MOLECULARANDCELLULARBIOLOGY,Sept.1994,p.5619-5627Vol.14,No.90270-7306/ - PDF document

MOLECULARANDCELLULARBIOLOGY,Sept.1994,p.
MOLECULARANDCELLULARBIOLOGY,Sept.1994,p.5619-5627Vol.14,No.90270-7306/94/$04.00+0CopyrightX1994,AmericanSocietyforMicrobiologySuppressionofaYeastCyclicAMP-DependentProteinKinaseDefectbyOverexpressionofSOKI,aYeastGeneExhibitingSequenceSimilaritytoaDevelopmentallyRegulatedMouseGeneMARYP.WARDANDSTEPHENGARRETT*DepartmentofMolecularCancerBiologyandDepartmentofBiochemistry,DukeUniversityMedicalCenter,Durham,NorthCarolina27710Received3March1994/Returnedformodification6April1994/Accepted25May1994SaccharomycescerevisiaecyclicAMP-dependentproteinkinase(Akinase)activityisessentialforgrowthandcellcycleprogression.DependenceonAkinasefunctioncanbepartiallyrelievedbytheinactivationofasecondkinaseencodedbythegeneYAK].Wehaveisolatedtwonewgenes,SOKIandSOK2(suppressorofkinase),asgenedosagesuppressorsoftheconditionalgrowthdefectofseveraltemperature-sensitiveAkinasemutants.OverexpressionofSOKI,likelesionsinYAK],alsorestoresgrowthtoastrain(tpkltpk2tpk3)lackingallAkinaseactivity.TheSOKIgeneisnotessential,butasokl::HIS3disruptionabrogatessuppressionofanAkinasedefectbyyak).TheseresultssuggestthatYaklandSokldefinealinearpathwaythatispartiallyredundantwiththatoftheAkinase.ActivationofSokl,bySOK1overexpressionorbyinactivationofthenegativeregulatorYakl,rendersacellindependentofAkinasefunction.TheimplicationsofsuchamodelareparticularlyintriguinginlightofthenuclearlocalizationpatternoftheoverexpressedSoklproteinandtheprimarysequencehomologybetweenSOKIandarecentlydescribed,developmentallyregulatedmousegene.IntheyeastSaccharomycescerevisiae,cyclicAMP(cAMP)-dependentproteinkinase(Akinase)activityisessentialforgrowthandcellcycleprogression.CellsdeficientinthisactivitystopgrowingandarrestinG1inamannersimilartothatobservedforwild-typecellsdeprivedofnutrients(16,20,21).Incontrast,mutationsyieldingelevatedAkinaseactivitypreventcellsfromarrestinginGIfollowingnutrientstarvationorheatshock(4).Suchmutationsalsocausesporulationdeficiency,lossofcarbohydratereserves,andsensitivitytovariousformsofstress,includingheatshockandnutrientstarvation.Together,thesephenotypeshavebeentakenasevidencethattheyeastAkinasepathwayplaysacentralroleinmediatingthegrowthandcellcyclearrestofstarvedcells(2).Low-levelactivitypromotesexitfromthemitoticcycleandentryintoGo,whereashigh-levelactivityprecludesaccesstoGo.YeastAkinaseactivityisregulatedbyacomplexsignaltransductionpathwaythatincludestheyeasthomologsofthemammalianrasproducts.S.cerevisiaecontainstwoRASgenes(RAS1andRAS2)whichencodefunctionallyredundant,mem-brane-associatedproteinsthatbindandhydrolyzeGTP(32).Intheiractive,GTP-boundstate,yeastRasproteinsactivateadenylatecyclase,whichisencodedbyasinglegene,CYRI(18).Likeitsmammaliancounterpart,yeastAkinaseisaheterotetramericproteinconsistingoftwocatalyticsubunitsandtworegulatorysubunits.Theyeastcatalyticsubunitsareencodedbythreefunctionallyredundantgenes,TPKI,TPK2,andTPK3(31),whereastheregulatorysubunitisspecifiedbyasinglegene,BCY1(4).BindingofcAMPtoBcylresultsinitsdissociationfromthecatalyticsubunitsandinstimulationofAkinaseactivity.ThegrowtharrestandcellcyclearrestofconditionalRasandAkinasemutantsareconsistentwiththenotionthatAkinasephosphorylationregulatesmanycellularprocesses.Tar-*Correspondingauthor.Mailingaddress:DepartmentofMolecularCancerBiology,Box3686,DukeUniversityMedicalCenter,Durham,NC27710.Phone:(919)684-6354.Fax:(919)684-8598.getsoftheyeastAkinasehavebeendescribedandincludeproteinsinvolvedinprocessessuchascarbohydratestorageandmetabolism,phospholipidmetabolism,andtranscriptionalregulation,aswellasfunctionsinvolvedinthesynthesisandbreakdownofcAMP(forreviews,seereferences1and2).Nevertheless,itremainsunclearwhetherAkinasephosphor-ylationoftheseknowntargetscaninfluencewhetherthecellexitsthecellcycleorcontinuesproliferation(3).PreviousattemptstoidentifydownstreameffectorsoftheRas/Akinasepathwayhaveexploitedclassicalandgenedosage(high-copy-number)suppressoranalyses(4,9,13,24,30).TwogenesexhibitingsignificanthomologytoknownproteinkinasegeneswereidentifiedinseparateselectionsforsuppressorsofconditionaldefectsintheAkinasepathway.Nullmutationsofonegene,YAK1,allowedstrainscompletelydeficientinAkinaseactivitytogrow;tpkltpk2tpk3YAK1+strainsareinviable,buttpkltpk2tpk3yaklstrainsgrow(13).TheseandotherresultsledustoproposethatYaklservedasanegativeregulatorofcellgrowth,inapathwayparalleltothatoftheAkinase,withoverlappingbutantagonisticeffects(14).Thesecondgeneexhibitingproteinkinasegenehomology,SCH9,wasisolatedasahigh-copy-numbersuppressorofatempera-ture-sensitivecdc25(Ts)mutation.AlthoughthemechanismbywhichSch9overproductionalleviatestheAkinasedefectisnotknown,thereciprocalsuppressionofnullmutationsinbothpathways(atpkstraingrowsinthepresenceofaSCH9high-copy-numberplasmid,andtheslow-growthdefectcausedbyansch9disruptionisalleviatedbyhighlevelsofAkinaseactivity)isconsistentwithamodelinwhichSch9andtheAkinasehavepartiallyoverlappingfunctions.Analternatemodel,basedinpartonthesignificantsequencesimilaritybetweenthetwokinases,suggeststhathyperactivationofeitherkinaseresultsintheabnormalphosphorylationofessentialkinasesubstratesoftheoppositepathway(15,30).SinceprevioussuppressorselectionswereconductedwithmutationsinRAS(5,8,13)andCDC25(30),weelec

tedtoisolatetemperature-sensitivetpk2mut
