INTRODUCTION

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

The microtubule cytoskeleton is required for a variety of cellular functions in eukaryotes, including the organization of cellular organelles, the establishment and maintenance of cell polarity, and chromosome movement in mitosis and meiosis. Microtubule dynamics and organization are controlled in part by proteins that associate with polymerized tubulin: microtubule-associated proteins and the motor proteins kinesin and dynein (reviewed in Hunter and Wordeman, 2000). The kinesin motor proteins constitute a superfamily of related molecules defined by a conserved motor domain of ~320 amino acids that displays ATPase activity and microtubule binding; together these produce motility. Kinesins are classified into subfamilies based on motor domain sequence similarities, whereas the sequences outside the motor domain often lack significant similarity, even within a subfamily (reviewed in Goldstein and Philp, 1999).

The fission yeast, S. pombe, is an attractive system for the study of kinesin function, given its relatively small complement of motors together with its suitability for detailed experimentation, e.g., one can efficiently manipulate its genes through molecular genetics. The study of motor enzymes in fission yeast is further facilitated by the elaborate and dynamic microtubule cytoskeleton found in these cells, which is readily visible by light microscopy (reviewed by Hagan, 1998). Interphase cells have an extensive array of cytoplasmic microtubules organized as 3 to 8 bundles that run essentially parallel to the long axis of the cell. These microtubules appear to be organized with their plus ends at the cell tips, while the minus ends congregate in a small region of interdigitation around the nucleus, which lies at the center of the cell (Drummond and Cross, 2000; Tran et al., 2001; reviewed by Hagan, 1998). Moreover, these microtubules undergo dynamic instability with parameters similar to those described for metazoan systems (Drummond and Cross, 2000; Tran et al., 2001). Unlike in budding yeast, these microtubules are essential for organizing vesicular organelles (Ayscough et al., 1993; Yaffe et al., 1996; Hagan and Yanagida, 1997) and maintaining cell polarity (reviewed in Mata and Nurse, 1998; Chang, 2001). The most dramatic change in microtubule organization occurs as the cells enter mitosis; cytoplasmic microtubules are completely disassembled and a mitotic spindle forms within the nucleus. On completion of mitosis, interphase microtubules reappear, first as a "postanaphase array" around the new septum in G₁ and then as the paraxial bundles described above.

The Schizosaccharomyces pombe genome encodes nine kinesin-related genes from seven subfamilies and one cytoplasmic dynein (Hagan and Yanagida, 1990; Pidoux et al., 1996; Yamamoto et al., 1999; Brazer et al., 2000; Browning et al., 2000; Troxell et al., 2001; pombe genome project). Among the kinesin subfamilies found in fission yeast, three have direct effects on microtubule dynamics and higher order microtubule organization (Goldstein and Philp, 1999). S. pombe expresses two members of the KAR3 subfamily, pkl1⁺ and klp2⁺. These promote microtubule disassembly and, together with dynein, are essential for meiosis (Troxell et al., 2001). S. pombe also expresses tea2⁺, a member of the KIP2 subfamily, which promotes microtubule growth (Cottingham and Hoyt, 1997). Deletion of tea2⁺ leads to unusually short microtubules and the mislocalization of the tip-specific marker Tea1p (Browning et al., 2000). This compromises the polarity of tea2 cells, leading to their "T" shape. In budding yeast, the activity of Kip2p is antagonized by two different microtubule-destabilizing kinesins, Kar3p and Kip3p (Cottingham and Hoyt, 1997; Huyett et al., 1998; Miller et al., 1998).

Here, we describe two fission yeast kinesins, klp5⁺ and klp6⁺, that are both homologs to KIP3 kinesin in Saccharomyces cerevisiae. Our data demonstrate that these fission yeast kinesins influence the behavior of microtubules and cell morphology by fostering microtubule disassembly. They are also essential for meiosis.

	MATERIALS AND METHODS

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

Strains and Cell Culture

All strains used in this study are listed in Table 1. The haploid strains 99 and 100 were generally used as wild type controls. Cell culture and genetic manipulations were performed with the use of standard techniques (Moreno et al., 1991). Growth of klp mutants was assayed on solid and liquid media over the range of temperatures commonly used to identify both cold and high temperature sensitivity in fission yeast (20, 25, 32, and 36°C). Growth rates were determined from cultures in yeast extract plus supplements (YES) medium during log phase with the use of the OD₅₉₅ at 20-min intervals. Temperature-sensitive mutants were incubated at 25°C for permissive temperature for growth, and at 36°C for restrictive growth. Cell transformations were carried out with the use of lithium acetate/sorbitol (Moreno et al., 1991) or polyethylene glycol (Elble, 1992)-based protocols. Strains were crossed on malt extract agar plates to induce meiosis, and the resultant spores were plated on YES medium to determine viability.

klp5⁺ and klp6⁺ Intron Mapping

The presence of small introns at the 5' end of both klp5⁺ and klp6⁺ was predicted from the DNA sequence provided by the S. pombe genome project and confirmed experimentally. The klp5⁺ intron was mapped by amplifying a cDNA clone from a S. pombe cDNA library (a generous gift of Drs. C. Norbury and B. Edgar, Imperial Cancer Research Fund, London, England) by polymerase chain reaction (PCR) with the use of primers KLP5-4 (gactaagaaatgtaacttggcaaatg) and KLP5A (ctgttgctgagtagcag). The klp6⁺ intron was confirmed by reverse transcriptase (RT) followed by PCR then a subsequent reaction of nested PCR, followed by DNA sequencing. RT-PCR was performed with reverse transcriptase and DNA polymerase following the manufacturer's specifications (Promega, Madison, WI). The substrate consisted of the poly(A⁺) fraction of RNA isolated from wild type cells with the use of standard methods (Sambrook and Russell, 2001). Parallel reactions were carried out with single primers, no added RNA, and no added RT, to control for primer artifacts or contaminating genomic DNA. RT-PCR first used primers KLP6-13 (cgatactgctatgaaagaagggtc) and KLP6-9 (gcaaacacagtggcattatatcc), followed with primers KLP6-13 and KLP6-14 (gatcaaatgcatatcgaacatc).

Computational Analysis of Sequences

Database searches were done with the BLAST algorithm (Altschul et al., 1997). Protein sequences were aligned with Clustal W (Thompson et al., 1994) then analyzed with the phylogenetic program PAUP, version 4.0, assuming maximum parsimony and with the use of a heuristic search method with stepwise addition (Sinauer Associates, Sunderland, MA). Coiled-coils predictions were done with the COILS program with the use of both MTK and MTIDK matrices, with and without the weighting option (Lupas et al., 1991). The ProPram tools (www.expasy.ch/cgi-bin./protparam) were used to determine molecular weights and theoretical isoelectric points (pI).

