The Saccharomyces cerevisiae Centromere Protein Slk19p Is Required for Two Successive Divisions During Meiosis

Corresponding author: William S. Saunders, Department of Biological Sciences, 258 Crawford Hall, University of Pittsburgh, Pittsburgh, PA 15260.

Communicating editor: M. D. ROSE

	ABSTRACT

TOP ABSTRACT MATERIALS AND METHODS RESULTS DISCUSSION LITERATURE CITED

Meiotic cell division includes two separate and distinct types of chromosome segregation. In the first segregational event the sister chromatids remain attached at the centromere; in the second the chromatids are separated. The factors that control the order of chromosome segregation during meiosis have not yet been identified but are thought to be confined to the centromere region. We showed that the centromere protein Slk19p is required for the proper execution of meiosis in Saccharomyces cerevisiae. In its absence diploid cells skip meiosis I and execute meiosis II division. Inhibiting recombination does not correct this phenotype. Surprisingly, the initiation of recombination is apparently required for meiosis II division. Thus Slk19p appears to be part of the mechanism by which the centromere controls the order of meiotic divisions.

MEIOSIS is a special type of cell division that generates haploid gametes to fulfill the needs of sexual reproduction. To achieve this end, a single round of DNA replication is followed by two successive rounds of chromosome segregation, reductional (meiosis I) and equational (meiosis II). While meiosis II is mechanically similar to mitosis, meiosis I is quite distinct from both mitosis and meiosis II. During anaphase of meiosis I, sister chromatids do not separate from each other. Instead they segregate as a unit away from the other homologous chromosome.

The two separate meiotic divisions are not interdependent, as demonstrated by single-division meiosis in Saccharomyces cerevisiae mutants, such as cdc5, cdc14, spo12, spo13, and cdc28 (KLAPHOLZ and ESPOSITO 1980B ; SHUSTER and BYERS 1989 ; SHARON and SIMCHEN 1990B ). In these mutants, chromosomes can undergo either a single reductional division or a single equational division. Although the two nuclear divisions can occur independently, in normal meiosis the progression is regulated so that equational division only happens after reductional division.

Substantial progress has been made to identify what determines whether chromosomes will undergo reductional or equational segregation (reviewed in SIMCHEN and HUGERAT 1993 ; MCKIM and HAWLEY 1995 ). By fusing two grasshopper spermatocytes at different meiotic stages, NICKLAS 1977 found that meiosis I chromosomes still segregated reductionally when micromanipulated into a meiosis II spindle. This result indicates that it is the chromosome, and not the spindle, or the general cell environment, which determines the segregational characteristic of the chromosome. The same conclusion was drawn from studies in S. cerevisiae. Studying mixed segregation in the single-division meiosis mutants cdc5, cdc14, and spo13 revealed that different chromosomes can have different segregational tendencies even within the same cell. For example, chromosome IV tends to segregate equationally, while chromosome XI tends to segregate reductionally in these mutants (SHARON and SIMCHEN 1990B ; HUGERAT and SIMCHEN 1993 ). The chromosomal determinant for segregational tendencies resides within the centromeric region as shown by centromere replacement experiments (SHARON and SIMCHEN 1990A ). When the centromere of chromosome XI was replaced with the centromere of chromosome IV, chromosome XI tended to segregate equationally like chromosome IV. However, the molecular mechanism that the centromere region uses to control whether chromosomes segregate reductionally or equationally is still unclear.

Centromeres perform multiple functions related to chromosome segregation during cell division. The most well characterized is attachment of the chromosome to the microtubules of the spindle, but centromeres also play less-well-defined roles in cell cycle control, spindle structure/assembly, and sister chromatid cohesion (reviewed in MAGUIRE 1990 ; BICKEL and ORR-WEAVER 1996 ; STRAIGHT 1997 ). During meiosis the cohesion between sister chromatids is lost at two steps. The first step happens at the metaphase I/anaphase I transition, during which the cohesion between sister chromosomal arms disappears but the cohesion at the centromeric region is maintained and is necessary for the proper bipolar spindle assembly during meiosis II. The second step happens at the metaphase II/anaphase II transition. During this step the cohesion between sister centromeres is also lost in order to separate sister chromatids. In S. cerevisiae CEN-ARS plasmids missing centromere DNA element I (CDEI), or chromosomes in cells lacking CDEI binding protein Cep1p, often missegregate, predominantly due to precocious sister chromatid segregation during meiosis I (CUMBERLEDGE and CARBON 1987 ; MASISON and BAKER 1992 ). Recently, the CDEIII element has also been shown to be important for centromere cohesion in S. cerevisiae (MEGEE and KOSHLAND 1999 ). In Drosophila melanogaster, the kinetochore protein MEI-S332 is required to maintain cohesion between sister centromeres after metaphase I/anaphase I transition. In its absence, a high level of meiosis II nondisjunction was observed (DAVIS 1971 ; KERREBROCK et al. 1992 , KERREBROCK et al. 1995 ; MOORE et al. 1998 ). These observations indicate that the centromere and its associated proteins are essential for regulating the stepwise loss of sister chromatid cohesion during meiosis.

As centromeres are a critical part of how the cell distinguishes between the different meiotic divisions, it is expected that some centromere proteins will play a key role in the ordering of meiosis I and II. We find here that the centromere protein Slk19p is required for this task. SLK19 was identified previously in a synthetic lethal screen to identify genes with functions similar to that of the kinesin motor KAR3. It encodes a potential coiled-coil protein of ~98 kD and shares no significant homology to previously characterized proteins. Like kar3 mutants, slk19 mutants have short spindles and more cytoplasmic microtubules than normally present during mitosis (ZENG et al. 1999 ). Immunofluorescence studies have shown that slk19 kar3 double mutants are unable to assemble or maintain spindles, suggesting that Slk19p and Kar3p work together to stabilize spindle structure. Here we report that Slk19p is essential during meiosis. In its absence, a single meiotic division takes place. Genetic analysis reveals that chromosomes in slk19 mutants predominantly undergo equational but not reductional segregation. Recombination is normal or slightly elevated and may be required for the meiosis I bypass. If recombination is inhibited, the slk19 mutants are unable to undergo either meiotic division. Thus Slk19p appears to be an essential part of the mechanism at the centromere that helps to ensure the proper ordering of meiosis I and II.

