High-Resolution Genetic Mapping With Ordered Arrays of Saccharomyces cerevisiae Deletion Mutants

Corresponding author: Charles Boone, Toronto, ON M5G 1L6, Canada., charlie.boone{at}utoronto.ca (E-mail)

Communicating editor: M. JOHNSTON

	ABSTRACT

TOP ABSTRACT MATERIALS AND METHODS RESULTS AND DISCUSSION LITERATURE CITED

We present a method for high-resolution genetic mapping that takes advantage of the ordered set of viable gene deletion mutants, which form a set of colinear markers covering almost every centimorgan of the Saccharomyces cerevisiae genome, and of the synthetic genetic array (SGA) system, which automates the construction of double mutants formed by mating and meiotic recombination. The Cbk1 kinase signaling pathway, which consists minimally of CBK1, MOB2, KIC1, HYM1, and TAO3 (PAG1), controls polarized morphogenesis and activation of the Ace2 transcription factor. Deletion mutations in the Cbk1 pathway genes are tolerated differently by common laboratory strains of S. cerevisiae, being viable in the W303 background but dead in the S288C background. Genetic analysis indicated that the lethality of Cbk1 pathway deletions in the S288C background was suppressed by a single allele specific to the W303 background. SGA mapping (SGAM) was used to locate this W303-specific suppressor to the SSD1 locus, which contains a known polymorphism that appears to compromise SSD1 function. This procedure should map any mutation, dominant or recessive, whose phenotype is epistatic to wild type, that is, a phenotype that can be scored from a mixed population of cells obtained by germination of both mutant and wild-type spores. In principle, SGAM should be applicable to the analysis of multigenic traits. Large-scale construction of ordered mutations in other model organisms would broaden the application of this approach.

SYNTHETIC genetic array (SGA) analysis automates the systematic construction of double mutants in the budding yeast Saccharomyces cerevisiae, enabling a comprehensive and high-throughput analysis of genetic interactions (TONG et al. 2001 ). In this system, a marked query mutation is first crossed to the set of 5000 viable haploid gene deletion strains and then a series of robotic arraying procedures allows selected growth of double-mutant meiotic progeny. Because these haploid double mutants are formed by meiotic recombination and because the set of viable gene deletions represents a colinear series of mapping markers covering almost every centimorgan of the yeast genome, SGA analysis also provides a method for high-resolution genetic mapping. Here, we apply the SGA mapping (SGAM) method to identify a suppressor of the lethality associated with defects in the Cbk1 signaling pathway.

Budding yeast Cbk1 is a member of the Cot-1/Orb6/Ndr1/Warts family of serine-threonine protein kinases, which may have conserved roles in the control of cell polarity and morphogenesis (YARDEN et al. 1992 ; JUSTICE et al. 1995 ; MILLWARD et al. 1995 ; VERDE et al. 1998 ; DURRENBERGER and KRONSTAD 1999 ; ZALLEN et al. 2000 ). Cells deleted for CBK1 have a number of morphological abnormalities, including defects in bipolar budding, polarized morphogenesis, and the separation of mother and daughter cell walls following cytokinesis (DORLAND et al. 2000 ; RACKI et al. 2000 ; BIDLINGMAIER et al. 2001 ). The cell separation defect is due to an inability to activate the Ace2 transcription factor, which localizes specifically to the daughter cell nucleus in a Cbk1-dependent manner (RACKI et al. 2000 ; BIDLINGMAIER et al. 2001 ; COLMAN-LERNER et al. 2001 ). Ace2 activates a daughter-cell-specific transcription program including CTS1, which encodes a chitinase required for cell separation. A conserved signaling pathway encompassing Cbk1 has been uncovered by disparate genetic screens (SULLIVAN et al. 1998 ; DORLAND et al. 2000 ; RACKI et al. 2000 ; BIDLINGMAIER et al. 2001 ; COLMAN-LERNER et al. 2001 ; DU and NOVICK 2002 ; B. NELSON and C. BOONE, unpublished results). This pathway is composed minimally of Cbk1 and a set of four other highly conserved proteins, Mob2, Tao3 (Pag1), Hym1, and Kic1, a kinase whose catalytic region resembles that of PAK-like kinases (B. NELSON and C. BOONE, unpublished results). In the W303 genetic background, deletion mutants of CBK1, MOB2, TAO3, HYM1, and KIC1 are all viable but associated with defects in polarized morphogenesis and cell separation similar to those described above for cbk1. In the S288C genetic background, deletion mutants of CBK1, MOB2, TAO3, HYM1, and KIC1 are not viable. Genetic analysis revealed that an allele of an unknown gene functions as a W303-specific suppressor of the lethality caused by abrogation of the Cbk1 signaling pathway. We used the SGAM method to map the suppressor of the Cbk1 pathway to the SSD1 locus.

	MATERIALS AND METHODS

TOP ABSTRACT MATERIALS AND METHODS RESULTS AND DISCUSSION LITERATURE CITED

Yeast strains and media:
Yeast strains used in this study are listed in Table 1. S288C-derived strains are congenic with BY4741 (BRACHMANN et al. 1998 ), except at indicated loci. W303-derived strains are congenic with W3031A (HIRSCH et al. 1991 ), except at the indicated loci. YPD is 2% peptone, 1% yeast extract, 2% glucose, and 2% agar. Sporulation medium is 2% agar, 1% potassium acetate, 0.1% yeast extract, and 0.05% glucose, supplemented with uracil, histidine, and leucine. SC medium is 0.2% amino acid drop-out mix, 0.17% yeast nitrogen base without amino acids and ammonium sulfate, 0.5% ammonium sulfate, 2% glucose, and 2% agar. Filter-sterilized solutions of L-canavanine (50 mg/liter; Sigma, St. Louis), G418 (200 mg/liter; GIBCO-BRL, Gaithersburg, MD), and clonNAT (100 mg/liter; Werner Bioagents) were added to cooled (<50°), autoclaved media where indicated. In cases where SC medium was supplemented with clonNAT or G418, the ammonium sulfate was replaced with 0.1% monosodium glutamate and the medium termed SC/MSG.

