genomic disruption of six budding yeast genes gives one drastic example of phenotype...

. 14: 655–664 (1998)

Genomic Disruption of Six Budding Yeast Genes gives
one Drastic Example of Phenotype Strain-dependence
ELIZABETH BILSLAND, MARIA DAHLEuN AND PER SUNNERHAGEN*

Department of Molecular Biology, Goteborg University, Sweden

Received 16 October 1997; accepted 26 November 1997

Using PCR to construct disruption cassettes, null alleles of six genes have been created in Saccharomyces cerevisiae.In a FY1679 background, no defects were detected in any of the haploid deletion mutants with respect to growth,gross morphology, or mating. A diploid FY1679-derived Äygl194c/Äygl194c homozygous disruptant displayedreduced sporulation. In contrast to the lack of phenotypic consequences of Äyol100w disruptions in the FY1679background, in the CEN.PK2 strain even a heterozygous disruption of the same gene caused striking effects, veryslow vegetative growth and highly impaired sporulation. Tetrad analysis showed YOL100w to be an essential genein this strain. A copy of the YGL194c or the YOL100w wild-type gene borne on a centromeric episomal plasmid wasintroduced into a corresponding disruption mutant strain, and in both cases was found to partially complement thedefects. ? 1998 John Wiley & Sons, Ltd.

Yeast 14: 655–664, 1998.

— Saccharomyces cerevisiae; PCR-based disruption; YOL113w; YOL100w; YOL107w; YOR267c;YGL196w; YGL194c

INTRODUCTION such analysis is genomic disruptions of genes of

Sweden. Tel: (+46) 31 773 3830; Fax: (+46) 31 773 3801;e-mail: Per. [email protected]/grant sponsor: European Commission (EUROFANprogramme)Contract/grant number: BIO4-CT95–0080

CCC 0749–503X/98/070655–10 $17.50

Saccharomyces cerevisiae was the first eukaryote tohave its genome entirely sequenced (Goffeau et al.,1997). The combination of complete sequenceinformation and ease of genetic manipulation hasmade this organism uniquely attractive to startanalysing the function of genes uncovered bygenomic sequencing, as well as of genes found byclassical genetics, by systematic approaches. Thework presented here is part of an internationalresearch effort, EUROFAN, with the aims ofcreating tools for and initiating such functionalanalysis on a large scale. One fundamental tool for

*Correspondence to: Per Sunnerhagen, Lundberg Laboratory,Goteborg University, P.O. Box 462, SE-405 30 Goteborg,

? 1998 John Wiley & Sons, Ltd.

unknown function, which can be produced withhigh speed and accuracy using PCR to createdisruption modules, and relying on the efficienthomologous recombination in S. cerevisiae.

Here we present the disruption of six genesfound by sequencing, and the initial phenotypiccharacterization of the resulting mutants.YOL100w, YOL113w and YOR267c are all pre-dicted to encode protein kinases (PKs). In terms ofsequence relatedness, YOL100w is close to DBF2,which encodes a PK implicated in nuclear division(Johnston et al., 1990), and is also similar to PKsregulated by cAMP. The predicted product ofYOL113w is similar to STE20p, a PK requiredfor activation of the pheromone response MAPcascade (Leberer et al., 1992). YOR267c is classi-fied in a PK subgroup so far not found inother organisms, where other members partici-pate in osmotolerance and regulation of nitrogen

Table 1. S. cerevisiae strains used in this study.

656 . .

Name Genotype Source or reference

FY1679 MATa/á ura3-52/ura3-52 his3-Ä200/+leu2-Ä1/+trp1-Ä63/+ (Winston et al., 1995)FSRN001 MATa/á yol113w(4, 1965)::kanMX4/+ This work

ura3-52/ura3-52 his3-Ä200/+leu2-Ä1/+trp1-Ä63/+FSRN002–05B MATa yol113w(4, 1965)::kanMX4 This work

ura3-52 leu2-Ä1 trp1-Ä63FSRN003–11C MATá yol113w(4, 1965)::kanMX4 This work

ura3-52 his3-Ä200FSRN004 MATa/á yol113w(4, 1965)::kanMX4/yol113w(4, 1965)::kanMX4 This work

ura3-52/ura3-52 his3-Ä200/+leu2-Ä1/+trp1-Ä63/+FSRN006 MATa/á yol100w(37, 2436)::kanMX4/+ This work

