suppressors of a saccharomyces cerevisiae pkcl mutation ... · posed to act in the pkcl pathway....

Copyright 0 1995 by the Genetics Society of America

Suppressors of a Saccharomyces cereviSiae pkcl Mutation Identify Alleles of the Phosphatase Gene PTCl and of a Novel Gene Encoding

a Putative Basic Leucine Zipper Protein

Kimberly N. Huang and Lorraine S. Symington

Institute of Cancer Research and Department of Microbiology, Columbia University College of Physicians nd Surgeons, New York, New York 10032

Manuscript received April 4, 1995 Accepted for publication September 15, 1995

ABSTRACT The PKCl gene product, protein kinase C, regulates a mitogen-activated protein kinase (MAPK)

cascade, which is implicated in cell wall metabolism. Previously, we identified the pkcl-4 allele in a screen for mutants with increased rates of recombination, indicating that PKCl may also regulate DNA metabolism. The pkcl-4 allele also conferred a temperature-sensitive (ts) growth defect. Extragenic suppressors were isolated that suppress both the ts and hyperrecombination phenotypes conferred by the pkcl-4 mutation. Eight of these suppressors fell into two complementation groups, designated KCSl and KCS2. KCSl was cloned and found to encode a novel protein with homology to the basic leucine zipper family of transcription factors. KCSZ is allelic with PTCl , a previously identified type 2C serine/ threonine protein phosphatase. Although mutation of either KCSl or PTCl causes little apparent phenotype, the k c s l A p t c l A double mutant fails to grow at 30". Furthermore, the ptcl deletion mutation is synthetically lethal in combination with a mutation in MPKl, which encodes a MAE'K homologue proposed to act in the PKCl pathway. Because PTCl was initially isolated as a component of the Hoglp MAPK pathway, it appears that these two MAE'K cascades share a common regulatory feature.

P ROTEIN kinase C (PKC) plays a critical role in the regulation of growth and proliferation in all eu-

karyotic cells (reviewed in NISHIZUKA 1992). A PKC homologue in the yeast S. cerevisiaeis encoded by the essential gene PKCl (LEVIN et al. 1990). Deletion of PKCl results in cell lysis (LEVIN and BARTLETT-HEUBUSCH 1992; LEVIN and ERREDE 1993). Two temperature-sensitive alleles of PKCl and one Ca2+-dependent allele were isolated by LEVIN and BARTLETT-HEUBUSCH (1992), and additional alleles were found to be allelic with cly15, a temperature-sensitive cell lysis mutant (PARAVICINI et al. 1992), and 4102, a hypo-osmolarity-sensitive mutant (SHIMIZU et al. 1994). Several components of a Pkclp- regulated kinase cascade have been identified in yeast. These were isolated based on their ability to suppress lysis in a strain deficient in either Pkclp or in another component of the Pkclp-governed cell wall metabolism pathway (LEE and LEVIN 1992; LEE et al. 1993a,b; IRIE et al. 1993). Epistatic analysis suggests that Pkclp activates Bcklp, which activates the mitogen-activated protein kinase kinase (MAPKK) homologues Mkklp and Mkk2p, which in turn activate Mpklp, a MAPK homologue. Members of the MAPK and MAPKK families have also been implicated in PKC-mediated responses in mammalian cells (KOLCH et al. 1993; LEVIN and ERREDE 1993).

Corresponding author: Lorraine S. Symington, Institute of Cancer Research and Department of Microbiology, Columbia University Col- lege of Physicians and Surgeons, 701 W. 168th St., New York, NY 10032. E-mail: [email protected]

Genetics 141: 1275-1285 (December, 1995)

A second MAPK cascade in yeast regulates the osmosensing pathway, which increases intracellular glycerol production in response to increases in extracellular osmolarity. HOG1 and PBS2 were identified as essential for viability on high-osmolarity medium and encode members of the MAPK and MAPKK gene families, respectively (BREWSTER et al. 1993). Phosphorylation of Hoglp in response to increased osmolarity is dependent on PBS2 (BREWSTER et al. 1993). Activation of this MAPK cascade is likely to be mediated by the Slnlp/ Ssklp protein complex. Slnlp, on the basis of its homology to histidine kinase sensory systems in bacteria, is proposed to recognize the osmolarity of the medium and regulate the phosphorylation state of Ssklp (MAEDA et al. 1994). Deletion of SLNl or a dominant negative mutation in SSKl results in constitutive activation of the Hoglp pathway and inviability. Three phosphatases were identified as potential negative regulators of this pathway based on their ability to suppress this lethality in a dosage-dependent manner. These include the PTPZ, PTCl and PTC3 gene products (MAEDA et al. 1994). PTP2 encodes a protein tyrosine phosphatase, and PTCl and PTC? encode type 2C serine/threonine protein phosphatases. PTCl was previously identified by mutations that were synthetically lethal in combination with a ptp l ptp2 double mutation, which removes the two known protein tyrosine phosphatases (MAEDA et al. 1993). PTCl was also identified as TPDI, based on its role in tRNA splicing (ROBINSON et al. 1994).

Previously we described a colony sectoring assay in

1276 K. N. Huang and L. S. Symington

which the rate of sectoring is proportional to the rate of recombination at two ade2 alleles in direct repeat (HUANG and SWINGTON 1994). Using this assay, an allele of PKCl was isolated which promoted a marked increase in the rate of sectoring and conferred a temperature-sensitive (ts) growth defect (HUANG and SW- INGTON 1994). We proposed that the regulation of cell wall metabolism and recombination were mediated through separate Pkclp-controlled pathways, based on two observations. First, the addition of osmotically sup- portive agents to the medium, such as 1 M sorbitol or 100 mM CaCl,, suppressed the growth defect but did not affect the hyperrecombination phenotype conferred by the pkcl-4 allele. Second, deletion of MPKl, which encodes a downstream component of the Pkclp cell wall metabolism pathway, had no effect on the rate of recombination. To identify regulatory components of the Pkclpdependent pathways controlling growth and recombination, we isolated second site mutations that suppressed both the ts and hyperrecombination phenotype conferred by the pkcl-4 allele or that suppressed either phenotype alone. Here we describe the mutations that suppressed both of these phenotypes.