Construction of Null Strains

The klp5⁺ and klp6⁺ null mutant strains were constructed with the use of a single-step gene replacement protocol and the selectable markers his3⁺ (Ohi et al., 1996) or ura4⁺ (Grimm et al., 1988) in the appropriate auxotrophic backgrounds (Table 1).

The klp5-null allele was constructed with the use of long flanking regions of klp5⁺ to target integration. The 5'-flanking 575 base pairs were PCR amplified with primers KLP-L1 (atagcgctcactagtcctct) and KLP-L2 (gccagtgggatttgtagctaagcttagaaaagagcgagaaacgcgt), which includes sequence overlapping the 5' end of ura4⁺. The 3'-flanking 520 base pairs were PCR amplified with primers KLP-L3 (gcgtttgttttcctaggcgaagctcacagcttgatcaactgctg), which contains sequence overlapping the 3'-end of ura4⁺, and KLP-L4 (tgtacattggaggtggcaga). The resulting PCR products were then used as long primers for a second PCR reaction, with the ura4⁺ gene as the substrate and the resulting product used in transformations.

The klp6-null allele was constructed with the use of long flanking PCR primers directly. The nutritional marker genes were PCR amplified with primers containing 75 bases identical to the 5'- and 3'-flanking regions klp6⁺. The primers were KLP6-URA4-5' (c a a g c t t c c c a t a c t t g t g t t c t t c t t a a t a g c t t c a c a c a a c t a a a a c a a a t t c a t t c c t a g a a t c a g t a t t a c g a t a c c c a c t g g c t a t a t g t a t g c a t t t g) with KLP6-URA4-3' (g a a g a a a a g c c a a t g a g g a g t t g a t g t t g t c c t t c c a a a a a a a a t t a t a c c a a c t a t t t g a g t g a a a a c c g a t c t c g t t g g t t t c c a a c a c c a a t g t t t a t a a c c a a g) and KLP6-HIS3-5' (c a a g c t t c c c a t a c t t g t g t t c t t c t t a a t a g c t t c a c a c a a c t a a a a c a a a t t c a t t c c t a g a a t c a g t a t t a c g a t a c t g c t t t g g a a a t g a a a g a c a t a t g g a g c) and KLP6-HIS3'3 (g a a g a a a a g c c a a t g a g g a g t t g a t g t t g t c c t t c c a a a a a a a a t t a t a c c a a c t a t t t g a g t g a a a a c c g a t c t c g t t g g c a c g g g t t a t a a t c c t t t a a a t t a g c g).

Diploid cells constructed from strains 99 and 100 (Table 1) were transformed with the PCR products, and Ura⁺ or His⁺ transformants were selected for growth on defined medium. Tetrad analysis showed that all four colonies in each tetrad were viable, whereas the marker genes segregated 2:2, indicating a single integration in a nonessential locus.

Homologous integration at the klp5⁺ locus was confirmed by PCR with the use of primers KLP5-1 (gactcaccaacattcatcctcaac) with URA4-1 (caagatagaatggatgtttgaaattaaacg) or URA4-E (catgctcctacaacattaccac). Homologous integration at the klp6⁺ locus was confirmed by PCR with the use of primers KLP6-1 (cgactatggttcatagatacatggatatg) and HIS3-3 (ctaattgcgcttgcattcc) or URA4-E.

The klp5, klp6, and klp5 klp6 double mutants are collectively referred to as "klp" when either the same experiment was done with all three strains or the same general conclusion is being drawn from experiments done with all three strains.

Construction of GFP-tagged Strains

A plasmid vector was constructed containing a [Glycine-Alanine]₂ spacer, three tandem copies of the Pk1 epitope (Southern et al., 1991), the green fluorescent protein (GFP) (F64L, S65T allele) (Cormack et al., 1996), the nmt1⁺ terminator sequence (Maundrell, 1993), and the ura4⁺ gene. A DNA molecule containing this construction of (GA)₂-(Pk1)₃-GFP-ura4⁺ was PCR amplified with the use of a 5' primer containing 75 bases identical to the 3' end of each respective Klp open-reading frame and 33 bases, in-frame, containing the Glycine-Alanine spacer and the [Pk1]₃ sequence. The 3' primers consisted of 81 bases identical to the 3' end of each klp, starting six base pairs 3' of the stop codon, and 25 bases corresponding to the 3' end of the ura4⁺ gene. The Klp-Pk1-GFP-ura4⁺ PCR product was used to transform diploids and Ura⁺ integrants identified. Homologous integration was confirmed by PCR with primers to the Klp sequence, GFP, and ura4⁺. The Klp-Pk1-GFP region from the selected integrant strains were PCR amplified and sequenced to confirm in-frame integration and to identify any PCR-generated sequence errors in the constructs. One missense mutation was found in the GFP portion of Klp5pGFP (agt-ggt; S2G) and one in the GFP portion of Klp6pGFP (ctt-ctc; L40P), but neither of these changes had a noticeable effect of the fluorescence properties of the GFP fusions.

The klp5-GFP-ura4⁺ fragment was amplified with primers 5'KLP5GFP (c t t c a t c t t t c a a a t c c a g c t a a c a t t a t t a g g a a a t c t t t a a g c a t g g c t g a a a a c g a a g a a g a g a a a g c c a c c g g a g g a g a g c t c a t g g g t a t t c c t a a c c c t t t g) and 3'KLP5GFP (c a t a t c a t c a a g c t t a t c c g t t t t t t t t t t t t a a a t a t a c c c a a c a g g a t a t t t a g a g g a t t c g t a t t t g a a t a t a c g g g t t c c a a c a c c a a t g t t t a t a a c c a a g). The klp6-GFP-ura4⁺ fragment was amplified with primers 5'KLP6GFP (c a a c c a g t a c g c c g t a t a t c g c t t g t t t c a c a a c c t t t a c a a a a a a c t g g c g g g a c t g a g a a t a c t c c t a a t g c t g g a g g a g a g c t c a t g g g t a t t c c t a a c c c t t t g) and 3'KLP6GFP (g a a g a a a a g c c a a t g a g g a g t t g a t g t t g t c c t t c c a a a a a a a a t t a t a c c a a c t a t t t g a g t g a a a a c c g a t c t c g t t g g g t t c c a a c a c c a a t g t t t a t a a c c a a g). The italicized sequences are from the klps, and the remaining are from Gly-Ala/Pk1 (5') or ura4⁺ (3') sequence. The klp5-GFP integration was confirmed with primers KLP5-1 (gactcaccaacattcatcctcaac) with GFP-HVEM3 (gtacataaccttcgggcatg) and KLP5-L4 (gtacattggaggtggcagac) with URA4-E. The klp6-GFP integration was confirmed with primers KLP6-3 (ctcattcttccaaatggccaac) with GFP-HVEM3 and KLP6-1 with URA4-E.

