F-actin–dependent manner to direct exocytic events.
Mutations in the S. cerevisiae exocyst members appear to block fusion of all post-Golgi vesicles with the plasma membrane. As a result, these mutants are unable to expand the cell surface and perish because of failure of all exocytic events (TerBush and Novick, 1995 ; TerBush et al., 1996 ; Finger and Novick, 1997 ; Roth et al., 1998 ). In contrast, S. pombe exocyst mutants are capable of polarized growth, cell surface expansion, and division septum assembly. S. pombe exocyst mutants appear to be specifically defective in cleavage of the division septum and cell separation. Given the differences in the phenotypes of exocyst mutants in the two yeasts, we have established the terminal phenotypes of the exocyst mutants using several approaches. We have investigated the terminal phenotypes of a sec8 temperature-sensitive mutant, a sec8 shut-off strain as well as the terminal phenotypes of germinated sec8-null mutant spores. We have also established that the phenotype of sec8-null mutant spores is not likely to be influenced by maternal carry-over of Sec8p from the heterozygous diploid. In addition, we have also shown that null mutations in sec6 and sec10 also result in a phenotype identical to that observed in the sec8 mutants. One possibility is that the exocyst is essential only for a subset of secretory events in S. pombe. This conclusion is similar to that obtained from studies in MDCK cells where it has been shown that the exocyst is only important for delivery of proteins to the basolateral membranes but not to the apical membranes (Grindstaff et al., 1998 ). Thus, the exocyst might be essential for delivery of proteins important for septum cleavage, but not for proteins involved in cell elongation and division septum assembly. It is possible that the lethality of the exocyst null mutants results from inappropriate cleavage of cell wall rather than the septum after prolonged incubation at the restrictive conditions. Alternatively, the exocyst might participate in all secretory events in wild-type S. pombe cells. In its absence, however, other pathways might substitute for the exocyst in some exocytic events. Previous studies have shown that additional mechanisms exist in budding yeast and mammalian cells for the delivery of proteins from the Golgi apparatus to the plasma membrane via early and recycling endosomes (Mallard et al., 1998 ; Brachet et al., 1999 ; Luo and Chang, 2000 ). Currently, it is unclear if transport from Golgi apparatus to the plasma membrane via endosomes requires exocyst function. In this model, the exocyst is rate limiting for the delivery of proteins important for septum cleavage and is redundant with other mechanisms important for targeting proteins required for polarized growth and division septum assembly. A third possibility is that in all the mutants that we have analyzed in this study, a low level of exocyst activity might persist that might be sufficient for cell elongation and division septum assembly but not for cell separation. A further investigation of these possibilities will require the isolation and characterization of a bank of temperature-sensitive mutant alleles of the various exocyst components, followed by detailed cell biological and biochemical characterization of these mutants using secretion assays. The identification of the contents of the 100-nm vesicles that accumulate in the exocyst mutants should also help unravel the cellular function of the exocyst in S. pombe.
The authors especially thank Prof. Nam-Hai Chua for his interest in this project and for several useful discussions about its design and improvement. The authors thank Drs. Keith Gull, Takashi Toda, Naweed Naqvi, and Ms. Srividya Rajagopalan for providing antibodies against tubulin, antibodies against Mok1p, plasmid JK210-GFP, and plasmid SK-ura4-nmt81, respectively. They also thank the members of the IMA-Electron Microscopy Facility (Mr. Qing Wen Lin and Ms. Yang Sun Chan) for expert assistance; Dr. Benedikt Kost and Mr. Desmond Kumar for their help with the confocal microscope; all members of the yeast laboratories, in particular, Drs. Snezhana Oliferenko, Naweed Naqvi, Ventris D' Souza; and Victoria Boulton, Mr. Kelvin Wong, Ms. Suniti Naqvi, and Mr. Fengwei Yu for thoughtful suggestions on the work and critical comments on the manuscript. This work was supported by research funds from the National Science and Technology Board, Singapore.
Article published online ahead of print. Mol. Biol. Cell 10.1091/mbc.01–11-0542. Article and publication date are at www.molbiolcell.org/cgi/10.1091/mbc.01–11-0542.