MATERIALS AND METHODS

Strains: The yeast strains used in this study (Table 1) are derivatives of W303-1A or W303-1B (THOMAS and ROTHSTEIN 1989). The kcs mutant strains YKH138 and YKH139 were de- rived from UV-induced mutagenesis of YKH29a. The kcsl deletion strain was made by introducing the HZS3gene between the EcoRV and Hind11 sites of KCSl (nucleotides 457 and 2731 of the coding sequence, respectively). This construct was introduced to the KCSl locus by one-step gene replacement (ROTHSTEIN 1983). The kcs2/ptcl deletion strains were made by introducing the LEU2 gene between the SnaBI and BamHI sites of KCS2/PTCl (nucleotide 118 of the coding region and nucleotide 7 in the 3' untranslated region, respectively) (MAEDA et al. 1993). A DNA fragment containing the kcs2/ ptclA::LEU2 construct was introduced to the KCS2/PTCl locus by one-step gene replacement (ROTHSTEIN 1983). For kcsl and kcs2/ptcl disruptions, transformants were subjected to Southern analysis to ensure integration at the correct locus (SOUTHERN 1975).

YKH40-3 was made by crossing DL523 to YKH12a. The resulting Leu+ segregants were backcrossed twice more to YKH12a or YKH12a to produce YKH40-3.

YKH10, 21, 146, 147, 149, 150,151,153, 161, 164,163, 162 and 165 are Ade' Ura- derivatives of YKH12a, 29a, 114, 141, 138, 139, 112, 142, 74, 113, 144, 140 and 145, respectively. These have lost the ade2 direct repeat recombination substrate.

Standard procedures were used for mating, sporulation and tetrad analysis (GUTHRIE and FINK 1991). Transformation of yeast was performed using the lithium acetate method (IT0 et al. 1983).

Media and culture conditions: Media for yeast growth were prepared as described by SHERMAN et al. (1986). YEPD medium (1% yeast extract, 2% Bacto-peptone, 2% dextrose, and 2% agar for plates) was used for nonselective growth. 100 mM CaC& or 1 M sorbitol was added to YEPD medium as indicated. Nutritional requirements were determined on synthetic medium lacking one amino acid or nucleic acid base (2% dextrose, 0.67% yeast nitrogen base, 2% agar). Eschm'chia coli

strains were grown at 37" in LB medium or LB supplemented with 100 pg of ampicillin per ml.

Determination of recombination rates: To determine the rate of adenine prototroph formation, cells were plated at SO" on YEPD plates. Seven colonies were picked from each plate, and dilutions were plated on medium lacking adenine to determine the number of Ade+ cells and on YEPD to determine the total number of cells in the colony. The median frequency was used to determine the rate of recombination per cell per generation (LEA and COUISON 1948). Each experiment was repeated three times; the median rate for each strain is given in Table 2.

Colony hybridization: A yeast genomic library (ROSE et al. 1987) was transformed into E. coli, and transformants were transferred to nitrocellulose filters. The colonies on the filters were lysed and probed for plasmids of interest using the proto- col of SAMBROOK et al. (1 989).

DNA sequence analysis: DNA sequencing was performed using the method of SANCER et al. (1977), using the Sequenase kit (United States Biochemical).

Northern analysis: Strains were grown at room temperature in YEPD medium to early log phase. The cultures were divided and a sample of each strain was concentrated by cen- trifugation and resuspended in YEPD containing 1 M glucose. The cultures were grown for 3 hr at 30" prior to RNA extrac- tion. RNA preparation and Northern analysis were performed as described (SHERMAN et al. 1986).

RESULTS

Isolation of mutants: To identify recombination mutants in yeast, we developed a colony sectoring assay that uses two heteroalleles of the ade2gene (STOTZ and LINDER 1990) in direct repeat orientation at the ADE2 locus (Fig- ure 1). The mutations in d e 2 cause the initial strain to form red colonies (REAUME and TATUM 1949; ROMAN 1956). If recombination between the two alleles produces a wild-type ADE2gene, a white sector will be formed within the red colony. Thus, the rate of sectoring indicates the rate of recombination within this strain, designated YKH12a (HUANG and SWINGTON 1994).

Using this assay, we previously identified an allele of PKCl (pkcl-4) that increased the rate of recombination in a temperature-dependent manner and conferred a temperature sensitive growth defect (HUANG and S Y " INGTON 1994). To identify components of the Pkclp- dependent pathways controlling cell wall metabolism and recombination, we initiated a genetic screen for second site mutations that suppressed either the hyperrecombination or ts phenotypes conferred by the pkcl- 4 allele. To isolate suppressors of the recombination phenotype, approximately 20,000 colonies of the p k c l - 4 strain YKH29a were plated on either YEPD o r YEPD supplemented with 100 mM CaCl,, mutagenized by W irradiation, and grown at 30". The addition of CaCl, to the medium suppresses the temperature-dependent growth defect of pkc l strains, but not the hyperrecombination phenotype. Colonies that displayed a decreased rate of sectoring were diluted into water and plated on medium lacking adenine to ensure that they were still capable of adenine prototroph formation, indicating that they retained the direct repeat construct. To iden- ti@ suppressors of the ts defect, YKH29a was mutagen-

