Ume6p is required for germination and early colony development of yeast ascospores


  • Editor: Terrance Cooper

  • Present address: Michael J. Mallory, Department of Biochemistry and Biophysics, University of Pennsylvania, Philadelphia, PA 19104, USA.

Correspondence: Randy Strich, UMDNJ-School of Osteopathic Medicine, Two Medical Center Drive, Stratford, NJ 08055, USA. Tel.: +1 856 566 6043; fax: +1 856 566 6366; e-mail:


Ume6p is a nonessential transcription factor that represses meiotic gene expression during vegetative growth in budding yeast. To relieve this repression, Ume6p is destroyed as cells enter meiosis and is not resynthesized until spore wall assembly. The present study reveals that spores derived from a ume6 null homozygous diploid fail to germinate. In addition, mutant spores from a UME6/ume6 heterozygote exhibited reduced germination efficiency compared with their wild-type sister spores. Analysis of ume6 spore colonies that did germinate revealed that the majority of cells in microcolonies following the first few cell divisions were inviable. As the colony developed, the viability percentage increased and achieved wild-type levels within approximately six cell divisions, indicating that the requirement for Ume6p in cell viability is transient. This function is specific for germinating spores as Ume6p has no or only a modest impact on the return to the growth ability of cells arrested at other points in the cell cycle. These results define a new role for Ume6p in spore germination and the first few subsequent mitotic cell divisions.


When deprived of nitrogen and a fermentable carbon source, the budding yeast Saccharomyces cerevisiae will withdraw from the cell cycle and initiate meiosis and spore morphogenesis (Honigberg et al., 1993). The spores remain quiescent until exposed to the appropriate growth signals, at which time they germinate and re-enter the mitotic cell cycle. However, spores evaluate growth signals differently than do mitotically dividing cells. Germination and subsequent cell division will occur optimally in the presence of specific growth stimuli such as glucose and ammonia. Other carbon sources that normally promote cell division in vegetative cells (e.g. lactose, pyruvate, ethanol) will not support germination or do so at lower efficiencies (Palleroni, 1961; Sussman & Halvorson, 1966; Rousseau & Halvorson, 1973; Donnini et al., 1986). These results suggest the existence of a separate (or modified) regulatory system directing the decision of a resting spore to re-enter cell division than the one controlling a vegetative cell.

Currently, few details are known concerning the molecular mechanisms governing spore quiescence or germination. Several studies have suggested that germination and return to growth is a complex, multistep process (Herman & Rine, 1997). For example, the growth arrest in yeast ascospores does not appear to be simply the response to the lack of nutrients that induced sporulation (Miller, 1989). The resting spore contains ample sources of carbon (trehalose, glycogen) and nitrogen (free amino acid pools) that can maintain vegetative cell growth, but will not stimulate germination. Initially, germination is marked by the breakdown of the specialized spore wall and storage carbohydrates (reviewed in Thevelein, 1984). Initially, translation was thought to occur before mRNA synthesis (Rousseau & Halvorson, 1973). However, more recent studies have found a robust transcription program that includes both mRNA synthesis (Joseph-Strauss et al., 2007) and degradation (Brengues et al., 2002). These studies revealed that spore germination is a dynamic process involving the activity of transcriptional activators, repressors and the mRNA decay machinery.

Ume6p is a transcription factor that represses early meiotic genes (EMG) required for the meiotic S phase and recombination (Bowdish & Mitchell, 1993; Strich et al., 1994). Although Ume6p is not required for mitotic cell division, it is required for meiosis with 95% of null mutants arresting before the first meiotic division (Strich et al., 1994; Steber & Esposito, 1995). UME6 mRNA levels do not vary during meiosis and spore formation (Strich et al., 1994; Primig et al., 2000). However, to relieve EMG repression, Ume6p is destroyed as cells enter the meiotic program via ubiquitin-mediated proteolysis (Mallory et al., 2007). The Ume6p levels remain below the limits of detection during meiosis, but revert to mitotic levels during spore wall assembly. This report identifies a role for Ume6p in spore germination and the first few divisions following germination that is not present in established mitotic cultures.