	MATERIALS AND METHODS

TOP ABSTRACT MATERIALS AND METHODS RESULTS DISCUSSION LITERATURE CITED

Yeast strains, plasmids, and media:
Strains and plasmids used in this study are listed in Table 1. All strains are derivatives of S288C. Media have been described (SHERMAN 1991 ). For sporulation, diploid cells with plasmids were first grown in selective media. Otherwise they were grown in YEPD. Cells from fresh overnight cultures (OD₆₀₀ > 2) were washed with sporulation media twice, resuspended in sporulation medium, and incubated at 26° with shaking for 2 days to allow them to sporulate.

Nomenclature:
The segregation of heterozygous markers in the two-spored asci (dyads) is represented by the format +: -. The + indicates the number of spores with the phenotype of the dominant allele, while - indicates the number of spores with the phenotype of the recessive allele.

Intergenic recombination:
Diploid strains WSY1302 (slk19) and WSY1303 (wild type) were sporulated. The resulting tetrads were dissected as described (SHERMAN and HICKS 1991 ). To unambiguously distinguish dyads from two spores derived from different asci, dyads were dissected as described (DAVIDOW et al. 1980 ), except that 0.5 mg/ml zymolyase 20T (U.S. Biological) in 1 M sorbitol was used to replace glusulase. The segregation of ura3 and can1 in the resulting colonies was scored based on their growth on -ura and canavanine plates. Map distance was determined using PERKINS 1949 formula as follows: Map distance = 50 x [single crossovers + 6 (four-strand double crossovers)]/total. As shown in Fig 4, dyads resulting from double crossovers and equational division cannot be distinguished from dyads resulting from equational division without recombination. Therefore, we assumed that the same frequencies of four-strand double crossovers occur in reductional division and equational division and calculated the total number of four-strand double crossovers as follows: Total four-strand double crossovers = drd + drd (ed/rd), where drd stands for the number of dyads resulting from four-strand double crossover and reductional division, ed for the number of dyads resulting from equational division, and rd for the number of dyads resulting from reductional division (HUGERAT and SIMCHEN 1993 ).

View larger version (120K): In this window In a new window Download PPT slide   Figure 1. (A) Sporulation in wild-type and slk19 mutants. Wild-type and slk19 diploids were sporulated at 26° for 2 days and viewed by DIC microscopy. Sporulation in wild-type cells usually results in the formation of asci containing four spores (tetrads). slk19 mutants produced almost exclusively asci with only two spores (dyads). The sporulation efficiencies are shown in Table 2. (B) Canavanine resistance test to determine the ploidy of the spores. HP1 and HP2 signify the two slk19 haploid parents and DP the diploid parent resulting from the cross of HP1 and HP2. Dyads 1–6 are six dyads resulting from the sporulation of DP. The left spore of dyad 1 was scored as a haploid and the rest of the spores as diploids based on this test. All the spores grew well on YEPD plates made at the same time (not shown).

View larger version (79K): In this window In a new window Download PPT slide   Figure 2. Microtubules in slk19 cells during meiosis. Cells in sporulation medium for 20 hr were fixed and treated for antitubulin immunofluorescence (MATERIALS AND METHODS). Spindles are visible as a bright fluorescent bar. Cytoplasmic MTs are visible as fainter lines at ends of spindles. Shown are four composite images of representative cells from the same samples. ben refers to sporulation in the presence of 5 µg/ml benomyl. The level of sporulation in these cultures is fairly high at ~50% but the synchrony is low and only ~20% of cells showed spindles at this timepoint.

View larger version (32K): In this window In a new window Download PPT slide   Figure 3. Schematic diagram showing the expected segregation patterns of heterozygous centromere-linked markers in dyads resulting from abnormal meiotic events. The meiotic products are shown in the squares with each oval representing one spore. The markers are indicated on the chromosome. + indicates the dominant allele and - indicates the recessive allele. No recombination events are shown. (A) Normal segregation in wild-type or single-division meiosis. (B) Two aberrant segregational events that could also result in 2:0 dyads or 1:1 dyads.

View larger version (49K): In this window In a new window Download PPT slide   Figure 4. Recombination and segregational events that produce the various types of dyads shown in Table 5. + represents the dominant allele while - represents the recessive allele. SCO stands for single crossing over, DCO for double crossing over, equational for equational segregation, and reductional for reductional segregation. (*)These dyads are phenotypically indistinguishable.

Heteroallelic recombination:
To determine the frequencies of heteroallelic recombination, diploid strains of wild-type (WSY1118) or slk19 (WSY1117) with ade2 heteroalleles (ade2-1/ade2-R) were constructed. Individual diploid colonies were inoculated into YEPD liquid and cultured at 30° with shaking until OD₆₀₀ reached 2.0. The cultures were then divided into two parts. One part was used directly to determine the mitotic recombination rates by spreading onto YEPD and -ade selective plates and counting the frequency of adenine prototrophy. The other part was washed with sporulation medium twice and resuspended in sporulation medium to induce meiosis and sporulation. The sporulation culture was vigorously aerated at 26° for 48 hr, at which time the cells were plated onto YEPD and -ade plates to determine the meiotic recombination frequencies.