Suppressors of a p k c l Defect

TABLE 1

Yeast strains used in this study

1277

Strain Genotype" Source

YKH12a YKH12a Y K H l O YKH29a YKH29a Y K H P l YKH74 YKH161 DL523 YKH40-3 YKH138 YKH149 YKH139 YKH150 YKH112 YKH151 SJRl3 YKHll3 YKH164 YKH114 YKH146 YKH144 YKH163 YKH159 YKH140 YKH162 YKH141 YKH147 YKH145 YKH165 YKH160 YKH142 YKH153 YKH148 YKH41

MA Ta MA Ta MATa ADE2 MATa pkcl-4 MATa pkcl-4 MATa pkcl-4 ADE2 MATa pkcl-2::HIS3 MATa pkcl-2::HIS3 ADE2 MATa pkclA::LEU2 leu2-3,112 tql-l ura3-52 his4 canlR MATa pkclA::LEU2 MATa kcsl-1 pkcl-4 MATa kcsl-1 pkcl-4 ADE2 MATa kcs2-1 pkcl-4 MATa kcs2-1 pkcl-4 ADE2 MATa kcs2-1 MATa kcs2-1 ADE2 MATa lys2 MATa kcslA::HIS3 MATa kcslA::HIS3 ADE2 MATa kcslA::HIS3 pkcl-4 MATa kcslA::HIS3 pkcl-4 ADE2 MATa kcslA::HIS3 pkcl-2::HIS3 MATa kcslA::HIS3 pkcl-2::HIS3 ADE2 MATa kcslA::HIS3 pkclA::LEU2 MATa ptclA::LEU2 MATa ptclA::LEU2 ADE2 MATa ptclA::LEU2 pkcl-4 MATa ptclA::LEU2 pkcl-4 ADE2 MATa ptclA::LEU2 pkcl-2::HIS3 MATa ptclA::LEU2 pkcl-2::HIS3 ADE2 MATa ptclA::LEU2 pkclA::LEU2 MATa kcslA::HIS3 pctlA::LEU2 MATa kcslA::HIS3 ptclA::LEU2 ADE2 MATa kcslA::HIS3 ptclA::LEU2 pkcl-4 MATa mpk1A::TRPl

HUANC and SYMINGTON (1994) HUANC and SYMINGTON (1994) This study HUANC and SYMINGTON (1994) HUANC and SYMINGTON (1994) This study HUANC and SYMINGTON (1994) This study D. LEVIN This study This study This study This study This study This study This study S. JINKS-ROBERTSON This study This study This study This study This study This study This study This study This study This study This study This study This study This study This study This study This study HUANG and SYMINGTON (1994)

~ ~~

"All strains are in the YKH12 backgroup (ad&n::URA3::ade2a leu2-3, 112his3-11, 15 can 1-100 ura3-l trpl-1) (HUANG and SYMINGTON 1994) except SJRl3 and DL523. which are unrelated strains.

ized by UV irradiation, and mutants that permitted growth at 37" on YEPD were selected. Mutants were selected from approximately 60,000 cells that would have survived the mutagenesis at the permissive temperature. The mutants identified for their ability to reduce the rate of sectoring were tested for the ability to suppress the ts defect, and those isolated at 37" were assayed for their effect on recombination rates.

TABLE 2

Rates of recombmation at the ADE2 locus in kcsl and ptcl strains at 30"

Relevant genotype Rate Ade+/cell/gen Relative rate

Wild type 5.52 X 1x Pkcl-4 4.05 X 1 0 - ~ 73 x Pkcl-4 k w l A 6.64 X 1.2x Pkcl-4 ptcl A 1.55 X 1 0 - ~ 2.8X kcslA 8.78 X 1.6X ptcl A 5.44 x 10-6 0.99x

A total of 26 mutants falling into 20 complementation groups were isolated. Sixteen of these mutants, compris- ing 16 complementation groups, suppressed only the hyperrecombination phenotype and not the ts growth defect conferred by the pkcl-4 allele. An additional 10 mutants defined four complementation groups, KCSl- 4, which suppressed both phenotypes. The KCSl complementation group, containing five mutant alleles, and the KCS2 complementation group, containing three mutant alleles, showed the strongest suppression and were characterized further. When backcrossed to YKH29a, kcsl and kcs2 mutations were determined to be recessive and to segregate as single gene traits. The kcs mutants were crossed to SJRl3, a strain with wild- type ADE2 and PKCl alleles, and the resulting diploids were dissected. Because many segregants from these crosses displayed the ts and hyperrecombination phenotypes of YKH29a, the kcs mutations were determined not to be linked to either the ade2 direct repeat construct or the PKCl locus.


Popout d & Gene Conversion

White, ura+

white, ura‘

Cloning of the KCSZ gene: From the crosses to SJRl3, which contains a wild-type TRPl gene, it was determined that both the kcsl and kcs2mutations were linked to the TRPl locus. Therefore, a chromosome walking strategy was used to identify a plasmid containing the KCSl gene. A cosmid clone containing approximately 40 kb of yeast DNA, including the TRPl locus (Ameri- can Type Culture Collection cosmid #70938) (Figure 2), was restriction mapped, and DNA fragments from different regions were isolated and used as probes to screen a yeast genomic library (ROSE et al. 1987) transformed into E. coli. Fragments from yeast plasmids that extended beyond the ends of the cosmid clone were

FIGURE 1.-The ade2 recombination s u b strate and pathways. Recombination between the two ade2 genes can produce a wild-type copy of ADE2. This can occur by a popout mechanism, resulting in loss of the interven- ing URA3 gene, or by gene conversion, in which the URA3 gene is retained ( HUANC and SYMINGTON 1994).

used to continue walking along the genome. A total of nine independent plasmids were obtained, which contained overlapping DNA inserts spanning 70 kb of chromosome N (Figure 2). Each of these was tested for the ability to confer temperature sensitivity to the kcsl pkcl-4 and the kcs2 pkcl-4 strains YKH149 and YKH150, respectively. One plasmid, #23, complemented the phenotype of the kcsl pkcl-4 double mutant, yet none complemented the kcs2 pkcl-4 phenotype.