Materials and methods

Yeast strains, media and plasmids

The yeast strains used in this study are listed in Table 1. Yeast strains were grown and induced to enter meiosis using a solid and a liquid medium as described previously (Cooper et al., 2000). The ume6-5 (ume6LEU2) and ume6Δ (ume6URA3) mutant alleles were described previously (Strich et al., 1994; Mallory et al., 2007). Both ume6Δ and ume6-5 mutant strains exhibited identical phenotypes with respect to EMG repression and sporulation efficiency. To place UME6 under the control of the CLB2 promoter, a 1 kbp BglII–PacI fragment containing the CLB2 promoter was inserted into the same sites of pFA6a-TRP1-pGAL (Longtine et al., 1998), replacing the GAL1 promoter. Oligonucleotides flanking the UME6 promoter were used to amplify the CLB2 promoter and the resulting fragment was used to replace the UME6 promoter using one-step transplacement. The correct insertion was verified by sequencing the amplified genomic sequence.

Table 1.   Yeast strains
  • *

    Alleles are homozygous in diploid strains unless stated otherwise.

JX192*MATa/MATαcan1-100 his4-519 leu2-3,112 lys2-1 trp1-1 ura3-1This study
JX193*MATa/MATαcan1-10 his4-519 leu2-3,11 lys2-1 trp1-1 ura3-1 UME6/ume6-5This study
RSY194*MATa/MATαcan1-100 his4-519 leu2-3,112 lys2-1 trp1-1 ura3 ume6-5This study
JX225*MATa/MATαade2/ADE2 ADE6/ade6 can1-100/CAN1 his3-11,15 leu2-3,112 lys2-1/LYS2 trp1-1 ura3 ume6Δ:URA3/UME6This study
RSY269MATαcan1-100 his4-519 leu2-3,11 trp1-1 ura3-1This study
RSY270MATacan1-100 his4-519 leu2-3,11 trp1-1 ura3 ume6-5This study
RSY271MATacan1-100 his4-519 leu2-3,11 trp1-1 ura3-1This study
RSY280MATαcan1-100 his4-519 leu2-3,11 trp1-1 ura3 ume6-5This study
JX414*MATa/MATαhoLYS2 lys2-1 ura3 ume6Δ∷URA3/UME6This study
RSY291MATaade2 ade6 can1-100 his3-11,15 leu2-3,11 trp1-1 ura3-1 ume6LEU2This study
RSY1295MATαade6 can1-100 his4-519 leu2-3,11 trp1-1 ura3 CLB2pro-UME6TRP1This study

Microcolony viability studies

Tetrads were dissected onto a rich medium and incubated for 24 h at 30 °C. The number of divisions the spore had accomplished was then determined by counting the number of cells in the microcolony (assuming exponential growth). For microcolonies with >20 cells, the number of generations was estimated. To determine the viability of individual cells within the microcolony, at least eight cells were manipulated away from the colony on the same plate. Following another 24-h incubation, the ability of the micromanipulated cells to continue cell division was determined microscopically. The genotype of each spore that formed a macrocolony was determined by replica plating onto medium selecting for the presence of the ume6-5 allele.

Return to growth assays

For nutritionally arrested cells, three independent isolates of RSY270 and RSY271 were maintained at the stationary phase in a liquid culture for 3 days at 30 °C. Cells were harvested by centrifugation, washed in water, lightly sonicated to disrupt clumps and counted using a hemocytometer. At least 300 cells from each isolate were plated in triplicate on a rich medium and incubated at 30 °C. Late G1 arrested cells were obtained using α-factor arrest/release protocols (Rita et al., 1991). Three independent log-phase cultures of RSY270 and RSY271 were treated with 15 μM α-factor (Sigma) for 3 h, at which time >90% of the cells exhibited the characteristic ‘shmoo’ morphology. Cells were counted and plated on rich agar plates and incubated at 30 °C. G2 arrest was accomplished by incubating log-phase cultures with 10 μM benomyl (DuPont) for 3 h at 30 °C (Guacci et al., 1997). Growth arrest was monitored both microscopically (>90% of the population containing large buds) and by OD650 nm. Cells were treated as described for the G1 arrest above. For nitrogen starvation studies, log-phase cultures (in triplicate) were grown to the mid-log phase in a complete minimal medium. The cells were harvested, washed in sterile water, then resuspended in minimal medium lacking nitrogen or amino acids and timepoints taken every 2 h for 12 h. The samples were lightly sonicated as before and then serially diluted (1 : 10) and spotted onto a rich medium. The plates were incubated for 2 days at 30 °C.