ura3-52/ura3-52 leu2-Ä1/+trp1-Ä63/+his3-Ä200/+FSRN007–02B MATa yol100w(37, 2436)::kanMX4 This work

ura3-52 trp1-Ä63FSRN008–02A MATá yol100w(37, 2436)::kanMX4 This work

ura3-52 his3-Ä200 leu2-Ä1FSRN009 MATa/á yol100w(37, 2436)::kanMX4/yol100w(37, 2436)::kanMX4 This work

ura3-52/ura3-52 his3-Ä200/+leu2-Ä1/+trp1-Ä63/+FSRN011 MATa/á yol107w(90, 1004)::kanMX4/+ This work

ura3-52/ura3-52 leu2-Ä1/+trp1-Ä63/+his3-Ä200/+FSRN012–14A MATa yol107w(90, 1004)::kanMX4 This work

ura3-52 trp1-Ä63FSRN013–08D MATá yol107w(90, 1004)::kanMX4 This work

ura3-52 his3-Ä200FSRN014 MATa/á yol107w(90, 1004)::kanMX4/yol107w(90, 1004)::kanMX4 This work

ura3-52/ura3-52 his3-Ä200/+leu2-Ä1/+trp1-Ä63/+FSRN016 MATa/á yor267c(11, 2161)::kanMX4/+ This work

ura3-52/ura3-52 leu2-Ä1/+trp1-Ä63/+his3-Ä200/+FSRN017–04C MATa yor267c(11, 2161)::kanMX4 This work

ura3-52 leu2-Ä1 trp1-Ä63FSRN018–08A MATá yor267c(11, 2161)::kanMX4 This work

ura3-52 his3-Ä200FSRN019 MATa/á yor267c(11, 2161)::kanMX4/yor267c(11, 2161)::kanMX4 This work

ura3-52/ura3-52 his3-Ä200/+leu2-Ä1/+trp1-Ä63/+FSRN021 MATa/á ygl196w(33, 574)::kanMX4/+ This work

ura3-52/ura3-52 leu2-Ä1/+trp1-Ä63/+his3-Ä200/+FSRN022–04C MATa ygl196w(33, 574)::kanMX4 This work

ura3-52 trp1-Ä63FSRN023–04B MATá ygl196w(33, 574)::kanMX4 This work

ura3-52 his3-Ä200 leu2-Ä1FSRN024 MATa/á ygl196w(33, 574)::kanMX4/ygl196w(33, 574)::kanMX4 This work

ura3-52/ura3-52 his3-Ä200/+leu2-Ä1/+trp1-Ä63/+FSRN026 MATa/á ygl194c(248, 1288)::kanMX4/+ This work

ura3-52/ura3-52 leu2-Ä1/+trp1-Ä63/+his3-Ä200/+FSRN027–06C MATa ygl194c(248, 1288)::kanMX4 This work

ura3-52 his3-Ä200 trp1-Ä63FSRN028–03B MATá ygl194c(248, 1288)::kanMX4 This work

ura3-52 leu2-Ä1 trp1-Ä63FSRN029 MATa/á ygl194c(248, 1288)::kanMX4/ygl194c(248, 1288)::kanMX4 This work

ura3-52/ura3-52 his3-Ä200/+leu2-Ä1/+trp1-Ä63/+CEN.PK2 MATa/á ura3-52/ura3-52 leu2-3,112/leu2-3,112 trp1-289/trp1-289

his3-Ä1/his3-Ä1K.-D. Entian

? 1998 John Wiley & Sons, Ltd. . 14: 655–664 (1998)

transport (Hunter and Plowman, 1997). YGL194c, cassettes by PCR was performed as described by

Table 1. S. cerevisiae strains used in this study.

657

also known as HOS2 (Rundlett et al., 1996) is aclose homolog of RPD3, which encodes a histonedeacetylase (De Rubertis et al., 1996). For theremaining two genes, YOL107w and YGL196w, noclear assignment to a functional class has so farbeen made on the basis of sequence similarity.