Canavanine reversion:
The ploidy of the spores was determined based on the frequencies of canavanine-resistant colonies after UV treatment. Diploid cells homozygous for slk19 and CAN1 were sporulated at 26° and the resulting dyads were dissected. Resulting individual colonies were serial diluted onto YEPD plates and incubated at 30° for 2 days. Cells were then replicated from YEPD plates onto canavanine plates and treated with UV light immediately after replica plating. The UV light was delivered from a UV crosslinker (FB-UVXL-1000, Fisher Scientific) with 9000 µJ/cm² energy. The UV-treated cells were then incubated at 30° for 2 days to allow canavanine-resistant colonies to grow.

Immunofluorescence microscopy:
To visualize microtubules (MTs) of cells from different meiotic stages, cells were cultured in the sporulation media for ~20 hr and fixed with 3.7% formaldehyde for 2 hr. Antitubulin indirect immunofluorescence was then performed using monoclonal antibody YOL 1/34 (Serotec, Oxford, UK) and rhodamine-conjugated secondary antibody (Jackson Immunoresearch, West Grove, PA) to visualize MT, and DNA-specific fluorescent dye 4,6-diamidino-2-phenylindole (DAPI; Sigma, St. Louis) to identify the position of the nucleus as described (PRINGLE et al. 1991 ; HOYT et al. 1992 ). The slides were examined with an Olympus BX60 epifluorescence microscope using a 100x oil immersion objective. Digital images were captured with a Hamamatsu Argus-20 CCD camera and image processor (Hamamatsu Co., Bridgewater, NJ). Fig 2 shows montages made by combining several representative samples using Adobe Photoshop software (Adobe Systems Inc., Mountain View, CA).

	RESULTS

TOP ABSTRACT MATERIALS AND METHODS RESULTS DISCUSSION LITERATURE CITED

Slk19p is required to obtain two successive divisions during meiosis:
To study the function of Slk19p during meiosis, a diploid strain homozygous for slk19 (ZENG et al. 1999 ) was sporulated. We found that while sporulation of wild-type cells usually led to the formation of tetrads with four spores enclosed in an ascus, sporulation of slk19 mutants usually led to the formation of dyads with two spores per ascus (Fig 1A). The sporulation efficiency of slk19 mutants is normal, with ~50% of the total cells forming asci compared to ~60% in wild-type cells. The frequency of dyads increased from ~11% of total asci for wild type to ~80% for slk19 mutants (see Table 4). Tetrads were rarely observed in slk19 mutants (<1%). The spore viability of the slk19 dyads is slightly lower than that of wild-type tetrads. While wild-type spores maintain >95% viability after dissection, only 70% of slk19 spores were alive among 239 dyads dissected.

View this table: In this window In a new window   Table 2. Segregation of heterozygous centromere-linked markers in slk19 dyads

View this table: In this window In a new window   Table 3. Coincident segregation of centromere-linked markers in slk19 dyads

Two types of dyads have been observed previously in S. cerevisiae. The first type is due to an ascospore packaging defect. Under these circumstances, two successive divisions take place but only two of the four nuclei are enclosed into spores (DAVIDOW et al. 1980 ). In this case, the spores are haploid. The second type of dyad is the result of a single meiotic division. Here two-spored asci are also seen but the spores are diploid. To investigate whether dyads of slk19 mutants were the result of an ascospore packaging defect or a single-division meiosis, we determined the ploidy of the spores. Typically only diploid cells can sporulate, and only MATa/MAT and not MATa/MATa or MAT/MAT diploids can sporulate. Dyads of slk19 mutants were dissected (MATERIALS AND METHODS) and the mating capability of viable spores was tested. Among 42 dyads with two viable spores, 13 dyads contained two spores capable of mating and 29 dyads contained two nonmating spores. Dyads with a mixture of mating and nonmating spores were not observed. The nonmating spores were able to sporulate and again produced dyads which themselves were able to sporulate to dyads (six spores from 3 dyads were tested). These observations indicate that stable diploid progenies are produced by meiosis in slk19 mutants. The ploidy of the spores was confirmed by determining the copy number of chromosome V using a canavanine mutation test. Haploid CAN1 cells have a much higher rate of mutation to canavanine resistance than do diploid CAN1/CAN1 cells. Only three spores out of the 42 dyads contained a single copy of CAN1 based on this analysis. The remaining were concluded to have at least two copies of CAN1 and thus two copies of chromosome V (Fig 1B).

The diploid spores are presumably a result of a single meiotic division. To confirm this, we also examined slk19 mutants in meiosis with antitubulin indirect immunofluorescence. Only one spindle per cell was observed, consistent with the occurrence of a single meiotic division (Fig 2). In conclusion, slk19 mutants form dyads after sporulation because there is a single meiotic division and not because there is an ascospore packaging defect.

The majority of chromosomes in slk19 mutants undergo an equational segregation:
The single meiotic division could be meiosis I, or meiosis II, or a mixture of the two. To address which kind of division was taking place, we studied the segregation pattern of three heterozygous centromere-linked markers. Centromere-linked markers remain associated with centromeres during meiosis and thus segregate with the same reductional and equational properties as chromosomes. For these experiments the markers chosen were trp1 linked to CEN4, ade1 linked to CEN1, and ura3 linked to CEN5. If chromosomes segregate reductionally (meiosis I), we expect to get 1:1 dyads in which one spore is able to grow on selective media, but the other spore cannot (Fig 3). If chromosomes segregate equationally (meiosis II), we expect to get 2:0 dyads in which both spores are able to grow on selective media. The majority of dyads in slk19 mutants were 2:0 dyads for each chosen marker, suggesting that most chromosomes segregate equationally (Table 2). As shown in Fig 3, Fig 2:0 dyads may result from a single normal equational division (Fig 3A), or a single equational segregation accompanied by chromosome loss, or incoordinate segregation (Fig 3B). But only 2:0 dyads resulting from a normal equational division will have both spores heterozygous for the chosen marker. To address whether the 2:0 dyads of slk19 mutants were a result of aberrant chromosome segregation during meiosis, we determined whether these slk19 spores were heterozygous for the tested marker. We transformed a plasmid containing wild-type SLK19 (pXZB9) into four spores from two 2:0 dyads for trp1 marker and allowed them to sporulate. Seven resulting tetrads for each spore were dissected (see MATERIALS AND METHODS) and the trp1 marker segregated 2:2 for each, indicating that those spores were heterozygous for trp1 and thus resulted from a normal equational division and not from aberrant chromosome segregation.