SGA analysis:
SGA analysis was carried out as described (TONG et al. 2001 ) to identify viable gene deletions that show synthetic growth defects with deletions in Cbk1 pathway genes. Five starting strains for SGA analysis were constructed. These starting strains were to contain a marked deletion in a Cbk1 pathway gene, the W303-specific suppressor allele, as well as a mating-type reporter construct (can1::MFA1pr-HIS3::MF1pr-LEU2). First, the viable cbk1::URA3 (Y3576), hym1::URA3 (Y1560), mob2::natR (Y3732), kic1::URA3 (Y3400), or tao3::URA3 (Y3577) deletion mutants in the W303 background were crossed to Y3068 (S288C background, MAT can1::MFA1pr-HIS3 ura30 leu20 his31 lys20) and MATa can1::MFA1pr-HIS3lys20 progeny carrying the URA3- or natR-marked deletion alleles were isolated by tetrad dissection. As these isolates were viable, they contained the W303-specific suppressor of deletions in the Cbk1 pathway genes. The W303/S288C hybrids were then backcrossed to another S288C strain to diminish the number of W303-specific alleles. Specifically, the W303/S288C hybrids were mated to the S288C strain Y3655 (MAT can1::MFA1pr-HIS3::MF1pr-LEU2 ura30 leu20 his31 lys20), and viable MAT can1::MFA1pr-HIS3::MFlpr-LEU2 progeny carrying the URA3- or natR-marked deletions were isolated by tetrad analysis. The resulting strains were one-fourth W303 and three-fourths S288C hybrids and were used as the MAT starting strains for the subsequent SGA analysis.

For SGA analysis, the MAT starting strains, cbk1::URA3 (Y3717), hym1::URA3 (Y3716), mob2::natR (Y3968), kic1::URA3 (Y3715), or tao3::URA3 (Y3714) were mated to 4700 individual MATa xxx::kanR S288C haploid deletion strains. Yeast arrays were manipulated with a CPCA robot (TONG et al. 2001 ). Diploids were selected on SC-Ura-Lys plates or YPD plates supplemented with G418 and clonNAT for 1 day at 30° and then sporulated for 5 days at room temperature. To select for MATa can1::MFA1pr-HIS3::MF1pr-LEU2 meiotic progeny, spores were germinated on SC-His-Arg medium supplemented with L-canavanine for 2 days at 30° and then transferred onto a fresh plate of the same medium for another day of growth at 30°. The resulting colonies were then transferred to SC/MSG-His-Arg supplemented with L-canavanine and G418 to select for MATa xxx::kanR meiotic progeny. Finally, cells were transferred to either SC/MSG-His-Arg-Ura supplemented with L-canavanine and G418 or SC/MSG-His-Arg supplemented with L-canavanine, G418, and clonNAT, to select for MATa double-mutant meiotic progeny. The final progeny were one-eighth W303 and seven-eighths S288C hybrids.

MATa double-mutant progeny with growth defects were identified by scoring colony size by eye on a scale of 1–5, where 1 is no colony and 5 is slightly decreased colony size. Gene deletions (xxx::kanR) that consistently exhibited reduced colony size in wild-type control screens were eliminated from consideration. That is, finding genetic interactions with gene deletions (xxx::kanR) that cause strong growth defects on their own is beyond the sensitivity of the current SGA methodology. SGA analysis was carried out three times on each starting strain and a synthetic interaction was scored if a gene deletion was synthetic lethal/sick in one or more of the three screens. Raw scores from SGA analysis are provided as supplemental data (available at http://www.genetics.org/supplemental/). While these data are expected to contain a substantial number of false positives that would not be confirmed by tetrad analysis (TONG et al. 2001 ), we anticipated that raw SGA data would be sufficient for mapping purposes.

To confirm some of the synthetic genetic relationships between the Cbk1 pathway deletions and gene deletions unlinked to SSD1, 31 deletions that appeared synthetic lethal in five out of five screens were mated to the cbk1::URA3 (Y3576) and hym1::URA3 (Y1560) starting strains and subjected to tetrad analysis. Seventeen deletions that showed synthetic interactions in four out of five screens and had known or suspected roles in cell polarity or structure were also subjected to tetrad analysis with cbk1::URA3 and hym1::URA3. Spores were separated and germinated on YPD media. Nine out of the 48 (19%) interactions observed by SGA analysis that were not linked to SSD1/YDR293C could be confirmed by tetrad analysis. The relatively high false-positive rate may be due to the large number of loci (five) that must segregate to generate viable Nat^R (or Ura⁺) Kan^R meiotic progeny, the sporulation defect associated with abrogation of the Cbk1 signaling pathway (B. NELSON and C. BOONE, unpublished results), and/or the relatively low stringency used in scoring synthetic growth defects with each Cbk1 pathway gene deletion (see above; a synthetic interaction needed to be observed in only one-third of the SGA screens to be scored).

Algorithm for detecting significant groups of linked genetic interactions in SGA data:
A list of gene deletions that were synthetic lethal/sick with at least four of the five Cbk1 pathway deletions vs. a list of 4700 kanR-marked deletion mutants contained in our arrays was queried. First, the marginal probability (P_m) of a synthetic interaction was calculated from the total number of interactions divided by the total number of deletion strains screened. At each gene represented by a deletion strain, the number of synthetic interactions in the region surrounding that gene (five represented deletions in each direction) was calculated. The probability of observing this many interactions by chance alone was determined from a binomial distribution, assuming the probability of an interaction at any given gene is P_m. A Bonferroni correction was employed to compensate for the increased false-positive rate that results from performing multiple statistical tests. The Bonferroni correction increases the stringency of each significance test by dividing the P-value threshold by the number of tests performed.

