Genetics, Vol. 159, 77-89, September 2001, Copyright © 2001
The Ras/PKA Signaling Pathway of Saccharomyces cerevisiae Exhibits a Functional Interaction With the Sin4p Complex of the RNA Polymerase II Holoenzyme
Susie C. Howarda,
Ya-Wen Changb,
Yelena V. Budovskayaa, and
Paul K. Hermana,b
a Department of Molecular Genetics, The Ohio State University, Columbus, Ohio 43210
b Program in Molecular, Cellular and Developmental Biology, The Ohio State University, Columbus, Ohio 43210
Corresponding author:
Paul K. Herman, Department of Molecular Genetics, The Ohio State University, 484 W. Twelfth Ave., Room 984, Columbus, OH 43210., herman.81{at}osu.edu (E-mail)
Communicating editor: F. WINSTON
 | ABSTRACT |
|---|
Saccharomyces cerevisiae cells enter into the G0-like resting state, stationary phase, in response to specific types of nutrient limitation. We have initiated a genetic analysis of this resting state and have identified a collection of rye mutants that exhibit a defective transcriptional response to nutrient deprivation. These transcriptional defects appear to disrupt the control of normal growth because the rye mutants are unable to enter into a normal stationary phase upon nutrient deprivation. In this study, we examined the mutants in the rye1 complementation group and found that rye1 mutants were also defective for stationary phase entry. Interestingly, the RYE1 gene was found to be identical to SIN4, a gene that encodes a component of the yeast Mediator complex within the RNA polymerase II holoenzyme. Moreover, mutations that affected proteins within the Sin4p module of the Mediator exhibited specific genetic interactions with the Ras protein signaling pathway. For example, mutations that elevated the levels of Ras signaling, like RAS2val19, were synthetic lethal with sin4. In all, our data suggest that specific proteins within the RNA polymerase II holoenzyme might be targets of signal transduction pathways that are responsible for coordinating gene expression with cell growth.
UPON nutrient deprivation, Saccharomyces cerevisiae cells cease mitotic division and can enter into a nondividing resting state, known as stationary phase (WERNER-WASHBURNE et al. 1993 
, WERNER-WASHBURNE et al. 1996 ). During this transition, yeast cells undergo a significant change in their overall physiology that results in an elevated resistance to a number of environmental stresses, including prolonged starvation and heat shock (WERNER-WASHBURNE et al. 1993 , WERNER-WASHBURNE et al. 1996 ). In addition, the stationary phase cell exhibits a markedly reduced rate of cellular metabolism. The rate of protein translation decreases more than 200-fold, whereas the total level of mRNA is reduced at least 35-fold (BOUCHERIE 1985 ; CHODER 1991 ; WERNER-WASHBURNE et al. 1996 ). Despite this general trend, several proteins do increase in relative abundance during stationary phase and are likely responsible for many of the properties associated with this resting state (WERNER-WASHBURNE et al. 1993 ; PADILLA et al. 1998 ).
The Ras protein signaling pathway appears to be a key regulator of stationary phase entry, as mutations that inactivate this pathway result in a constitutive stationary phase-like arrest (MATSUMOTO et al. 1983 ; IIDA and YAHARA 1984 ; BROACH 1991 ). In contrast, elevated levels of Ras signaling prevent the acquisition of stationary phase characteristics upon nutrient deprivation (TODA et al. 1985 ). Although the S. cerevisiae Ras proteins have multiple effectors (MORISHITA et al. 1995 ; MOSCH et al. 1996 ), the pathway involving cAMP and the cAMP-dependent protein kinase (PKA) is the most important for these effects on stationary phase biology (WERNER-WASHBURNE et al. 1993 ). The yeast Ras proteins, Ras1p and Ras2p, bind directly to adenylyl cyclase, Cyr1p, and stimulate the production of cAMP (FIELD et al. 1990 ; SUZUKI et al. 1990 ). This, in turn, results in elevated levels of PKA activity and the increased phosphorylation of proteins presumably important for cell proliferation (TODA et al. 1987B ; BROACH 1991 ). Although several PKA substrates have been characterized, the identification of Ras/PKA targets relevant for growth control remains an area of keen interest (REINDERS et al. 1998 ; THEVELEIN and DE WINDE 1999 ).
The entry into stationary phase is accompanied by broad changes in the patterns of gene expression that are controlled, in part, by the Ras/PKA pathway (WERNER-WASHBURNE et al. 1993 ; DERISI et al. 1997 ). However, it is not yet known precisely how Ras activity affects the transcriptional apparatus. In S. cerevisiae, as in other eukaryotes, RNA polymerase (pol) II is present as a large holoenzyme complex that contains the 12-subunit polymerase, the Mediator coactivator complex, the Srb8-11 protein complex, and several general transcription factors (KOLESKE and YOUNG 1995 ; LEE and YOUNG 2000 ; MYERS and KORNBERG 2000 ). This holoenzyme is actively recruited to promoters in vivo as a result of specific interactions between Mediator subunits and DNA-bound transactivators (PTASHNE and GANN 1997 ; KEAVENEY and STRUHL 1998 ). Therefore, there are two a priori targets for the Ras effects on RNA pol II activity: the various transcription factors bound at the individual promoters and the regulatory proteins associated with the RNA pol II holoenzyme. Indeed, several studies suggest that the Ras pathway regulates the activity of specific transcriptional regulators, like Msn2p and Msn4p (GORNER et al. 1998 ; THEVELEIN and DE WINDE 1999 ). However, to date, there have been few reports of signaling pathways directly targeting components within the RNA pol II holoenzyme (JIANG et al. 1998 ; KUCHIN et al. 2000 ; CHANG et al. 2001 ).
We are interested in the control of stationary phase biology and have identified a collection of mutants that exhibit a defective transcriptional response to nutrient deprivation (CHANG et al. 2001 ). These rye mutants were originally isolated on the basis of defects in the expression pattern of YGP1. The YGP1 gene is induced specifically upon nutrient deprivation and this induction has been used as a marker for the ensuing entry into stationary phase (DESTRUELLE et al. 1994 ; RIOU et al. 1997 ; CHANG et al. 2001 ). In the rye mutants, YGP1 and related genes are expressed at an elevated level during mitotic growth (CHANG et al. 2001 ). These transcriptional defects appear to disrupt the control of normal growth as the rye mutants are unable to enter into a normal stationary phase upon nutrient deprivation (CHANG et al. 2001 ). Interestingly, three of the RYE genes encode Srb proteins that comprise part of the Srb complex associated with the RNA pol II holoenzyme (CHANG et al. 2001 ). These observations suggested that the RNA pol II holoenzyme could be a target of signaling pathways responsible for coordinating yeast cell growth with nutrient availability.