tedtoisolatetemperature-sensitivetpk2mutants[tpkltpk2(Ts)tpk3mutants]andusethemtoisolatesecond-sitesuppressors.We56195620WARDANDGARRETMTABLE1.YeaststrainsusedinthisstudyStrainGenotypeSourceorreferenceS7-7AxS7-5AMATa/AM4Tattpkl::URA3/TPKJtpk2::HIS3/TPK2tpk3::TRPI/TPK3ura3-52/ura3-5231leu2-3,112/leu2-3,112his3/his3trpl/trplade8/ade8SGY356MATottpkl::URA3TPK2tpk3::TRPlbcyl::LEU2ura3-52his3leu2-3,112trplade8ThisstudySGP3MATotrasl::HIS3RAS2his3ura3-52leu2-3,112trplade813SGP34MTarasl::HIS3ras2-34-URA3(Ts)his3ura3-52leu2-3,112trplade813MWY63SGY356tpk2-63(Ts)ThisstudyMWY65SGY356tpk2-65(Ts)ThisstudyMWY123MWY63yakl::HIS3ThisstudyMWY131MWY65yakl::HIS3ThisstudySGY398MWY63tpkl::ADE8Thisstudy1029MATa/A4Tatura3/ura3leu2-3,112/leu2-3,112his3/his3lys2/lys2trpl/TRP1ade2IADE213MWY2641029sokl::HIS3/SOKJThisstudyMWY285MWY63sokl::HIS3ThisstudyMWY273MWY63yakl::ADE8ThisstudyMWY313MWY63sokl::HIS3yakl::ADE8ThisstudySGP406AL4Taleu2-3,112trplura3-52his3tpkl::URA3tpk2::HIS3tpk3::TRPIyakl::LEU213RS13.58A-1MATatpklWlatpk2::HIS3tpk3::TRP1bcyl::LEU2ura3-52his3leu2-3,112trplade83aWIisthenotationusedbyCameronetal.(3)forthewimpallelesofTPK1,TPK2,andTPK3.reasonedthatbydoingso,andtherebyeliminatingthemajorityofsuppressorsalreadyisolated(CDC25,RASIandRAS2,andthethreeTPKgenes,etc.),wewouldidentifyrareorweaksuppressorsthatmightoriginallyhavebeenoverlooked.Thiscommunicationdescribestheisolationandcharacterizationofseveralindependent,temperature-sensitivemutationsintheTPK2gene,aswellastheisolationoftwogenedosagesuppressorsoftheresultingconditionalgrowthdefect.Char-acterizationofoneofthehigh-copy-numbersuppressorshasidentifiedanovelgene,SOKI,thatexhibitssequencesimilaritytoadevelopmentallyregulatedmousegene.TestsofepistasisareconsistentwithamodelinwhichactivationofSokl,bySOK]overexpressionorbyinactivationofYakl,resultsinsuppressionoftheAkinasedefect.MATERIALSANDMETHODSMediaandgrowthconditions.Mediaused,includingyeastrichandminimalmediaaswellasbacterialmedia,werepreparedasdescribedpreviously(7,13).Yeastcellswereheatshockedbyreplicatingpatchestoprewarmedagarandplacingtheagarplatesinashallowwaterbathsetto55°C.After10minattheelevatedtemperature,theplateswereincubatedat23or30°Cforseveraldays.Physiologicalcharacterizationofthetemperature-sensitiveAkinasemutantsforglycogenaccumu-lationandgrowtharrestwascarriedoutbyessentiallythesamemethodsdescribedpreviously(13).GenedosagesuppressorsoftheAkinaseconditionalmutantswereisolatedbytrans-formingtwotemperature-sensitivetpk2(Ts)strains,MWY63andMWY65,toadenineprototrophyandtemperatureresis-tance(34.5and36°C,respectively)withahigh-copy-numberplasmidlibrarybasedontheADE82,umvectorYEpADE8(14).Strainsandplasmids.Yeaststrainsarelisted,withrefer-enceswhereappropriate,inTable1.BacterialstrainsMC1066[A(lac)X74galUgalKstrAhsdRtrpC9830leuB6pyrF::TnS]andDH5[F'/endA4hsdRl7(rKMK)supE44thi-1recAlgyrArelAlA(lacZYA-argF)U169(A80dlac(lacZ)M15)]havebeendescribedpreviously(6,35).Thehigh-copy-numberyeastvectorYEpADE8wasconstructedbyTodaandCameronandhasbeendescribedpreviously(14).