Random spore analysis:
Twenty-one strains carrying kanR-marked deletions in genes surrounding SSD1, in SSD1 itself, and in four open reading frames (ORFs) lying at a great distance from SSD1 on chromosome IV were mated to the mob2::natR (Y3968) SGA starting strain. The heterozygotes were sporulated for 10+ days at room temperature and subjected to random spore analysis. Spores were released from asci by a 30-min incubation at 30° with 0.6 mg/ml zymolyase in 1 M sorbitol, followed by 1 min of vigorous vortexing. The separated spores were diluted in distilled water and spread on two SC/MSG-His-Arg plates supplemented with L-canavanine and clonNAT to select for the germination of MATa can1::MFA1pr-HIS3::MF1pr-LEU2 mob2::natR spores. Spores that germinate and form colonies must contain the W303-specific suppressor of mob2. Plates were incubated at 30° for 3 days. A total of 50–400 colonies were obtained per plate for each strain. Each plate was replicated onto SC/MSG-His-Arg supplemented with L-canavanine, clonNAT, and G418 and incubated for 2 days at 30° to determine what percentage of viable MATa mob2::natR spore colonies were Kan^R.

Sequencing of ssd1-d:
An 4-kb fragment containing the entire SSD1 open reading frame, as denoted in the Saccharomyces genome database (SGD), was amplified from genomic DNA derived from wild-type W3031A. PCR was performed with the Expand long template PCR system (Roche, Indianapolis) using the primers oMT1183/SSD1-UP (5' GTCACTTTAATATCGCAAAACAG) and oMT1182/SSD1-DOWN (5' GGATACTGAGGGGTGAAGC). Fifteen individual PCR reactions were pooled, the fragment purified with a Qiaquick PCR purification kit (QIAGEN, Chatsworth, CA), and the coding strand sequenced (ACGT Corporation). When compared to the SSD1 sequence at SGD, which was derived from a S288C strain carrying the SSD1-v allele (GOFFEAU et al. 1996 ), no differences were detected in base pairs 1–2093 of the open reading frame. A single base-pair change was detected at nucleotide 2094, a C (SGD) to G (W303) mutation that creates a premature stop codon. The truncated ssd1-d gene product has a predicted molecular weight of 77 kD. An identical mutation was found in the ssd1-d allele of a S288C-derived strain, JO371 (H. FRIESEN and B. ANDREWS, unpublished data).

Restoration of cbk1 lethality by SSD1-v:
cbk1::kanR (Y1748) and congenic wild-type (W3031A) strains in the W303 genetic background were transformed with plasmid P3765, a derivative of YEp24 (2µ, URA3) that houses a genomic fragment containing the CBK1 gene. P3765 rescues both the "crusty" colony and cell-clumping phenotypes of Y1748 (data not shown). Individual Ura⁺ transformants were subsequently transformed with either plasmid YEp13 (2µ, LEU2) or plasmid MTP251, a derivative of YEp13 that houses a 6-kb genomic fragment containing SSD1-v. Three Ura⁺ Leu⁺ transformants were picked and inoculated into SC-Ura-Leu and grown to saturation at 30°. Cultures were serially diluted (four 10-fold dilutions) in SC-Ura-Leu and then 5 µl of culture and each dilution were spotted onto SC-Leu and SC-Leu containing 0.1% 5-fluoro-orotic acid (5-FOA). Plates were photographed after 2 days at 30°.

	RESULTS AND DISCUSSION

TOP ABSTRACT MATERIALS AND METHODS RESULTS AND DISCUSSION LITERATURE CITED

Mapping a common suppressor of lethality associated with cbk1, mob2, tao3, hym1, and kic1 deletion mutations:
In principle, any allele leading to an epistatic phenotype, one identifiable within a mixed population of cells resulting from the germination of both mutant and wild-type spores, can be mapped with the SGAM methodology. Such phenotypes include but are not limited to: colony growth in an otherwise inviable genetic and/or environmental context, expression of chromophores, and morphogenetic phenotypes such as filamentous growth.

Consider the case of a haploid strain in which a suppressor mutation (s) in an unidentified gene (S) rescues the lethality associated with deletion of an essential gene (query::natR) marked with the dominant natR selectable marker (Fig 1). The viable double-mutant starting strain (query::natR s) is crossed systematically to each of the 5000 viable haploid deletion strains (xxx::kanR) and the double-deletion meiotic progeny (query::natR xxx::kanR) are selected. If, for a given kanR-marked deletion strain, the wild-type allele (S) and the deletion (xxx::kanR) are unlinked, 50% of the double-deletion meiotic progeny (query::natR xxx::kanR) will contain the suppressor mutation (s) and germinate to form a colony. In strains where the wild-type allele (S) is tightly linked to the deleted gene (xxx::kanR), however, the low frequency of recombination between the suppressor allele (s) and the deleted gene (xxx::kanR) will limit the recovery of viable double-deletion meiotic progeny. The general chromosomal location of the suppressor allele (s) can thereby be identified by the failure to observe colony growth for a linked set of double-deletion strains. In essence, to map the location of the mutation, SGAM employs a genome-wide and systematic set of two point crosses between the mutated allele and the marked deletion mutations.

View larger version (26K): [in this window] [in a new window] [Download PPT slide]   Figure 1. Logic behind SGA mapping. In the example shown, SGA analysis is used to map the location of a mutant suppressor allele (s, open red box), which suppresses the lethality of a natR-marked gene deletion mutation (query::natR, green box). As described in MATERIALS AND METHODS, SGA analysis combines the natR-marked deletion mutation with 5000 viable kanR-marked deletion mutations (xxx::kanR, blue boxes) through mating, meiotic recombination, and germination of haploid MATa query::natR xxx::kanR spore progeny. In deletion strains where the wild-type allele (S, solid red box) is tightly linked to the kanR-marked deletion (far-right yeast), the low frequency of recombination between the suppressor allele (s) and the kanR-marked deletion will limit the recovery of viable NatR KanR double-deletion progeny.