Ume6p is required for efficient spore colony formation

Ume6p negatively regulates genes (e.g. IME2, SPO13) that are involved in early meiotic processes (Bowdish & Mitchell, 1993; Strich et al., 1994). Ume6p-dependent repression is removed when it is destroyed as cells enter meiosis (Mallory et al., 2007). Although the Ume6p levels remained below the limits of detection during the meiotic nuclear divisions, they returned to vegetative concentrations as the cells formed spores. These observations suggested a possible role for Ume6p late in the sporulation process. To address this question, a homozygous diploid strain (JX194) harboring an insertion allele (ume6-5) for both copies of UME6 was sporulated on solid medium. As described previously (Strich et al., 1994), the sporulation efficiency of the mutant diploid was significantly reduced compared with the wild type (Table 2). Of the few tetrads obtained, the outer spore wall and 4′-6-diamidino-2-phenylindole staining nuclei appeared to be normal compared with wild type (data not shown). However, dissection analysis revealed that none of the spores derived from 20 ume6-5 tetrads were able to form colonies (Table 2). Microscopic examination revealed that the spores remained as single cells (data not shown). As Ume6p is not required for normal mitotic cell division (Strich et al., 1994), these results suggest that either Ume6p is required for spore germination or that the spores fail to germinate due to aberrant meioses.

Table 2.   Requirement of Ume6p for spore colony formation
Strain% Asci*Tetrads dissected% CFS total% CFS UME6% CFS ume6
  • *

    % asci was calculated as the sum of two-, three- and four-spored cells in the sporulated culture divided by the total cells counted (n≥200).

  • % colony-forming spores (CFS) were calculated by dividing the number of spore colonies observed after a 5-day incubation by the number of spores dissected.

  • % CFS for UME6 and ume6 mutant spores as determined by the presence of the ume6 disruption marker LEU2 or URA3. For spores that did not form colonies, genotypic designations were assigned assuming a 2 : 2 segregation of the UME6 locus. Numbers are normalized to 100% of each genotype (UME6 or ume6) expected from each tetrad. NA, not applicable.

JX414 (SK1)
JX225 (W303)

To address this question, the experiments were repeated with a UME6/ume6-5 heterozygote. The rationale behind this experiment is that if Ume6p levels return only at the onset of spore wall assembly, spores harboring the ume6-5 allele may possess reduced amounts of Ume6p. Tetrad analysis of the UME6/ume6-5 diploid (JX193) revealed a 33% reduction in colony-formation ability of the resulting spores compared with the wild-type control (Table 2, Fig. 1). Further examination revealed that less than half of the spores carrying the ume6-5 allele were able to form colonies. In contrast, the colony-forming ability of the UME6 spores derived from the heterozygote were similar to that observed for spores from the homozygous wild-type strain (97% vs. 100%). To control for possible strain variations, these experiments were repeated in the W303 (JX225) and SK1 (JX414) backgrounds. These studies revealed variations in ume6Δ spore viability with respect to these strains. Mutant ume6Δ spores derived from the W303 background were more likely to form colonies (45%) than from the SK1 strain (17%). However, mutant spores were still less likely to form colonies than their wild-type sister spores. These results indicate that there are strain differences with respect to the viability of ume6Δ spores. Finally, to test whether the spore colony-forming ability of ume6 mutant spores was due to the presence of an extragenic suppressor, three ume6-5 spore colonies were mated to a wild-type strain and the resulting diploids were sporulated. Tetrad dissection analysis revealed a similar loss in colony-forming ability in the ume6-5 spores as observed previously (data not shown), indicating that ume6-5 spore colony-forming ability is not due to a second site suppressor. These results indicate that Ume6p is required for normal colony-forming ability of yeast spores. Furthermore, these results indicate that the viability defect associated with the ume6 allele is spore autonomous, i.e. the viability of the wild-type spores in the same ascus was not affected.