While most chromosomes segregate equationally, we only observed ~68% of the dyads showing 2:0 phenotype for all three markers (Table 3), indicating that at most 68% of the cells undergo a single equational division of the entire genome. The rest of the dyads showed a 1:1 phenotype for at least one of the three markers. The 1:1 phenotype may be a result of a single reductional segregation, or an aberrant segregation, or a rare meiotic recombination between centromere and linked marker followed by a single equational segregation (Fig 3). The expected number of 1:1 dyads resulting from meiotic recombination followed by a single equational division was calculated based on the distance between the marker locus and the centromere (Table 2). The observed frequency was much higher than expected based on recombination alone, suggesting that most of these 1:1 dyads were not due to recombination, but more likely from a single reductional segregation or an aberrant segregation event. The cells rarely go through reductional division for the entire genome, as we observed only 4.6% of dyads showing 1:1 for all three markers (Table 3). The rest of the dyads showing 1:1 phenotypes for at least one of the markers were probably from a mixed division with some chromosomes segregating equationally and some chromosomes segregating reductionally. The chromosomes did not behave equivalently. For example, chromosome I has a higher tendency than chromosome V to segregate reductionally as evidenced by the observation that more 1:1 dyads were observed for the ade1 marker than for the ura3 marker, even though ade1 is closer to the centromere than ura3.

In conclusion, chromosomes predominantly segregate equationally during meiosis upon loss of Slk19p. Consistent with the previous finding that different chromosomes may have different segregational properties within the same cell, we have also observed different segregational tendencies for different chromosomes.

An excess of cytoplasmic microtubules is not the direct cause of the meiosis I bypass:
Immunofluorescence with antitubulin antibodies revealed that like vegetatively growing slk19 cells, meiotic slk19 cells have an increased amount of cytoplasmic microtubules (Fig 2). These results indicate that Slk19p also influences microtubule numbers and distribution in meiosis. This excess is not an inevitable result of a single equational division, since we did not observe this phenotype in spo13 mutants, which are known to have a single equational division (Fig 2; KLAPHOLZ and ESPOSITO 1980B ). Among 200 spindles examined, the average number of cytoplasmic microtubules was three per spindle for slk19 mutants and one per spindle for spo13 mutants. MOENS and RAPPORT 1971 have shown using electron microscopy that yeast meiotic cells lose most of their cytoplasmic microtubules when they enter meiosis I. The functional significance of this loss is not clear, but it may be an important part of meiotic progression. To test whether the excess of cytoplasmic microtubules is the direct cause of the meiosis I bypass, we let slk19 mutants sporulate in the presence of benomyl, a microtubule depolymerization drug (DAVIDSE 1986 ). The amount of cytoplasmic microtubules was substantially reduced, as shown by immunofluorescence with antitubulin antibody (Fig 2, a montage of representative cells from the same sample is shown), but we found no increase in the frequency of tetrads (Table 4). Cytoplasmic microtubules can also be reduced by genetic deletion of the nonessential -tubulin TUB3 gene (SCHATZ et al. 1986 ) or the kinesin-related KIP2 gene (HUYETT et al. 1998 ). slk19tub3 and slk19kip2 mutants also showed no increase in the frequency of tetrad formation compared to slk19 alone (not shown). These results indicate that the excess of cytoplasmic MTs is not likely to be the direct cause of the meiosis I bypass in slk19 mutants.

slk19 mutants exhibit normal or slightly elevated meiotic recombination rates:
Proper meiosis I segregation depends on a stable bipolar spindle, which in turn relies on a stable attachment between homologous chromosomes. Recombination is responsible for most of the association between homologues. In the absence of recombination, a high level of meiosis I nondisjunction is observed (JOHN 1990 ). To test whether slk19 mutants have normal levels of meiotic recombination, we examined two types of recombination, intergenic recombination and heteroallelic recombination. For intergenic recombination, wild type and slk19 diploid strains heterozygous for ura3 and can1 were sporulated and the resulting tetrads (wild type) or dyads (slk19) were dissected as described (MATERIALS AND METHODS) and scored for the segregation of markers (Table 5; the meiotic events producing various types of dyads are shown in Fig 4). The genetic distance between can1 and ura3 in the slk19 background was calculated to be 64.2 cM, which is higher than the value of 37.3 cM observed in our wild-type background, or the 43 cM registered with the Saccharomyces Genome Database, suggesting that slk19 diploids might have an increased frequency of recombination (see MATERIALS AND METHODS for a description of the calculation of map distance). To further examine recombination frequencies, we measured the intragenic recombination between two ade2 alleles, ade2-1 and ade2-R. Recombination between these heteroalleles is mostly due to gene conversion (LICHTEN et al. 1987 ) and will generate a wild-type ADE2 allele, resulting in adenine prototrophy. For this analysis we measured both mitotic and meiotic recombination by determining the frequency of Ade⁺ colonies following growth on YEPD and sporulation media. The number of Ade⁺ colonies varied widely for different cultures following vegetative growth, most likely due to differences in the time of occurrence of the Ade⁺ recombinants (Table 6). However, in all cases the frequency of mitotic recombination was <3% of the meiotic frequency. The frequency of meiotic recombination was similar in wild-type and slk19 diploids, with slightly elevated levels observed for the slk19 diploids (Table 6).