MATERIALS AND METHODS

Yeast strains and culture conditionsThe origins of all yeast strains used are given in

Table 1. Rich vegetative medium was YPD (1%yeast extract, 2% peptone, 2% glucose) or YPG(1% yeast extract, 2% peptone, 3% glycerol). G418-resistant transformants were selected on YPDcontaining 0·2 mg/ml G418. Synthetic vegetativemedium was SC containing 2% glucose, and sup-plemented with amino acids, uracil and adenine, asappropriate (Sherman, 1991). Sporulation mediumwas 1% potassium acetate.

Genomic disruptionsThe principle for the PCR-based gene disrup-

tions used here is shown in Figure 1. For identitiesand positions of homology of PCR primers, seeTable 2 and Figure 2. Synthesis of disruption


Wach (1996), with modifications. For the firstreaction, two standard PCR reactions were setup with primers L1+L2 or L3+L4 with genomicS. cerevisiae FY1679 DNA as a template. For thesecond reaction, the Expand= Long TemplatePCR system from Boehringer Mannheim, contain-ing a mixture of Taq and Pwo DNA polymerases,was used. Typically, 1 ìl (containing about 50 ngof PCR product) each of the first PCR reactionswas added and 0·3–0·5 ìg of NotI-digested plas-mid, pFA6-KanMX4 (Wach et al., 1994) or pFA6-HIS3MX6 (Wach et al., 1997). The second PCRwas then performed in a reaction volume of 50 ìl,containing 1 ì each of primers L1 and L4, 200 ìof each dNTP, 2·25 m-MgCl2, 50 m-Tris/HClpH 9·2, 16 m-(NH4)2SO4, and 0·75 ìl of enzymemix. A Perkin Elmer Cetus DNA Thermal Cyclerwas programmed to initially denature the samplesat 94)C for 60 s, then to perform 5 cycles consistingof 15 s at 94)C, 60 s at 50)C, and 90 s at 68)C, andfinally to perform 25 cycles consisting of 15 s at94)C, 30 s at 54)C, and 90 s at 68)C.

The product of a 50 ìl PCR reaction wasethanol precipitated and one half was used withoutfurther purification to transform S. cerevisiaeFY1679 to G418 resistance with a lithium acetate-based protocol (Gietz et al., 1992). To verify that

. 14: 655–664 (1998)

Name Genotype Source or reference

CSRN005 MATa/á yol113w(4, 1965)::kanMX4/+ This workura3-52/ura3-52 leu2-3,112/leu2-3,112 trp1-289/trp1-289

his3-Ä1/his3-Ä1CSRN010 MATa/á yol100w(37, 2436)::kanMX4/+ This work

ura3-52/ura3-52 leu2-3,112/leu2-3,112 trp1-289/trp1-289his3-Ä1/his3-Ä1

CSRN015 MATa/á yol107w(90, 1004)::kanMX4/+ This workura3-52/ura3-52 leu2-3,112/leu2-3,112 trp1-289/trp1-289

his3-Ä1/his3-Ä1CSRN020 MATa/á yor267c(11, 2161)::kanMX4/+ This work


CSRN025 MATa/á ygl196w(33, 574)::kanMX4/+ This workura3-52/ura3-52 leu2-3,112/leu2-3,112 trp1-289/trp1-289

his3-Ä1/his3-Ä1CSRN030 MATa/á ygl194c(248, 1288)::kanMX4/+ This work


FSRN strains were derived from FY1679, CSRN strains from CEN.PK2.

transformants represented integration events into YPD, YPG or SC, at three different temperatures:

658 . .

the genome by homologous recombination at thedesired locus, genomic DNA was prepared andanalysed by Southern blotting. The disrupted locuswas detected with a probe from the KanMX4module made by PCR using the primers 5*-CTGCAGGTCGACGGATCC-3* and 5*-CATCGATGAATTCGAGCTCG-3*. The respective wild-type loci were detected using probes generated byPCR and the respective L1+L2, L3+L4, orL1+L4 primers.

Phenotypic analysisAfter verification of the correctness of the

genomic integrations, the heterozygous diploiddisruptants derived from FY1679 were sporulatedand the tetrads dissected. The resulting haploiddisruptants were then subjected to growth tests on


15)C, 30)C and 37)C. Their mating proficiency wasalso investigated by crossing mutants with appro-priate auxotrophic markers and scoring formationof prototrophic diploids. By crossing haploid dis-ruptants, diploid homozygous disruptants werealso made. These were tested for sporulation pro-ficiency by microscopic inspection of the frequencyof tetrads in sporulation medium. By dissection ofsuch tetrads it was verified that four viablespores were produced and that all four wereG418-resistant.