Figure 1.

 Ume6p is required for spore colony formation. (a) Tetrad dissection of the JX193 diploid (UME6/ume6-5) following a 3-day incubation at 30°C. The panel on the right indicates wild type (+) and ume6-5 (−) mutants. Solid boxes indicate no colony formation. The tetrads (numbers) and spores (letters) are indicated. (b) Growth rates of cultures derived from UME6 spore clones 6C (JX269), 6D (JX271) and ume6-5 spore clones 6A (RSY270), 6B (RSY280).

Ume6p is required for normal spore colony development

To further investigate the role of UME6 in spore colony formation, the growth characteristics of wild type and mutant spores were compared. The UME6 spores derived from a UME6/ume6-5 heterozygote produced colonies of consistent size and shape following a 2-day incubation (Fig. 1a). However, ume6-5 spore colonies were of variable sizes. Some asci produced wild-type and mutant colonies of similar size (e.g. tetrad 14). However, most of the ume6-5 spore colonies were smaller than their wild-type sisters (e.g. tetrads 6 and 20). This reduced colony size may be explained if the individual ume6-5 mutants exhibited a slower growth rate. A previous study found that ume6 mutants in a similar strain background grew approximately 10% slower compared with the wild type (Strich et al., 1994). To test this possibility, the spore colonies from tetrad #6 were picked, colony purified and inoculated into rich liquid medium. The analysis of the growth rates for these cultures revealed generation times for the ume6-5 strains of 2.4 and 3.2 h for 6A and 6B, respectively (Fig. 1b). The wild-type control cultures (6C and 6D) doubled every 2.4 and 2.7 h. These results suggest that the differences observed in spore colony size are probably not due to a variation in the growth rates among individual spore colonies.

Ume6p does not regulate bud emergence kinetics in spores

The analysis described above indicated that growth rates were not responsible for the disparity in ume6 spore colony size. An alternative possibility is that ume6-5 spores germinate and reinitiate mitotic cell division slower than wild-type spores. This model would predict that colony size would be a function of the appearance of the first bud. To examine the return to growth kinetics of ume6-5 and UME6 spores, 11 tetrads derived from JX193 were dissected on rich medium and the spores were followed microscopically to detect the appearance of the first and second bud. No buds were observed for the first 3 h. By 4 h, buds were observed and scored. The percentage of the spores that formed their first bud at each timepoint was calculated from the total number of spores monitored. The genotypes (wild type or ume6-5) of the resulting colonies were determined by replica plating to leucine dropout medium. For spores that did not grow into colonies, their genotype was designated assuming a 2 : 2 segregation pattern. No significant difference was observed in the appearance of the first bud (Fig. 2a) or the second bud (data not shown) between the wild type and the mutant. The significance of the oscillation observed in bud emergence for both spore genotypes is not known. Next, the relationships between bud emergence and colony size were directly examined. Eight tetrads derived from JX193 were dissected and the germination status of the spores was examined microscopically 8 h later. As expected, no distinction was observed between the germination kinetics of wild type and ume6-5 spores (Fig. 2b, bottom panel). The plate was incubated for an additional 2 days and then photographed and the ume6-5 spores were identified by the presence of the LEU2 marker. Although no difference was observed in germination kinetics, ume6-5 spores still produced smaller colonies compared with the wild type (Fig. 2b, top panel). Taken together, these results indicate that the small colony phenotype observed with the ume6-5 mutation is not due to delayed germination.

Figure 2.

 Bud emergence kinetics for UME6 and ume6-5 spores. (a) Wild-type and ume6-5 mutant spores were monitored microscopically following dissection to determine when the first bud appeared for each spore. The percentage of the population that produced their initial bud for a given timepoint is plotted. For UME6, n=22, ume6-5, n=19. (b) Germination kinetics and colony size. The germination status of spores dissected from eight tetrads was examined microscopically 14 h following dissection. The number in each box indicates the number of cells found at this timepoint. ‘B’ indicates the presence of a bud. Open boxes indicate UME6 spores, shaded indicate ume6-5 spores and solid boxes indicate spores that did not form colonies.

