Large-Scale Functional Genomic Analysis of Sporulation and Meiosis in Saccharomyces cerevisiae

Corresponding author: William S. Saunders, Department of Biological Sciences, University of Pittsburgh, Pittsburgh, PA 15260., wsaund{at}pitt.edu (E-mail)

Communicating editor: M. JOHNSTON

	ABSTRACT

TOP ABSTRACT MATERIALS AND METHODS RESULTS DISCUSSION LITERATURE CITED

We have used a single-gene deletion mutant bank to identify the genes required for meiosis and sporulation among 4323 nonessential Saccharomyces cerevisiae annotated open reading frames (ORFs). Three hundred thirty-four sporulation-essential genes were identified, including 78 novel ORFs and 115 known genes without previously described sporulation defects in the comprehensive Saccharomyces Genome (SGD) or Yeast Proteome (YPD) phenotype databases. We have further divided the uncharacterized sporulation-essential genes into early, middle, and late stages of meiosis according to their requirement for IME1 induction and nuclear division. We believe this represents a nearly complete identification of the genes uniquely required for this complex cellular pathway. The set of genes identified in this phenotypic screen shows only limited overlap with those identified by expression-based studies.

DURING meiosis cells exit the vegetative cell cycle and enter a linear divisional and differentiation pathway. The regulation of this process is probably best described in the budding yeast Saccharomyces cerevisiae (PRINGLE and HARTWELL 1981 ). The decision to begin meiosis in this organism is controlled by two known major inductional mechanisms (MALONE 1990 ). One is tuned to nutritional starvation, and the other is tuned to the presence of the a1/2 heterodimer, found only in diploids formed by the joining of haploids of opposite mating type. Initiation of meiosis requires expression of many specialized genes. Transcriptional induction of these genes is controlled in large part by the early meiotic-specific transcription factor IME1 (KASSIR et al. 1988 ). Ime1p is required for the induction of most early meiotic genes except IME4 (SHAH and CLANCY 1992 ) and is essential for all of the major downstream events of meiosis and sporulation.

Meiosis begins with a prolonged S phase during which a single replication of the genome occurs (PRINGLE and HARTWELL 1981 ). S phase is followed by a very specialized meiotic prophase when homologous chromosomes pair and exchange DNA through recombination. Following recombination, meiotic cells undergo two rounds of division without further DNA replication, leading to the formation of four haploid nuclei. In S. cerevisiae, each of the nuclei is encapsulated in a hardened cell wall formed by growth of a double-membrane prospore wall adjacent to the outer plaque of the meiosis II spindle pole bodies (MOENS and RAPPORT 1971 ). This capsule then differentiates into a multilayered spore wall by deposition of materials between the double layers of the prospore membrane (LYNN and MAGEE 1970 ). With normal division, four viable ascospores, each containing one haploid nucleus, become enclosed within the residual cell wall.

Several major genetic screens have been performed to identify the various contributors to the meiotic pathway in S. cerevisiae. These include screens for reduced sporulation (ESPOSITO and ESPOSITO 1969 ), for resistance to chemical agents (DAWES and HARDIE 1974 ), for reduced spore viability (ROCKMILL and ROEDER 1988 ), and for variations in spore wall components (BRIZA et al. 1990 ), as well as recent genomic-based screens (RABITSCH et al. 2001 ; BRIZA et al. 2002 ). The study of the genes identified in these screens has defined the pathway for meiotic activation and regulation in more detail in S. cerevisiae than in any other organism. More than 150 genes that have known roles in meiosis and sporulation have been identified by these traditional genetic screens (KUPIEC et al. 1997 ). However, as shown here, many meiosis/sporulation genes remain to be identified and characterized.

	MATERIALS AND METHODS

TOP ABSTRACT MATERIALS AND METHODS RESULTS DISCUSSION LITERATURE CITED

Strains and medium:
All experiments were performed in the S288C background. Established protocols for media synthesis were used (GUTHRIE and FINK 1991 ). To test for growth on glycerol medium, the deletion mutants were resuspended from a nutrient-rich YEPD plate into water and transferred dropwise without dilution to a plate containing 2.5% glycerol as the primary carbon source.

Handling the deletion mutants and sporulation test:
The mutant strains were received from Research Genetics (Birmingham, AL) in 96-well microtiter dishes on dry ice. The frozen cell pellets were thawed on ice and a portion was transferred to a YEPD plate and grown overnight at 26°. From the YEPD plate, a liquid culture was started in 300 µl of YEPD in 24-well microtiter plates (Becton Dickinson, Lincoln Park, NJ) and grown overnight in a 22° shaking water bath.

The sporulation test was performed by pelleting the cells in the 24-well microtiter plates by centrifugation in a Beckman 5810R centrifuge with an A-2-MTP rotor at 4000 rpm for 5 min at 22°. The supernatant was removed by aspiration and the cell pellet was washed twice by centrifugation with 300 µl of water, repelleted, and resuspended in sporulation medium (1 g/liter of yeast extract, 10 g/liter potassium acetate, 0.5 g/liter dextrose, 20 mg/liter leucine, 40 mg/liter uracil supplemented with 50 mg/ml uracil, and 250 mg/ml leucine). The 24-well microtiter plates were then shaken for 3 days at 22°. The cells were then examined for sporulation by DIC microscopy with an Olympus BX microscope with a x100 oil immersion objective.