On the basis of this logic, a number of advantages of the SGAM approach are evident:

1. Although the mutation is mapped by linkage to nonessential genes, it may lie in either nonessential or essential genes. To map mutations in essential genes, the allele must not cause lethality at the screening temperature.

2. Both dominant and recessive mutations can be mapped.

3. As the size of the linked group will be inversely proportional to the frequency of meiosis in the parent diploid colonies, the assay could be fine tuned by modulating sporulation efficiency. For instance, hampering sporulation by manipulating temperature and/or incubation times should extend the linked region.

4. In principle, multigenic traits in which several alleles are required to confer the mutant phenotype can be mapped, as linked groups lacking the mutant phenotype will form around each requisite allele.

5. Sequencing of the gene(s) lying within the center of the linked group should reveal the mutation. Once the candidate gene is mapped, subsequent molecular genetic analysis can then be carried out to prove that the mapped mutation is coincident with the suppressor mutation.

We applied the SGAM method to study of the Cbk1 signaling pathway. Deletion of CBK1, MOB2, HYM1, KIC1, or TAO3 genes in the S288C genetic background, in which the complete set of yeast deletion strains is constructed, is lethal, whereas the equivalent deletion in the W303 background is not (WINZELER et al. 1999 ; DU and NOVICK 2002 ; B. NELSON and C. BOONE, unpublished results). Tetrad analysis of crosses between the viable W303 deletion mutants and an S288C strain revealed that 50% of the spores carrying a deletion mutation would germinate to form a colony, indicating that a single W303-based suppressor allele rescued the lethality of each deletion (data not shown). To map the suppressor allele, we first created appropriate starting strains for SGA analysis by crossing W303 strains that carried URA3- or natR-marked deletions of CBK1, MOB2, HYM1, KIC1, and TAO3 with an S288C strain that carried the SGA mating-type-specific haploid selection reporter (can1::MFA1pr-HIS3::MF1pr-LEU2). SGA analysis was carried out with cbk1, mob2, hym1, kic1, and toa3 as query mutations against 4700 viable xxx::kanR deletion strains, and colony growth defects in the resultant double-deletion meiotic progeny were scored. We performed the strain manipulation robotically; however, SGA analysis can be conducted easily by hand pinning, as described previously (TONG et al. 2001 ).

Because it was likely that a single W303-derived suppressor mutation rescued the lethality of each deletion, we pooled the resultant genetic interactions and examined those that occurred for at least 4 of the 5 Cbk1 pathway deletions. In total, 31 deletions interacted with all five of the pathway members and 85 deletions interacted with four of the pathway members (see supplementary table at http://www.genetics.org/supplemental/ for results of all screens). Using a novel algorithm (see MATERIALS AND METHODS), we identified one chromosomal region that had a statistically significant enrichment of genetic interactions, which presumably mapped the location of the W303-derived suppressor (Fig 2 and Fig 3). In this colinear set of 48 essential and nonessential genes, spanning an 89.6-kb (28 cM) region of chromosome IV (MAPS), we identified 13 genetic interactions. This same region could also be identified from the primary data collected for three of the individual screens (hym1, kic1, and tao3; see Fig 3), but pooling the data from all of the screens provides a method for removing false positives associated with the high-throughput analysis.

View larger version (64K): [in this window] [in a new window] [Download PPT slide]   Figure 2. (A) SGA analysis with gene deletions of Cbk1 pathway components revealed dispersed synthetic interactions as well as an apparent set of colinear synthetic interactions on chromosome IV. Starting strains deleted for CBK1, MOB2, HYM1, KIC1, and TAO3 were subjected to SGA analysis vs. a near complete set of 4700 kanR-marked, viable haploid deletion strains. Gene deletions included in this set are represented as blue lines and cover nearly the whole genome from chromosome I to XVI. Gene deletions that consistently exhibited synthetic interactions with deletions in Cbk1 pathway genes, found in at least four out of five screens, are represented as red lines and were found scattered throughout the genome. Centromeres are represented as a solid black circle. A set of genetic interactions in an 90-kb region of chromosome IV (boxed) was identified as statistically significant by an algorithm that measures the density of synthetic interactions over a sliding window of 11 genes (see MATERIALS AND METHODS). The boxed region localizes the W303-specific suppressor of Cbk1 pathway deletions. (B) Magnification of the boxed 90-kb region of chromosome IV. Genes are represented as in A. Essential genes are marked with * and are not represented by haploid deletion mutants. Gene deletions that consistently caused slow growth in SGA analysis vs. wild-type control strains were beyond the sensitivity of the assay and are marked with #.

View larger version (71K): [in this window] [in a new window] [Download PPT slide]   Figure 3. A linked region of inviable or slow-growing double mutants surrounded the SSD1 locus when SGA analysis was carried out with deletions of Cbk1 pathway components. In the final step of SGA analysis, double-mutant meiotic progeny were selected by growth on SC/MSG-His-Arg-Ura plates supplemented with L-canavanine and G418 (see MATERIALS AND METHODS). Of the 16 plates of arrayed double-mutant meiotic progeny, one contained the ssd1::kanR strain and most of the deletions surrounding the SSD1 locus in the top half of the plate (dashed lines); this plate and region are shown for a wild-type control screen (Y2806) and a tao3::URA3 (Y3714) screen. Each deletion strain is represented twice on the plate, located immediately diagonal to one another. In almost all cases, the size of the two colonies is identical, although there are exceptions (e.g., only one of the ssd1::kanR colonies grew up in the tao3::natR screen, probably because of a pinning error). The ssd1::kanR strain and deletions linked to the SSD1 locus are outlined; red indicates tao3::URA3 xxx::kanR double deletions (top) that showed reduced colony size relative to the xxx::kanR deletion alone (wild-type control screen, middle) while blue indicates roughly equal colony size in the two screens. The presence of blue circles, indicating equal growth in the presence and absence of TAO3, in the middle of the linked group of synthetic genetic interactions (red circles) can be explained (i) by growth defects, even in the presence of TAO3 (e.g., gcn2, atp5), which prevent the synthetic genetic interaction from being observable (see Fig 2 legend), and (ii) by weak or uncertain interactions (e.g., hnt2). The identities of each of the xxx::kanR deletions is shown at the bottom. The deletion strains are arrayed on the plates in the same order as the corresponding genes on chromosome IV. Arrows indicate the topology of the genes with respect to the SSD1 locus, with the arrowhead pointing away from the SSD1 locus. Because Fig 2 presents the cumulative data from SGA analysis carried out against deletions in all five components of the Cbk1 signaling pathway, the synthetic interactions listed in Fig 2 differ somewhat from those observed in this single screen vs. tao3::URA3.