View this table: In this window In a new window   Table 5. Intergenic recombination is elevated in slk19 strains

View this table: In this window In a new window   Table 6. Mitotic and meiotic heteroallelic recombination in wild-type and slk19 mutants

Loss of SPO11 cannot rescue the meiotic bypass of slk19 mutants:
If the recombination attachment between homologues cannot be resolved at the metaphase I/anaphase I transition, the intact bivalent (the attached meiotic homologues) could segregate to one pole during meiosis I. This may result in disomic spores and genetically appear as a meiosis I bypass. Since recombinant dyads were observed, the recombination attachment must eventually be resolved, but this may not occur until the second meiotic division, with loss of the cohesion between the arms of sister chromatids, which has been proposed as a mechanism to maintain chiasmata (MOORE and ORR-WEAVER 1998 ). To examine the influence of recombination on the slk19 meiotic phenotype, we measured sporulation in slk19spo11 double mutants. Spo11p is the catalytic subunit of the double-strand break-associated protein (KEENEY et al. 1997 ). spo11 mutants have been shown previously to be able to bypass the requirement of recombination for meiotic nuclear division (KLAPHOLZ et al. 1985 ; BISHOP et al. 1992 ). In its absence, cells skip recombination and undergo two successive divisions with low spore viability. Interestingly, we found that slk19spo11 double mutants failed to perform either meiotic division. No nuclear division was visible by DAPI staining and no spores were observed by differential interference contrast (DIC) microscopy. Therefore, lack of recombination is apparently unable to rescue the meiosis I bypass of slk19 mutants. Furthermore, this result implies that at least the beginning step of recombination is actually required for slk19 mutants to skip meiosis I or to undergo meiosis II.

slk19 mutants utilize a different mechanism from spo13 mutants to bypass meiosis I:
The meiotic phenotype of slk19 mutants is similar to that of spo13 mutants. Both are competent for meiotic recombination but skip meiosis I (KLAPHOLZ and ESPOSITO 1980B ). But unlike slk19 spo11 double mutants, spo11 spo13 double mutants are able to sporulate and produce dyads with viable spores (KLAPHOLZ et al. 1985 ). Spo13p has been proposed to act as a regulatory protein to slow down progression into meiosis I in order to allow sufficient time to assemble the meiosis I spindle apparatus (MCCARROLL and ESPOSITO 1994 ). This conclusion is supported by the observation that the meiosis I bypass in spo13 mutants can be partially rescued by delaying meiotic progression, for example, by sporulating at low temperature or in the presence of sublethal concentration of hydroxyurea (HU). HU is a DNA synthesis inhibitor (TIMSON 1975 ) and has been shown to be able to delay the cell cycle through the RAD9-dependent pathway (WEINERT and HARTWELL 1988 ). The different meiotic phenotypes of slk19 spo11 and spo11 spo13 diploids suggest that slk19 mutants utilize a different mechanism from spo13 to bypass meiosis I. To confirm this, we examined sporulation of slk19 mutants at 20° or in the presence of 5 mM HU. The meiotic defect of slk19 mutants could not be improved by these treatments (Table 7 and data not shown). This result further suggests that slk19 mutants bypass meiosis I under different circumstances than spo13 mutants.

View this table: In this window In a new window   Table 7. Effect of hydroxyurea on the sporulation of slk19 and spo13 mutants

slk19 mutant diploids begin the first meiotic division and spore morphogenesis at the same time as the wild-type cells:
Since the slk19 mutant diploids undergo only a single meiotic division, they might be expected to sporulate more rapidly than wild-type cells, which must divide twice prior to spore wall formation. Alternatively, if slk19 mutants attempt both meiotic divisions or if the timing of sporulation is independent of meiotic division, the rates of sporulation may be similar. To see which case is true for slk19 mutants, isogenic slk19 and wild-type diploid cells were sporulated and samples at various timepoints were stained with DAPI to visualize the DNA. The number of binucleate and tetranucleate cells and asci for each sample were counted. Interestingly, we found that slk19 mutants started their meiotic division at nearly the exact time as the first meiotic division of wild-type cells (Fig 5A). The second meiotic division in wild-type cells occurred ~3 hr later than the first meiotic division. Surprisingly, we also found that although only a single meiotic division occurred in slk19 mutants, both wild-type and slk19 mutant cells form asci at the same rate (Fig 5B). This can be explained by two possible mechanisms. First there is a delay between the completion of meiotic division and the spore wall formation in slk19 mutants. Alternatively, there may be an internal clock that regulates the timing of spore wall formation that is independent of meiotic division.

View larger version (20K): In this window In a new window Download PPT slide   Figure 5. Timing of the sporulation events in wild-type and slk19 mutant diploids. Isogenic wild-type and slk19 mutant diploids were sporulated at 26°. Samples were taken throughout sporulation and the percentage of bi- and tetranucleate cells (by DAPI staining) and asci was determined. A total of 400 cells were counted for each sample. (A) Percentage of binucleate and tetranucleate cells vs. time in sporulation. (B) Percentage of asci vs. time in sporulation.

	DISCUSSION

TOP ABSTRACT MATERIALS AND METHODS RESULTS DISCUSSION LITERATURE CITED

We have identified a gene in S. cerevisiae that is required if the diploid cell is to complete two successive meiotic divisions. When mutants homozygous for slk19 sporulate, they produce mostly two-spore dyads. The dyads are a result of a single equational division as demonstrated by the following observations: (1) Nonmating spores of the dyads can sporulate, indicating that they are diploid. (2) Mutational frequencies indicate spores of the dyads contain at least two copies of chromosome V on which the CAN1 gene is located, consistent with being diploid. (3) Antitubulin indirect immunofluorescence revealed only a single spindle per cell during meiosis. (4) The segregation pattern of centromere-linked markers indicates that most chromosomes segregate equationally but not reductionally. Therefore Slk19p is essential for the reductional division from a diploid to a haploid genome. Recombination in slk19 mutants was normal or slightly elevated over wild-type cells, consistent with previous observations that meiotic recombination is not necessarily associated with reductional division and that equational division does not rely on prior reductional division (KLAPHOLZ and ESPOSITO 1980B ). These observations confirm the independent and nonlinear organization of the meiotic pathway.