Microscopy

Cells were grown to mid log phase (1-3 × 10⁵cells/ml) for microscopy. Experiments with GFP-tagged genes were generally done at 25°C, and others were done at the temperatures noted. Fixed cells were prepared for tubulin and DNA staining with the use of a double aldehyde method (Hagan and Hyams, 1988), and antibodies were applied as previously described (West et al., 1998). DNA staining was also done with cells fixed by adding 1/10 volume of cell culture to methanol at 20°C for 2-15 min. Cells were collected by centrifugation, resuspended in phosphate-buffered saline (pH 7.4) containing 1 µg/ml 4',6-diamino-2-phenylindole (DAPI) (Sigma, St. Louis, MO) and mounted on glass slides.

Live cells were analyzed by visualizing microtubules with GFP--tubulin, Klp proteins with their GFP fusions, and DNA with Hoescht 33342 (Sigma). The GFP--tubulin plasmid pDQ105 (Ding et al., 1998) was transformed into cells and the resultant strains grown in defined medium (Moreno et al., 1991) containing 5 µg/ml thiamine (Sigma) to limit levels of expression (Maundrell, 1993) of the GFP--tubulin. Cells were collected by centrifugation (5-10 ml) at 3000 rpm for 3 min in a Beckman Coulter CS-6 centrifuge, and resuspended in 5 ml of YES, pH 7.5, containing 2 µg/ml Hoescht 33342 for ~30 min. Cells were then mounted on glass coverslips, and images were collected on a Zeiss Axiophot2 fluorescence microscope with a 100× Plan-APO objective lens (numerical aperture 1.4). Images were captured with a Cooke SensiCam charge-coupled device camera and processed with the use of the SlideBook software package (Intelligent Imaging Innovations, Denver, CO) as follows. Generally, 6 to 12 focal planes spaced at 200- or 300-nm intervals along the Z-axis were collected, with 2 × 2 binning (voxel size 67 nm), with the use of exposure times of ~300 ms in the DAPI channel (Hoescht33342 or DAPI) and 500-1000 ms in the fluorescein isothiocyanate channel (GFP) per focal plane. The images were deconvolved with the use of either a No Neighbor (short Z-series) or Nearest Neighbor (long Z-series) algorithm, with a subtraction factor of 0.6-0.9. A two-dimensional image was then rendered with the use of an orthographic projection of pixel maxima along lines parallel to the optical axis. The resulting image was prepared as figures with the use of CorelDraw (Ottawa, Ont).

	RESULTS

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

klp5⁺, klp6⁺, and KIP3 Constitute a Kinesin Subfamily

We have characterized two of the nine kinesins of fission yeast, designated klp5⁺ (kinesin-like protein) (accession no. Z97211; right arm of chromosome 2, centromere proximal) and klp6⁺ (accession no. ALO23587; left arm of chromosome 2, telomere proximal). Several different sequence alignment algorithms (see MATERIALS AND METHODS) indicated that Klp5p and Klp6p are more closely related to each other than to any other proteins, and each is equally similar to the S. cerevisiae motor KIP3. For example, an alignment of Klp5p, Klp6p, Kip3p and several other kinesins with the use of the phylogenetic program PAUP (version 4.0), groups klp5⁺ and klp6⁺ together with KIP3 but separates them from other kinesins (Figure 1A). Although these two kinesins are not genetically linked, they are physically linked on the same chromosome and contain short introns that are identically placed at their 5' ends, as confirmed by PCR and DNA sequencing (see MATERIALS AND METHODS). The klp5⁺ intron splits the 11th codon (g/gttagt... aaatag/tc) and is 55 base pairs in length. The klp6⁺ intron also splits the 11th codon (g/gtaaga... atcaag/tt) and is 39 base pairs in length.

View larger version (22K): [in this window] [in a new window]   Figure 1.   Sequence alignments and structural features of the KIP3 subfamily of kinesins. (A) Bootstrap analysis produced a phylogenetic tree made with S. pombe Klp5p and Klp6p, S. cerevisiae Kip3p and Kip2p, Drosophila melanogaster KLP67A, and Caenorhabditis elegans LF224KLP (see MATERIALS AND METHODS). The numbers represent bootstrap values for 100 replicas. (B) Generalized domain map for the KIP3 subfamily. The amino acid that marks the approximate boundary between each domain is marked on the scale below the diagram. See below for the exact positions. (C) BESTFIT alignment of the amino-terminal leader portion of KIP3 kinesins. Identical amino acids are boxed in dark gray, and similar amino acids in light gray (beginning at amino acid 3). (D) COILS (Lupas et al., 1991) prediction for Klp5p with the use of a window of 28. A coiled-coil probability of 1.0 is calculated for Klp5p amino acids 396-436, corresponding to the "neck" in B. A similar graph is generated for Klp6p, with a probability of 0.91 for amino acids 404-440 (our unpublished data). (E) BESTFIT alignment of the "tail" boxes for Klp5p and Klp6p. The position of these amino acids is indicated as boxes in the tail of B (each box begins at Klp5p, amino acid 554, then 691; Klp6p, amino acid 548, then 671).

Sequence analyses of the predicted proteins identified distinguishing motifs and generated a models their domain structures (Figure 1B). Klp5p is predicted to contain 883 amino acids and a molecular mass of 99 kDa, whereas Klp6p is 784 amino acids (aa) with a predicted molecular mass of 88 kDa. As with many other kinesins, the similarity seen is largely confined to ~320 amino acids of the motor domain. The motor domains of Klp5p and Klp6p are 66% identical/76% similar, whereas those of both Klp5p and Klp6p are 62% identical/73% similar with Kip3p.

The motor domain of each polypeptide is located near its amino-terminal end, but there is a short nonmotor sequence at the amino terminus (Figure 1B). This amino-terminal region is variable in size (Klp5p, ~90 aa; Klp6p, ~110 aa) and is unique to each protein, with the exception of a conserved sequence of 18 aa (Figure 1C). A similar sequence is also found in UNC-104 kinesin subfamily members (for Internet attachments to sequence files see the Kinesin Home Page; http://www.blocks.fhcrc.org/%7Ekinesin), but the significance of this similarity is unknown.