The rye mutants identified in the original genetic selection defined eight complementation groups, and more than half of the mutants fell into the rye1 group. In general, the rye1 mutants exhibited the most severe defects in YGP1 expression and stationary phase entry. In this report, the RYE1 gene is characterized and shown to encode Sin4p, a component of the yeast Mediator. Thus, a second complex in the RNA pol II holoenzyme appears to be important for proper growth control in S. cerevisiae. In addition, sin4 mutations exhibited specific genetic interactions with alterations that affected signaling through the Ras/PKA pathway. In all, the data suggested that Ras/PKA signaling might influence gene expression by modulating the activities of proteins associated with the RNA pol II holoenzyme.
| MATERIALS AND METHODS |
|---|
Growth media:
Standard Escherichia coli growth conditions and media were used throughout this study (MILLER 1972 ). Yeast YPAD, 5-fluoroorotic acid (5-FOA), and SC growth media were as described (SIKORSKI and BOEKE 1991 ; KAISER et al. 1994 ). YM-glucose medium refers to a yeast minimal medium containing 0.67% yeast nitrogen base (DIFCO), 2% glucose, and those growth supplements required for cell proliferation. Bromcresol purple (BCP)-sucrose medium was as described, except that 75 µg of Antimycin A1 were top-spread onto the plates immediately before use (ROBINSON et al. 1988 ; CHANG et al. 2001 ).
Plasmid constructions:
The YGP1-SUC2 fusion plasmid, pYGP1-SUC2, was described previously (DESTRUELLE et al. 1994 ; CHANG et al. 2001 ). The expression of this reporter is driven by YGP1 promoter sequences present within the 350 bp immediately upstream of the YGP1 initiation codon (CHANG et al. 2001 ). The MET3-RAS2val19 plasmids were constructed as follows. The MET3 promoter region was cloned as a 550-bp SalI-EcoRV fragment from the pHAM8 plasmid (kindly provided by Dr. H. Mountain) into pRS403 to form pPHY440. RAS2val19 was then cloned as a 1.3-kb BamHI fragment from pJW83.1 (kindly provided by Dr. J. Whistler) into pPHY440 to form pPHY446. This RAS2val19 fragment contained the coding sequences and transcriptional terminator but lacked the RAS2 promoter. A MET3-RAS2 plasmid, pPHY442, was constructed in a similar fashion. The MET3-RAS2val19 hybrid gene from pPHY446 was then subcloned into pRS416 to form pPHY796. The RAS2val19 plasmid, PHY453, contains the RAS2val19 allele cloned into pRS415. The pRS plasmids were described previously (SIKORSKI and HIETER 1989 ; CHRISTIANSON et al. 1992 ).
Yeast strain constructions and genetic methods:
The strains used in this study are listed in Table 1. Unless otherwise noted, the strains were from our lab collection or were derived during the course of this work. Standard yeast genetic methods were used for the construction of all strains (KAISER et al. 1994 ). The isolation of the rye mutants was described previously (CHANG et al. 2001 ). Gene disruptions were constructed with a PCR-based deletion protocol (BAUDIN et al. 1993 ). The cyr1-99 allele was isolated as an extragenic suppressor of the sin4 RAS2val19 synthetic lethality. MET3-RAS2val19 sin4 cells (PHY1649) were plated to YM-glucose minimal medium lacking methionine at a density of 
3 x 107 cells/plate and incubated for 4 days at 30°. One suppressor was analyzed further and was found to contain an allele of CYR1 that was designated as cyr1-99.
For the stationary phase experiments, yeast cells were grown in a YM-glucose minimal medium at 30°. The cultures were typically inoculated at a density of 0.1 OD600 units/ml. Under these conditions, the cells underwent the diauxic shift after a little more than 1 day of growth and generally entered into stationary phase after 4 days of growth (CHANG et al. 2001 ). For the MET3-RAS2val19 experiments, strains carrying this inducible construct were typically grown to mid-log in YM-glucose minimal medium containing 500 µM methionine. The cells were then collected by centrifugation and resuspended in the same growth medium lacking methionine to induce expression from the MET3 promoter.
Cloning of RYE1/SIN4:
The RYE1 gene was cloned by plasmid complementation of the severe flocculation phenotype exhibited by rye1 mutants. When grown in liquid medium, the rye1 strains formed a single cluster of cells at the bottom of the culture vessel; the medium above this cluster was almost devoid of turbidity. A yeast genomic DNA library constructed in the pSB32 plasmid was introduced into the rye1-1 mutant, PHY1454 (SPENCER et al. 1990 ). The transformed cells were separated into 16 equal aliquots and all but one was used to inoculate 6 ml of YM-glucose medium lacking leucine; the pSB32 plasmid is marked with the wild-type LEU2 gene. The remaining aliquot was plated to solid medium to determine the transformation efficiency for the experiment. The liquid cultures were incubated overnight at 30°. A 150-µl aliquot was removed from each culture and used to inoculate a fresh 6 ml of the same medium. Following a second overnight incubation, 1 of the 15 cultures exhibited a significant degree of turbidity in the medium above the clumps of rye1-1 cells. A 50-µl aliquot of this turbid culture was plated to a solid medium and the single colonies formed were analyzed. All 38 colonies analyzed gave rise to nonflocculating cultures. The library plasmids present in three of these strains were isolated and characterized further. The plasmids were found to be identical and each corrected the other phenotypes associated with the rye1-1 mutant. Comparisons between the plasmid sequences and genomic databases were performed with the assistance of analysis programs available at the Saccharomyces Genome Database.
To ensure that the cloned gene corresponded to the rye1 locus altered in the original mutants, a rye1 null mutant was crossed to the original rye1-1 strain. These rye1
/rye1-1 diploids were sporulated and the meiotic progeny of 36 tetrads were characterized for both their growth on sucrose and their propensity to flocculate. For each tetrad, we observed that all progeny were Suc+ and flocculated when grown in liquid culture. These data indicated that the cloned gene represented the genomic locus that was altered in the original rye1-1 strain.