Test for ploidy of deletion strains:
The mutant strains were taken from a YEPD plate, suspended in water, and transferred dropwise without dilution to a YEPD plate spread with a fresh layer of either the a or the -mating-type tester strains. After overnight growth at 30°, velvet squares were used to transfer the yeast mixture to a minimal media plate to select for diploids (GUTHRIE and FINK 1991 ). Growth on the selective plate indicated a high level of haploids in the deletion parent strain, most likely due to incomplete selection when the diploid mutant strain was first created. In most of these cases, the diploid was further selected by streaking the original deletion mutant strain for single colonies on plates lacking methionine and lysine, which should allow growth of only the diploid. Single colonies of Met⁺ Lys⁺ cells were isolated and retested for sporulation and ploidy. If nonmating strains could be readily isolated, the sporulation results from these isolates are given. Otherwise, the strain was not evaluated.

Test for correct open reading frame deletion:
The tested strains were deletions of YOL051w, YHR124w, YLL005c, YHL023c, YPL232w, YBR279w, YKL167c, YLR377c, YJR094c, and YDL226c. Two sets of conformational primers with the sequence chosen from the deletion project web site (http://www-sequence.stanford.edu/group/yeast_deletion_project/deletions3.html) were used. The first pair of primers, designated A and B, are unique to each open reading frame (ORF), where A is from regions 200–400 bases upstream of a particular ORF, and B is from regions within the particular ORF. This pair of primers gives a PCR product only if that ORF is intact. The second pair of primers, designated A and kanB, are unique to a deletion of the particular ORF, where A is as described above and kanB is specific for kanMX4 cassette. The second pair of primers gives a PCR product only if deletion of the specific ORF is present. In all 10 strains tested, a PCR product was observed only when primers A and kanB were used.

Nuclear division assay:
Mutant strains were induced to sporulate by growing the cells in 1–2 ml of YEPD liquid medium in glass tubes, washing twice in water by centrifugation, and resuspending in sporulation medium in glass tubes. The cells were fixed at 24–96 hr by mixing equal volumes of culture medium and 3.7% formaldehyde in PBS. The mixture was immediately pelleted and resuspended in 3.7% formaldehyde in PBS for 2 hr at room temperature or 15 hr at 4°. The cells were stained with 4',6-diamidino-2-phenylindole (Sigma, St. Louis) at 1 µg/ml in PBS and examined by epifluorescence microscopy with an Olympus BX microscope and photographed with an Orca camera (Hamamatsu, Bridgewater, NJ).

	RESULTS

TOP ABSTRACT MATERIALS AND METHODS RESULTS DISCUSSION LITERATURE CITED

An international consortium of laboratories have collaborated to individually delete most of the annotated ORFs in the S. cerevisiae genome as described previously (GIAEVER et al. 2002 ). Homozygous diploids were constructed by the consortium by mating independently mutated haploid strains. We have used this diploid deletion mutant collection to screen for genes required for meiotic division and sporulation.

Our primary screen was differential interference contrast microscopic inspection for the formation of visible spores following growth in liquid sporulation medium (MATERIALS AND METHODS). Under these conditions, a majority of mutant strains were able to sporulate at levels of 50% of all cells, with 30% of the cells forming full tetrads. This is comparable to the sporulation levels for the S288C wild-type background from which the collection was derived and for auxotrophic mutants in the collection such as deletion of ADE4 (not shown). The phenotypes for the mutant collection were divided into five categories of sporulation levels. The "normal" category was indistinguishable from wild type. A score of "low" was given if sporulated cells could be readily observed but at a lower frequency than that in wild type. "Very low" indicates that spores were only rarely seen, and "none" indicates the absence of spores. In addition, some mutants had high levels of sporulation but reduced numbers of spores per ascus. These mutants are listed as "low-4." No mutants had significant numbers of asci with more than four spores. Thirty-eight strains grew poorly or not at all and were not included in this analysis.

Each mutant with a sporulation defect was retested at least one additional time, and more if needed, until a consistent phenotype was observed. If normal sporulation was observed in any of these trials, that mutant was placed in the normal category. Three additional tests were performed to eliminate false identification of mutant phenotypes:

1. To confirm that the expected ORF was deleted and to verify that strain mixing did not occur during shipping and handling, the identity of the deleted ORF of 10 randomly chosen nonsporulating mutants was tested by PCR analysis (MATERIALS AND METHODS). Five strains were chosen from each of the two releases and each strain was chosen from a different microtiter plate. All 10 of the chosen mutants contained a replacement of the expected ORF with the marker gene KanMX4 (not shown).

2. To distinguish the mutants unable to utilize nonfermentable carbon sources, all sporulation-defective mutants containing deletions of novel ORFs were tested for growth on plates containing glycerol as the sole carbon source (YEPG). Twenty-five novel mutant strains were unable to grow on YEPG and are assumed to be unable to sporulate due to respiratory defects. It should be noted that some of these may be defective for glycerol utilization rather than for respiration. After the phenotypes were confirmed, the identity of each of the ORFs was determined and a comparison to known phenotypes was made by reference to the SGD and YPD (Table S1, supplemental material at http://www.genetics.org/supplemental/). One hundred eighteen genes that were known or suspected to be required for respiratory growth are shown in yellow in Table S1 and were not studied further. Deletion mutants for the genes previously characterized, but not shown to be involved in respiration in the phenotypic description of the SGD and YPD, were tested for growth on YEPG. Any of the mutants that failed to grow on YEPG plates were not included in subsequent analyses.