Immediately carboxy terminal from the motor domain, these kinesins are predicted to have a single turn of coiled-coil (see MATERIALS AND METHODS) (Figure 1D). The position, length, and sequence of this region are analogous to the 7 coil of kinesin heavy chain, which has been shown to be sufficient for dimerization and critical in specifying plus-end directed motility (reviewed in Sack et al., 1999).

The carboxy-terminal half of each of these kinesins constitutes the tail domain, but there are no significant structural predictions for this domain. Each tail is also of variable length (Klp5p, ~450 aa; Klp6p, ~360 aa) and its sequence is unique to each protein. The theoretical isoelectric points for the tails of Klp5p (pI = 5.6) and Klp6p (pI = 10.6) further exemplify their differences. There is, however, a short region of 34 amino acids, designated the "tail box," which is highly conserved among just these two fission yeast Klps (Figure 1E). This sequence does not show any significant similarity with other proteins, although the lysine rich region bares some resemblance to several unrelated DNA binding proteins.

klp5⁺ and klp6⁺ Are not Essential for Vegetative Growth

The functions of klp5⁺ and klp6⁺ were investigated through the construction of null mutants. The entire open-reading frame for each kinesin was deleted, and the resultant strains, designated klp5 and klp6, were analyzed for growth under various conditions (see MATERIALS AND METHODS). The growth of klp5 and klp6 strains on solid medium was virtually indistinguishable from wild type (Figure 2). They were, however, slightly darker on plates containing the exclusion stain Phloxin B (Sigma), indicating a subtle loss of viability (our unpublished data). The lack of a significant growth defect in these klp strains was confirmed by examining growth in liquid culture. The klp6 strain did exhibit a slightly longer doubling time at both moderate and high temperatures (Table 2). Surprisingly, the klp5 klp6 double mutant was also viable and not observably compromised (Figure 2 and Table 2). Rather, growth of klp5 klp6 more closely resembled that of the klp5 single mutant, so the klp5⁺ deletion is epistatic to the klp6⁺ deletion. Throughout our analyses of these two kinesin genes, we repeatedly observed the strongest phenotype in the klp6 strain, although the phenotypic differences among the klp mutants are subtle.

View larger version (43K): [in this window] [in a new window]   Figure 2.   The klp5, klp6, and klp5 klp6 mutants are viable. Strains were grown in liquid culture to early log phase then plated as fivefold serial dilutions onto agar and grown at 32°C for 3 d.

View this table: [in this window] [in a new window]   Table 2.  Growth rates for the klp mutants Cultures were grown in YES medium from early to late log phase and the OD595 recorded at 20-min intervals. Doubling times were derived from three independent experiments, and the variations are given as the standard error of the mean.

klp5 and klp6 Cells Have Long Microtubules

Several kinesins alter the dynamics of microtubules in cells, either by increasing or decreasing tubulin assembly (reviewed in Hunter and Wordeman, 2000). The loss of a motor enzyme with either of these activities can have a distinct effect on cells, and it can therefore be studied in several ways, including changes in cell morphology and microtubule organization, interactions with tubulin mutants, and altered sensitivity to microtubule drugs. We have used all these assays to probe microtubule behavior in the klp alleles and found that the microtubules are stabilized by the absence of either Klp5p or Klp6p.

Genetic interactions between klp and tubulin were examined by constructing double and triple mutants between klp strains and mutants in both - and -tubulin. The klp mutants were synthetically lethal with the cold-sensitive, thiabendazole (TBZ)-sensitive, -tubulin mutant nda2-K52 (Toda et al., 1984), because double and triple mutants were not identified at the permissive (32°C) or restrictive (20°C) temperatures for nda2-K53. No interaction was apparent between the cold-sensitive, TBZ-resistant, -tubulin mutant nda3-311 (Hiraoka et al., 1984), and the klp strains (our unpublished data).

The klp mutants grew comparatively well in the presence of the microtubule-destabilizing drug TBZ, as demonstrated by plating assays (Figure 3). Wild-type cells failed to grow in TBZ concentrations higher than 25 µg/ml at 32°C, but klp strains continued to grow in 75 µg/ml. Comparable differences in resistance to TBZ among wild type and klp strains were also seen at 25 and 36°C, although the lower temperature increased the sensitivity of all strains to TBZ. No differences in TBZ resistance were observed among the three klp strains. The ability of the klp strains to withstand elevated concentrations of a microtubule poison suggests that microtubules are more stable in the absence of either of these two kinesins than in wild type cells.

View larger version (18K): [in this window] [in a new window]   Figure 3.   klp mutants are resistant to the microtubule-destabilizing drug TBZ. Strains were grown in liquid medium to early log phase then plated in fivefold serial dilutions onto agar plates containing the indicated concentration of TBZ and incubated at 32°C for 3 d.

If the microtubules in klp cells are indeed hyperstable then their length or organization would likely be altered. To test this possibility, the microtubules in living cells were examined directly with the use of GFP--tubulin. Wild-type interphase cells contain a modest array of straight microtubules, which generally run essentially from one end of the cell to the other (Figure 4A; reviewed in Hagan, 1998). The klp strains, on the other hand, frequently contained microtubules of sufficient length to curl around the ends of the cells (Figure 4, B-D, arrows). As described for TBZ resistance, differences between wild type and klp were apparent, but no significant difference in microtubule length or organization was observed among the klp strains. The tendency for klp microtubules to grow to the ends of cells and bend, together with the resistance of these strains to TBZ, is consistent with the hypothesis that both Klp5p and Klp6p promote microtubule disassembly.

View larger version (85K): [in this window] [in a new window]   Figure 4.   klp mutants strains have longer, bent microtubules. Microtubules were visualized with GFP--tubulin by deconvolution microscopy (see MATERIALS AND METHODS). (A) Wild-type cells with straight microtubules. (B) klp5 cells with long, curved microtubules at the ends (arrow). (C) klp6 cells with long, curled microtubules (arrow). (D) klp5 klp6 cells are similar to the single mutants. Bars, 5 µm.

cdc25⁺ Function Modulates Microtubule Behavior in the Absence of klp5⁺ and klp6⁺

Although defects in cytoplasmic microtubules often result in defects in cell shape, the klp cells have an apparently normal cell morphology, both during log phase growth and as they emerge from stationary phase (our unpublished data). However, abnormal cell morphology was observed in certain genetic backgrounds that significantly increase cell size. The strongest effect was seen with the temperature-sensitive cell cycle progression mutant cdc25-22 (Figure 5). At restrictive temperature this allele arrests cell cycle progression at` the G₂/M boundary, resulting in cells that are >3 times longer than wild type (Fantes, 1979). Even at permissive temperature cell cycle progression is somewhat delayed, resulting in cells that are ~50% longer than wild type. The cdc25-22 cells grow with a straight, symmetric shape at both permissive and restrictive temperatures (Figure 5, A and B). The klp, cdc25-22 strains, on the other hand, become progressively bent as they get longer at restrictive temperature, forming "C"-shaped cells (compare Figure 5, B and D). After 8 h of arrest, virtually all of the cells had an abnormal morphology. Although the bending was pronounced, it was generally symmetric, and the cells septate in the middle of the C after they were released from G₂ arrest and divided (our unpublished data).