Ume6p is required to maintain mitotic cell division following spore germination

The analysis of ume6-5 spores that could form colonies did not provide an explanation for the variation in mutant spore viability or the variation in colony size. Therefore, the studies were then focused on the ume6-5 spores that did not form macroscopic colonies. Over half of ume6-5 spores from JX193 failed to form a colony (Table 2). Microscopic examination revealed that 76% of these mutant spores had executed approximately one to four rounds of cell division (as inferred by the estimated number of cells in the microcolony), but failed to continue dividing. This observation was paradoxical to the previous finding that UME6 is a nonessential gene (Strich et al., 1994). Specifically, it was not clear why cells that were able to undergo one round of cell division could not continue to form a colony. One possibility was that Ume6p performed a function during the re-entry into mitotic cell division that was not required for established logarithmic growth. To test this possibility, the heterozygote JX193 was again dissected onto rich medium. After 24 h, eight cells from UME6 and ume6-5 microcolonies that had undergone approximately two, three, six and nine generations (as calculated by the number of cells in a microcolony assuming exponential growth) were micromanipulated away from their siblings and reincubated for 24 h. The genotype of each colony was determined later by the presence of the ume6-5 allele. Individual cells from ume6-5 microcolonies executing two or fewer cell divisions after 24 h were not able to continue growth in the six examples that were followed. These cells predominantly arrested as a single large cell or as equal-sized mothers and daughters. In addition, these mothers and daughters were easily separated, suggesting that they had undergone septation. In S. cerevisiae, the relative position in the cell cycle for an individual cell can be approximated by the presence and size of the bud (Pringle & Hartwell, 1981). Therefore, these cells may have ceased cell division in the G1 (or G0) stage of the cell cycle.

In ume6-5 microcolonies able to execute three cell divisions within 24 h, approximately 90% of the cells were unable to continue growth (closed arrows, Fig. 3a). The one cell in this example that was able to continue dividing (open arrow) eventually formed a macroscopic colony. Microcolonies that had concluded approximately six (Fig. 3b) or nine (Fig. 3c) divisions contained an increasing percentage of viable cells. As shown in Fig. 3d, a linear relationship was observed between the number of cells in a colony and the percentage of viable cells. As expected, cells derived from wild-type microcolonies were >95% viable regardless of the colony age (data not shown). These results indicate that UME6 is required for the efficient transition between resting spore to mitotic division. Moreover, these findings provide an explanation for the small colony phenotype. Colonies initially producing a high percentage of inviable cells would exhibit slower development. However, this requirement appears to be transient as the ume6-5 microcolonies able to execute approximately 10 cell divisions exhibit nearly wild-type growth rates.

Figure 3.

 Ume6p is transiently required for cell division following germination. Eight cells were micromanipulated from a ume6-5 microcolony executing approximately three (a), six (b) or nine (c) generations at 24 h after dissection. The images were taken after another 24-h incubation, when the ability of the individual cells to continue growth was determined. Closed arrowheads indicate cells that failed to continue growth; open arrowheads indicate cells that continued growth. The mother colony in (c) expanded into the manipulated single cell colonies, allowing the analysis of five manipulated cells in this example. Scale bar=30 μM. (d) The percentage of cells continuing growth was plotted vs. the number of cells in the microcolony 24 h after dissection. At least three microcolonies were assayed for each point. The bars indicate the range at each point; the data points indicate the average.