The meiotic phenotype of slk19 mutants is similar to that of the previously described spo12 and spo13 mutants (KLAPHOLZ and ESPOSITO 1980B ). Each has a single equational division meiosis and normal or elevated levels of meiotic recombination. Spo13p has been proposed to act as a timing regulatory factor to slow down the meiotic progression to allow sufficient time for assembly of the meiosis I spindle apparatus (MCCARROLL and ESPOSITO 1994 ). But unlike spo13 mutation, slk19 mutation cannot allow spo11 mutants to form viable spores, and the meiotic defect of slk19 mutants cannot be improved by the addition of HU or by lowering the sporulating temperature. These results suggest that the meiosis I bypass in slk19 mutants is fundamentally different from that of spo13 mutants. slk19 mutants are more similar to spo12 mutants in that both require recombination for their single meiotic division. Both slk19 spo11 double mutants and spo12 spo11 double mutants are unable to sporulate (ESPOSITO and KLAPHOLZ 1981 ). But the underlying mechanism for the meiotic defect in spo12 mutants is also unclear.

We propose two models to explain the slk19 meiotic phenotype. The first is that Slk19p may be uniquely required for spindle stability in meiosis I. Thus Slk19p has a similar role in meiosis I as mitosis, but becomes essential in meiosis I. Slk19p could become essential due to a difference in the nature of the kinetochore-microtubule attachment. In meiosis I each sister kinetochore pair is attached to only a single spindle pole. Bipolar attachment is important to generate tension and stabilize the microtubule-kinetochore association (reviewed in NICKLAS 1997 ). Slk19p has also been proposed to stabilize kinetochore microtubules in mitosis (ZENG et al. 1999 ) and in meiosis I, without the benefit of the tension associated with bipolar attachment, this activity may become critical. However, the apparent low level of chromosome loss in viable slk19 spores may argue against this model. Slk19p could also become essential due to a change in activity of Kar3p. Previous work has shown that Slk19p and Kar3p work together during mitosis to assemble and maintain spindles (ZENG et al. 1999 ). During meiosis I Kar3p appears to be required for homologous pairing (BASCOM-SLACK and DAWSON 1997 ), an event not observed in mitotic division or meiosis II. If Kar3p becomes diverted to other tasks during meiosis I, Slk19p could be expected to become essential for spindle stability at this division.

A second model proposes that what appears to be a single meiosis II division in slk19 mutants is in fact an aberrant meiosis I. This model can explain why slk19 mutants start their single meiotic division at the same time as the first meiotic division of wild-type cells. The delay between the completion of meiotic division and the spore wall formation in slk19 mutants could result from the failed attempt at a second meiotic division in the slk19 mutants. Chromosomes may segregate equationally during meiosis I if for example, Slk19p is required to hold the sister centromeres together in meiosis I. In slk19 mutants the sister centromeres may separate prematurely in meiosis I, giving the appearance of an equational division. The sister chromatids may also missegregate equationally during meiosis I if Slk19p is required to regulate the timing of sister kinetochore differentiation. Substantial experimental evidence suggests that sister kinetochores undergo functional (GOLDSTEIN 1981 ; RUFAS et al. 1989 ) and spatial differentiation during meiosis I (reviewed in NICKLAS 1997 ), which limits their ability to bind to both poles. If the timing of sister kinetochore differentiation depends on Slk19p then mutational loss could lead to precocious sister kinetochore differentiation. As a result, sister chromatids may bind microtubules from opposite poles and segregate equationally during meiosis I.

We may be able to distinguish between these two models by electron microscopy. The spindle pole bodies during meiosis I and mitosis have a similar structure, while the meiosis II spindle pole body has an outer plaque that is larger and denser (MOENS and RAPPORT 1971 ; PETERSON et al. 1972 ). This observation was used by Moens and co-workers to determine that ATCC4117 yeast cells, from which the original spo12-1 and spo13-1 mutations were isolated (KLAPHOLZ and ESPOSITO 1980A ), initiated some aspect of meiosis I even though only equational division is observed genetically (MOENS 1974 ; MOENS et al. 1977 ). Similarly, the recent advent of green fluorescent protein-labeled centromeres may allow us to determine if centromeres divide prematurely in slk19 mutants (KLEIN et al. 1999 ).

Although the underlying mechanism for the single division meiosis of slk19 mutants is not clear, the observation that loss of SLK19 leads to a single equational division is the first identification of a centromere protein that influences the choice between equational and reductional division in meiosis.

	ACKNOWLEDGMENTS

The authors thank Drs. Giora Simchen of the Hebrew University of Jerusalem, Israel, for the spo13 mutants and ade2 heteroalleles and Rochelle Esposito of the University of Chicago for the spo11 mutants. The work was funded through American Cancer Society award CB-171 to W.S.S.

Manuscript received August 5, 1999; Accepted for publication February 18, 2000.

	LITERATURE CITED

TOP ABSTRACT MATERIALS AND METHODS RESULTS DISCUSSION LITERATURE CITED

BASCOM-SLACK, C. A. and D. S. DAWSON, 1997  The yeast motor protein, Kar3p, is essential for meiosis I. J. Cell Biol. 139:459-467[Abstract/Free Full Text].

BICKEL, S. E. and T. L. ORR-WEAVER, 1996  Holding chromatids together to ensure they go their separate ways. Bioessays 18:293-300[Medline].