View larger version (66K): [in this window] [in a new window]   Figure 5.   Cell cycle arrest mutant cdc25-22 has a bent-cell shape and particularly long, bent microtubules in the klp backgrounds. (A-D) Differential interference contrast images of cdc25-22 and cdc25-22, klp5 cells. (A) cdc25-22 cells at permissive temperature. (B) cdc25-22 cells at restrictive temperature for 6 h. (C) cdc25-22, klp5 cells at permissive temperature. (D) cdc25-22, klp5 cells at restrictive temperature for 6 h. (E-H) Microtubules were visualized with GFP--tubulin by deconvolution microscopy. (E) cdc25-22 cells at permissive temperature. Inset is a higher magnification of the tip of one cell (arrow) with straight microtubules. (F) cdc25-22 cells at restrictive temperature for 6 h. (G) klp5 cdc25-22 cells at permissive temperature. Inset is a higher magnification of the tip of one cell (arrow) with microtubules curled around the end. (H) klp5 cdc25-22 cells at restrictive temperature for 6 h. Curled microtubules are present but not always at the end of the cell (arrows). Bars, 5 µm.

The organization of the microtubule cytoskeleton in these cells was determined with GFP--tubulin. The cdc25-22 cells had long, straight microtubules at both permissive and restrictive temperature, although the microtubules often failed to reach the ends of the longer cells (Figure 5, E and F). The klp, cdc25-22 cells, in contrast, had especially long microtubules that wrapped around the ends of the cells (compare Figure 5, E and G), even at permissive temperature. At restrictive temperature, these microtubules sometimes curled before they reached the ends of the cells (Figure 5H, arrows). This phenotype was the most pronounced in the klp5, klp6, cdc25-22 triple mutants.

The extreme elongation of microtubules in these cells could be due either to a general effect from increased length, or to a more direct, regulatory interaction between the klp5⁺ and klp6⁺ genes and cdc25⁺. To distinguish between these possibilities, we constructed klp strains that grew longer than wild type as a result of factors independent of cdc25⁺, and we also examined interactions between klp and genes directly related to the function of cdc25⁺.

The cdc10-V50 mutant arrests cell cycle progression at START in G₁ at 36°C, but the cells continue to elongate at one end, becoming long and straight, similar to cdc25-22 (Marks et al., 1992). The microtubule arrays in cdc10-V50 grown at either permissive or restrictive temperature did not look different from wild type. The klp, cdc10-V50 cells, on the other hand, became bent at restrictive temperature (Figure 6, A and B). These cells were similar to the klp cdc25-22 mutants, except they were generally bent at just one end, forming "J-" rather than C-shaped cells. The microtubules in klp, cdc10-V50 cells sometimes bundled, but they were not long and curled as in cdc25-22 (Figure 6, C and D).

View larger version (129K): [in this window] [in a new window]   Figure 6.   Cell morphology and microtubule organization in cell cycle mutants and diploids. (A) Bright field image of klp5, cdc10-V50 cells at permissive temperature. (B) Bright field image of klp5, cdc10-V50 cells at restrictive temperature for 6 h. (C) Microtubule labeling in klp5 cdc10-V50 cells at permissive temperature. (D) Microtubule staining in klp5 cdc10-V50 cells at restrictive temperature for 6 h. (E) Bright field image of klp5/klp5 diploid cells at 25°C. (F) Bright field image of klp5/klp5 diploid cells at 36°C. (G) Microtubule labeling in klp5/klp5 diploid cells at 25°C. (H) Microtubule staining in klp5/klp5 diploid cells at 36°C. (I-L) Microtubule labeling in wee1-50 backgrounds: wee1-50 cells at permissive temperature (I); wee1-50 cells at restrictive temperature for 6 h (J); klp5 wee1-50 cells at permissive temperature (K); klp5 wee1-50 cells at restrictive temperature for 6 h (L). (M) Bright field image of klp5, cdc2-33 cells at permissive temperature. (N) Bright field image of klp5, cdc2-33 cells at restrictive temperature for 4 h. (O) Microtubule labeling of klp5, cdc2-33 cells at permissive temperature. (P) Microtubule labeling of klp5, cdc2-33 cells at restrictive temperature for 4 h. Bars, 10 µm.

Fission yeast diploids cells are ~85% longer than haploids (Nurse and Thuriaux, 1980), and they have microtubule arrays similar to those seen in haploid cells (reviewed in Hagan, 1998). The klp/klp homozygous diploid cells were appropriately long and slightly bent, with an occasional curled microtubule at 25°C, but at 36°C these effects became pronounced (Figure 6, E-H).

Together, these results indicate that the normal shape of fission yeast cells is disrupted when either klp5⁺ or klp6⁺ is deleted and the cells are sufficiently long. This effect is enhanced by growth at elevated temperature. The effect of klp on microtubule structure varies depending on the genetic background, with the greatest effect observed in cdc25-22 cells.

We asked whether cdc25⁺ might play a more direct role in microtubule/motor regulation by generating double mutants with klp and other genes related to cdc25⁺ function. The wee1⁺ kinase acts antagonistically to the cdc25⁺ phosphatase, inhibiting the cyclin-dependent kinase cdc2⁺, and thus restraining entry into mitosis. The lose-of-function allele, wee1-50, produces cells that are small (Fantes, 1981; reviewed in Forsburg and Nurse, 1991) but have wild type microtubule arrays at both permissive and restrictive temperature (Figure 6, I and J). The klp, wee1-50 mutants showed no obvious shape abnormality at permissive temperature, but at restrictive temperature they were 15% shorter than wee1-50 alone at the same temperature. At both temperatures, the microtubules in klp, wee1-50 cells tended to curl at the ends of the cells (Figure 6, K and L).