Subsequent subcloning and sequence analysis of plasmid #23 determined that the minimal complementing region contained a previously unidentified gene (Fig-

R R R R R R R R R R R R R RRR RRR R I I I 1 1 1 I I 1 I 1 I I I 1 I II I I

0 I 0 0 /v PTC1 CEN4 TRPl KCSI

t I 70938

10 kb c”-------l

FIGURE 2.-Cloning of the KCSl gene. The centromeric region of chromosome N i s shown. Fragments of cosmid 70938 were used to screen a yeast genomic library and identify seven of the nine overlapping plasmid clones shown. Fragments from yeast plasmids that extended beyond the ends of the cosmid clone were used to continue walking along the genome. Plasmid #23 conferred temperature-sensitivity to the kcsl pkcl-4 strain. The open reading frames of PTCl, TRPl and KCSl are indicated by open boxes. The centromere is indicated by a darkened box. R, EcoRI site.

Suppressors of a pkc l Defect

A W M C l C C W T C Q Q T A C C C C A Q A T ~ ~ l l T l ' A m ~ m M A T P K S V P E G L Q V S E K K N N P D T L S S S L S S F I L S N H E E P A I K

c ~ ~ T A ~ C ~ ~ ~ ~ A ~ ~ R H T F K V K T Y S T L S Q S L R Q E N V N N R S N E K K P Q Q F V P H S E S I

TAQCQMTCMT

C T A T A M T A ~ T ~ ~ ~ ~ ~ m A T ~ ~ A N A Q D G V P D M D E D T G N E T I N N D N H G C H L D S Q K N M I I K S L A Y

1279

35

155

175

395

515

635

755

875

995

1115

la35

1355

1175

1595

1715

1835

1955

1075

1195

2315

1435

1555

2675

1795

2915

3035

3155

3275

3395

FIGURE 3.-Nucleotide and predicted amino acid sequence of the KCSl gene. Nucleotide 1 is the A of the initiating methionine. Boxes have been placed around leucine residues that could participate in the formation of two leucine zippers. The asterisk indicates the termination codon. Components of a possible transcription termination consensus sequence are underlined (ZARET and SHERMAN 1982).

ure 3). This coding region was partially replaced with indicating that this coding region contains the KCSl the HZS3 gene, and the resulting construct was used to gene (data not shown). replace the wild-type locus in the pkcl-4 strain YKH29a. Analysis of the KCSl sequence predicts that it en- The resulting strain is no longer ts or hyperrecombino- codes a protein 1050 amino acids in length, with a pre- genic and fails to complement a kcsl pkcl-4 mutant, dicted molecular weight of 119 kD. The sequence is


Y EPD CaCI, first leucine zipper element, the sequence NKPCAL DLK is found, and upstream of the second, the sequence most closely matching the consensus is QFAR- VLAR.

Cloning the KC32 gene: None of the genomic clones

perature sensitivity to the kcs2 pkcl-4 strain. An alterna- tive cloning strategy was enabled by the observation that kcs2 mutants were unable to grow on medium containing 100 mM CaC& (Figure 4). This phenotype was

34" described above and shown in Figure 2 conferred tem-

F l c ; r ' K ~ . 4.-(:;dcium sensitivity o l ' p / c l strains. From top: YKHIO (wild type). YKH2I (pkr l -4) , YKH14i ( p t c l A : : I J W 2 p k c I - 4 ) , YKH 162 (/)/CIA :: IJZJ2) . Serial dilutions of each strain werc grown on YEPD medium (A) or YEPD supplemented with 100 mu CaCI, (B) at 34".

most closely related to the basic leucine zipper (bZIP) family of transcription factors (reviewed in RAXEVANIS and VINSON 1993; COHEN and PARRY 1994). Kc.YI is unusual in that it contains two potential leucine zipper motifs, although the presence of two glycine residues in the last seven residues of the first putative leucine zipper region suggests that this segment may not participate in the formation of an alpha-helical structure. In addition to two putative leucine zipper elements, the KCSI gene product contains a number of basic residues N-terminal to these elements. It should be noted that the consensus sequence NXX(A/X) (A/X)XXXR, which is conserved in most but not all bZIP proteins ( BAXEVANIS and VINSON 1993), is only loosely matched in the predicted sequence of Kcslp. Upstream of the

specific for the addition ofCaCIP, ratherthan a general sensitivity to increased osmolarity, as the kcs2 mutants were able to grow on plates containing 1 M glucose, 0.5 M NaCl, or 1 M sorbitol (data not shown). To isolate the gene altered in the kcs2 mutant, a yeast genomic library (ROSE et al. 1987) in the vector YCp50 was introduced to YKH151, and transformants were selected on YEPD plates containing 100 mM CaCl?. Colonies were subsequently tested for uracil prototrophy to test for the presence of plasmids. Three transformants were isolated and found to contain the same plasmid. This plasmid also conferred temperature sensitivity to the kcs2 pkcl-4 strain YKH150. Subsequent subcloning and sequence analysis showed that the minimal complementing fragment contained the previously identified PTC1 gene (MAEDA et al. 1993). The PTC1 gene was deleted, and crosses using this deletion confirmed link- age to the TRPI locus, as previously reported (MAEDA et al. 1993; ROBINSON et al. 1994). Our analysis suggests that PTC1 is approximately 8-10 kb from the centromere on the left arm of chromosome N. This estimate - - c

FIGUKF. 5.-kcsl and p t c l mutations reduce the rate of Ade' sectoring in pkcl-4 and pkcl-2 strains. Recombination between the two ndc.2 alleles can produce a wild-type ALE2 gene, resulting in the formation of a white sector in the red colony. The rate of sectoring is therefore proportional to the rate of recombination in these strains. (A) W 1 2 a (wild type); (B) YKH29a (pkcl-4);