Enzyme assays:
Invertase assays were performed as described, where one unit of activity is equivalent to the release of 1 nmol glucose/30 min/OD600 unit of cells (JOHNSON et al. 1987 ). ß-Galactosidase assays were performed as described and the units of activity refer to the amount of o-nitrophenol released per minute per OD600 unit of cells (AUSUBEL et al. 1995 ). All ß-galactosidase assays were performed in triplicate and the standard errors were typically <15%.
Stationary phase characteristics:
Stationary phase viability assays were performed on cultures that were grown for 7 to 10 days in YM-glucose medium. Cells were collected by centrifugation and resuspended in distilled water at a concentration of 1 OD600 unit/ml. The suspensions were subjected to a series of fivefold dilutions and 200 µl of each suspension was placed into a well of a microtitre plate. These suspensions were then plated with a 48-prong replicating block to YPAD medium. The plates were incubated for 3 days at 30° and the relative number of survivors was determined for each strain analyzed.
The heat-shock sensitivity of the appropriate cultures was tested after 4 days of growth in minimal medium at 30°. For these assays, 200-µl aliquots of the cultures were placed into a microcentrifuge tube and incubated at 50° for 30 min. Dilutions of the cultures were plated to YPAD medium before the initiation of the heat shock and at 10-min intervals thereafter. These plates were then incubated for 3 days at 30°. The relative survival rate was determined by comparing the number of colonies formed by the cultures after heat shock to the number formed by the original culture.
RNA analyses:
Total RNA was prepared from yeast cells by a hot phenol extraction method described previously (AUSUBEL et al. 1995 ). For Northern analyses, 20 µg of total RNA per lane was loaded onto a formaldehyde-agarose gel and subjected to electrophoretic separation. The gel was blotted to nylon membranes that were then hybridized with the appropriate 32P-labeled probes (AUSUBEL et al. 1995 ). Typically, these probes were 0.7- to 1.0-kb PCR fragments that were prepared with the oligolabeling kit (Amersham). To ensure uniform loading for the stationary phase RNA experiments, rRNA levels were assessed visually after staining the gel with ethidium bromide.
| RESULTS |
|---|
RYE1/SIN4 function was required for the regulation of YGP1:
All of the rye1 mutants tested exhibited an elevated level of YGP1 expression during the log phase of growth (Fig 1; data not shown). For these assays, the cells contained a YGP1-SUC2 hybrid reporter that was used in the original rye mutant selection (CHANG et al. 2001 ). The SUC2 gene encodes invertase, an enzyme that is necessary for yeast cell growth on sucrose (JOHNSTON and CARLSON 1992 ). In wild-type cells, this reporter was expressed at very low levels during log phase growth and was induced >150-fold upon entry into stationary phase (Fig 1A; see DESTRUELLE et al. 1994 ; CHANG et al. 2001 ). This induction was relatively specific to stationary phase entry as no significant increase in expression was observed in response to a variety of other stresses, including heat shock and increased osmolarity (CHANG et al. 2001 ). In the rye1-1 mutant, the log phase level of this YGP1 reporter was 57 times higher than that seen in an isogenic wild-type strain (Fig 1A and Fig B). This elevated expression of YGP1-SUC2 allowed rye1 mutants to grow on sucrose-containing media, whereas RYE strains were phenotypically Suc- (Fig 1C). The rye1 mutations also resulted in elevated levels of expression from the endogenous YGP1 locus (CHANG et al. 2001 ). Therefore, the RYE1 gene was required for the normal regulation of YGP1 expression.
The rye1 mutants exhibited a number of additional phenotypes, including temperature-sensitive (ts) growth defects, a severe propensity to flocculate, and a partial inositol auxotrophy. The wild-type RYE1 locus was cloned by complementation of this flocculation defect (see MATERIALS AND METHODS) and the identified gene was found to complement all of the phenotypes ascribed to rye1 mutations. Interestingly, DNA sequence analysis revealed that RYE1 was identical to the previously identified gene, SIN4. The SIN4 gene encodes a protein that functions as both a positive and negative regulator of RNA pol II transcription (JIANG and STILLMAN 1992 ; CHEN et al. 1993 ; JIANG et al. 1995 ). Previous work has shown that Sin4p is associated with the RNA pol II holoenzyme and that Sin4p may exert its effects by altering chromatin structure (JIANG and STILLMAN 1992 ; LI et al. 1995 ). We constructed a null allele of the RYE1/SIN4 locus and found that this deletion resulted in a spectrum of phenotypes similar to that observed with other rye1 alleles (Fig 1; data not shown). Since the SIN4 gene has been extensively characterized, we refer to the RYE1/SIN4 locus as SIN4 for the remainder of this report.
Sin4p is part of a complex within the Mediator that contains Gal11p, Hrs1p, and Med2p (LI et al. 1995 ; MYERS et al. 1999 ). Mutations in the genes encoding these proteins typically have overlapping but distinct defects in RNA pol II transcription (MYERS and KORNBERG 2000 ). Therefore, the expression of the YGP1-SUC2 reporter was examined in gal11, hrs1, and med2 mutants. We found that the reporter levels were significantly elevated in gal11 and hrs1 mutants but not in med2 mutants (Fig 2A). In addition, reporter levels were not elevated in cells lacking Med1p, a Mediator component not thought to be associated with the Sin4p module (Fig 2A). Finally, strains defective in both the Srb complex and the Sin4p module, such as sin4 srb10 mutants, did not exhibit any additional YGP1 expression defects (data not shown). Thus, as observed with other promoters, individual Mediator components appeared to make distinct contributions to the normal expression pattern of the YGP1-SUC2 reporter (HAN et al. 1999 ; MALIK and ROEDER 2000 ; MYERS and KORNBERG 2000 ).