Both Cdc25p and Wee1p act on the kinase encoded by cdc2⁺, the master regulator of mitosis (reviewed in Forsburg and Nurse, 1991). We therefore looked for an effect of the temperature-sensitive allele cdc2-33 on cell morphology and/or microtubule organization in the klp backgrounds. Both cell morphology and microtubule organization in the cdc2-33 parental strain resembled that of cdc25-22, and they also elongated at restrictive temperature (our unpublished data). The klp, cdc2-33 cells did tend to bend, but their microtubules did not curl more than was seen in wild type cells (Figure 6, M-P).

These experiments reveal that cdc25⁺ can modulate microtubule behavior in a way that is normally masked by the expression of Klp5p and Klp6p, suggesting that this phosphatase may act on microtubule proteins to modulate tubulin dynamics.

Klp5p and Klp6p Localize to Cytoplasmic Microtubules and Mitotic Spindles

To better understand the functions of Klp5p and Klp6p in microtubule organization, each of these proteins has been localized in vivo, with the use of GFP fused to the carboxy terminus by integration of appropriate sequences at the klp5⁺ and klp6⁺ loci. This strategy assures that expression of each chimeric protein is under the control of its endogenous promoter (see MATERIALS AND METHODS). The Klp5p-GFP and Klp6p-GFP strains showed normal growth on plates and produced viable spores in meiotic crosses, indicating that the fusions did not significantly compromise motor function. Live cells were examined through the cell cycle, and the same pattern of localization was observed for both Klp5p and Klp6p.

The Klp-GFP proteins localized to cytoplasmic microtubules in interphase cells, with no detectable preference for a subset of microtubules or bias toward one microtubule end or the other (Figure 7, A and B). As the cells entered mitosis, the Klp-GFP proteins localized to the mitotic spindle and the astral microtubules (Figure 7, C and D). Immediately upon exit from mitosis, the Klp-GFP proteins relocalized to cytoplasmic microtubules, initially to the postanaphase array, and subsequently to the normal interphase microtubules (Figure 7).

View larger version (53K): [in this window] [in a new window]   Figure 7.   Klp5p and Klp6p localize to cytoplasmic microtubules in interphase and spindles in mitosis. Klp5-GFP and Klp6-GFP were visualized in live cells (green) with DNA staining (Hoescht 33342, blue) by deconvolution microscopy. (A) Klp5p-GFP localized to cytoplasmic microtubules in interphase. (B) Klp6p-GFP localized to cytoplasmic microtubules in interphase. (C) Klp5p-GFP localized to a mitotic spindle in the nucleus, and to the "astral" microtubules in the cytoplasm (arrows). (D) Klp6p-GFP localized to a mitotic spindle in the nucleus. (E) Klp5p-GFP redistributes to the interphase microtubules as they are formed at the conclusion of mitosis. (F) Klp6p-GFP redistributes to interphase microtubules after mitosis. These first appear as the "post anaphase array" near the region of septation. Bar, 5 µm.

To determine whether the localization of either Klp5p or Klp6p is dependent on expression of the other, we crossed each Klp-GFP strain into the reciprocal klp strain and examined the distribution of fluorescence. Both of these double mutant strains showed the same localization of Klp5p and Klp6p in their reciprocal klp background as was seen in the wild type background (our unpublished data).

Together, these data demonstrate that both Klp5p and Klp6p localize to cytoplasmic microtubules in interphase and to both the mitotic spindle and astral microtubules in mitosis. The localizations of Klp5p and Klp6p were indistinguishable, although each protein localized independently of the other.

Interactions with Other Kinesins

Complex interactions among kinesin subfamily members have been described in several systems (reviewed in Goldstein and Philp, 1999). Of particular interest for our studies of klp5⁺ and klp6⁺ are the antagonistic interactions between budding yeast KIP3 and KIP2 (Cottingham and Hoyt, 1997; Miller et al., 1998), and the aspects of overlapping function between KIP3 and KAR3 (DeZwaan et al., 1997; Cottingham et al., 1999). Similar interactions in fission yeast have been investigated by constructing double, triple, and quadruple mutants between the klp strains and the relevant fission yeast homologs.

The KIP2 homolog tea2⁺ has previously been characterized; null mutants are viable but have short microtubules and often take on a "T" morphology (Browning et al., 2000). The klp, tea2 double and triple mutants were as viable as the parental single and double mutant strains. Further, the T morphology phenotype described for tea2 cells was neither enhanced nor rescued in klp backgrounds (our unpublished data). These data suggest a lack of significant functional interaction among these kinesins in fission yeast.

S. pombe has two KAR3 homologs, pkl1⁺ and klp2⁺ (Pidoux et al., 1996; Troxell et al., 2001). Neither of these kinesins is essential, either individually or together, but null mutants have altered sensitivities to TBZ and severe meiotic phenotypes. The four possible triple mutant combinations, and the klp5, klp6, pkl1, klp2 quadruple mutant were constructed and analyzed for growth. There were no obvious growth defects in any of the triple mutants, and the quadruple mutant showed only a slight growth defect at 36°C (our unpublished data). These data suggest a lack of significant functional redundancy among the members of the KIP3 and KAR3 kinesin subfamilies in fission yeast.

klp5⁺ and klp6⁺ Are Essential for Meiosis

Kinesins (Meluh and Rose, 1990) and dynein (Yamamoto et al., 1999; Troxell et al., 2001) are important for meiosis in several organisms, including yeasts, so we asked whether klp5⁺ and klp6⁺ play a role in meiosis of fission yeast. S. pombe cells initiate a meiotic life cycle when they are starved for nitrogen and when haploid cells of opposite mating types are present. The meiotic cycle culminates with the formation of an ascus containing four symmetric, haploid spores (Figure 8A). This configuration was observed in 90% of homozygous wild type zygotic asci. However, the klp5 and klp6 strains showed only ~3% wild type asci (four symmetric spores) (n = ~1000) in crosses homozygous for either klp allele. Heterozygous crosses between these klp alleles and wild type yielded a lower frequency of abnormal asci (30-60% normal; n = 300). A wide range of morphological anomalies was seen in klp asci, but the occurrence of asci that contained one single, large spore in klp5 crosses suggests a defect in the first meiotic division (Figure 8B). The klp6 homozygous crosses produced some one-spore asci, but pinched asci were also seen, which may represent a failure to complete conjugation (Figure 8C).

View larger version (49K): [in this window] [in a new window]   Figure 8.   klp strains have defects in meiosis as indicated by the number and morphology of spores in meiotic asci. (A) Differential interference contrast image of normal asci from a homozygous wild type cross, each contains four symmetric spores. (B) Ascus from a klp5 homozygous cross with a single, large spore (arrow). (C) Ascus from a klp6 homozygous cross with several small spores (arrowheads) and incompletely fused (arrow) mating cells. Bar, 5 µm.