3. All of the sporulation mutants were tested for ploidy by crossing the presumptive diploid strains to haploid mating-type tester strains (MATERIALS AND METHODS). Fifteen mutants were apparently haploid and were not studied further.

The mutants with a consistent sporulation phenotype and normal growth on YEPG and YEPD plates are interpreted to be the genes important for sporulation in S. cerevisiae and are shown in Fig 1. By this analysis, 154 (3.6% of total) of the tested genes are essential for sporulation, with very few, if any, spores visible. One hundred eighty-one (4.2%) play an important but nonessential role, including 48 (1.1%) that are required to efficiently produce all four spores.

View larger version (61K): In this window In a new window Download PPT slide   Figure 1. Genes found to be required for sporulation. Genes are listed alphabetically followed by the ORF names for unpublished or uncharacterized loci. The level of sporulation observed in the mutant is shown to the right of the name. Included are genes previously shown to be required for sporulation or meiosis (red) and genes previously characterized but without a meiosis or sporulation phenotype listed in the SGD or YPD (green). All of the strains with deletions of these genes and ORFs were found to be able to grow on YEPG except deletions of the known sporulation/meiosis genes (red), which were not tested. When two ORFs are listed, they overlap in position and the functional gene is not known.

To further define the role of the novel genes in the meiotic/sporulation pathway we tested mutants of 56 novel and unpublished ORFs, including most of those with severe sporulation defects, for induction of the early meiotic inducer IME1 (MITCHELL et al. 1990 ) and for meiotic nuclear division (Fig 2; Table 1). The selected strains were transformed with a plasmid containing a ß-galactosidase reporter gene under the control of the IME1 promoter. Thirteen mutants were unable to induce ß-galactosidase expression, showing that these ORFs are required for full activation of this early meiotic inducer. The bulk of the novel mutants (27) were able to induce IME1, but were unable to undergo nuclear division. Thus, these genes were not required for initiation of meiosis, but were required for essential steps in the middle stages of meiosis. Twelve mutants were able to induce IME1 and to divide meiotically, but were blocked for spore wall formation. Note that three of the mutants (deletions of YOR338W, YAL068C, and YNL296W) failed to induce detectable expression of IME1 yet still showed high levels of nuclear division. Since IME1 expression is essential for all but the earliest stages of meiosis (MITCHELL et al. 1990 ), some low level of IME1 induction may still be occurring in these strains. Alternatively, mutations in these genes may bypass the requirement for IME1 induction. Further analysis will be required to determine what role the gene products play in activation of IME1 expression. However, it is important to note that a "0" in our assay may not represent the complete absence of IME1 expression. Higher than normal IME1 expression levels could be due to lack of IME2 induction, which is known to downregulate IME1 (SMITH and MITCHELL 1989 ). Four of the mutants, deletions of YAL035W, YNL170W, YPL183W-A, and YBL100C, did not give clear and reproducible IME1 expression profiles.

View larger version (82K): In this window In a new window Download PPT slide   Figure 2. Examples of IME1 induction. Top, selected mutants transformed with an IME1-lacZ reporter. MCK1 is required for IME1 induction and shows the phenotype expected if IME1 expression is absent. IME1 and IME2 are required for later steps of meiosis, after IME1 induction. These mutants show the expression expected of meiotic mutants that are able to induce IME1 to wild-type levels. YAL061W is a representative essential sporulation gene and the deletion mutant shows an IME1 expression phenotype consistent with an arrest after IME1 induction. Bottom, various examples of IME1 expression phenotypes found among the tested strains.

On the basis of these phenotypes, we have named 14 of the most essential sporulation loci. The genes required for IME1 induction were named EMI1-5 (Early Meiotic Induction). Those that were not required for entry into meiosis, but were required for meiotic nuclear division, were named RMD1-12 (Required for Meiotic nuclear Division). Genes essential for spore wall formation were named SWF1 (Spore Wall Formation) and SWF5.

	DISCUSSION

TOP ABSTRACT MATERIALS AND METHODS RESULTS DISCUSSION LITERATURE CITED

The sporulation genes of S. cerevisiae:
We have used a collection of 4323 single-gene deletion mutants in nonessential genes to identify 334 sporulation-essential genes. Included within this group of genes are 140 previously identified meiosis/sporulation genes and 78 novel or unpublished ORFs. We also found that 115 genes that had been previously characterized, but did not show a sporulation phenotype on the SGD or YPD, were actually essential for full sporulation. For some of these genes the sporulation defect will be a novel finding, and in others it may be known but not included as a phenotype in these databases. Fifty-three genes were not screened because of weak growth or apparent haploidy. Most of the remaining 1944 genes and predicted ORFs in the S. cerevisiae genome are essential for life and by definition are required for steps other than the meiotic sporulation pathway. Thus, we believe the data presented here define a nearly complete collection of sporulation genes in this pathway. This represents one of the first large-scale genetic characterizations of a complex cellular pathway in a eukaryotic organism.

That being said, many questions concerning the function of these genes obviously remain, including where in the sporulation pathway they exert their influence and in what order the gene products are active. We and others are in the process of applying functional genomic techniques to further define the role of these gene products in the sporulation/meiotic pathway. It is also important to emphasize that many genes essential for a normal meiotic sporulation pathway, including inhibition of sporulation in rich medium, recombination and accurate chromosome segregation, complete spore wall synthesis, and formation of viable spores, may not be included in this screen since these defects often do not block the formation of visible spores.