The above data indicated that Sin4p and associated proteins were required for the efficient repression of YGP1 during the log phase of growth. We also examined the roles of two additional transcriptional regulators, Tup1p and Ure2p, in the regulation of YGP1. Tup1p is a negative regulator of the transcription of many genes in yeast, and mutations in SIN4 often weaken Tup1-mediated repression (WAHI and JOHNSON 1995 ; FRIESEN et al. 1998 ; LEE et al. 2000 ). Ure2p is required for the full repression of many genes important for N2 metabolism (MAGASANIK 1992 ). These genes include a family of asparaginase enzymes that exhibit sequence similarity to Ygp1p (BON et al. 1997 ). We found that deletion of either TUP1 or URE2 did not result in a Rye- phenotype, and thus neither gene product was required for the repression of the YGP1-SUC2 reporter (Fig 2B).
rye1/sin4 mutants did not enter a normal stationary phase upon nutrient deprivation:
The entry into stationary phase appears to involve a tightly regulated program of gene expression (CHODER 1991 ; WERNER-WASHBURNE et al. 1993 , WERNER-WASHBURNE et al. 1996 ). Whereas some genes are induced at the earliest stages of nutrient deprivation, the expression of others increases at later stages. Moreover, disruptions of this transcriptional program result in a failure to enter into a normal stationary phase (CHANG et al. 2001 ). The above data with YGP1 clearly indicated that such a disruption might be associated with a loss of SIN4 function. Indeed, sin4 mutations also affected the expression patterns of other genes normally expressed during nutrient limitation. For example, the CTT1 and HSP12 genes are normally expressed at very low levels during log phase but are significantly induced during the diauxic shift (PRAEKELT and MEACOCK 1990 ; DERISI et al. 1997 ). In sin4 mutants, the log phase level of each of these mRNAs was elevated more than fivefold (Fig 1B; data not shown). Therefore, we tested whether sin4 mutants were able to enter into a normal stationary phase upon nutrient deprivation.
One of the hallmarks of a stationary phase yeast cell is the ability to survive for extended periods of time under nutrient-limiting conditions (WERNER-WASHBURNE et al. 1993 ). In contrast, mutants that fail to enter into a normal stationary phase rapidly lose viability upon nutrient deprivation (CANNON and TATCHELL 1987 ; TODA et al. 1987A ; WERNER-WASHBURNE et al. 1993 ). Therefore, we assessed the relative number of survivors in stationary phase cultures of wild-type and sin4 strains. As expected, wild-type cells remained viable after 8 days of growth in minimal medium (Fig 3A). The number of survivors after 8 days of growth was not significantly different than that observed after 2 days. In contrast, after 8 days, the sin4 cultures had at least 200-fold fewer survivors than the wild type (Fig 3A). Thus, sin4 mutants were unable to survive a prolonged period of nutrient limitation.
View larger version (19K):
In this window
In a new window
Download PPT slide
|
Figure 3.
SIN4 function was required for the entry into a normal stationary phase. (A) The rye1 mutants exhibited decreased viability following an extended period of nutrient deprivation. Wild-type (PHY1184) and rye1 (PHY1264) strains were grown at 30° for either 2 or 8 days in YM-glucose minimal medium. Cells from these cultures were collected by centrifugation, washed, and resuspended in distilled water at a concentration of 1 OD600 unit/ml. Fivefold serial dilutions of these suspensions were then plated to YPAD medium and incubated for 3 days at 30°. The number of colonies formed was a measure of the number of survivors in the original cultures. (B) Stationary phase cultures of sin4/rye1 mutants were sensitive to a brief heat shock. The indicated strains were grown at 30° for 4 days in YM-glucose minimal medium and then subjected to a 50° heat shock for 20 min. The percentage of cells surviving this treatment are shown. The strains analyzed were wild type (PHY1184), rye1 (PHY1264), and sin4 (PHY1575). The RAS2val19 control is the wild-type strain (PHY1220) carrying a plasmid with the dominant RAS2val19 allele. The standard errors were typically <10% and each value represents the average of three independent experiments. (C) sin4 mutants exhibited defects in the stationary phase repression of ACT1. The steady-state level of ACT1 mRNA in stationary phase cultures of wild-type (PHY1184) and sin4 (PHY1264) strains was measured by Northern blot analysis. Total RNA was prepared from cultures grown for 1, 4, or 6 days at 30° in minimal medium. Twenty micrograms of total RNA was loaded for each sample. As a control for loading, rRNA levels were assessed by visual inspection after staining the gel with ethidium bromide.
|
|
The response of sin4 mutants to nutrient deprivation was examined further by assessing two additional properties normally associated with the stationary phase of growth. Stationary phase cells generally exhibit an elevated resistance to a number of environmental stresses, including heat shock (WERNER-WASHBURNE et al. 1993 ). Therefore, stationary phase cultures of wild-type and sin4 cells were subjected to a 50° heat shock for 20 min. We found that the sin4 cultures were significantly more sensitive than wild type to this heat-shock regimen (Fig 3B). Finally, we examined the general decrease in RNA pol II transcription that occurs upon stationary phase entry. In general, stationary phase levels of most mRNAs are significantly lower than that observed during log phase growth (CHODER 1991 ). This repression phenomenon has been best characterized for the ACT1 locus and hence ACT1 mRNA levels were assessed in the sin4 cultures. Once again, the sin4 mutants did not respond normally to nutrient deprivation and contained significantly elevated levels of ACT1 mRNA after 6 days of growth (Fig 3C). Altogether, these data indicated that Sin4p was required for stationary phase entry in S. cerevisiae.
Mutations that elevate the level of Ras signaling were synthetically lethal with sin4:
The Ras/PKA signaling pathway appears to negatively regulate YGP1 expression because decreased levels of Ras activity result in increased levels of the YGP1-SUC2 reporter (CHANG et al. 2001 ). Therefore, the rye mutants could identify targets of the Ras pathway that are important for this transcriptional control. This possibility was tested by asking whether elevated levels of Ras signaling would suppress rye mutant phenotypes. Indeed, the presence of a dominant hyperactive allele of RAS2, known as RAS2val19 (KATAOKA et al. 1984 ), was able to suppress the YGP1 misexpression phenotype associated with rye4 mutants (data not shown). However, in sin4 cells, the presence of RAS2val19 instead resulted in a severe synthetic growth defect.
This unexpected growth defect associated with sin4 RAS2val19 mutants was demonstrated in three independent assays. The first was a transformation-based assay where we found that a plasmid containing RAS2val19 could not be stably introduced into sin4 cells. The frequency of transformation with sin4 mutants for plasmids with RAS2val19 was 5000-fold less than that for control plasmids. For the second assay, an inducible allele of RAS2val19 was constructed by placing the RAS2val19 coding sequences under the control of the yeast MET3 promoter. This promoter is active when cells are grown in media lacking methionine and is repressed by the presence of methionine in the growth medium (CHEREST et al. 1987 ; MOUNTAIN et al. 1991 ). The introduction of this MET3-RAS2val19 construct into sin4 cells resulted in a severe growth defect specifically in media lacking methionine (Fig 4A). Note that the growth rates of the sin4 and RAS2val19 single mutants were very similar to that of the isogenic wild-type control (Fig 4A). Finally, sin4 RAS2val19 mutants were unable to lose a plasmid that contained the wild-type SIN4 locus (Fig 4B). In contrast, this SIN4 plasmid was readily lost from either wild-type or both single mutant cells (Fig 4B). A similar synthetic growth defect with RAS2val19 was exhibited by all other sin4 alleles, including the null. These growth defects were not specific to the genetic background used for these studies because sin4 mutations in five different backgrounds were all found to be sensitive to RAS2val19. Thus, elevated levels of Ras signaling caused a severe growth defect in sin4 cells.