Spore viability was greatly reduced in klp homozygous crosses, whereas heterozygous crosses of either klp5 or klp6 with wild type showed nearly wild type spore viability (Table 3A). In heterozygous crosses between klp5 and klp6, spore viability was intermediate between that observed for wild type and homozygous klp crosses (Table 3A). Equivalent results were obtained when the parental strains were either heterozygous for klp5⁺ and klp6⁺ in cis (klp5 klp6 × wt) or trans (klp5 × klp6).

View this table: [in this window] [in a new window]   Table 3.  klp spore viability (A) Spore viability from zygotic asci. Random spores from each cross were isolated, plated on YES medium, and colonies counted after 5 d. (B) Spore viability for azygotic asci. Stable diploids were isolated from each indicated cross, grown vegetatively to form colonies, and then induced to sporulate (see MATERIALS AND METHODS). Spore viability was then determined as described for zygotic crosses. Spores (1000) were plated for each cross.

To distinguish between defects in the initial formation of a diploid zygote versus subsequent meiotic DNA segregation, stable klp diploid strains were isolated, grown vegetatively for several generations, and then induced to sporulate. The relative frequency of stable diploids isolated from klp5 crosses was not significantly different than from wild-type crosses, but klp6/klp6 diploids were less frequently found. These results suggest that zygote formation is not significantly affected in klp5 but is compromised in klp6. Spore viability was not increased in any of the klp strains by first selecting for stable diploid cells, indicating that meiotic defects subsequent to karyogamy result in loss of spore viability (Table 3B).

	DISCUSSION

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

KIP3 Kinesin Subfamily

We propose a new subfamily of kinesin-like proteins that includes the fission yeast genes klp5⁺ and klp6⁺ and the group's founding member, KIP3, from S. cerevisiae. There are several shared characteristics that define members of this new kinesin subfamily. First and foremost, there is considerable sequence similarity in their motor domains and in their general domain structure and organization (Figure 1). Kip3p, Klp5p, and Klp6p also all localize to cytoplasmic and spindle microtubules (Figure 7) (DeZwaan et al., 1997). Furthermore, Klp5p and Klp6p share with Kip3p an activity that fosters microtubule disassembly, as evidenced by the unusually long and/or robust microtubules found in null mutants (Figures 3-6) (Cottingham and Hoyt, 1997; DeZwaan et al., 1997; Miller et al., 1998). Microtubule-disassembling activity has also been described for the KinI kinesin subfamily (XKCM1, Walczak et al., 1996; MCAK, Maney et al., 1998; Desai et al., 1999) and for Kar3p (Endow et al., 1994). There is, however, no obvious sequence similarity between members of the KIP3 subfamily and the other kinesins with "exotubulase" activity. It is possible that each kinesin subfamily uses a different mechanism to promote microtubule disassembly or that there is conservation in protein structure that is not apparent from the amino acid sequence.

Although all the KIP3 family members share the feature of promoting microtubule disassembly, their roles in the physiology of each yeast cell are distinct in several important ways. First, the klp mutants of fission yeast show disrupted patterns of DNA segregation along the mitotic spindle but nuclear positioning appears normal (West et al., 2001). KIP3, on the other hand, is part of a pathway required for proper nuclear migration to the bud neck early in mitosis, but movement of the chromosomes along the spindle appears normal (DeZwaan et al., 1997). Second, klp5⁺ and klp6⁺ are not essential for vegetative growth, even in the absence of KAR3 family members pkl1⁺ and klp2⁺. KIP3 is likewise nonessential, but either KIP3 or KAR3 (Meluh and Rose, 1990) (together with at least one BimC family member) must be present for viability, suggesting that budding yeast requires at least one microtubule-destabilizing motor protein for survival (DeZwaan et al., 1997; Saunders et al., 1997; Cottingham et al., 1999). S. pombe may not share this requirement for a destabilizing motor, or this function may be accomplished by other proteins expressed in the fission yeast cell. Third, both klp5⁺ and klp6⁺ are necessary for proper microtubule organization and cell morphology in several genetic backgrounds. As discussed below, this phenotype is particularly sensitive to the activity of cell cycle regulatory phosphatase cdc25⁺. However, no such functions have been reported for KIP3 in budding yeast. Finally, klp5⁺ and klp6⁺ are both essential for meiosis, as discussed below, whereas KIP3 is not (Cottingham and Hoyt, 1997; Miller et al., 1998).

Microtubules and Cell Morphology

The establishment and maintenance of cell polarity and morphology is a microtubule-dependent process in fission yeast such that perturbations in microtubule organization can produce defects in cell shape and polarity (reviewed in Hagan, 1998; Chang, 2001). Previous work has demonstrated that shortening microtubule length, either by mutations in tubulin or tubulin-Cofactor genes (Umesono et al., 1983; Grishchuk and McIntosh, 1999; Radcliffe et al., 1999), by the addition of TBZ (Sawin and Nurse, 1998), or by deletion of microtubule-associated proteins (mal3⁺, Beinhauer et al., 1997; tea2⁺, Browning et al., 2000; tip1⁺, Brunner and Nurse, 2000; dis1⁺ and mtc1⁺, Nakaseko et al., 2001) leads to the formation of T-shaped cells. Results presented here suggest that an increase in microtubule length can cause bending at the growing ends of the cells, resulting in C- or J-shaped cells. This conclusion is supported by similar results reported for null mutants of moe1⁺, a component of the Ras1 signaling pathway (Chen et al., 1999), and components of the -tubulin complex (Paluh et al., 2000; Vardy and Toda, 2000). The changes in cell shape manifested in the klp mutants described here occurred in genetic backgrounds and growth conditions that made the cells especially long, including several cell cycle arrest mutants and diploid strains. There may be a heightened sensitivity to changes in microtubule organization when the distance between the cell's ends is significantly increased. The T phenotype of tea2 cells is also enhanced when the cells are longer (Browning et al., 2000).