Plasmid construction

Figure 1. General strategy for long flanking homology PCR-based genedisruption (Wach, 1996). (A) First PCR reaction. Primers L1+L2 andL3+L4 are used to synthesize 200–400 bp products of sequence flanking thegene to be disrupted, using genomic yeast DNA as a template. L2 and L3are hybrid primers which also contain homology to the marker cassette, inthis case the KanMX module. (B) Second PCR reaction. The products of thefirst PCR prime synthesis from the ends of the marker cassette by virtue ofthe cassette-homologous part of the L2 and L3 primers. (C) Homologousrecombination in vivo. The region between the 3* end of the genome-homologous part of the L2 and L3 primers (marked by the inner twovertical arrows) is deleted in the mutant, and replaced by the markercassette.

Disruption cassettes were cloned by ligatingthe PCR product to pUG7 (U. Guldener and J.Hegemann, in preparation) linearized with EcoRVand transforming Escherichia coli to ampicillin and

. 14: 655–664 (1998)

Table 2. Sequences of oligonucleotides used for generation of genomic disruptions.

659

kanamycin resistance; selection was performed on purification to transform S. cerevisiae strain

Primer Homologies Sequence

EBL1A YOL113w (103964–103983) 5*-GGCAAACAGTAATGTTAAGG-3*EBL2A KanMX4 (38–14) 5*-GGGGATCCGTCGACCTGCAGCGTAC-

YOL113w (104327–104308) CATCGAAAGATATTCTTTCG-3*EBL3A KanMX4 (1498–1522) 5*-AACGAGCTCGAATTCATCGATGATA-

YOL113w (106290–106315) TGAAATTTAGCTTTTTTTTATTTAGG-3*EBL4A YOL113w (106667–106648) 5*-GCGAGTCTTCCTCGATTTTG-3*EBL1B YOL100w (128654–128676) 5*-CCCAAATACGAAAAGTCTAAGGC-3*EBL2B KanMX4 (38–14) 5*-GGGGATCCGTCGACCTGCAGCGTAC-

YOL100w (129271–129250) CCTAGGGCTCATGGAATTATCC-3*EBL3B KanMX4 (1498–1522) 5*-AACGAGCTCGAATTCATCGATGATA-

YOL100w (131672–131691) TGCAATCAAAACAGCTCACC-3*EBL4B YOL100w (131990–131972) 5*-TAGTTCCCATTTCCAAGGG-3*EBL1C YOL107w (111913–111931) 5*-CGCCCTCTTAACTCCTTCG-3*EBL2C KanMX4 (38–14) 5*-GGGGATCCGTCGACCTGCAGCGTAC-

YOL107w (112188–112167) GTAATAAATTTAGTAGCATCAGG-3*EBL3C KanMX4 (1498–1522) 5*-AACGAGCTCGAATTCATCGATGATA-

YOL107w (113105–113125) GGAAGAGCGTATGGTAAACCC-3*EBL4C YOL107w (113652–113634) 5*-GCAGCGTAGTGAGTGCCAC-3*EBL1D YOR267c (825240–825221) 5*-CCGCTTAAAGAAGTTTTTCC-3*EBL2D KanMX4 (38–14) 5*-GGGGATCCGTCGACCTGCAGCGTAC-

YOR267c (824851–824871) GATTAGGCATCTTTATACGAG-3*EBL3D KanMX4 (1498–1522) 5*-AACGAGCTCGAATTCATCGATGATA-

YOR267c (822699–822681) GGTTTGCCGCCATTGCTGC-3*EBL4D YOR267c (822385–822405) 5*-GATGTTTCTTGCTTTTGGAGC-3*MDL1E YGL196w (130224–130241) 5*-CCCTGTTGCGTTCCAGAG-3*MDL2E KanMX4 (38–14) 5*-GGGGATCCGTCGACCTGCAGCGTAC-