( G ) YKH145 ( p / c J A :.'LEU2 pkcl-2). All strains were grown on YEPD plates at 30". (C) YKH114 (k~~lA: :HI .S3f ikc l -4 ) ; (D) YKH141 (pk lA: :LEU2pkc l -4 ) ; (E) YKH74 (pkcl-2); (F) YKH144 (kmlA::HIS3pkcl -2) ;

Suppressors

Sorbitol YEPD

24"

34"

FKX.KI;. (i.--kc:sI and p / c l mutations supprcss the temperature sensitivity o f the pkcl-4 strain. From top: YKH10 (wild type), YKH21 ( p k c l - 4 ) , YKH146 (kcs lA::HIS3 pkcl -4) , YKH147 (ptcIA::I-I:'U2 pltcl-4), YKH164 (kcs lA: :HIS3) , and YKH 162 ( p / c l A ::IJW2). Serial dilutions of each strain were grown on YEPD plates with 1 M sorbitol (A, C and E) or YEPD plates without sorbitol (B, D and F ) at 24" (A and B), 34" (C and D), or 37" ( E and F ) .

is made by combining information from the genome sequencing project (a genomic clone with accession #I248432 contains the IYI'CI gene), the chromosome maps compiled by L. RILES and M. OLSON (personal communication), and the YCp50 genomic clones gener- ated in this study.

The ptclA mutation suppressed both the ts and hyperrecombination phenotypes conferred by the j1kc1-4 allele, and the ptclA strain failed to complement kcs2 mutants. We shall henceforth refer to KCS2 as IYTCI.

Because the disruption alleles of KCSI and PTC1 had the same phenotypes as the original alleles, all subsequent experiments were performed using the disrup tion strains.

Rates of recombination at the ade2 locus: As shown in Figure 5, mutation of KCSI or K l ' C I reduced the rate of sectoring of the pkcl-4 and pkcl-2 strains to a p proximately wild-type levels. The rate of recombination at the ADK2 locus was determined by measuring the frequency of adenine prototroph formation. The kcsl and ptcl disruption mutations suppressed the rate of recombination in the pkrl-4 background to near wild- type levels, but did not affect recombination in the wild- type background (Table 2). This suggests that the KCSI and IYI'CI gene products are involved in the fIkcl-depen-

of a p k c l Defect 1981

dent stimulation of recornhination rather than the con- trol of recombination in general.

Allelespecific suppression of the ts phenotype: As shown in Figure 6, a kcslA or ptr1.A mutation s u p presses the ts growth defect of the pkcl-4 strain at 34". Although ptc1A 11kc1-4 and krslA pkrl-4 mutants grow as patches and are identifiable when replica plated to a plate incubated at 37", they fail to grow well as single colonies at this temperature (Figure 6) . For the j I t r . 1 j1kc1-4 double mutant this was not unexpected, since a p t c l mutation alone results in a ts phenotype at 37" (MAEDA Pt nl. 1993; ROBINSON Pt nl. 1994). Mutation o f KCSI or PTC1 failed to suppress the inviability of the pkcl deletion mutant on medium lacking an osmotic agent (Figure 7). In addition, the ptcl mutation only weakly suppressed the growth defect conferred by the @l-Z allele at 32". and neither the krsl nor ptcl mutation suppressed the growth defect of the pkcl-2 strain at 34" (Figure 8). Because the activity of the pkrl-4 and ~1kcl-2-encoded kinases are likely to decrease with increasing temperature, these results indicate that kcsl and p tc l mutations are able to compensate for reduced Pkclp kinase activity, yet cannot compensate for loss of the kinase.

To determine if a kcsl or ptcl mutation could s u p press the ts phenotype associated with mutation of another component of the Pkclp cascade, ka lA and ptclA mutants were crossed to an mpkl deletion strain. The resulting diploids were sporulated and tetrads were dissected. From the haploid progeny, it was determined that a kcslA mutation failed to suppress the ts phenotype of the mpklA mutant. Furthermore, it was determined that a p tc l mpkl double mutant was inviable. When a ptclA::LEU2 haploid was crossed to an m p k l A :: TRPI haploid, no Leu+ Trp' segregants were recovered from 23 tetrads analyzed. However, 17 Leu- Trp- spores were obtained, indicating that the two loci are unlinked.

kcsl A::HZS3 ptclA::LEU2 cells display a severe growth defect: To determine the phenotype of the kcslA ptclA double mutant, the F.Cl gene was dis-

Sorbitol Y EPD

24"

FI(;I.KE 7.-M.lutation of p/cI or k1:c.I liils 10 supprw the inviability caused by deletion o f j ) k c l . From top: Y K H 1 0 (wild type), YKH40-3 (/1kclA::Il:U2),YKH159 ( k c r l A : : I - I I S 3 ~ ) k c l A : : I ~ U ~ , YKHIfO (p/clA::I~,'U2pkckclA::I~U2), YKH164 (kcTlA::HLS$, and YKH162 ( p / r l A : : I L U 2 ) . Serial dilutions of each strain were grown on a W,PD plate with 1 M sorbitol (A) or a YEPD plate without sorbitol (B) at 24".

12x2 I(. N. Hrlang and I.. S. Synington

Sorbitol Y EPD

24"

32"

34"

PTP2 transcript levels are reduced in mpkl and pkcl- 4 strains: I t has previouslv been reported that a ptr l mutation results in a synthetic growth defect in combination with ptf12, which encodes a protein tyrosine phosphatase ( MAEDA pt nl. 1994). PTP2 would therefore be a possible target for regulation by the KCSl gene product. Failure to induce expression of 1'77'2 in a kcslA strain would result in a krsl ptr l synthetic growth defect similar to that observed in a ptp2 p f c l strain.