Many of the novel ORFs overlap with the sequences of known genes or other ORFs in the genome. Since overlapping functional genes are very rare, or nonexistent, in S. cerevisiae, we assume that the phenotypes may represent the loss of function of the other gene or ORF found on the opposite DNA strand. In some cases, the overlapping gene is known or suspected to be required for sporulation. Thus, we suggest that the phenotypes due to loss of the ORFs YBL100C, YGR226C, YJL175W, YKL118W, YMR158W-A, YOR135C, YOR199W, YOR333C, YLR235C, YPL205C, and YPR053C are most likely due to loss of portions of the known meiotic genes on the opposite strand of DNA, which overlap with these hypothetical ORFs. In other cases, the overlapping gene has not been shown to be required for sporulation, and in some cases the deletion of the overlapping ORF in our hands does not cause sporulation defects. Thus, while deletions of the ORFs YBR174C, YDL041W, YGL218W, YLR358C, YML009W-B, YMR306C-A, YNL170W, YNL296W, YDR417C, YDR442W, and YDR455C all give a sporulation phenotype, further analysis will be required to determine if these sequences represent functional genes. In addition, a few of the deleted ORFs overlap with larger, uncharacterized ORFs. We suggest that the phenotypes from deletion of YDL118W, YLR269C, and YDR048C may be due to the loss of the larger overlapping ORFs on the opposite DNA strand, YDL119C, YLR270W, and YDR049W, respectively. In each case, further analysis will be required to define the functional transcriptional unit in the case of these overlapping ORFs.

Functional classes for the sporulation genes:
Fig 3 shows how the previously characterized genes (red and green in Fig 1) fit into functional groups. Classification was made by analysis of mutant phenotype at the SGD and YPD and selective review of the literature. Fig 3A shows 252 known genes, which we found to be required for full sporulation. Surprisingly, only 17% of the identified genes appear to be primarily specialized for meiosis/sporulation, having little or no known function during vegetative growth. Thus, while meiosis and sporulation have many specialized features, only a small minority of the genes required for this transformation have functions unique to meiosis. Nearly equally well represented are genes primarily involved in vacuolar function (15%) and transcription (13%). The significance of both of these categories is expected. The vacuole is important for survival under the starvation conditions required to induce sporulation (VAN DEN HAZEL et al. 1996 ). During sporulation, 1000 genes show increased or decreased expression (CHU et al. 1998 ; PRIMIG et al. 2000 ), most likely explaining the large number of transcription-related proteins we find important for sporulation.

View larger version (64K): In this window In a new window Download PPT slide   Figure 3. Pie charts showing functional groupings of the genes required for sporulation in S. cerevisiae. Data include both known sporulation genes and previously characterized genes not described as essential for sporulation in the SGD and YPD phenotypic descriptions. (A) Genes that are required for full sporulation but not for life or growth on nonfermentable carbon sources. (B) Genes that are essential for even low levels of sporulation.

Fig 3B shows the 124 genes whose absence produces a severe sporulation defect (mutants with none or very low phenotypes in Fig 1). Again, a majority are involved in vacuolar function (23%) or transcription (15%) or are specialized for meiosis/sporulation (23%). Genes primarily functioning in the secretory pathway (10%) also become prominent. This may reflect the essential role of this pathway in prospore membrane synthesis when membrane vesicles are targeted to the growing tips. A notable difference between these groups is that genes involved primarily in protein synthesis or RNA processing, while making up 8.0% of the genes required for full sporulation, are rarely essential for sporulation (0.88%). We can add 78 ORFs to the list of genes essential for full sporulation. A majority of the novel ORFs are required between the beginning of meiotic induction and nuclear division. Others play roles in the initiation of meiosis and spore wall formation. Additional analysis will be required to narrow the mutant phenotypes further.

Efficiency and accuracy of this screen:
There are two general issues to consider regarding the usefulness of this type of large-scale functional genomic screen: the efficiency and the accuracy. The screens were performed blind with the identity of the deleted ORFs initially distinguished only by their position on the microtiter plate. Only after the phenotypes were established was the identity of the locus known by the investigator. By determining our success at finding known genes required for sporulation we can estimate our efficiency at identifying unknown sporulation genes. To make this comparison we searched the SGD for genes previously shown to be required for sporulation and verified the phenotypes in the appropriate literature references. Seventy-seven sporulation-essential phenotypic citations were found. Of these 77 mutants, we found 68 to be sporulation defective in our double-blind screen. For the remaining 9 mutants, a sporulation deficiency was reported previously that was inconsistent with the data presented here. Null mutations in two genes (RIM9 and LIF1) are reported to reduce sporulation to between 47 and 67% of wild-type levels (KLAPHOLZ et al. 1985 ; LI and MITCHELL 1997 ; HERRMANN et al. 1998 ; JIAO et al. 1999 ) but had no effect on sporulation in our hands. This relatively low reduction in sporulation may not be readily detected in our large-scale, nonquantitative assay. Also, some differences between these previous studies and ours in regard to time in sporulation medium or temperature could contribute to the differences in sporulation levels. Loss of any of seven genes (UMP1, FEN1, INO2, SDS3, GSG1, PHO2, and SNF8) was previously shown to produce severe or complete loss of sporulation (BERBEN et al. 1988 ; HAMMOND et al. 1993 ; EL-SHERBEINI and CLEMAS 1995 ; YEGHIAYAN et al. 1995 ; VANNIER et al. 1996 ; BYRD et al. 1998 ; RAMOS et al. 1998 ) but sporulated at 69% of wild-type levels in our hands. Furthermore, ubr1 mutants were reported previously to have a deficiency of four-spore tetrads (BARTEL et al. 1990 ) but were normal in our hands. Reasons for the discrepancy between our data and the established literature remain to be determined, but could include differences in strain backgrounds or experimental error. Thus, we believe our efficiency at detecting sporulation genes is 88%, including both severe and mild phenotypes.