BISHOP, D. K., D. PARK, L. XU, and N. KLECKNER, 1992  DMC1: a meiosis-specific yeast homolog of E. coli recA required for recombination, synaptonemal complex formation, and cell cycle progression. Cell 69:439-456[Medline].

CUMBERLEDGE, S. and J. CARBON, 1987  Mutational analysis of meiotic and mitotic centromere function in Saccharomyces cerevisiae.. Genetics 117:203-212[Abstract/Free Full Text].

DAVIDOW, L. S., L. GOETSCH, and B. BYERS, 1980  Preferential occurrence of nonsister spores in two-spored asci of Saccharomyces cerevisiae: evidence for regulation of spore-wall formation by the spindle pole body. Genetics 94:581-595[Abstract/Free Full Text].

DAVIDSE, L. C., 1986  Benzimadazole fungicides: mechanism of action and biological impact. Annu. Rev. Phytopathol. 24:43-65.

DAVIS, B. K., 1971  Genetic analysis of a meiotic mutant resulting in precocious sister-centromere separation in Drosophila melanogaster.. Mol. Gen. Genet. 113:251-272[Medline].

ESPOSITO, R. E., and S. KLAPHOLZ, 1981 Molecular Biology of the Yeast Saccharomyces, Life Cycle and Inheritance. Cold Spring Harbor Laboratory Press, Cold Spring Harbor, NY.

GOLDSTEIN, L. S., 1981  Kinetochore structure and its role in chromosome orientation during the first meiotic division in male D. melanogaster.. Cell 25:591-602[Medline].

HOYT, M. A., L. HE, K. K. LOO, and W. S. SAUNDERS, 1992  Two S. cerevisiae kinesin-related gene-products required for mitotic spindle assembly. J. Cell Biol. 118:109-120[Abstract/Free Full Text].

HUGERAT, Y. and G. SIMCHEN, 1993  Mixed segregation and recombination of chromosomes and YACs during single-division meiosis in spo13 strains of Saccharomyces cerevisiae.. Genetics 135:297-308[Abstract].

HUYETT, A., J. KAHANA, P. SILVER, X. ZENG, and W. S. SAUNDERS, 1998  The Kar3p and Kip2p motors function antagonistically at the spindle poles to influence cytoplasmic microtubule numbers. J. Cell Sci. 111:295-301[Abstract].

JOHN, B., 1990 Meiosis. Cambridge University Press, Cambridge, UK.

KEENEY, S., C. N. GIROUX, and N. KLECKNER, 1997  Meiosis-specific DNA double-strand breaks are catalyzed by Spo11, a member of a widely conserved protein family. Cell 88:375-384[Medline].

KERREBROCK, A. W., W. Y. MIYAZAKI, D. BIRNBY, and T. L. ORR-WEAVER, 1992  The Drosophila mei-S332 gene promotes sister-chromatid cohesion in meiosis following kinetochore differentiation. Genetics 130:827-841[Abstract].

KERREBROCK, A. W., D. P. MOORE, J. S. WU, and T. L. ORR-WEAVER, 1995  Mei-S332, a Drosophila protein required for sister-chromatid cohesion, can localize to meiotic centromere regions. Cell 83:247-256[Medline].

KLAPHOLZ, S. and R. E. ESPOSITO, 1980a  Isolation of SPO12-1 and SPO13-1 from a natural variant of yeast that undergoes a single meiotic division. Genetics 96:567-588[Abstract/Free Full Text].

KLAPHOLZ, S. and R. E. ESPOSITO, 1980b  Recombination and chromosome segregation during the single division meiosis in SPO12-1 and SPO13-1 diploids. Genetics 96:589-611[Abstract/Free Full Text].

KLAPHOLZ, S., C. S. WADDELL, and R. E. ESPOSITO, 1985  The role of the SPO11 gene in meiotic recombination in yeast. Genetics 110:187-216[Abstract/Free Full Text].

KLEIN, F., P. MAHR, M. GALOVA, S. B. BUONOMO, and C. MICHAELIS et al., 1999  A central role for cohesins in sister chromatid cohesion, formation of axial elements, and recombination during yeast meiosis. Cell 98:91-103[Medline].

LICHTEN, M., R. H. BORTS, and J. E. HABER, 1987  Meiotic gene conversion and crossing over between dispersed homologous sequences occurs frequently in Saccharomyces cerevisiae.. Genetics 115:233-246[Abstract/Free Full Text].

MAGUIRE, M. P., 1990  Sister chromatid cohesiveness: vital function, obscure mechanism. Biochem. Cell Biol. 68:1231-1242[Medline].

MASISON, D. C. and R. E. BAKER, 1992  Meiosis in Saccharomyces cerevisiae mutants lacking the centromere-binding protein CP1. Genetics 131:43-53[Abstract].

MCCARROLL, R. M. and R. E. ESPOSITO, 1994  SPO13 negatively regulates the progression of mitotic and meiotic nuclear division in Saccharomyces cerevisiae.. Genetics 138:47-60[Abstract].

MCKIM, K. S. and R. S. HAWLEY, 1995  Chromosomal control of meiotic cell division. Science 8:1595-1601.

MEGEE, P. C. and D. KOSHLAND, 1999  A functional assay for centromere-associated sister chromatid cohesion. Science 285:254-257[Abstract/Free Full Text].

MOENS, P. B., 1974  Modification of sporulation in yeast strains with two-spored asci (Saccharomyces, Ascomycetes). J. Cell Sci. 16:519-527[Abstract/Free Full Text].

MOENS, P. B. and E. RAPPORT, 1971  Spindles, spindle plaques, and meiosis in the yeast Saccharomyces (Hansen). J. Cell Biol. 50:344-361[Abstract/Free Full Text].