YGL196w (130584–130562) GCGGCAGAATTGACTGCTTTCAC-3*MDL3E KanMX4 (1498–1522) 5*-AACGAGCTCGAATTCATCGATGATA-

YGL196w (131127–131151) CGAAGGCATTGTCAACGATGTTTGG-3*MDL4E YGL196w (131455–131436) 5*-CTCAAAGATCTTTTCGAAGG-3*MDL1F YGL194c (141816–141797) 5*-GGTGTGTCGCCAATGTACAG-3*MDL2F KanMX4 (38–14) 5*-GGGGATCCGTCGACCTGCAGCGTAC-

YGL194c (141482–141505) GTAGTAATTCGTCTCTGGTAGCGC-3*MDL3F KanMX4 (1498–1522) 5*-AACGAGCTCGAATTCATCGATGATA-

YGL194c (140440–140417) CAGAAGCAGACAGTTCAAATAGGC-3*MDL4F YGL194c (140063–140083) 5*-CGATAATGCTGCGGGAGAAGC-3*

Nucleotide coordinates are given with the 5* end of the oligonucleotide sequence first. For yeast genes, coordinates refer to therespective whole chromosomes; for the KanMX4 module, the first C of the unique PvuII site of pFA6-kanMX4 is defined asnucleotide 1, and numbering then proceeds in the direction of transcription of the kanR gene.

2#YT plates (1·6% tryptone; 1% yeast extract;0·5% NaCl; 1·6% agar) containing 100 ìg/ml ofampicillin and 35 ìg/ml of kanamycin. After veri-fication by restriction that the full-length cassettehad indeed been inserted, approximately 1 ìg ofplasmid DNA from crude minilysates was digestedwith BsgI, releasing the entire insert withoutsequences non-homologous to the yeast genome atthe ends. This was then used without further


CEN.PK2 to G418 resistance as described above.For each gene, a centromeric construct contain-

ing the entire coding sequence and flanking DNAwas also made by subcloning a fragment fromcosmids (gifts from B. Dujon and H. Tettelin)containing genomic DNA from S. cerevisiae strainFY1679, into pRS416 (Sikorski and Hieter, 1989).The details of these plasmid constructs are givenin Table 3.

. 14: 655–664 (1998)

660 . .

Figure 2. Maps of genomic disruptions. Position and orientation ofdisrupted ORFs (shaded arrows) are shown relative to neighbouring ORFs(open arrows) and elements; this information was derived from theXCHROMO program at the Martinsried Institute for Protein Sequences(http://speedy. mips. biochem. mpg. de/mips/yeast/). Jagged lines at the endsof ORFs indicate that only part of the ORF is shown. Vertical arrows showpositions of homology between the chromosome and oligonucleotides usedfor disruption, as defined in Figure 1. The region between the inner twoarrows is deleted in the mutants. Note different scale in the different panels,as indicated by scale bars. (A) Disruption of YOL113w and YOL107w. (B)Disruption of YOL100w. (C) Disruption of YOR267c. (D) Disruption ofYGL196w. (E) Disruption of YGL194c.

MicroscopyFor staining with 4,6-diamino-2-phenylindole

added to the pellet. Partial digestion of cell wallswas at 37)C for 10 min with gentle shaking. After

(DAPI), cells were iodine-fixed (Jimenez et al.,1992), the suspension was homogenized and thecells were collected by centrifugation. Then, 15 ìlof a solution of 5 mg/ml lysing enzymes (fromTrichoderma harzianus; Sigma) in 1 -sorbitol was


centrifugation, the pellet was washed in 1 -sorbitol. Cells were then resuspended in 15 ìl of2 ìg/ml DAPI in 1 -sorbitol and incubated for atleast 5 min in the dark. Fluorescence microscopywas at 100#magnification.

. 14: 655–664 (1998)

RESULTS AND DISCUSSION When disrupting these six genes in the S.

Table 3. Plasmids generated in this study.

661

Methodology

The first PCR reaction, amplifying sequencesdirectly from the S. cerevisiae genome, wasstraightforward and robust. In the second PCR,the ratio between the amount of the substrates, i.e.the products of the first PCR reactions and thedigested plasmid pFA6-kanMX4, was found to becritical. When between 0·3 and 0·5 ìg of wholeplasmid DNA was added, this reaction wasreproducibly successful.