Northern analysis was performed to determine pTp2 transcript levels in wild tye , krsl, p tc l , mpkl, and pkc l - 4 strains. Because ITP2 has been implicated as a component of the Hoglp MAPK pathwav, which regulates the response to osmotic stimulation, cells were grown in either YEPD or in YEPD supplemented with 1 b[ glucose. As shown in Figure 10, although the level of PIP2 mRNA was not reduced in a krsl mutant strain, the level was markedly reduced in both the mpkl strain and in the pkrl-4 strain, suggesting that expression of pTP2 is dependent on components of the Pkclp MAPK cascade. In addition, expression of ITP2 was increased slightly when cells were grown in the presence of 1 M

glucose. The role of the PTP2 gene product in the Pkclp

MAF'K cascade: Because I ' K I and l T P 2 were both identified as negative regulators of the Hoglp MAPK cascade, and our work suggested that the ITC1 gene product similarly regulates the Pkclp MAPK cascade, we disrupted the PIP2 gene in the pkcl-4 background to determine if this would affect either the ts or hyperrecombination phenotype conferred by the pkrl-4 allele. It was determined that deletion of l T P 2 affected neither of these phenotypes (data not shown).

DISCUSSION

The isolation of the pkcl-4 allele in a screen for hyperrecombination mutants identified a novel role for Pkclp, as the regulation of DNA metabolism was a previously unidentified role for protein kinase C in any system (HL~,~NG and SYMINGTON 1994). We proposed that failure to phosphorylate one or more proteins involved in DNA metabolism results in the formation of recombinogenic DNA lesions and the observed hyperrecombination phenotype. The isolation of compensa-

Y EPD Y EPD Sorbitol

FIGLIKE 9.-A IrcslA p/rlA double mutant grows only at 24". From top: YKH10 (wild type). YKH164 (/tcslA::HIS?),M(Hl(~2 (jltrlA::IJ:lJ2), and YKHl.53 (/tcslA::HIS? fltr1A::IJXJZ). Serial dilutions of each strain were grown on YEPD plates (A and R) or a YEPD plate with 1 &I sorbitol ( C ) at 24" (A) or 30" (R and C).

24" 30" 36'

Suppressors of a pkcl Defect 1283

YEP + Y EPD 1 M glucose

1 2 3 4 5 1 2 3 4 5 "

PTP2

FIGURI:. 10.-Northern analysis of pTP2 transcript levels. Cells were grown at 30" in either YEPD or YEPD with 1 Y glucose. Ten pg of total RNA was loaded in each lane. Strains analyzed include YKH12a (wild type, lane l ) , RH113 (kc.sIA::HIS3, lane 2), YKH140 (p1clA::lIU2, lane 3), YKH41 ( m p k l A : : T R P l , lane 4), and YKH29a (pkcl-4, lane 5). The blot was first probed with a fragment of the " 2 gene, then stripped and reprobed with a fragment of the ACT1 gene.

tory mutations in the phosphatase gene PTC1 supports this model. We suggest that Ptclp is involved in the dephosphorylation of specific targets of Pkclp or the Pkclp kinase cascade. The ability of ptcl mutations to suppress both the ts and hyperrecombination phenotypes conferred by the pkcl-4 allele suggests that there are at least two substrates for Pkclp in the cell that are also both targets of the Ptclp phosphatase. We propose that a p tc l defect compensates for a kinase with reduced activity. This is supported by our observation that mutation of PTC1 does not suppress the growth defect of the pkcl null mutant. In addition, a plcl mutation fails to suppress the ts phenotype of a pkc l -2 strain at 34", yet weakly suppresses the growth defect at 32" and sup presses the hyperrecombination defect conferred by the pkcl-2 allele at 30". It is likely that the activity of the kinase encoded by the pkcl-2 allele decreases with increasing temperature, and the ability of a p tc l mutation to compensate for the defects conferred by the pkcl-2 allele is similarly temperature dependent.

The KCSI gene encodes a previously unidentified protein most closely related to the bZIP family of transcription factors. The target(s) of Kcslp are not known at this time. One possible role for Kcslp is to regulate the expression of a phosphatase gene. This is consistent with our observation that, as with ptcl mutations, a kc.sl mutation suppresses the pkcl growth defect in an allelespecific manner. Our data show that expression of the phosphatase gene PTP2 is not altered in a hcsl strain. A second possible target for regulation by Kcslp is the PTC1 gene. This would be consistent with the ability of kcsl and p tc l mutations to suppress the phenotypes conferred by the different pkcl alleles in a similar fash- ion. However, PTC1 expression could not be solely controlled by Kcslp since a kcsl mutation does not display many of the phenotypes displayed by the p tc l mutant, such as sensitivity to CaCI,,. In addition, PTC1 could

not be the sole target of Kcslp, since the kcslA::HZS3 p t c l A : : M U 2 double mutant has a more severe phenotype than the p t c l A : : Z E U 2 single mutant. Northern analysis confirmed that Kcslp is not required for the expression of PTC1 (not shown).

I t is interesting that PTC1 has been isolated in con- nection with two MAF'K cascades. In this study, we re- port the isolation of three ptcl alleles that suppress a pkc l defect. Previously, PTC1 was found to act as a dos- agedependent suppressor of an slnl deletion mutant (MAEDA et nl. 1994). Deletion of SLN1 results in lethality due to the constitutive activation of the Hoglp/Pbs2p MAPK cascade. It is proposed that Ptclp acts to dephosphorylate Hoglp and/or Pbs2p, resulting in downreg- dation of the cascade. The isolation of PTC1 as an extragenic suppressor of a defect in a second MAPK cascade could indicate a role for Ptcl in the dephosphorylation of the MAPK/MAPKK homologues encoded by M P K l , M K K l or MKK2. Again, it could be proposed that dephosphorylation by Ptclp reduces the activity of components of the MAPK cascade. A model for this is shown in Figure 11.