To ensure the accuracy of our screen, we eliminated several sources of false positives within the data, including respiratory-deficient strains, poor growing strains, and haploid contaminants. We also showed that in all of the strains examined the expected locus was deleted. However, we have not shown that the phenotype was due to the deleted locus. To estimate the frequency of random false-positive errors we can ask how often we picked a known nonessential gene by mistake. For comparison we checked auxotrophic mutants in biosynthetic pathways known to be dispensable for sporulation. Twenty-six mutants with deletion of genes in the adenine, histidine, leucine, lysine, tryptophan, and uracil pathways are included in the tested collection (Table S1). None of these were found to be required for sporulation in this blind study. Thus the frequency of random false-positives is apparently <4% (1/26).

Comparison with other genomic-based screens:
The Eurofan set of 624 mutants was screened for dityrosine spore wall components and, secondarily, for sporulation and nuclear division (BRIZA et al. 2002 ). Of the 16 sporulation mutants they found that were also in our collection, we identified 12 and failed to identify 4. Deletions of genes SEP7/SHS1, PHO91, YDL121C, and YJR003C had little or no effect on sporulation in our hands. As with the examples above, further evaluations will be required to resolve these discrepancies.

RABITSCH et al. 2001 have taken a slightly different approach to identify genes primarily involved in meiotic chromosome segregation. They deleted 301 loci in genes that showed increased expression during meiosis I (CHU et al. 1998 ; PRIMIG et al. 2000 ). Eighteen were required for sporulation. Of these, 14 were in the collection we used, and all 14 were also found by us to be required for sporulation.

Comparing phenotypic and expression-based genetic screens:
As functional genomic assays like this one are completed, it will be useful to compare the genes identified as essential with the genes showing the greatest transcriptional regulation. We can compare our phenotypic analysis to the sporulation expression data of CHU et al. 1998 . These investigators have shown that 500 ORFs increase expression levels during sporulation. Do those genes that show an increase in expression represent the genes most likely to be essential? Of 78 YEPG+ mutants in our study that were blocked for sporulation, 39% of the deleted ORFs showed increased expression during sporulation in wild-type strains, 21% showed a decrease in expression, and 40% showed no change (or both an increase and a decrease) during sporulation. For comparison we checked a random sample of 50 novel ORFs whose deletion we show did not produce a sporulation defect. Fourteen percent had increased expression during meiosis, 38% had decreased expression, and 48% had no change of expression. Thus, we see 39% of essential sporulation genes, but only 14% of random sporulation-positive genes, show increased expression during sporulation. Analysis of the Eurofan collection produced a similar result, showing 35% of the sporulation-essential genes to have a sporulation-specific induction (BRIZA et al. 2002 ). So there appears to be a correlation between requirement and increased expression for sporulation.

However, there was surprisingly little total overlap between the genes essential for sporulation and those showing the most increase in expression. Of the 84 genes in our study that showed a >7-fold induction at any time during sporulation, only 21% gave a sporulation phenotype when deleted (CHU et al. 1998 ). Of the genes with a >15-fold induction, only 24% produced any sporulation phenotype when deleted. Thus, loss of highly expressing sporulation genes in most cases produced no sporulation phenotype. Perhaps most significantly, only 10.5% of the genes required for normal sporulation were found in the collection of 200 genes with >7-fold induction of expression at any time during sporulation. This comparison indicates that expression-based profiling and phenotypic analysis are yielding different sets of sporulation genes. Similarly, previous studies have shown that there is little correlation between the expression of different genes and phenotypic requirement for those same genes for growth in selected media (WINZELER et al. 1999 ; GIAEVER et al. 2002 ).

Meiosis and sporulation are complicated pathways and expression profiles may be expected to vary depending on exactly when the protein is needed. Therefore, considering expression at specific time points may produce greater overlap with the phenotypic data. RABITSCH et al. 2001 used this approach by first identifying genes upregulated at least fourfold higher during meiosis than during the first 2 hr of sporulation conditions. Additional analysis, including comparison with the expression profiles of known meiotic genes, yielded 301 candidates, but even in this case only 11% showed defects in meiosis or spore formation when deleted. This is higher than our recovery of 7.6% (327 sporulation mutants among 4323 candidates). In comparison, it seems that expression profiling will increase efficiency at a cost of reduced total yield. Most recently, the use of clustering algorithms has had the greatest success predicting function by combining data from different expression arrays. In 23 of 42 functional sets, transcriptional coregulation accurately predicted function in 30% of the test cases (WU et al. 2002 ).