Restricting UME6 expression in ascospores reduces viability

Our data suggest that Ume6p is required for spore germination. However, the sporulated UME6/ume6-5 diploid reduced, but did not eliminate, the ability of the ume6-5 spore to form colonies. These results suggest two possibilities. The loss of Ume6p function in a spore may result in partial loss in germination ability. Alternatively, the presence of the wild-type allele may allow Ume6p to be present in a genetically null spore. The amount of Ume6p packaged in the ume6Δ spore would then influence the germination ability of the spore. To test these models, we took advantage of the finding that the B-type cyclin CLB2 is transcribed poorly during meiosis (Primig et al., 2000), but is induced shortly after the exposure of spores to glucose (Joseph-Strauss et al., 2007). Therefore, we placed genomic UME6 under the control of the CLB2 promoter (see Materials and methods). This construct was functional as determined by its ability to repress a spo13-lacZ reporter gene during vegetative growth (Fig. 4a). This strain was mated to a ume6-5 strain and the subsequent diploid was sporulated in either liquid or solid medium. After 24 h, the liquid sporulation cultures were examined and found to contain 54% asci, comparable to the W303 background (66%, Table 2) from which it was derived. Tetrads were dissected (N=33) and viable spore colonies were assayed for the presence of the CLB2pro-UME6 (Trp+) or ume6-5 (Leu+) alleles by replica plating. We obtained 77% of the expected number of Trp+ spore colonies (see Fig. 4b for the representative plate) similar to the percentage observed for UME6 spores in the W303 heterozygote (Table 2). However, only 16% of the ume6-5 spores derived from the CLB2pro-UME6/ume6-5 heterozygote formed colonies. This value is threefold lower than the ume6Δ spores from the W303 heterozygote (45%, Table 2). These findings indicate that the reduction of Ume6p levels in sporulating cells has a negative effect on ume6Δ mutant spore colony formation.

Figure 4.

CLB2pro-UME6 promotes germination. (a) Haploid UME6 (RSY269), ume6-5 (RSY291) and CLB2pro-UME6 (RSY1295) strains transformed with pBW2 harboring a spo13-lacZ reporter gene were assayed for β-galactosidase expression using X-gal containing top agar (see Materials and methods). X-gal cleavage (dark color) indicates β-galactosidase expression. (b) Tetrads resulting from sporulated CLB2pro-UME6/ume6-5 diploid (RSY1295 × RSY291) were dissected onto rich medium and incubated for 3 days at 30°C. The left panel shows an image of a representative plate, with tetrads (numbers) and spores (letters) indicated. Subsequent genotyping of spore colonies is indicated on the right. +, CLB2proUME6; Δ, ume6-5. Closed squares indicate spores that did not form colonies.

Ume6p is required for efficient return to growth from stationary-phase arrest

The findings presented above suggest that Ume6p is required for the successful transition from quiescent spore to mitotic cell division. To determine whether this finding represents a general requirement for Ume6p following arrest at other points in the cell cycle, wild-type (RSY271) and ume6-5 (RSY270) haploid strains were subjected to growth arrest in the early G1, late G1 or G2 stages of the cell cycle (see Materials and methods for details). Early G1 (or G0) arrest was accomplished through nutrient deprivation by growing triplicate cultures to saturation density in rich liquid medium and maintaining these cultures at 30 °C for 3 days. Haploids were used to prevent the cultures from entering meiosis under these starvation conditions. The cells were harvested, counted and their viability was assayed by plating onto rich medium. These experiments revealed that ume6-5 mutants exhibited a twofold reduction in their plating efficiency (Table 3). To determine whether this loss in plating efficiency was dependent on the cells being growth arrested or whether ume6-5 cultures always contain a significant subpopulation of inviable cells, the plating efficiency of log-phase RSY271 and RSY270 cultures was determined. No difference was observed in the colony-forming ability between these strains (Table 3), indicating that cell cycle arrest is required for the observed return to growth defect. These results suggest that Ume6p is required for the efficient return to the growth of starvation-induced G0-arrested cells.

Table 3.   Requirement of Ume6p for return to growth following cell cycle arrest
StrainLog phase (% CFU)*Stationary phase (% CFU)α-Factor arrest (% CFU)Benomyl arrest (% CFU)
3 days100 days
  • *

    The log phase represents 5.6–6.5 × 106 cells mL−1 in rich liquid medium. The % CFU was calculated as the number of colonies observed divided by the number of cells plated as determined by direct counting.

  • The stationary phase indicates cells grown to a density of 2–3 × 108 cells mL−1 in rich liquid medium, followed by an additional 2-day incubation at 30°C. Percentages presented are averages from three independent cultures plated in triplicate. The 100-day arrest was conducted by placing the 3-day stationary culture at 4°C for an additional 97 days.