Mutations in SSD1 suppress the lethality associated with deletions in Cbk1 pathway genes:
Random spore analysis enabled us to determine the frequency of recombination between the putative Cbk1 pathway suppressor and the kanR-marked gene deletions in the mapped region of chromosome IV. The mob2::natR SGA starting strain (Y3968) was mated to each kanR-marked deletion strain in this region. Viable MATa mob2::natR meiotic progeny were selected for and replica plated onto media containing G418 to determine the percentage of viable MATa mob2::natR xxx::kanR recombinants (Fig 4A). For gene deletions such as dpp1::kanR and sur2::kanR, <10% of the viable MATa mob2::natR meiotic progeny also grew on the G418 medium, as expected if a W303-based suppressor was linked to these deletion alleles (Fig 1). In contrast, for ssd1::kanR, 50% of the viable MATa mob2::natR meiotic progeny also grew on the G418 medium. This result suggests that ssd1::kanR suppresses the lethality associated with the mob2::natR deletion. Indeed, unlike other deletions in this region, ssd1::kanR showed no synthetic interactions with any of the Cbk1 pathway members by SGA analysis (Fig 2 and Fig 3 and supplementary data at http://www.genetics.org/supplemental/).

View larger version (71K): [in this window] [in a new window] [Download PPT slide]   Figure 4. Mutations in SSD1 suppress the lethality associated with deletion of genes coding for Cbk1 pathway components. (A) Random spore analysis was carried out as described in MATERIALS AND METHODS to determine the frequency of recombination between the W303-specific suppressor of mob2::natR and kanR-marked deletions on a section of chromosome IV in the vicinity of the SSD1 locus. The percentage of recombinant progeny was plotted vs. position on chromosome IV, with the end of the ORF closest to the central SSD1 locus determining the plotted base pair. As kanR-marked deletions approached SSD1, the percentage of recombinant progeny approached zero. At loci far removed from SSD1, the percentage of recombinant progeny returned to 50%. The 50% recombinant progeny seen for ssd1::kanR is actually due to suppression of the lethality associated with mob2::natR by ssd1::kanR. (B) Tetrad analysis confirmed that ssd1::kanR suppresses the lethality associated with deletion of HYM1. When a viable W303-derived hym1::URA3 is crossed to a wild-type S288C strain, a single suppressor segregates with the viability of the hym1::URA3; i.e., 50% of the hym1::URA3 spores germinate to form a colony (top). When the W303-derived hym1::URA3 was crossed to an S288C-derived ssd1::kanR strain, all hym1::URA3 spores were viable (bottom). This result demonstrates that ssd1::kanR suppresses the hym1-associated lethality and that the W303-derived suppressor of the hym1-associated lethality is tightly linked to the SSD1 locus. Identical results were obtained for cbk1, mob2, tao3, and kic1 (data not shown). (C) When expressed in the W303 background, SSD1-v restored the lethality associated with CBK1 deletion. Congenic wild-type and cbk1 strains were sequentially transformed with a URA3-marked YEp24 plasmid containing the CBK1 gene (P3765) and either a LEU2-marked YEp13 plasmid carrying SSD1-v (MTP251) or the YEp13 empty vector. These strains were grown to saturation in SC-Leu-Ura, serially diluted (four 10-fold dilutions), and the culture and each dilution were spotted on SC-Leu and SC-Leu containing 5-FOA, a drug that is toxic to yeast expressing the URA3 gene. In the presence of SSD1-v, the cbk1 strain could not form colonies on SC-Leu containing 5-FOA.

When cbk1, mob2, hym1, kic1, and toa3 strains in the W303 background were crossed to ssd1 in the S288C background, tetrad analysis revealed that all of the spores containing a Cbk1 pathway gene deletion were viable (Fig 4B and data not shown). Therefore, not only does the deletion of SSD1 suppress the loss of the Cbk1 pathway, but also the W303-derived suppressor must be either within or tightly linked to the SSD1 locus. While this work was in progress, DU and NOVICK 2002 identified mutations within SSD1 as suppressors of cbk1 and tao3 by a transposon insertion mutagenesis approach.

SSD1 encodes a putative RNA-binding protein implicated in the regulation of cell polarity and cell integrity that appears to physically interact with Cbk1 (RACKI et al. 2000 ; HO et al. 2002 ). It has been well established that SSD1 is polymorphic in laboratory strains of S. cerevisiae: most S288C strains carry the dominant SSD1-v allele, while W303 strains usually harbor the recessive ssd1-d allele. The suffixes "v" in SSD1-v and "d" in ssd1-d are derived from SUTTON et al. 1991 and refer to the fact that deletion mutations in the SIT4 gene are viable in the presence of SSD1-v but dead in the presence of ssd1-d. The opposite relationship appears to be true for deletion mutations in the Cbk1 signaling pathway, which appear to be viable in the presence of ssd1-d but dead in the presence of SSD1-v. We sequenced the ssd1-d allele in the W303 background and compared it to the SSD1-v allele sequenced as part of the yeast genome project (GOFFEAU et al. 1996 ). A point mutation at base-pair 2094 (C to G) of the SSD1 open reading frame creates a premature stop codon in the W303 ssd1-d allele, resulting in a truncated protein with a predicted molecular weight of 77 kD (data not shown). These findings are in general agreement with the observation that ssd1-d is expressed as a truncated 83-kD protein (UESONO et al. 1997 ).