In summary, data from this and other studies suggest that the phenotypic and expression-based studies will yield different sets of genes involved in sporulation and other pathways in budding yeast. Given the unique cellular processes occurring during sporulation and the well-known transcriptional activation pathways, this is an unexpected result. There are many possible explanations. For example, small subtle changes in expression may be all that are needed for regulation of some of the meiotic genes. Another possibility, suggested by the data presented here, is that most of the sporulation-essential genes function during vegetative growth and therefore are not expected to be preferentially transcribed during sporulation. Whatever the explanation, it appears that these two types of screens will yield different sets of genes involved in sporulation and other pathways.

	ACKNOWLEDGMENTS

The authors thank Anna Blumental-Perry and Gioria Simchen for the IME1-lacZ reporter plasmid and also Alex Sica for help with the nuclear division and IME1 assays. We also thank Janet Reing, Diane Hoffelder, Li Luo, Nicole Kotchey, TeoDora Dumitrescu, and Ceyda Acilan for critical reading of the manuscript. We also thank the SGD curators, especially Anand Sethuraman, for their helpful suggestions in dealing with overlapping ORFs. This work was made possible by the support of March of Dimes grant FY200C-58.

Manuscript received May 8, 2002; Accepted for publication October 10, 2002.

	LITERATURE CITED

TOP ABSTRACT MATERIALS AND METHODS RESULTS DISCUSSION LITERATURE CITED

BARTEL, B., I. WUNNING, and A. VARSHAVSKY, 1990  The recognition component of the N-end rule pathway. EMBO J. 9:3179-3189.[Medline]

BERBEN, G., M. LEGRAIN, and F. HILGER, 1988  Studies on the structure, expression and function of the yeast regulatory gene PHO2. Gene 66:307-312.[Medline]

BRIZA, P., M. BREITENBACH, A. ELLINGER, and J. SEGALL, 1990  Isolation of two developmentally regulated genes involved in spore wall maturation in Saccharomyces cerevisiae. Genes Dev. 4:1775-1789.[Abstract/Free Full Text]

BRIZA, P., E. BOGENGRUBER, A. THUR, M. RUTZLER, and M. MUNSTERKOTTER et al., 2002  Systematic analysis of sporulation phenotypes in 624 non-lethal homozygous deletion strains of Saccharomyces cerevisiae. Yeast 19:403-422.[Medline]

BYRD, C., G. C. TURNER, and A. VARSHAVSKY, 1998  The N-end rule pathway controls the import of peptides through degradation of a transcriptional repressor. EMBO J. 17:269-277.[Medline]

CHU, S., J. DERISI, M. EISEN, J. MULHOLLAND, and D. BOTSTEIN et al., 1998  The transcriptional program of sporulation in budding yeast. Science 282:699-705. (erratum: Science 282: 1421).[Abstract/Free Full Text]

DAWES, I. W. and I. D. HARDIE, 1974  Selective killing of vegetative cells in sporulated yeast cultures by exposure to diethyl ether. Mol. Gen. Genet. 131:281-289.[Medline]

EL-SHERBEINI, M. and J. A. CLEMAS, 1995  Cloning and characterization of GNS1: a Saccharomyces cerevisiae gene involved in synthesis of 1,3-beta-glucan in vitro. J. Bacteriol. 177:3227-3234.[Abstract/Free Full Text]

ESPOSITO, M. S. and R. E. ESPOSITO, 1969  Genes controlling meiosis and spore formation in yeast. Genetics 78:215.

GIAEVER, G., A. M. CHU, L. NI, C. CONNELLY, and L. RILES et al., 2002  Functional profiling of the Saccharomyces cerevisiae genome. Nature 418:387-391.[Medline]

GUTHRIE, C., and G. FINK, 1991 Guide to Yeast Genetics and Molecular Biology (Methods in Enzymology, Vol. 194). Academic Press, New York.

HAMMOND, C. L., P. ROMANO, S. ROE, and P. TONTONOZ, 1993  INO2, a regulatory gene in yeast phospholipid biosynthesis, affects nuclear segregation and bud pattern formation. Cell Mol. Biol. Res. 39:561-577.[Medline]

HERRMANN, G., T. LINDAHL, and P. SCHAR, 1998  Saccharomyces cerevisiae LIF1: a function involved in DNA double-strand break repair related to mammalian XRCC4. EMBO J. 17:4188-4198.[Medline]

JIAO, K., S. A. BULLARD, L. SALEM, and R. E. MALONE, 1999  Coordination of the initiation of recombination and the reductional division in meiosis in Saccharomyces cerevisiae. Genetics 152:117-128.[Abstract/Free Full Text]

KASSIR, Y., D. GRANOT, and G. SIMCHEN, 1988  IME1, a positive regulator gene of meiosis in S. cerevisiae. Cell 52:853-862.[Medline]

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]

KUPIEC, M., B. BYERS, R. ESPOSITO and A. MITCHELL, 1997 Meiosis and sporulation in Saccharomyces cerevisiae, pp. 889–1036 in The Molecular and Cellular Biology of the Yeast Saccharomyces cerevisiae, edited by J. PRINGLE, J. BROACH and E. JONES. Cold Spring Harbor Laboratory Press, Cold Spring Harbor, NY.