Cloning of KanMX-containing disruptioncassettes into plasmid pUG7 in some cases wasmore complicated than expected. For three out ofthe six genes, the frequency of correct constructsamong AmpR KanR E. coli transformants wasbetter than 75%, but for the remaining three genesit was less than 1 in 24. Preliminary analysis of anumber of aberrant plasmids indicated that amajority of those carried a 3*-terminally truncatedkanamycin resistance gene. Apparently, such rear-ranged plasmids still confer kanamycin resistanceto E. coli. For generation of large numbers ofdisruption cassettes, subcloning into the plasmidmay become the limiting step.


Name VectorYeast selectable

marker

Amount of remaining genomic codingsequence after disruption (bp)*

5* end 3* end

pYORC YOL113w pUG7 KanMX4 0 0pYORC YOL100w pUG7 KanMX4 33 806pYORC YOL107w pUG7 KanMX4 86 22pYORC YOR267c pUG7 KanMX4 7 116pYORC YGL196w pUG7 KanMX4 29 0pYORC YGL194c pUG7 KanMX4 244 68

Name VectorYeast selectable

markerRestriction enzymes usedfor cutting genomic DNA

Amount of flanking DNA (bp)*

5* end 3* end

pYCG YOL113w pRS416 URA3 NdeI/KpnI 401 461pYCG YOL100w pRS416 URA3 ApaI/EcoR1 539 140pYCG YOL107w pRS416 URA3 AhdI/ClaI 342 358pYCG YOR267c pRS416 URA3 HindIII/HindIII 274 715pYCG YGL196w pRS416 URA3 HindIII/HindIII 798 251pYCG YGL194c pRS416 URA3 HindIII/KpnI 249 285

*Exclusive of start and stop codons.

cerevisiae genome in two strain backgrounds, foreach gene four primary transformants (two each inthe FY1679 and CEN.PK2 backgrounds) wereanalysed by Southern blotting, making a total of24 independent integration events. In all cases, thedisruption cassette was found to have integrated atthe correct locus (not shown), confirming the con-clusion that PCR-based genomic disruption usinglong (200–400 bp) homology is a highly accurateand efficient method in this organism (Wach,1996).

From a methodological perspective, anotheroccurrence deserves mentioning. One of theoriginal Äyol107w/YOL107w disruptants in theFY1679 background showed a 2:0 segregationupon dissection; however, lethality did not co-segregate with G418 resistance as expected if thisgene were essential (not shown). Southern analysisshowed this integration to be correct. We thendissected another Äyol107w/YOL107w hetero-zygous disruptant (FSRN011) derived fromFY1679 and one CEN.PK2-derived disruptant. Inboth cases all four spores were viable and appearedwild-type in all aspects studied. We concludethat the aberrant transformant displaying 2:0

. 14: 655–664 (1998)

segregation was the result of an independent

662 . .

Figure 3. Phenotypes of Äyol100w disruptions in the FY1679and CEN.PK2 backgrounds. (A) Growth of Äyol100w haploidmutants and Äyol100w/YOL100w heterozygous diploidmutants (YPD medium, 30)C for 3 days). I, FY1679 (wild-type); II, FSRN006 (Äyol100w/YOL100w); III,FSRN007–02B(Äyol100w); IV, FSRN009 (Äyol100w/Äyol100w); V, CSRN010(Äyol100w/YOL100w); VI, CEN.PK2 (wild-type). (B) Tetradmorphology as seen by DAPI fluorescence. Cells were pre-grown in SC medium and then put in sporulation medium for 1day (CEN.PK2) or 5 days (CSRN010). Left panel, CSRN010(Äyol100w/YOL100w). Right panel, CEN.PK2. (C) Tetrad dis-sections. Left panel, dissection of CSRN010 (Äyol100w/YOL100w derived from CEN.PK2). Right panel, dissection ofFSRN006 (Äyol100w/YOL100w derived from FY1679).

and the CSRN010 Äyol100w/YOL100w hetero-

spontaneous recessive lethal mutation which hadoccurred in a particular cell lineage. When moreexperience has been gathered from systematicfunctional analysis of the S. cerevisiae genome, itwill be possible to tell how frequently cases similarto this occur, and if they are in any way induced bythe transformation procedure.

Phenotype of disruptantsVegetative growth for all haploid FY1679-

derived strains deleted in one of the six genes wasnormal on all media and temperatures tested; alsono obvious morphological mutant phenotypeswere observed. No differences between the a or ámating types were observed. Mating proficiencywas examined and was likewise found to be intact.Sporulation proficiency of the diploid homozygousdisruptants was unaffected except in one case,FSRN029 (Äygl194c/Äygl194c), which sporulatedwith a reduced efficiency.