View larger version (50K):
In this window
In a new window
Download PPT slide
|
Figure 4.
The presence of RAS2val19 resulted in a severe synthetic growth defect specifically in sin4 mutants. (A) sin4 RAS2val19 double mutants exhibited a synthetic growth defect. The indicated strains were grown on YM-glucose minimal media that contained either 0 or 500 µM methionine for 3 days at 30°. The strains analyzed were wild type (PHY1837), sin4 (PHY1647), RAS2val19 (PHY1834), and sin4 RAS2val19 (PHY1649). The RAS2val19 strains contained RAS2val19 under the control of the methionine-repressible promoter from the yeast MET3 gene. Therefore, RAS2val19 was expressed only on media lacking methionine. (B) Growth of RAS2val19 sin4 double mutants required the presence of the pSIN4 plasmid. The indicated strains containing the pSIN4 plasmid were grown for 3 days at 30° on either YM-glucose minimal (+pSIN4) or 5-FOA (-pSIN4) media. The 5-FOA medium selects against the URA3-marked pSIN4 plasmid; therefore, only those strains that can be cured of this plasmid will exhibit growth on 5-FOA medium. The strains analyzed were wild type (PHY1220 with pRS415), sin4 (PHY1454 with pRS415), RAS2val19 (PHY1220 with pPHY453), and sin4 RAS2val19 (PHY1454 with pPHY453).
|
|
In the above experiments, most of the sin4 cells were likely in stationary phase before they were tested for growth with RAS2val19. Therefore, it was a formal possibility that the double mutant growth defect was specific to stationary phase cells and that our assays were measuring a defect in stationary phase exit. However, this possibility was ruled out since the induction of RAS2val19 in log phase sin4 cultures also resulted in a severe growth arrest (Fig 5A). This arrest occurred with very rapid kinetics as the sin4 cells did not undergo even a single round of division following the induction of RAS2val19 expression. Moreover, the elevated levels of Ras signaling caused the sin4 cells to rapidly lose viability; <0.01% of the sin4 cells remained viable 2 hr after the induction of RAS2val19 (Fig 5B). Therefore, elevated levels of Ras signaling caused a rapid growth arrest and subsequent cell death specifically in sin4 mutants.
Since Sin4p is involved in transcriptional regulation, we tested whether the sin4 RAS2val19 lethality was due to a general defect in mRNA production. For this analysis, the steady-state levels of multiple mRNAs were assessed in wild-type and sin4 cells at 0, 2, and 4 hr after induction of RAS2val19 expression. At these latter two time points, the sin4 cultures contained very few, if any, viable cells (Fig 5B). Nonetheless, the levels of each of the mRNAs tested remained unchanged throughout the course of this experiment (Fig 6). More importantly, the levels in wild-type and sin4 cells were essentially identical (Fig 6). Thus, the loss of sin4 RAS2val19 viability was not correlated with a global defect in RNA pol II transcription. Instead, the observed synthetic lethality might be due to transcriptional defects at a subset of essential genes.
View larger version (88K):
In this window
In a new window
Download PPT slide
|
Figure 6.
RAS2val19 sin4 double mutants did not exhibit a global defect in mRNA production. Yeast strains were grown at 30° to mid-log in YM-glucose minimal medium containing 500 µM methionine. The cells were then transferred to media lacking methionine to induce the expression from the MET3-RAS2val19 construct. Cells were collected at 0, 2, and 4 hr after the shift and total RNA was prepared as described in MATERIALS AND METHODS. The steady-state levels of ACT1, ANC1, CDC9, RAD23, and RP51A mRNA were then measured by Northern RNA blot analysis. Twenty micrograms of total RNA were loaded for each sample. The strains analyzed were wild type (PHY1837), sin4 (PHY1647), RAS2val19 (PHY1834), and sin4 RAS2val19 (PHY1649).
|
|
The RAS2val19 sin4 synthetic lethality required the cAMP/PKA pathway:
The S. cerevisiae Ras proteins have been shown to function through at least three different effectors: the cAMP/PKA pathway, a MAP kinase pathway important for pseudohyphal growth, and a poorly defined third effector that is required for the exit from mitosis (GIBBS and MARSHALL 1989 ; MORISHITA et al. 1995 ; MOSCH et al. 1996 ). The importance of the PKA pathway for the sin4 RAS2val19 lethality was indicated by two independent lines of investigation. In the first, sin4 mutations were combined with null alleles of BCY1 and the viability of the double mutants was tested. BCY1 encodes the inhibitory subunit of PKA, and deletion of BCY1 results in elevated levels of PKA activity without affecting the Ras proteins. In this experiment, the BCY1 locus was deleted in sin4 cells that carried the wild-type SIN4 gene on a plasmid. The resulting sin4 bcy1 double mutant was unable to lose the SIN4 plasmid indicating that sin4 mutations are synthetically lethal with alterations that elevate the levels of PKA activity (Fig 7A).
The second strategy used to test the importance of the Ras/PKA pathway involved lowering the level of cAMP produced in sin4 RAS2val19 cells. This was accomplished by introducing two different cyr1 alleles, cyr1-230 and cyr1-99, into the above double mutant. CYR1 encodes the yeast adenylyl cyclase, and cyr1 mutations would be expected to lower the cellular levels of cAMP and thus PKA activity (MATSUMOTO et al. 1982 ). Indeed, both of these cyr1 alleles were found to restore near normal growth rates to the sin4 RAS2val19 double mutant (Fig 7B). Therefore, the lethal effects of RAS2val19 in sin4 mutants were due to the elevated levels of PKA present.