Our observation that the p t c l A m p k l A double mutation is synthetically lethal was unexpected. Since MPKl and PTC1 were isolated as positive and negative factors in this cascade, respectively, it is odd that they have a combinatorial defect. If the sole role of Ptclp in the Pkclp kinase cascade is to dephosphorylate components of the cascade, deletion of PTC1 in an m p k l A background should cause no greater defect than either single deletion. It is possible that Ptclp acts on an additional substrate that is important for cell wall mainte- nance. Alternatively, this effect could be related to the apparent role of Mpklp in the regulation of PTP2gene

DNA

Protein pKC1 + Metabolism

1 T BCK1

PTC1

MKKl I PTC1

MKK2 .;/ I

1 MPKl

FIGURE 11.-A potential model for the regulation of the Pkclp-mediated recombination and cell wall metabolism cascades by Ptclp. PKCl, RCKl, MKKI, MKK2 and MPKl all encode putative serine-threonine protein kinases that have been proposed to act sequentially in a kinase cascade as dia- grammed (FIELDS and THORNER 1991; LEE and LEVIN 1992; IRIE ef a/. 19%; LEE el a/. 1993b).


expression. Since a ptp2 mutation results in a synthetic growth defect in combination with a ptcl mutation, it would be expected that deletion of a gene encoding a positive regulator of FTP2, such as MPKI, would have a similar effect. Because the phenotype of the ptcl mpkl double mutant is more severe than the ptcl ptp2 double mutant, it is likely that deletion of MPKl results in additional defects as well.

It is interesting that expression of pTp2 is dependent on the PKCI and W K l genes, Because we find no evidence for regulation of the Pkclp MAPK cascade by Ptp2p, our data suggest that there is “crosstalk” between the Hoglp and Pkclp MAPK cascades. We propose that one role of Mpklp is to activate pTp2 transcription, thereby inducing a repressor of the competing Hoglp MAPK pathway. There are at least two additional examples of phosphatases whose expression is regulated by MAPK cascades. In mammalian cells, mitogenic stimulation triggers the activation of MAP kinases through the phosphorylation of both tyrosine and threoine residues. In mouse 3T3 cells, the phosphatase MKF”1 (3CH134) is a dual specificity phosphatase that dephosphorylates ~ 4 2 ” ~ (SUN et al. 1993). Expression of the gene encoding MKP-1 is induced within minutes following mitogenic stimulation (LAU and NATHANS 1987), suggesting that one role of the MAPK pathway is to activate the phosphatase that negatively regulates this cascade. This provides a mechanism of attenuation for the mitogenic signal. A second dual specificity phosphatase is encoded by the MSG5 gene of S. cereuisiae. The MSG5 gene product acts to dephosphorylate Fus3 (DOI et al. 1994), the MAPK homologue induced in response to pheromone stimulation. Like the gene encoding MKP-1, MSG5 transcription is induced concurrently with activation of the MAPKpathway that it regulates (DOI et al. 1994). Unlike MSG5 and MKF”1, Ptp2p is the first phosphatase whose synthesis appears to be regulated by a MAPK cascade that it does not modify.

We thank S. JINKS-ROBERTSON, D. LEVIN and T. MAEDA for plasmids and strains. Supported by grants from the Irma T. Hirschl Trust and from the National Institutes of Health (GM41784). K.N.H. was supported in part by a predoctoral fellowship from the National Insti- tutes of Health (CA09503). L.S.S. is a scholar of the Leukemia Society of America.

LITERATURE CITED

BAXEVANIS, A. D., and C. R. VINSON, 1993 Interactions of coiled coils in transcription factors: where is the specificity? Curr. Opin. Genet. Dev. 3: 278-285.

BREWSTER, J. L., T. DE VALOIR, N. D. DWYER, E. WINTER, and M. C. GUSTIN, 1993 An osmosensing signal transduction pathway in yeast. Science 259: 1760-1763.

COHEN, C., and D. A. D. PARRY, 1994 a-helical coiled coils: more facts and better predictions. Science 263: 488-489.

Dol, K., A. GARTNER, G. AMMERER, B. ERREDE, H. SHINKAWA et al., 1994 MSG5, a novel protein phosphatase promotes adaptation to pheromone response in S. cereuisiae. EMBO J. 13: 61-70.

FIELDS, F. O., and J. THORNER, 1991 Genetic suppression analysis of the function of a protein kinase C (PKCI gene product) in

Cold Spring Harbor Symp. Quant. Biol. 56: 51-60. Saccharomyces cereuisiaecell cycle progression: the SKCd mutations.

GUTHRIE, C., and G. R. FINK, 1991 Guide to Yeast Genetics and Molecu- lar Biology. Academic Press, San Diego, CA.

HUANG, K. N., and L. S. SYMINGTON, 1994 Mutation of the gene encoding protein kinase C 1 stimulates mitotic recombination in Saccharomyces cereuisiae. Mol. Cell. Biol. 14: 6039-6045.

IRIE, K., M. TAKASE, K. S. LEE, D. E. LEVIN, H. ARAKI et al., 1993 MKKl and MKK2, which encode Saccharomyces cereuisiae mitogen- activated protein kinase-kinase homologs, function in the pathway mediated by protein kinase C. Mol. Cell. Biol. 13: 3076- 3083.

ITO, H., Y. FUKUDA, K. MURATA and A. KIMURA, 1983 Transforma- tion of intact yeast cells treated with alkali cations. J. Bacteriol. 153: 163-168.

KOL.CH, W., G. HEIDECKER, G. KOCHS, R. HUMMEI., H. VAHIDI PI al., 1993 Protein kinase Ca activates RAF-1 by direct phosphorylation. Nature 364: 249-252.