Among the six gene disruptions studied here,there was one case of remarkable differencebetween phenotypes observed in the two yeaststrains. In the FY1679 background, all parametersstudied for Äyol100w were wild-type, includingin the diploid homozygous Äyol100w/Äyol100wstrain FSRN009 (Table 1, Figure 3A). However,when YOL100w was disrupted in the diploidCEN.PK2 strain, the heterozygous Äyol100w/YOL100w mutants grew significantly slower thanthe parental strain (Figure 3A). Six independenttransformants were tested, with an identical out-come. We estimate that, on agar, such hetero-zygous mutants took about three times longer togrow than the wild-type parent. In liquid medium,the difference was even greater; the doubling timewas then four- to five-fold that of the wild-type.This was true both for rich (YPD) and selective(SC) medium as well as for growth at 37)C (datanot shown). When attempting to study the pheno-type of haploid Äyol100w mutants in this back-ground, we discovered that these heterozygousÄyol100w/YOL100w mutants were blocked insporulation under standard conditions. We thenasked whether YOL100w is essential in this back-ground, and generated a second disruption cassettecontaining HIS3 as the selectable marker, in placeof the KanMX4 module. This was done using thesame primer set, but using pFA6-HIS3MX6 asthe substrate. The generated disruption cassettewas used to transform both wild-type CEN.PK2


zygous mutant. Histidine prototrophs were readilyobtained from CEN.PK2 and PCR analysisshowed that they represented correct disruptionevents (not shown). However, only two His+

transformants were seen after two attemptsto transform CSRN010. On restreaking, bothtransformants turned out to be G418-sensitive,

. 14: 655–664 (1998)

indicating that they represented recombinationevents at the disrupted yol100w::KanMX4 locus

highly overlapping functions of this gene pair.

REFERENCES

663

. 14: 655–664 (1998)

rather than the wild-type locus.To see to what extent the observed phenotypes

would be corrected by the wild-type gene borneon a centromeric plasmid, we transformedFSRN029 with pYCG YGL194c and CSRN010with pYCG YOL100w, respectively. The growthrate was only slightly affected, in thatCSRN010[pYCG YOL100w] was found togrow marginally faster than CSRN010[pRS416].Increased sporulation was observed inFSRN029[pYCG YGL194c] upon transfor-mation, however not to the wild-type level.When pre-cultured in SC medium, bothCSRN010[pYCG YOL100w] and CSRN010[pRS416] did produce a low percentage of tetrads.These were mostly of abnormal morphology,including two-spored tetrads (Figure 3 B). Ondissection, it was found that only one to twospores from each tetrad were viable (Figure 3C).All viable spores gave rise to G418-sensitive col-onies with wild-type growth rate. We conclude thatYOL100w is an essential gene in the CEN.PK2background, and that YOL100w expression frompYCG YOL100w was insufficient to rescuelethality in Äyol100w haploids.

Thus, for both genes, there was incompletecomplementation of mutant phenotypes. Sporula-tion was only partially restored in both cases.pYCG YOL100w only marginally improved thegrowth rate of the Äyol100w/YOL100w hetero-zygous strain CSRN010, and could not rescue thelethality of a Äyol100w deletion in a haploidCEN.PK2 derivative. Among the plausible expla-nations for this is that insufficient amounts of5*-flanking DNA was included in the plasmidconstructs. Differences between chromatin struc-ture in chromosomes versus plasmids might alsocontribute.

For YOL100w, there is a very closely relatedPK-encoding gene YDR490c, which has also beenidentified by genomic sequencing (Hunter andPlowman, 1997). The degree of identity betweenthe amino acid sequences of the predicted geneproducts is 74% over the length of the catalytic PKdomain. One possible explanation for the pheno-typic differences between Äyol100w mutants in theFY1679 and the CEN.PK2 backgrounds is thatthe YDR490c gene might be misfunctional inCEN.PK2. It is noteworthy that neither of thesegenes has previously been found by classicalgenetics; one reason for this, in turn, could be


Future investigations will reveal if disruption ofboth genes will yield synthetic phenotypes. Evenso, it is remarkable that the Äyol100w/YOL100wheterozygous CSRN010 strain displayed suchmarked phenotypes. Since a gene deletion isunlikely to be dominant, this means that genedosage is critical in this case, and that a reductionto half the normal concentration of gene product issufficient to cause these effects. This may providean explanation for the reduced complementationseen on introduction of the plasmid-borne wild-type gene, the expression of which may beinadequate if the phenotype is very sensitive toprotein concentration.