Although PKA is likely to phosphorylate a number of substrates important for S. cerevisiae growth, few of these potential targets have been identified (REINDERS et al. 1998 ; THEVELEIN and DE WINDE 1999 ). However, recent data indicate that two related transcription factors, Msn2p and Msn4p, might be targets of the Ras/PKA signaling pathway (GORNER et al. 1998 ; SMITH et al. 1998 ). Msn2p and Msn4p are required for the transcription of a number of genes induced during the cellular response to environmental stress (MARCHLER et al. 1993 ; SCHMITT and MCENTEE 1996 ). Interestingly, the deletion of both MSN2 and MSN4 suppresses the otherwise lethal loss of all three genes encoding catalytic subunits of the yeast PKA (SMITH et al. 1998 ). These observations led to the proposition that Ras/PKA activity was negatively regulating Msn2p/Msn4p function. Therefore, we tested whether the sin4 RAS2val19 lethality observed here was mediated by Msn2p and/or Msn4p. If these two Msn proteins are the primary target of Ras signaling, deletion of both MSN2 and MSN4 should also be synthetic lethal with sin4. However, the triple msn2 msn4 sin4 mutant was viable and exhibited a wild-type growth rate (Fig 7C). In addition, the loss of MSN2 and MSN4 did not suppress the sin4 RAS2val19 lethality (data not shown). Therefore, the Ras effects on sin4 growth appear to involve PKA targets other than these two transcription factors.
Mutations that affect the Sin4p module of the Mediator were synthetic lethal with RAS2val19:
Our data indicate that sin4 mutants have several phenotypes in common with mutants that possess high levels of Ras signaling activity. In addition to the stationary phase defects described above, sin4 and RAS2val19 mutants exhibited similar ts growth defects, inositol auxotrophy, and flocculation phenotypes. One potential explanation for these similarities is that Sin4p is a negative regulator of some aspect of Ras signaling. However, several observations indicated that this possibility was unlikely. First, the levels of both Ras proteins, and of intracellular cAMP, were very similar in wild-type and sin4 cells (data not shown). In addition, sin4 mutations were not able to suppress the growth defects associated with mutations that lower the level of Ras signaling, such as cdc25-1, ras2-23, and cyr1-230 (data not shown). Therefore, Sin4p did not appear to be a negative regulator of Ras protein expression or signaling activity.
To further examine the interaction between Ras/PKA signaling activity and Sin4p, we tested whether other mutations affecting the Mediator and RNA pol II were influenced by the presence of elevated levels of Ras signaling. These experiments indicated that the synthetic lethality with RAS2val19 was relatively specific to mutations affecting the Sin4p-containing module of the Mediator. Mutations in GAL11, MED2, HRS1, and RGR1 were all synthetically lethal with RAS2val19 (Fig 8). RGR1 encodes a protein that is thought to link the Sin4p module to the remainder of the Mediator complex (LI et al. 1995 ). In contrast, mutations that affected other Mediator components (med1 and med9), the Srb complex (srb9, srb10, and srb11), the Snf/Swi chromatin remodeling complex (swi2), or RNA pol II (rpb1 and rpb5) were relatively insensitive to changes in Ras signaling activity (Fig 8). SNF2/SWI2 was tested because sin4 mutations have been shown to suppress transcriptional defects associated with the loss of this gene (JIANG and STILLMAN 1992 ; SONG et al. 1996 ). In all, these genetic data suggest the existence of a functional interaction between the Ras/PKA signaling pathway and the Sin4p module of the Mediator complex.
View larger version (30K):
In this window
In a new window
Download PPT slide
|
Figure 8.
Mutations that affected the Sin4p module of the RNA pol II holoenzyme exhibited a synthetic lethal interaction with RAS2val19. Strains with the indicated genotype were transformed with either a control vector, pRS416 (RAS2), or a plasmid containing the MET3-RAS2val19 construct, pPHY796 (RAS2val19). The strains were plated to YM-glucose minimal media lacking methionine and incubated for 2 to 3 days at 30°. (Top) The RAS2val19 growth defect was specific to sin4/rye1 mutants. The strains analyzed were wild type (PHY1220), sin4/rye1 (PHY1454), srb11/rye2 (PHY1456), srb9/rye3 (PHY1459), rye4 (PHY1469), and srb10/rye5 (PHY1470). (Bottom) The growth effects of RAS2val19 on various mutations affecting RNA pol II activity. The strains analyzed were wild type (PHY1220), sin4 (PHY1575), gal11 (PHY1669), med2 (PHY2130), hrs1 (PHY2533), rgr1 (PHY1829), rpb1-1 (PHY1081), rpb5-9 (PHY2203), med1 (PHY2129), med9 (PHY2512), swi2 (PHY1719), and swi2 sin4 (PHY1720).
|
|
Some sin4 phenotypes were suppressed by mutations that lower Ras signaling activity:
The presence of elevated levels of Ras signaling activity had a very dramatic effect on cells lacking the SIN4 gene. Interestingly, we found that lowering Ras/PKA activity also influenced specific sin4 phenotypes. For example, the presence of a cyr1 mutation suppressed the flocculation phenotype associated with particular sin4 mutants (Fig 9). Moreover, cyr1 mutations also suppressed the CTS1 expression defect caused by the loss of Sin4p. CTS1 encodes an endochitinase, and CTS1 expression is decreased 2.5- to 3-fold in sin4 mutants (KURANDA and ROBBINS 1991 ; JIANG et al. 1995 ). This CTS1 expression defect was corrected in cyr1 sin4 double mutants (Fig 9). Thus, both raising and lowering Ras/PKA signaling activity had profound effects on sin4 phenotypes.
View larger version (45K):
In this window
In a new window
Download PPT slide
|
Figure 9.