92 ± 1.153 ± 331 ± 9100 ± 733 ± 16
92 ± 3.181 ± 165 ± 1498 ± 521 ± 2

Reduced plating efficiency following nutritional deprivation can either be due to the failure to arrest properly, the inability to re-enter cell division or both. For example, cells carrying constitutively activated RAS2val19 alleles do not arrest correctly and exhibit reduced viability upon starvation (Tatchell et al., 1985). Two phenotypes are commonly associated with an aberrant growth arrest phenotype. First, these cells will stop dividing randomly throughout the cell cycle (Toda et al., 1985). Normally, cells arrest growth in G1 in response to nutrient deprivation. Second, prolonged maintenance in a growth-arrested state increasingly reduces viability compared with wild-type cells. However, growth-arrested ume6-5 cells are predominantly large and unbudded, indicative of G1 (data not shown). Moreover, the plating efficiency of ume6-5 cells held for 100 days at 4 °C showed only a modest reduction in viability (31%) compared with the wild type (65%, Table 3). In addition, starving ume6Δ mutant cultures for nitrogen alone for 12 h did not adversely affect cell viability compared with the wild-type control (Fig. 5). These results suggest that ume6 mutants are most likely defective in their ability to return to growth, not growth arrest.

Figure 5.

 Ume6p is not required for viability following nitrogen starvation. Haploid UME6 (RSY269) and ume6-5 (RSY270) strains were grown to mid-log phase in minimal complete medium. The cultures were harvested, washed and transferred to minimal medium lacking a nitrogen source. At the times indicated after the shift to a nitrogen-depleted medium, cells were taken, serially diluted (1 : 10) and plated on rich solid medium. The plates were incubated for 2 days at 30°C and then the image was collected.

Ume6p is not required to reinitiate the cell cycle following late G1 or G2 arrest

To examine the requirement of Ume6p for cells to continue cell division following blocks at other points in the cell cycle, triplicate wild-type (RSY271) and ume6-5 (RSY270) log-phase cultures were arrested in late G1 with the mating pheromone α-factor or in G2 with benomyl, an inhibitor of microtubule assembly. Growth arrest was monitored microscopically for the expected cell morphology (i.e. shmoo formation for α-factor arrest, large budded cells for the benomyl block; see Materials and methods for details). Arrested cultures were plated on a rich solid medium lacking either compound. No significant differences in the plating efficiency were observed between the ume6-5 and the wild-type strains (Table 3). Both strains exhibited a significant, but equal reduction in the plating efficiency following benomyl arrest due to the toxic effects of this drug. These results indicate that Ume6p is not generally required for cells to resume mitotic cell division following cell cycle arrest, but instead is specific for cells returning to the cell cycle upon nutrient stimulation.


The requirement for Ume6p in spore colony formation appears to be different from those generally observed in other studies. Three factors have been described that influence spore viability. First, mutations that disrupt mitotic cell division prevent spore colony formation. However, loss of Ume6p function produces only a modest growth defect in established cultures. Second, mutations causing spore wall defects reduce viability when the spore is challenged with enzymatic digestion or other cellular stresses (Briza et al., 1990; Krisak et al., 1994). The finding that ume6-5 mutant spores can undergo two to three rounds of division before arresting suggests that the spores are intact. In addition, spore colony formation did not occur even when intact tetrads from ume6/ume6 diploids were examined. Finally, spore viability is reduced due to aberrant chromosome segregation during meiosis (Shonn et al., 2000). A previous study found that recombination/gene conversion at one locus was reduced about fourfold in a ume6 mutant (Steber & Esposito, 1995). Therefore, the loss of viability in spores derived from the ume6/ume6 diploid could be attributed (at least in part) to the recombination defect. However, this explanation does not account for the observation that ume6 spores derived from the heterozygote also display a two- to fivefold reduction in colony-forming ability. Because the ume6-5 allele is recessive, the presence of wild-type UME6 should promote normal development. This idea is consistent with the finding that no reduction in sporulation levels (Table 2) or kinetics (data not shown) is observed in heterozygous diploids compared with homozygous wild-type controls. Second, if problems did occur during meiosis due to haplo-insufficiency in the heterozygote (e.g. aberrant chromosome segregation), the wild-type spores should also exhibit a reduced viability because a defect in meiosis would not be spore autonomous. However, no differences in viability were observed in UME6 spores derived from either UME6/ume6-5 or UME6/UME6 diploids. These results argue that Ume6p functions during the transition between spore quiescence and the establishment of normal logarithmic growth.