ConclusionsAt first it may seem surprising that not more

phenotypes were detected in this set of six genomicdisruptants. However it must be kept in mind thatonly quite crude measures of mutant phenotypeshave been looked for in this work. The chiefpurpose of this category of effort is to createtools for further functional analysis, which willinevitably require specialized knowledge about thegene under study. This work also stresses theimportance of the strain background in geneticanalysis in general. In particular, with thegeneration of cloned disruption cassettes it isstraightforward to disrupt a gene in severalgenetic backgrounds, which will yield a morecomprehensive picture of mutant phenotypes.

ACKNOWLEDGEMENTS

This work was supported by the European Com-mission (contract no. BIO4-CT95–0080) in theframe of the EUROFAN programme.

De Rubertis, F., Kadosh, D., Henchoz, S., et al. (1996).The histone deacetylase RPD3 counteracts genomicsilencing in Drosophila and yeast. Nature 384,589–591.

Gietz, D., St. Jean, A., Woods, R. A. and Schiestl, R. H.(1992). Improved method for high efficiency transfor-mation of intact yeast cells. Nucl. Acids Res. 20, 1425.

Goffeau, A., Aert, R., Agostini-Carbone, M. L., et al.(1997). The yeast genome directory. Nature 387(suppl.), 1–105.

Hunter, T. and Plowman, G. D. (1997). The proteinkinases of budding yeast: six score and more. TrendsBiochem. Sci. 22, 18–22.

Jimenez, G., Yucel, J., Rowley, R. and Subramani, S.(1992). The rad3+ gene of Schizosaccharomyces pombe

Sherman, F. (1991). Getting started with yeast. Meth.Enzymol. 194, 3–21.

664 . .

is involved in multiple checkpoint functions andin DNA repair. Proc. Natl. Acad. Sci. USA 89,4952–4956.

Johnston, L. H., Eberly, S. L., Chapman, J. W., Araki,H. and Sugino, A. (1990). The product of the Sac-charomyces cerevisiae cell cycle gene DBF2 hashomology with protein kinases and is periodicallyexpressed in the cell cycle. Mol. Cell. Biol. 10,1358–1366.

Leberer, E., Dignard, D., Harcus, D., Thomas, D. Y.and Whiteway, M. (1992). The protein kinasehomologue Ste20p is required to link the yeastpheromone response G-protein beta gamma subunitsto downstream signalling components. EMBO J. 11,4815–4824.

Rundlett, S. E., Carmen, A. A., Kobayashi, R.,Bavykin, S., Turner, B. M. and Grunstein, M. (1996).HDA1 and RPD3 are members of distinct yeasthistone deacetylase complexes that regulate silencingand transcription. Proc. Natl. Acad. Sci. USA 93,14503–14508.


Sikorski, R. S. and Hieter, P. (1989). A system of shuttlevectors and yeast host strains designed for efficientmanipulation of DNA in Saccharomyces cerevisiae.Genetics 122, 19–27.

Wach, A. (1996). PCR-synthesis of marker cassetteswith long flanking homology regions for genedisruptions in S. cerevisiae. Yeast 12, 259–265.

Wach, A., Brachat, A., Alberti-Segui, C., Rebischung,C. and Philippsen, P. (1997). Heterologous HIS3marker and GFP reporter modules for PCR-targetingin Saccharomyces cerevisiae. Yeast 13, 1065–1075.

Wach, A., Brachat, A., Pohlmann, R. and Philippsen, P.(1994). New heterologous modules for classical orPCR-based gene disruptions in Saccharomycescerevisiae. Yeast 10, 1793–1808.

Winston, F., Dollard, C. and Ricupero-Hovasse, S. L.(1995). Construction of a set of convenient Sac-charomyces cerevisiae strains that are isogenic toS288C. Yeast 11, 53–55.

. 14: 655–664 (1998)

genomic disruption of six budding yeast genes gives one drastic example of phenotype...

Documents