Decreased signaling through the Ras/PKA pathway suppressed the flocculation and CTS1 transcription defects associated with sin4 mutants. The steady-state levels of CTS1 and ACT1 mRNA in mid-log cultures of the indicated strains were measured by Northern RNA blot analysis. Twenty micrograms of total RNA were loaded for each sample. The strains analyzed were wild type (PHY1721), sin4 (PHY1722), cyr1 (PHY1447), and cyr1 sin4 (PHY2115). The relative degree of flocculation observed for the following strains is indicated below the RNA blot: wild type (PHY1220), sin4 (PHY1454), cyr1 (PHY2121 with pSIN4), and cyr1 sin4 (PHY2121). These strains were grown to mid-log in YM-glucose minimal medium containing 500 µM methionine.
|
|
Previous work has shown that sin4 mutations affect the expression of a number of genes in yeast (JIANG and STILLMAN 1992 ; CHEN et al. 1993 ; COVITZ et al. 1994 ; WAHI and JOHNSON 1995 ). Therefore, we tested whether lowering Ras signaling activity would suppress other transcriptional defects associated with sin4 mutants. For these experiments, three different reporter genes were analyzed in wild-type, sin4, cyr1, and cyr1 sin4 strains. The expression of each of these reporter genes, PHO5::lacZ, HIS4::lacZ, and Ty1::lacZ, is altered in sin4 mutants (JIANG and STILLMAN 1992 , JIANG and STILLMAN 1995 ). However, in contrast to the above results with CTS1, we found that cyr1 mutations had no significant effect on the sin4 defects associated with these three reporters (data not shown). Therefore, Ras/PKA signaling appears to affect only specific phenotypes associated with the loss of SIN4 function.
| DISCUSSION |
|---|
We are interested in understanding the mechanisms regulating stationary phase biology in S. cerevisiae. To this end, we have identified a collection of rye mutants that exhibit defects in the transcriptional response to nutrient deprivation (CHANG et al. 2001 ). These transcription defects appear to disrupt yeast cell growth as all of the rye mutants characterized to date are unable to enter into a normal stationary phase. In this report, we examined the rye1 mutants and found that these mutants also exhibited multiple stationary phase defects. The RYE1 gene was cloned and was found to encode Sin4p, a component of the yeast Mediator complex associated with the RNA pol II holoenzyme. Sin4p is part of a subcomplex that also contains Gal11p, Hrs1p, and Med2p (LI et al. 1995 ; MYERS et al. 1999 ). Mutations that inactivated Gal11p and Hrs1p also resulted in a Rye- phenotype, indicating that this Sin4p module is generally important for the regulation of genes like YGP1. Finally, a specific genetic interaction was observed between this Sin4p module of the Mediator and the Ras/PKA signaling pathway. We found that elevated levels of Ras/PKA signaling resulted in a synthetic lethality specifically with mutations that compromised the Sin4p module. Mutations affecting other components of the Mediator or RNA pol II itself were not significantly affected by changes in Ras/PKA signaling activity. Altogether, the data presented here and in a previous study indicate that transcriptional regulators within the RNA pol II holoenzyme are required for the normal control of cell growth in S. cerevisiae (CHANG et al. 2001 ).
Two distinct complexes within the RNA pol II holoenzyme are required for proper stationary phase entry:
Wild-type cells respond to nutrient deprivation by undergoing an orderly series of changes in gene expression (CHODER 1991 ; DERISI et al. 1997 ). Our studies with the rye mutants suggest that the proper execution of this transcriptional program is necessary for stationary phase entry. Mutations in SIN4 and the other RYE genes disrupt the transcriptional response to nutrient deprivation and prevent cells from entering into a normal stationary phase (CHANG et al. 2001 ). In rye mutants, genes normally induced upon nutrient limitation, such as YGP1, are instead expressed at elevated levels during log phase growth (CARLSON 1997 ; HOLSTEGE et al. 1998 ; CHANG et al. 2001 ). Moreover, subsequent nutrient deprivation does not appear to result in a significant induction of several of these genes (our unpublished data). Our current model is that these transcription defects result in a failure to undergo a normal growth arrest during nutrient deprivation. As a result, the rye mutants fail to assume many of those characteristics normally associated with the stationary phase of growth.
Interestingly, all of the RYE genes characterized thus far have been found to encode transcriptional regulators associated with the RNA pol II holoenzyme. The RYE2, RYE3, and RYE5 genes were previously shown to encode the Srb11p, Srb9p, and Srb10p proteins, respectively (CHANG et al. 2001 ). These proteins are all components of the Srb complex associated with the RNA pol II holoenzyme (CARLSON 1997 ). Like Sin4p, these SRB gene products are also required for the proper entry into stationary phase (CHANG et al. 2001 ). These results therefore implicate two distinct complexes within the RNA pol II holoenzyme in the control of yeast cell growth: the Srb complex and the Sin4p module of the Mediator. However, the key question that remains is whether the activities of these two complexes are indeed regulated during stationary phase entry. The answer appears to be yes for the Srb complex as the stability of both Srb10p and Srb11p decreases upon nutrient limitation (COOPER et al. 1997 ; HOLSTEGE et al. 1998 ). Thus, the Srb complex may be inactivated by the physical removal of particular members of this complex (HOLSTEGE et al. 1998 ; CHANG et al. 2001 ). This inactivation would result in the increased expression of those genes required during nutrient limitation. However, it is not yet known if either the stability or activity of proteins within the Sin4p module are similarly influenced by nutrient availability.
A novel mode of transcriptional control:
The specificity of the genetic interactions observed here suggests that a functional relationship exists between the Ras/PKA signaling pathway and the Sin4p module of the Mediator. One interesting possibility is that Ras/PKA signaling influences RNA pol II activity to ensure that gene expression is properly coordinated with nutrient availability and cell growth. Clearly, the key to understanding this relationship is the identification of the PKA substrate responsible for the above genetic interactions. However, the nature of this target has remained elusive. Our genetics suggests that the relevant PKA substrate is likely not any of the known components of the Sin4p module. This assertion follows from observations that null alleles of SIN4, GAL11, HRS1, and MED2 are all synthetic lethal with RAS2val19; deletion of the relevant target should render the resulting strain insensitive to the effects of elevated Ras/PKA activity. Moreover, none of the proteins in the Sin4p module contain a consensus site for PKA phosphorylation. The best-characterized consensus for the S. cerevisiae PKA enzyme fits the format of R-R-x-S/T-B, where x indicates any amino acid and B indicates a residue with a hydrophobic side-chain (DENIS et al. 1991 ). Other potential candidates for this PKA target include proteins encoded by those genes, such as KIN28, SPT20, and SRB5, that have been shown to exhibit a genetic interaction with sin4 mutations (VALAY et al. 1995 ; ROBERTS and WINSTON 1997 ; CHANG et al. 1999 ). However, none of these candidates possesses an obvious PKA consensus site either. Thus, we feel that classical genetic approaches may represent the best way to identify this PKA substrate. For this reason, we have initiated a search for extragenic suppressors of the sin4 RAS2val19 synthetic lethality.