The klp-mediated changes in morphology are also enhanced by higher growth temperatures, as evidenced by the phenotype of klp homozygous diploid cells grown at 36 versus 25°C. This result is consistent with the hypothesis that the phenotype arises as a result of hyperstabilized microtubules, because higher temperatures generally favor microtubule assembly. A similar temperature dependence has been reported for cell morphology defects observed in cells treated with TBZ (Sawin and Nurse, 1998). Because the cell cycle mutants used here to produce long cells are all temperature sensitive, it is not possible to distinguish clearly between the effects of temperature and cell length. Nonetheless, several observations argue that the effects we report are both temperature dependent and enhanced by cell length. Normal cell shapes are observed in both klp, cdc25-22 and klp diploid cells at 25°C, despite the presence of abnormal microtubules, suggesting a temperature-dependent effect on cell morphology. On the other hand, the normal cell shape seen in klp haploid cells at any temperature argues that increased cell length also contributes to the bent cell phenotype. Moreover, the bent-cell phenotype observed in all the cell cycle arrest mutants was greater as the cells continued to get longer in the arrest.

The formation of C-shaped cells in the cdc25-22 background and of J-shaped cells in the cdc10-V50 background is likely to be the consequence of monopolar (cdc10-V50) versus bipolar (cdc25-22) cell growth. The switch between these states, termed New-End Take Off, occurs at the beginning of G₂ (Mitchison and Nurse, 1985). Thus, the arrest points of cdc10-V50 (G₁) and cdc25-22 (late G₂) are on opposite sides of this event.

Examination of the microtubule cytoskeleton in the bent cells with klp5 and klp6 genotypes indicates that the interplay between cell elongation and the microtubule cytoskeleton is not a simple matter of microtubule geometry. First, the organization of the microtubules in the long klp, cdc25-22 and klp diploid cells is distinct from either klp cdc10-V50 or klp cdc2-33 cells, and from previously published ban mutants (Verde et al., 1995), although cell shape is similar in all of these cases. Second, the cdc25-22 mutation had a profound effect on the organization of the microtubules in the klp background at permissive temperature, but cell morphology was not changed. Finally, the T morphology observed in tea2 cells is not rescued in tea2 klp double and triple mutants, although the short-microtubule phenotype was at least partially rescued in these mutants (our unpublished data). The effect of cdc25⁺ function on microtubule organization is discussed below, but the results discussed here indicate that certain kinds of changes in microtubules do not affect a cell's polar organization. Meanwhile, different changes in microtubule arrangement can produce the same net result on cell morphology.

cdc25⁺ and Microtubule Organization

Even a subtle disruption in the function of cdc25⁺ leads to a profound rearrangement of the microtubule cytoskeleton in the absence of either klp5⁺ or klp6⁺ (Figure 5). Cells carrying the temperature-sensitive allele cdc25-22 together with the klp mutations had extended microtubule arrays that often curled around the ends of the cells, even at permissive temperature. Several lines of evidence indicate that this effect is not simply the result of the increased length of the cdc25-22 cells. First, the effect was not observed in either cdc2-33 or cdc10-V50 cells, although they are similar in size to cdc25-22. Second, curled microtubules were also observed in klp, wee1-50 cells, which are shorter than wild-type cells. Third, the microtubules are also not likely to be curling simply due to a physical barrier presented by the ends of the cells, because the microtubules often curled before they reached the ends of the cells. Moreover, the shorter klp, wee1-50 cells have a less severe phenotype than that observed in cdc25-22 cells, contrary to what might be expected if hyperstable microtubules were confined to a smaller space. These results favor a regulatory interaction between the microtubules and cdc25⁺ that is normally masked by the presence of klp5⁺ and klp6⁺. The interaction between klp and both wee1-50 and cdc25-22 may seem enigmatic because these mutants have opposite effects on cell size and are antagonistic to each other in their regulatory pathway (reviewed in Forsburg and Nurse, 1991). It is possible, however, that any alteration in the balance between Cdc25p and Wee1p activity leads to aberrant microtubules.

It is also possible that the phenotype observed in klp, cdc25-22 mutants is dependent on cells being in G₂, and not on cdc25⁺ activity, alone. This could explain why the klp, cdc10-V50 mutants do not have the same phenotype. However, this seems unlikely because the majority of the cdc2-33 cells are also arrested in G₂, and their phenotype more closely resembled that of cdc10-V50. Furthermore, the klp, wee1-50 cells have a shortened G₂ stage, and yet they also have the bent microtubule phenotype. At present it is impossible to rule out other G₂-specific factors because the substrate for either cdc25⁺ or wee1⁺ that affects microtubule behavior remains unknown. Because the phenotype is present in null alleles of klp5⁺ and klp6⁺, it is likely that the potential substrates revealed here include proteins other than these two kinesins. The absence of the curled microtubules in the klp, cdc2-33 mutants suggests that the mechanism is independent of MPF activity. Rather, it seems likely that cdc25⁺ is interacting with an unidentified gene whose activity is normally buffered by klp5⁺ or klp6⁺. A role for cdc25⁺ in regulating microtubule behavior is also indicated by interactions between the cdc25-22 allele and several other kinesins, including klp2⁺ (Sweezy and McIntosh, personal communication), and tea2⁺ (Browning et al., 2000). Tubulin metabolism involves many different classes of genes, so the candidates for such a locus are numerous.

Meiosis

Our data indicate that both Klp5p and Klp6p play a major role in the production of viable spores from meiosis. Crosses among klp5⁺ and klp6⁺ null mutants produced zygotic asci with abnormal morphologies and extremely low spore viability. Azygotic asci induced from stable diploid cells displayed a similar loss in spore viability. The similarity in phenotypes observed between zygotic and azygotic spore formation suggests that these mutations lead to meiotic defects downstream from conjugation and karyogamy. The precise point(s) of meiotic failure remains to be determined. The intermediate phenotype in heterozygous crosses between klp5 and klp6, versus heterozygous crosses between either klp5 or klp6 and wild type, does suggest an unusual, interallelic form of haplo-insufficiency for these two kinesins in meiosis.

The essential role for both klp5⁺ and klp6⁺ in the meiotic life cycle contrasts strongly with the behaviors found for vegetative growth. klp5⁺ and klp6⁺ are the fifth and sixth motor proteins demonstrated to have a major role in fission yeast meiosis (cut7⁺, Hagan and Yanagida, 1990; dhc1⁺, Yamamoto et al., 1999; pkl1⁺ and klp2⁺, Troxell et al., 2001), whereas only cut7⁺ is essential for mitosis (Hagan and Yanagida, 1990). This indicates that fission yeast meiosis is very sensitive to the normal functioning of the microtubule cytoskeleton compared with vegetative growth. It remains to be determined whether the activity of each kinesin, per se, is distinct in these two life cycle stages, or whether the differences between meiosis and mitosis arise from differences in the mechanical requirements or functional redundancies among the motor proteins used to achieve these mechanics, or even from the checkpoints that operate in each life cycle stage.