LI, W. and A. P. MITCHELL, 1997  Proteolytic activation of Rim1p, a positive regulator of yeast sporulation and invasive growth. Genetics 145:63-73.[Abstract]

LYNN, R. R. and P. T. MAGEE, 1970  Development of the spore wall during ascospore formation in Saccharomyces cerevisiae. J. Cell Biol. 44:688-692.[Free Full Text]

MALONE, R. E., 1990  Dual regulation of meiosis in yeast. Cell 61:375-378.[Medline]

MITCHELL, A. P., S. E. DRISCOLL, and H. E. SMITH, 1990  Positive control of sporulation-specific genes by the IME1 and IME2 products in Saccharomyces cerevisiae. Mol. Cell. Biol. 10:2104-2110.[Abstract/Free Full Text]

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

PRIMIG, M., R. M. WILLIAMS, E. A. WINZELER, G. G. TEVZADZE, and A. R. CONWAY et al., 2000  The core meiotic transcriptome in budding yeasts. Nat. Genet. 26:415-423.[Medline]

PRINGLE, J. R., and L. H. HARTWELL, 1981 The Saccharomyces cerevisiae cell cycle, pp. 97–142 in The Molecular Biology of the Yeast Saccharomyces: Life Cycle and Inheritance, edited by J. N. STRATHERN, E. W. JONES and J. R. BROACH. Cold Spring Harbor Laboratory Press, Cold Spring Harbor, NY.

RABITSCH, K. P., A. TOTH, M. GALOVA, A. SCHLEIFFER, and G. SCHAFFNER et al., 2001  A screen for genes required for meiosis and spore formation based on whole-genome expression. Curr. Biol. 11:1001-1009.[Medline]

RAMOS, P. C., J. HOCKENDORFF, E. S. JOHNSON, A. VARSHAVSKY, and R. J. DOHMEN, 1998  Ump1p is required for proper maturation of the 20S proteasome and becomes its substrate upon completion of the assembly. Cell 92:489-499.[Medline]

ROCKMILL, B. and G. S. ROEDER, 1988  RED1: a yeast gene required for the segregation of chromosomes during the reductional division of meiosis. Proc. Natl. Acad. Sci. USA 85:6057-6061.[Abstract/Free Full Text]

SHAH, J. C. and M. J. CLANCY, 1992  IME4, a gene that mediates MAT and nutritional control of meiosis in Saccharomyces cerevisiae. Mol. Cell. Biol. 12:1078-1086.[Abstract/Free Full Text]

SMITH, H. E. and A. P. MITCHELL, 1989  A transcriptional cascade governs entry into meiosis in Saccharomyces cerevisiae. Mol. Cell. Biol. 9:2142-2152.[Abstract/Free Full Text]

VAN DEN HAZEL, H. B., M. C. KIELLAND-BRANDT, and J. R. WINTHER, 1996  Review: biosynthesis and function of yeast vacuolar proteases. Yeast 12:1-16.[Medline]

VANNIER, D., D. BALDERES, and D. SHORE, 1996  Evidence that the transcriptional regulators SIN3 and RPD3, and a novel gene (SDS3) with similar functions, are involved in transcriptional silencing in S. cerevisiae. Genetics 144:1343-1353.[Abstract]

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]

WU, L. F., T. R. HUGHES, A. P. DAVIERWALA, M. D. ROBINSON, and R. STOUGHTON et al., 2002  Large-scale prediction of Saccharomyces cerevisiae gene function using overlapping transcriptional clusters. Nat. Genet. 31:255-265.[Medline]

YEGHIAYAN, P., J. TU, L. G. VALLIER, and M. CARLSON, 1995  Molecular analysis of the SNF8 gene of Saccharomyces cerevisiae. Yeast 11:219-224.[Medline]

This article has been cited by other articles:

				A. Auger, L. Galarneau, M. Altaf, A. Nourani, Y. Doyon, R. T. Utley, D. Cronier, S. Allard, and J. Cote Eaf1 Is the Platform for NuA4 Molecular Assembly That Evolutionarily Links Chromatin Acetylation to ATP-Dependent Exchange of Histone H2A Variants Mol. Cell. Biol., April 1, 2008; 28(7): 2257 - 2270. [Abstract] [Full Text] [PDF]

				L. Lariviere, M. Seizl, S. van Wageningen, S. Rother, L. van de Pasch, H. Feldmann, K. Strasser, S. Hahn, F. C.P. Holstege, and P. Cramer Structure-system correlation identifies a gene regulatory Mediator submodule Genes & Dev., April 1, 2008; 22(7): 872 - 877. [Abstract] [Full Text] [PDF]

				T. Iwaki, H. Iefuji, Y. Hiraga, A. Hosomi, T. Morita, Y. Giga-Hama, and K. Takegawa Multiple functions of ergosterol in the fission yeast Schizosaccharomyces pombe Microbiology, March 1, 2008; 154(3): 830 - 841. [Abstract] [Full Text] [PDF]

				Y. Suda, H. Nakanishi, E. M. Mathieson, and A. M. Neiman Alternative Modes of Organellar Segregation during Sporulation in Saccharomyces cerevisiae Eukaryot. Cell, November 1, 2007; 6(11): 2009 - 2017. [Abstract] [Full Text] [PDF]

				B. L. Pike and J. Heierhorst Mdt1 Facilitates Efficient Repair of Blocked DNA Double-Strand Breaks and Recombinational Maintenance of Telomeres Mol. Cell. Biol., September 15, 2007; 27(18): 6532 - 6545. [Abstract] [Full Text] [PDF]