The possibility that proteins within the RNA pol II holoenzyme might be direct targets of particular signaling pathways is very intriguing. The proteins within the Srb complex and the Sin4p module of the Mediator appear to control the transcription of distinct subsets of genes (CARLSON 1997 ; PTASHNE and GANN 1997 ; HAN et al. 1999 ; MYERS and KORNBERG 2000 ). By directly targeting components within these complexes, the cell could bring about rather large changes in gene expression with a single regulatory event. For example, by modulating Sin4p activity, the cell would be able to coordinately control the expression of all promoters affected by this transcriptional regulator. This type of a control mechanism clearly would be more efficient than one where each individual promoter was regulated independently. The ability to effect rather global changes in gene expression would be very useful in those instances where cells undergo significant changes in their overall physiology, such as during the entry into a G0-like resting state. Although no example of this regulatory mechanism has yet been described, several recent reports have hinted at this type of transcriptional control (HOLSTEGE et al. 1998 ; KUCHIN et al. 2000 ; CHANG et al. 2001 ). The further characterization of the rye mutants could therefore provide important insights into the manner in which signaling pathways control gene expression.
| ACKNOWLEDGMENTS |
|---|
We thank Drs. A. Aguilera, D. Balciunas, M. Carlson, F. Estruch, L. Myers, H. Mitsuzawa, H. Mountain, Y.-J. Kim, D. Stillman, C. Trueblood, J. Whistler, F. Winston, N. Woychik, and R. Young for providing yeast strains and plasmids used in this study. This work was supported by grants from the American Cancer Society, the Ohio Cancer Research Associates, and the National Science Foundation.
Manuscript received May 15, 2001; Accepted for publication June 18, 2001.
| LITERATURE CITED |
|---|
AUSUBEL, F. M., R. BRENT, R. E. KINGSTON, D. D. MOORE, J. G. SEIDMAN et al., 1995 Current Protocols in Molecular Biology. John Wiley & Sons, New York.
BALCIUNAS, D., C. GALMAN, H. RONNE, and S. BJORKLUND, 1999 The Med1 subunit of the yeast mediator complex is involved in both transcriptional activation and repression. Proc. Natl. Acad. Sci. USA 96:376-381. [erratum: Proc. Natl. Acad. Sci. USA 96(6): 3330].[Abstract/Free Full Text].
BAUDIN, A., O. OZIER-KALOGEROPOULOS, A. DENOUEL, F. LACROUTE, and C. CULLIN, 1993 A simple and efficient method for direct gene deletion in Saccharomyces cerevisiae. Nucleic Acids Res. 21:3329-3330[Free Full Text].
BON, E. P., E. CARVAJAL, M. STANBROUGH, D. ROWEN, and B. MAGASANIK, 1997 Asparaginase II of Saccharomyces cerevisiae. GLN3/URE2 regulation of a periplasmic enzyme. Appl. Biochem. Biotechnol. 63–65:203-212.
BOUCHERIE, H., 1985 Protein synthesis during transition and stationary phases under glucose limitation in Saccharomyces cerevisiae. J. Bacteriol. 161:385-392[Abstract/Free Full Text].
BROACH, J. R., 1991 RAS genes in Saccharomyces cerevisiae: signal transduction in search of a pathway. Trends Genet 7:28-33[Medline].
CANNON, J. F. and K. TATCHELL, 1987 Characterization of Saccharomyces cerevisiae genes encoding subunits of cyclic AMP-dependent protein kinase. Mol. Cell. Biol. 7:2653-2663[Abstract/Free Full Text].
CARLSON, M., 1997 Genetics of transcriptional regulation in yeast: connections to the RNA polymerase II CTD. Annu. Rev. Cell Dev. Biol. 13:1-23[Medline].
CHANG, M., D. FRENCH-CORNAY, H. Y. FAN, H. KLEIN, and C. L. DENIS et al., 1999 A complex containing RNA polymerase II, Paf1p, Cdc73p, Hpr1p, and Ccr4p plays a role in protein kinase C signaling. Mol. Cell. Biol. 19:1056-1067[Abstract/Free Full Text].
CHANG, Y. W., S. C. HOWARD, Y. V. BUDOVSKAYA, J. RINE, and P. K. HERMAN, 2001 The rye mutants identify a role for Ssn/Srb proteins of the RNA polymerase II holoenzyme during stationary phase entry in Saccharomyces cerevisiae. Genetics 157:17-26[Abstract/Free Full Text].
CHEN, S., R. W. WEST, JR., S. L. JOHNSON, H. GANS, and B. KRUGER et al., 1993 TSF3, a global regulatory protein that silences transcription of yeast GAL genes, also mediates repression by alpha 2 repressor and is identical to SIN4. Mol. Cell. Biol. 13:831-840[Abstract/Free Full Text].
CHEREST, H., P. KERJAN, and Y. SURDIN-KERJAN, 1987 The Saccharomyces cerevisiae MET3 gene: nucleotide sequence and relationship of the 5' non-coding region to that of MET25. Mol. Gen. Genet. 210:307-313[Medline].
CHODER, M., 1991 A general topoisomerase I-dependent transcriptional repression in the stationary phase in yeast. Genes Dev. 5:2315-2326[Abstract/Free Full Text].
CHRISTIANSON, T. W., R. S. SIKORSKI, M. DANTE, J. H. SHERO, and P. HIETER, 1992 Multifunctional yeast high-copy-number shuttle vectors. Gene 110:119-122[Medline].
COOPER, K. F., M. J. MALLORY, J. B. SMITH, and R. STRICH, 1997 Stress and developmental regulation of the yeast C-type cyclin Ume3p (Srb11p/Ssn8p). EMBO J. 16:4665-4675[Medline].
COVITZ, P. A., W. SONG, and A. P. MITCHELL, 1994 Requirement for RGR1 and SIN4 in RME1-dependent repression in Saccharomyces cerevisiae. Genetics 138:577-586[Abstract].
<
TOP
ABSTRACT
MATERIALS AND METHODS
), sin4 (PHY1647; 
), RAS2val19 (PHY1834; 
), and sin4 RAS2val19 (PHY1649; 
). All growth was carried out at 30°. (B) The expression of RAS2val19 resulted in a rapid cell death in sin4 mutants. Yeast strains were grown as described in A. Following the shift to media lacking methionine, cells were collected at the indicated intervals, diluted in water, and plated to YPAD media. These plates were incubated for 3 days at 30°. The number of colonies present was a measure of the number of viable cells in the original cultures. The relative number of survivors was determined by calculating the number of colonies formed per OD600 unit and normalizing these values to that obtained for the wild-type strain at time zero.