Because Ume6p is not required for vegetative growth, it is not clear why newly germinated ume6 mutants were unable to continue growth even following an initial round of cell division (see Fig. 4). A previous study identified genes that were repressed during spore germination (Joseph-Strauss et al., 2007). Many of these genes were involved in oxidative phosphorylation that were repressed as spores germinated on dextrose medium. In addition, components of the 26S proteasome were also downregulated. These genes could be potential targets of Ume6p-dependent repression. Comparison of the list of downregulated genes with those loci repressed by Ume6p (Williams et al., 2002) revealed only three overlaps in these data sets. One is Pre10p, a component of the 26S proteasome. It could be envisioned that aberrant upregulation of proteasome activity could have a detrimental impact on cell division. However, the remaining proteasome subunits are not regulated by Ume6p and are therefore still repressed upon germination. Therefore, it seems unlikely that aberrant expression of one subunit of this complex is important when the remaining subunits are still repressed. The other two genes are Sol4p, a 6-phosphogluconolactonase that is part of the pentose-phosphate shunt, and Cox5A, a component of cytochrome c oxidase. The derepression of these genes would not be expected to induce growth arrest on a rich glucose medium. In addition, other genes repressed by Ume6p (e.g. CAR1, CAR2, INO1) were not downregulated as spores germinated. Therefore, no potential targets of Ume6p repression that would regulate spore germination were identified.

A similar defect in spore germination, but not mitotic cell division, was observed, with mutants lacking the ubiquitin-conjugating enzyme Ubc1p (Seufert et al., 1990). Ubc1p is involved in endoplasmic reticulum-associated protein degradation in response to stress (for a review, see Kostova et al., 2007). There are several similarities between Ume6p and Ubc1p. First, UBC1 mRNA is induced late in meiosis (Cho et al., 1998), coincident with the return of Ume6p levels. In addition, both Ubc1p and Ume6p exert negative regulation through protein destruction and transcriptional repression, respectively. As discussed earlier, yeast spores decide to re-enter the cell cycle based on the quality of the environmental nutrients. To control this decision, a regulatory system may be in place to prevent germination in response to suboptimal conditions. Therefore, one possible role for Ume6p and Ubc1p would be to downregulate this pathway(s) in response to rich growth signals. Failure to do so may divert the cell back into a spore-like quiescence even in the presence of optimal growth conditions. This model is consistent with the role of Ume6p in repressing genes involved in arginine catabolism (CAR1 and CAR2) in the presence of a rich nitrogen source (Park et al., 1992; Strich et al., 1994). Joseph-Strauss et al. (2007) proposed a two-step process for spore germination. The first step, triggered by dextrose, initiates changes in the spore wall and subcellular localization of septins. The second step requires the presence of nitrogen and allows transition into the mitotic cell cycle. This model would be consistent with Ume6p functioning in the second step to promote late germination events and re-entry into the mitotic cell cycle.


Ume6p is a nonessential protein that represses the transcription of diverse loci including EMG. In this report, we find that Ume6p promotes the colony formation of germinating yeast ascospores. However, this requirement is transient as no significant effect on cell viability is observed in established (>10 generations) ume6 mutant cultures. Finally, this role is specific for spores re-entering mitotic cell division as Ume6p is not required for efficient return to growth of cells arrested in G1 or G2, although a small reduction in viability is observed in cells arrested for an extended period of time in G0. Our findings reveal a new role for Ume6p in the transition of a resting spore to mitotic cell division.


The authors thank K.F. Cooper and M. Law for critical reading and helpful discussions. This work was supported in part by grants from the National Institutes of Health General Medicine (GM086788) and the National Cancer Institute (CA099003).