				F. Chalmel, A. D. Rolland, C. Niederhauser-Wiederkehr, S. S. W. Chung, P. Demougin, A. Gattiker, J. Moore, J.-J. Patard, D. J. Wolgemuth, B. Jegou, et al. The conserved transcriptome in human and rodent male gametogenesis PNAS, May 15, 2007; 104(20): 8346 - 8351. [Abstract] [Full Text] [PDF]

				H. Nakanishi, Y. Suda, and A. M. Neiman Erv14 family cargo receptors are necessary for ER exit during sporulation in Saccharomyces cerevisiae J. Cell Sci., March 1, 2007; 120(5): 908 - 916. [Abstract] [Full Text] [PDF]

				C. Nouet, M. Bourens, O. Hlavacek, S. Marsy, C. Lemaire, and G. Dujardin Rmd9p Controls the Processing/Stability of Mitochondrial mRNAs and Its Overexpression Compensates for a Partial Deficiency of Oxa1p in Saccharomyces cerevisiae Genetics, March 1, 2007; 175(3): 1105 - 1115. [Abstract] [Full Text] [PDF]

				E. H. Williams, C. A. Butler, N. Bonnefoy, and T. D. Fox Translation Initiation in Saccharomyces cerevisiae Mitochondria: Functional Interactions Among Mitochondrial Ribosomal Protein Rsm28p, Initiation Factor 2, Methionyl-tRNA-Formyltransferase and Novel Protein Rmd9p Genetics, March 1, 2007; 175(3): 1117 - 1126. [Abstract] [Full Text] [PDF]

				X. H. Hu, M. H. Wang, T. Tan, J. R. Li, H. Yang, L. Leach, R. M. Zhang, and Z. W. Luo Genetic Dissection of Ethanol Tolerance in the Budding Yeast Saccharomyces cerevisiae Genetics, March 1, 2007; 175(3): 1479 - 1487. [Abstract] [Full Text] [PDF]

				J. A. Hill and S. P. Otto The Role of Pleiotropy in the Maintenance of Sex in Yeast Genetics, March 1, 2007; 175(3): 1419 - 1427. [Abstract] [Full Text] [PDF]

				S. Komili and F. P. Roth Genetic interaction screens advance in reverse Genes & Dev., January 15, 2007; 21(2): 137 - 142. [Full Text] [PDF]

				A. Gattiker, C. Niederhauser-Wiederkehr, J. Moore, L. Hermida, and M. Primig The GermOnline cross-species systems browser provides comprehensive information on genes and gene products relevant for sexual reproduction Nucleic Acids Res., January 12, 2007; 35(suppl_1): D457 - D462. [Abstract] [Full Text] [PDF]

				J. Li, S. Agarwal, and G. S. Roeder SSP2 and OSW1, Two Sporulation-Specific Genes Involved in Spore Morphogenesis in Saccharomyces cerevisiae Genetics, January 1, 2007; 175(1): 143 - 154. [Abstract] [Full Text] [PDF]

				A. Gonzalez, A. Ruiz, R. Serrano, J. Arino, and A. Casamayor Transcriptional Profiling of the Protein Phosphatase 2C Family in Yeast Provides Insights into the Unique Functional Roles of Ptc1 J. Biol. Chem., November 17, 2006; 281(46): 35057 - 35069. [Abstract] [Full Text] [PDF]

				J. P. Gerke, C. T. L. Chen, and B. A. Cohen Natural Isolates of Saccharomyces cerevisiae Display Complex Genetic Variation in Sporulation Efficiency Genetics, October 1, 2006; 174(2): 985 - 997. [Abstract] [Full Text] [PDF]

				K. L. Auld, A. L. Hitchcock, H. K. Doherty, S. Frietze, L. S. Huang, and P. A. Silver The Conserved ATPase Get3/Arr4 Modulates the Activity of Membrane-Associated Proteins in Saccharomyces cerevisiae Genetics, September 1, 2006; 174(1): 215 - 227. [Abstract] [Full Text] [PDF]

				G. Lesage and H. Bussey Cell Wall Assembly in Saccharomyces cerevisiae Microbiol. Mol. Biol. Rev., June 1, 2006; 70(2): 317 - 343. [Abstract] [Full Text] [PDF]

				P. M. Kane The Where, When, and How of Organelle Acidification by the Yeast Vacuolar H+-ATPase Microbiol. Mol. Biol. Rev., March 1, 2006; 70(1): 177 - 191. [Abstract] [Full Text] [PDF]

				J. E. Connolly and J. Engebrecht The Arf-GTPase-Activating Protein Gcs1p Is Essential for Sporulation and Regulates the Phospholipase D Spo14p Eukaryot. Cell, January 1, 2006; 5(1): 112 - 124. [Abstract] [Full Text] [PDF]

				A. M. Neiman Ascospore Formation in the Yeast Saccharomyces cerevisiae Microbiol. Mol. Biol. Rev., December 1, 2005; 69(4): 565 - 584. [Abstract] [Full Text] [PDF]

				C. M. McDonald, K. F. Cooper, and E. Winter The Ama1-Directed Anaphase-Promoting Complex Regulates the Smk1 Mitogen-Activated Protein Kinase During Meiosis in Yeast Genetics, November 1, 2005; 171(3): 901 - 911. [Abstract] [Full Text] [PDF]