MOENS, P. B., M. MOWAT, M. S. ESPOSITO, and R. E. ESPOSITO, 1977  Meiosis in a temperature-sensitive DNA-synthesis mutant and in an apomictic yeast strain (Saccharomyces cerevisiae). Philos. Trans. R. Soc. Lond. B Biol. Sci. 277:351-358[Medline].

MOORE, D. P. and T. L. ORR-WEAVER, 1998  Chromosome segregation during meiosis: building an unambivalent bivalent. Curr. Top. Dev. Biol. 37:263-299[Medline].

MOORE, D. P., A. W. PAGE, T. T. TANG, A. W. KERREBROCK, and T. L. ORR-WEAVER, 1998  The cohesion protein MEI-S332 localizes to condensed meiotic and mitotic centromeres until sister chromatids separate. J. Cell Biol. 140:1003-1012[Abstract/Free Full Text].

NICKLAS, R. B., 1977  Chromosome distribution: experiments on cell hybrids and in vitro. Philos. Trans. R. Soc. Lond. B Biol. Sci. 277:267-276[Medline].

NICKLAS, R. B., 1997  How cells get the right chromosomes. Science 275:632-637[Abstract/Free Full Text].

PERKINS, D. D., 1949  Biochemical mutants in the smut fungus Ustilago maydis. Genetics 34:607-626[Free Full Text].

PETERSON, J. B., R. H. GRAY, and H. RIS, 1972  Meiotic spindle plaques in Saccharomyces cerevisiae.. J. Cell Biol. 53:837-841[Free Full Text].

PRINGLE, J. R., A. E. M. ADAMS, D. G. DRUBIN, and B. K. HAARER, 1991  Immunofluorescence methods for yeast. Methods Enzymol. 194:565-602[Medline].

RUFAS, J. S., C. MAZZELLA, J. A. SUJA, and C. G. DE LA VEGA, 1989  Kinetochore structures are duplicated prior to the first meiotic metaphase. A model of meiotic behavior of kinetochores in grasshoppers. Eur. J. Cell Biol. 48:220-226.

SCHATZ, P. J., L. PILLUS, P. GRISAFI, F. SOLOMON, and D. BOTSTEIN, 1986  Two functional alpha-tubulin genes of the yeast Saccharomyces encode divergent proteins. Mol. Cell. Biol. 6:3711-3721[Abstract/Free Full Text].

SHARON, G. and G. SIMCHEN, 1990a  Centromeric regions control autonomous segregation tendencies in single-division meiosis of Saccharomyces cerevisiae.. Genetics 125:487-494[Abstract].

SHARON, G. and G. SIMCHEN, 1990b  Mixed segregation of chromosomes during single-division meiosis of Saccharomyces cerevisiae.. Genetics 125:475-485[Abstract].

SHERMAN, F., 1991  Getting started with yeast. Methods Enzymol. 194:3-21[Medline].

SHERMAN, F. and J. HICKS, 1991  Micromanipulation and dissection of asci. Methods Enzymol. 194:21-37[Medline].

SHUSTER, E. O. and B. BYERS, 1989  Pachytene arrest and other meiotic effects of the start mutations in Saccharomyces cerevisiae.. Genetics 123:29-43[Abstract/Free Full Text].

SIMCHEN, G. and Y. HUGERAT, 1993  What determines whether chromosomes segregate reductionally or equationally in meiosis? BioEssays 15:1-8[Medline].

STRAIGHT, A. F., 1997  Cell cycle: checkpoint proteins and kinetochores. Curr. Biol. 7:R613-R616[Medline].

TIMSON, J., 1975  Hydroxyurea. Mutat. Res. 32:115-132[Medline].

WEINERT, T. A. and L. H. HARTWELL, 1988  The RAD9 gene controls the cell cycle response to DNA damage in Saccharomyces cerevisiae.. Science 241:317-322[Abstract/Free Full Text].

ZENG, X., J. A. KAHANA, P. A. SILVER, M. K. MORPHEW, and J. R. MCINTOSH et al., 1999  Slk19p is a centromere protein that functions to stabilize mitotic spindles. J. Cell Biol. 146:415-426[Abstract/Free Full Text].

This article has been cited by other articles:

				D. D'Amours and A. Amon At the interface between signaling and executing anaphase--Cdc14 and the FEAR network Genes & Dev., November 1, 2004; 18(21): 2581 - 2595. [Abstract] [Full Text] [PDF]

				T. V. Reddy, J. Kaur, B. Agashe, V. Sundaresan, and I. Siddiqi The DUET gene is necessary for chromosome organization and progression during male meiosis in Arabidopsis and encodes a PHD finger protein Development, December 15, 2003; 130(24): 5975 - 5987. [Abstract] [Full Text] [PDF]

				B. Agashe, C. K. Prasad, and I. Siddiqi Identification and analysis of DYAD: a gene required for meiotic chromosome organisation and female meiotic progression in Arabidopsis Development, March 10, 2003; 129(16): 3935 - 3943. [Abstract] [Full Text] [PDF]

				M. A. Shonn, R. McCarroll, and A. W. Murray Spo13 protects meiotic cohesin at centromeres in meiosis I Genes & Dev., July 1, 2002; 16(13): 1659 - 1671. [Abstract] [Full Text] [PDF]

				R. M. Q. Shanks, R. J. Kamieniecki, and D. S. Dawson The Kar3-Interacting Protein Cik1p Plays a Critical Role in Passage Through Meiosis I in Saccharomyces cerevisiae Genetics, November 1, 2001; 159(3): 939 - 951. [Abstract] [Full Text] [PDF]

				R. Mercier, D. Vezon, E. Bullier, J. C. Motamayor, A. Sellier, F. Lefevre, G. Pelletier, and C. Horlow SWITCH1 (SWI1): a novel protein required for the establishment of sister chromatid cohesion and for bivalent formation at meiosis Genes & Dev., July 15, 2001; 15(14): 1859 - 1871. [Abstract] [Full Text] [PDF]