To confirm that the ssd1-d allele present in the W303 genome suppresses loss of the Cbk1 signaling pathway, we used a plasmid loss assay. cbk1 and congenic wild-type cells in the W303 background were transformed with a plasmid containing the CBK1 gene and maintained with the URA3 selectable marker. The plasmid rescued several cbk1 phenotypes (see MATERIALS AND METHODS). These strains were then transformed either with a LEU2-marked plasmid housing the SSD1-v allele or with the empty vector. The four Ura⁺ Leu⁺ genetic combinations were cultured and plated on SC-Leu and SC-Leu containing 5-FOA. On both types of plates, the absence of leucine maintained the LEU2-marked plasmids, but on the plates containing 5-FOA, cells possessing URA3-marked plasmids were selected against. In wild-type transformants, in either the presence or the absence of SSD1-v, some of the plated cells formed colonies on SC-Leu containing 5-FOA (Fig 4C). These cells were able to grow in the absence of the URA3-marked plasmid. Similarly, cbk1 cells transformed with the LEU2-marked empty vector formed colonies on SC-Leu containing 5-FOA. However, cbk1 cells transformed with the LEU2-marked vector housing SSD1-v did not form colonies on SC-Leu medium containing 5-FOA (Fig 4C), indicating that in the presence of SSD1-v, cells lacking CBK1 (as carried by the URA3 plasmid) were not viable. When introduced into the W303 background, the SSD1-v allele restored the lethality associated with deletion of CBK1.

Synthetic lethality with genes encoding Cbk1 pathway components:
The SGA screens used for mapping the Cbk1 pathway suppressor also provided many candidate synthetic genetic interactions at loci unlinked to SSD1 (Fig 2A). Here, we directly examined a subset of the strongest synthetic genetic interactions (see MATERIALS AND METHODS). Tetrad analysis confirmed that nine deletions unlinked to SSD1 are important for growth in haploid cbk1::URA3 ssd1-d cells and hym1::URA3 ssd1-d cells (Table 2). Four of these genes are involved in reorganization of the cortical actin cytoskeleton (BEM1, BEM2, BEM4, and SAC7) and play key roles in bud emergence and development. Another gene, NBP2, has been generally implicated in cytoskeletal organization because nbp2 is synthetically lethal with several genes involved in spindle orientation and actin assembly (TONG et al. 2001 ). Moreover, the Nbp2 ortholog in the fission yeast Schizosaccharomyces pombe, Skb5, appears to bind and activate the PAK-like kinase Shk1(YANG et al. 1999 ) and shk1 shows synthetic lethality with a deletion of the gene encoding the S. pombe Cbk1 ortholog, Orb6 (VERDE et al. 1998 ). Two other genes (GAS1 and KRE1) are directly involved in cell wall maintenance and polarized growth through ß-1,6-glucan crosslinking and assembly. Finally, two of the genes (BST1 and GOS1) are involved in vesicular transport. All of these genes function directly or indirectly in polarized cell growth, which could explain their essential role in ssd1-d cells defective for Cbk1 pathway signaling.

View this table: [in this window] [in a new window]   Table 2. Gene deletions that show synthetic growth defects with cbk1 ssd1-d and hym1 ssd1-d

Ordered arrays as mapping tools:
Ordered arrays of marked yeast deletion strains provide an inherently powerful tool for high-resolution genetic mapping. When combined with SGA methodology, this mapping method can be automated and carried out in high throughput. Although the test case shown here demonstrated the mapping of the suppressor of an essential gene deletion mutation, SGAM can be applied to any allele that leads to an epistatic phenotype that S288C strains do not normally display. These phenotypes include suppression of conditional alleles of essential genes, cytotoxic drug resistance, and filamentous growth. SGAM may be of particular use for rapid mapping of dominant mutations, which can be challenging to clone with standard techniques. Furthermore, we expect that the rapid exploration of all null phenotypes, as allowed by the yeast deletion sets, will eventually lead researchers to a fuller exploration of genetic space, as provided by screens for dominant alleles. In theory, several alleles responsible for complex, multigenic quantitative traits, such as the high-temperature growth phenotype of pathogenic S. cerevisiae, could be mapped in a single round of SGA analysis (MCCUSKER et al. 1994 ; STEINMETZ et al. 2002 ). To further the mapping process, the unique oligonucleotide bar-code tags built into each yeast deletion strain may be exploited for quantitative analysis of growth phenotypes of meiotic progeny by hybridization to DNA microarrays (SHOEMAKER et al. 1996 ).

Other methods for rapid mapping of mutations have been devised. For instance, thousands of markers provided by the allelic variation between different strains of S. cerevisiae have been detected with high-density oligonucleotide arrays and exploited for mapping novel mutations (WINZELER et al. 1998 ; STEINMETZ et al. 2002 ). When the mutant phenotype is amenable (see above), mapping with SGA analysis may be preferable to mapping with oligonucleotide arrays because of the potential for higher resolution. In the first stage of SGA mapping, a starting strain is crossed to 4700 deletion stains to identify a chromosomal region(s) centered around the mutation. In the example presented here, a 89.6-kb (28-cM) region was detected (Fig 2). This mapping resolution is similar to that obtained in test applications of high-density oligonucleotide arrays (WINZELER et al. 1998 ), which detected 11- to 64-kb regions surrounding four known loci. However, in the second stage of SGA mapping, random spore analysis of the meiotic progeny derived from deletion strains in the identified chromosomal region pinpoints the gene(s) tightly linked to the mutation. While S. cerevisiae genetics is particularly suited for automation and large-scale genetic analysis, the construction of ordered arrays of gene deletion mutants would enable this type of approach to be applied to other model organisms (COELHO et al. 2000 ), especially those with high rates of meiotic recombination.