LAU, L. F., and D. NATHANS, 1987 Expression of a set of growth- related immediate early genes in BALB/c 3T3 cells: coordinate regulation with c-fos or c-fos. Proc. Natl. Acad. Sci. USA 84: 1182- 1186.

LEA, D. E., and C. A. COULSON, 1948 The distribution of the num- bers of mutants in bacterial populations. J. Genet. 49: 264-284.

LEE, K. S., and D. E. LEVIN, 1992 Dominant mutations in a gene encoding a putative protein kinase (BCKI) bypass the require- ment for a Saccharomyces cereuisiaeprotein kinase C: homolog. Mol. Cell. Biol. 12: 172-182.

LEE, K. S., L. K. HINES and D. E. LEVIN, 1993a A pair of functionally

L

redundant yeast genes (PPZI and PPZ2) encoding type 1-related protein phosphatases function within the PKCI-mediated pathway. Mol. Cell. Biol. 13: 5843-5853.

,EE, K. S., K. IRIE, Y. GOTOH, Y. WATANABE, H. ARAKI et nl., 1993b A yeast mitogen-activated protein kinase homolog (Mpklp) me- diates signalling by protein kinase C. Mol. Cell. Biol. 13: 3067- 3075.

.EVIN, D. E., and E. BARTLETFHEURUSCH, 1992 Mutants in the S. cerevisine PKCI gene display a cell cycle-specific osmotic stability defect. J. Cell Biol. 116: 1221-1229.

.EVIN, D. E., and B. ERREDE, 1993 A multitude of MAP kinase activation pathways. 1. NIH Res. 5: 49-52.

LEVIN, D. E., F. 0 . FIELDS, R. KUNISAWA, J. M. BISHOP, and J. THORNER, 1990 A candidate protein kinase C gene, PKCI, is required for the S. cereuisiar cell cycle. Cell 62: 213-224.

MAEDA, T., A. Y. TSAI and H. SAITO, 1993 Mutations in a protein tyrosine phosphatase gene (lTP2) and a protein serine/threonine phosphatase gene ( P T C I ) cause a synthetic growth defect in Saccharomyces cereuisiae. Mol. Cell. Biol. 13: 5408-5417.

MAEDA, T., S. M. WURGLER-MURPHY and H. SAITO, 1994 A two-component system that regulates an osmosensing M A P kinase cascade in yeast. Nature 369: 242-245.

NISHIZUKA, Y., 1992 Intracellular signaling by hydrolysis of phospho- lipids and activation of protein kinase C. Science 258: 607-614.

PARAVICINI, G., M. COOPER, L. F1u~nl.1, D. J. SMITH, J.-L. CARPENTIER

functional PKCZ gene product. Mol. Cell. Biol. 12: 4896-4905. et al., 1992 The osmotic integrity of the yeast cell requires a

REAUME, S. E., and E. L. TATUM, 1949 Spontaneous and nitrogen mustard-induced nutritional deficiencies in Saccharomycrs rrrpui.v iae. Arch. Biochem. 22: 331-338.

ROBINSON, M. K., W. H. VAN ZYL, E. M. PHIZICKY, and J. R. BROACII, 1994 TPDI of Saccharomyces rereuisiap encodes a protein phosphatase PC-like activity implicated in tRNA splicing and cell sepa- ration. Mol. Cell. Biol. 14 3634-3645.

ROMAN, H., 1956 Studies of gene mutation in Saccharomyces. Cold Spring Harbor Symp. Quant. Biol. 21: 175-185.

ROSE, M. D., P. NOVICK, J. H. THOMAS, D. BOTSTEIN, and G. R. FINK, 1987 A Saccharomyces cereuisiae genomic plasmid bank based on a centromere-containing shuttle vector. Gene 60: 237-243.

ROTHSTEIN, R. J., 1983 One-step gene disruption in yeast. Meth. Enzymol. 101: 202-211.

SAMBROOK, J., E. F. FRITSCH and T. MANIATIS, 1989 Molecular cloning: a laboratg. manual. Cold Spring Harbor I>aboratory Press, Cold Spring Harbor, NY.

SANGER, F., S. NICKLEN and A. R. COLJISON, 1977 DNA sequencing with chain-terminating inhibitors. Proc. Natl. Acad. Sci. USA 7 4 5463-5467.

SHERMAN, F., G. R. FINK and J. B. HICKS, 1986 Methods in yeast genetics. Cold Spring Harbor Laboratories, Cold Spring Harbor, N Y .

Suppressors of a pkc l Defect 1285

SHIMIZU, J., K. YODA and M. YAMASAKI, 1994 The hypo-osmolarity- sensitive phenotype of the Saccharomyces mmisiae hpo2 mutant is due to a mutation in PKCI, which regulates expression of & glucanase. Mol. Gen. Genet. 242 641-648.

SOUTHERN, E. M., 1975 Detection of specific sequences among DNA fragments separated by gel electrophoresis. J. Mol. Bio. 98: 503-517.

STOTZ, A,, and P. LINDER, 1990 The ADE2 gene from Saccharomyces cumisim sequence and new vectors. Gene 95: 91-98.

SUN, H., C. H. CHARLES, L. F. LAU, and N. K. TONKS, 1993 MKP-1

(3CH134), an immediate early gene product, is a dual specificity phosphatase that dephosphorylates MAP kinase in vivo. Cell 75:

THOMAS, B. J., and R. ROTHSTEIN, 1989 Elevated recombination

ZARET, K. S., and F. SHERMAN, 1982 DNA sequence required for

487-493.

rates in transcriptionally active DNA. Cell 5 6 619-630.

efficient transcription termination in yeast. Cell 28: 563-573.

Communicating editor: F. WINSTON

suppressors of a saccharomyces cerevisiae pkcl mutation ... · posed to act in the pkcl pathway....

Documents