	FOOTNOTES

¹ These authors contributed equally to this work.

	ACKNOWLEDGMENTS

The authors thank Howard Bussey for access to yeast deletion strains, Helena Friesen for discussions on ssd1-d sequencing, and Joe Horecka for comments on the manuscript. This work was supported by a Canadian Institutes of Health Research (CIHR) special genomics project grant to B.A., M.T., and C.B.; P.J. holds a CIHR Doctoral Research Award, and B.N. holds a Natural Sciences and Engineering Research Council of Canada (NSERC) graduate student fellowship.

Manuscript received June 10, 2002; Accepted for publication August 20, 2002.

	LITERATURE CITED

TOP ABSTRACT MATERIALS AND METHODS RESULTS AND DISCUSSION LITERATURE CITED

BIDLINGMAIER, S., E. L. WEISS, C. SEIDEL, D. G. DRUBIN, and M. SNYDER, 2001  The Cbk1p pathway is important for polarized cell growth and cell separation in Saccharomyces cerevisiae.. Mol. Cell. Biol. 21:2449-2462.[Abstract/Free Full Text]

BRACHMANN, C. B., A. DAVIES, G. J. COST, E. CAPUTO, and J. LI et al., 1998  Designer deletion strains derived from Saccharomyces cerevisiae S288C: a useful set of strains and plasmids for PCR-mediated gene disruption and other applications. Yeast 14:115-132.[Medline]

COELHO, P. S., A. KUMAR, and M. SNYDER, 2000  Genome-wide mutant collections: toolboxes for functional genomics. Curr. Opin. Microbiol. 3:309-315.[Medline]

COLMAN-LERNER, A., T. E. CHIN, and R. BRENT, 2001  Yeast Cbk1 and Mob2 activate daughter-specific genetic programs to induce asymmetric cell fates. Cell 107:739-750.[Medline]

DORLAND, S., M. L. DEEGENAARS, and D. J. STILLMAN, 2000  Roles for the Saccharomyces cerevisiae SDS3, CBK1 and HYM1 genes in transcriptional repression by SIN3.. Genetics 154:573-586.[Abstract/Free Full Text]

DU, L. L. and P. NOVICK, 2002  Pag1p, a novel protein associated with protein kinase Cbk1p, is required for cell morphogenesis and proliferation in Saccharomyces cerevisiae.. Mol. Biol. Cell 13:503-514.[Abstract/Free Full Text]

DURRENBERGER, F. and J. KRONSTAD, 1999  The ukc1 gene encodes a protein kinase involved in morphogenesis, pathogenicity and pigment formation in Ustilago maydis.. Mol. Gen. Genet. 261:281-289.[Medline]

GOFFEAU, A., B. G. BARRELL, H. BUSSEY, R. W. DAVIS, and B. DUJON et al., 1996  Life with 6000 genes. Science 274:546-563. 567.[Abstract/Free Full Text]

HIRSCH, J. P., C. DIETZEL, and J. KURJAN, 1991  The carboxyl terminus of Scg1, the G alpha subunit involved in yeast mating, is implicated in interactions with the pheromone receptors. Genes Dev. 5:467-474.[Abstract/Free Full Text]

HO, Y., A. GRUHLER, A. HEILBUT, G. D. BADER, and L. MOORE et al., 2002  Systematic identification of protein complexes in Saccharomyces cerevisiae by mass spectrometry. Nature 415:180-183.[Medline]

JUSTICE, R. W., O. ZILIAN, D. F. WOODS, M. NOLL, and P. J. BRYANT, 1995  The Drosophila tumor suppressor gene warts encodes a homolog of human myotonic dystrophy kinase and is required for the control of cell shape and proliferation. Genes Dev. 9:534-546.[Abstract/Free Full Text]

MCCUSKER, J. H., K. V. CLEMONS, D. A. STEVENS, and R. W. DAVIS, 1994  Genetic characterization of pathogenic Saccharomyces cerevisiae isolates. Genetics 136:1261-1269.[Abstract]

MILLWARD, T., P. CRON, and B. A. HEMMINGS, 1995  Molecular cloning and characterization of a conserved nuclear serine(threonine) protein kinase. Proc. Natl. Acad. Sci. USA 92:5022-5026.[Abstract/Free Full Text]

RACKI, W. J., A. M. BECAM, F. NASR, and C. J. HERBERT, 2000  Cbk1p, a protein similar to the human myotonic dystrophy kinase, is essential for normal morphogenesis in Saccharomyces cerevisiae.. EMBO J. 19:4524-4532.[Medline]

SHOEMAKER, D. D., D. A. LASHKARI, D. MORRIS, M. MITTMANN, and R. W. DAVIS, 1996  Quantitative phenotypic analysis of yeast deletion mutants using a highly parallel molecular bar-coding strategy. Nat. Genet. 14:450-456.[Medline]

STEINMETZ, L. M., H. SINHA, D. R. RICHARDS, J. I. SPIEGELMAN, and P. J. OEFNER et al., 2002  Dissecting the architecture of a quantitative trait locus in yeast. Nature 416:326-330.[Medline]

SULLIVAN, D. S., S. BIGGINS, and M. D. ROSE, 1998  The yeast centrin, Cdc31p, and the interacting protein kinase, Kic1p, are required for cell integrity. J. Cell Biol. 143:751-765.[Abstract/Free Full Text]

SUTTON, A., D. IMMANUEL, and K. T. ARNDT, 1991  The Sit4 protein phosphatase functions in late G1 for progression into S phase. Mol. Cell. Biol. 11:2133-2148.[Abstract/Free Full Text]

TONG, A. H., M. EVANGELISTA, A. B. PARSONS, H. XU, and G. D. BADER et al., 2001  Systematic genetic analysis with ordered arrays of yeast deletion mutants. Science 294:2364-2368.[Abstract/Free Full Text]

UESONO, Y., A. TOH-E, and Y. KIKUCHI, 1997  Ssd1p of Saccharomyces cerevisiae associates with RNA. J. Biol. Chem. 272:16103-16109.[Abstract/Free Full Text]

VERDE, F., D. J. WILEY, and P. NURSE, 1998  Fission yeast orb6, a ser/thr protein kinase related to mammalian rho kinase and myotonic dystrophy kinase, is required for maintenance of cell polarity and coordinates cell morphogenesis with the cell cycle. Proc. Natl. Acad. Sci. USA 95:7526-7531.[Abstract/Free Full Text]

WINZELER, E. A., D. R. RICHARDS, A. R. CONWAY, A. L. GOLDSTEIN, and S. KALMAN et al., 1998  Direct allelic variation scanning of the yeast genome. Science 281:1194-1197.[Abstract/Free Full Text]

WINZELER, E. A., D. D. SHOEMAKER, A. ASTROMOFF, H. LIANG, and K. ANDERSON et al., 1999  Functional characterization of the S. cerevisiae genome by gene deletion and parallel analysis. Science 285:901-906.[Abstract/Free Full Text]

YANG, P., R. PIMENTAL, H. LAI, and S. MARCUS, 1999  Direct activation of the fission yeast PAK Shk1 by the novel SH3 domain protein, Skb5. J. Biol. Chem. 274:36052-36057.[Abstract/Free Full Text]

YARDEN, O., M. PLAMANN, D. J. EBBOLE, and C. YANOFSKY, 1992  cot-1, a gene required for hyphal elongation in Neurospora crassa, encodes a protein kinase. EMBO J. 11:2159-2166.[Medline]

ZALLEN, J. A., E. L. PECKOL, D. M. TOBIN, and C. I. BARGMANN, 2000  Neuronal cell shape and neurite initiation are regulated by the Ndr kinase SAX-1, a member of the Orb6/COT-1/warts serine/threonine kinase family. Mol. Biol. Cell 11:3177-3190.[Abstract/Free Full Text]

This article has been cited by other articles:

				D. Liko, M. G. Slattery, and W. Heideman Stb3 Binds to Ribosomal RNA Processing Element Motifs That Control Transcriptional Responses to Growth in Saccharomyces cerevisiae J. Biol. Chem., September 7, 2007; 282(36): 26623 - 26628. [Abstract] [Full Text] [PDF]

				A. Wang, S. Lane, Z. Tian, A. Sharon, I. Hazan, and H. Liu Temporal and Spatial Control of HGC1 Expression Results in Hgc1 Localization to the Apical Cells of Hyphae in Candida albicans Eukaryot. Cell, February 1, 2007; 6(2): 253 - 261. [Abstract] [Full Text] [PDF]

				S. Seiler, N. Vogt, C. Ziv, R. Gorovits, and O. Yarden The STE20/Germinal Center Kinase POD6 Interacts with the NDR Kinase COT1 and Is Involved in Polar Tip Extension in Neurospora crassa Mol. Biol. Cell, September 1, 2006; 17(9): 4080 - 4092. [Abstract] [Full Text] [PDF]

				M. J. Brauer, C. M. Christianson, D. A. Pai, and M. J. Dunham Mapping Novel Traits by Array-Assisted Bulk Segregant Analysis in Saccharomyces cerevisiae Genetics, July 1, 2006; 173(3): 1813 - 1816. [Abstract] [Full Text] [PDF]

				J. A. Daniel, B. E. Keyes, Y. P. Y. Ng, C. O. Freeman, and D. J. Burke Diverse Functions of Spindle Assembly Checkpoint Genes in Saccharomyces cerevisiae Genetics, January 1, 2006; 172(1): 53 - 65. [Abstract] [Full Text] [PDF]

				C. Kurischko, G. Weiss, M. Ottey, and F. C. Luca A Role for the Saccharomyces cerevisiae Regulation of Ace2 and Polarized Morphogenesis Signaling Network in Cell Integrity Genetics, October 1, 2005; 171(2): 443 - 455. [Abstract] [Full Text] [PDF]

				W. P. Voth, A. E. Olsen, M. Sbia, K. H. Freedman, and D. J. Stillman ACE2, CBK1, and BUD4 in Budding and Cell Separation Eukaryot. Cell, June 1, 2005; 4(6): 1018 - 1028. [Abstract] [Full Text] [PDF]

				F. Malagon, A. H. Tong, B. K. Shafer, and J. N. Strathern Genetic Interactions of DST1 in Saccharomyces cerevisiae Suggest a Role of TFIIS in the Initiation-Elongation Transition Genetics, March 1, 2004; 166(3): 1215 - 1227. [Abstract] [Full Text] [PDF]

				L. Schneper, A. Krauss, R. Miyamoto, S. Fang, and J. R. Broach The Ras/Protein Kinase A Pathway Acts in Parallel with the Mob2/Cbk1 Pathway To Effect Cell Cycle Progression and Proper Bud Site Selection Eukaryot. Cell, February 1, 2004; 3(1): 108 - 120. [Abstract] [Full Text] [PDF]

				B. Nelson, C. Kurischko, J. Horecka, M. Mody, P. Nair, L. Pratt, A. Zougman, L. D.B. McBroom, T. R. Hughes, C. Boone, et al. RAM: A Conserved Signaling Network That Regulates Ace2p Transcriptional Activity and Polarized Morphogenesis Mol. Biol. Cell, September 1, 2003; 14(9): 3782 - 3803. [Abstract] [Full Text] [PDF]

				D. L. Mason, M. P. Mallampalli, G. Huyer, and S. Michaelis A Region within a Lumenal Loop of Saccharomyces cerevisiae Ycf1p Directs Proteolytic Processing and Substrate Specificity Eukaryot. Cell, June 1, 2003; 2(3): 588 - 598. [Abstract] [Full Text] [PDF]

This Article Abstract Full Text (PDF) Supplemental Data Alert me when this article is cited Alert me if a correction is posted Services Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Jorgensen, P. Articles by Boone, C. Search for Related Content PubMed