Tocircumventthere-peatedisolationofplasmidscontainingTPKI,TPK2,andTPK3,ahigh-copy-numberlibrarywasconstructedbydigestingchromosomalDNAofatpk-deficientstrain(tpkl::URA3tpk2::HIS3tpk3::TRP1yakl::LEU2)(SGP406)topartialcom-pletionwithSau3Aandligatingsize-selectedfragmentsintothesingleBamHIsiteofYEpADE8.Eightseparatepoolsoftransformants�(8,000coloniesperpool)werecollectedandtestedforthefractionofplasmidswithinsertsaswellasforinsertsize.Physicalanalysisofplasmidsisolatedfromrandombacterialcoloniesshowedthat�80%(10of12)containedinserts,withanaverageinsertsizeof14kb.ThebacterialvectorpBSK+hasbeendescribedpreviously(StratageneProd-uctCatalogue).PlasmidspMW25andpMW42areclassIsuppressors,whereasplasmidpMW43isaclassIIsuppressor.DNAmanipulations.PlasmidDNAwaspreparedfromEscherichiacolibythealkalilysismethod(19).Allenzymeswereusedaccordingtothespecificationsoftheirsuppliers(NewEnglandBiolabsorBethesdaResearchLaboratories),andcloningtechniqueswereasdescribedpreviously(19).TheHIS3disruptionofSOK1wasconstructedbydigestingplasmidpMW22(YEpADE8containinga6-kbBglIISOKIfragmentattheBamHIsite;seeFig.3,plasmidC)withNcoI,fillingintheendswithKlenowfragment,ligatingonBamHIlinkers,andtheninsertingthe1.7-kbBamHIfragmentofHIS3.Theresultingsokl::HIS3plasmidwasdesignatedpMW26.Theyakl::HIS3andyakl::ADE8disruptionshavebeendescribedpreviously(13).ThesequenceofSOK1wasdeterminedbyamodificationofthemethodofSangeretal.(28),usingdouble-strandedplasmidDNAcontainingrandomdeletionsofSOK1.RandomdeletionsofSOKIinpBSK+wereconstructedbydigestingpMP11andpMP13withKpnIandXhoIandthensequentiallydigestingwithexonucleaseIIIandS1nucleaseaccordingtotheinstructionsofthesupplier(Promega).TheSOK1genewasepitopetaggedbyconvertingthesingleHindlIlsite(bp312inFig.4)withinthe5'endofthecodingregiontoaNotIsiteandinsertinga112-bpNotlDNAfragment(GTEP)containingthreerepeatsofa27-codonsequencespecifyingthehemagglutinin(HA)peptide,YPYD-VPDYA(11,26a).TheNotlsitewascreatedbycuttingSOK1withHindlIl,fillinginthe5'overhangwithKlenowfragment,andinsertingNotllinkers(8-mersfromNewEnglandBiolabs).OneoftheclonescontainingthecorrectlinkerswasdigestedwithNotI,dephosphorylatedwithcalfintesti

nalphosphatase,andthenligatedwiththeGTEP
nalphosphatase,andthenligatedwiththeGTEPfragment.PlasmidscontainingasingleinsertweretestedbydigestionwithseveralrestrictionMOL.CELL.BIOL.YEASTAKINASESUPPRESSOR5621TPK2tpkIW,~~~~s_~~~~~~~~~~i_.I_.._TPK2tDklWYEp35SYEp35Stpk2-63tpk2-65tpk2-63tpk2-65tpk2-63tpk2-6523°CHS-_23°C36°CFIG.1.CharacterizationofconditionalAkinasemutants.Growthwastestedbyreplicatingpatchestoagarthatwaseitherincubatedat23°C,heatshocked(HS)for10minandreturnedto23°C,orincubatedat36°C.Strains:TPK2,SGY356(TPK2bcyl::LEU2);tpklw,RS13.58A-1(tpklw'bcyl::LEU2);tpk2-63,MWY63[tpk2-63(Ts)bcyl::LEU2];tpk2-65,MWY65[tpk2-65(Ts)bcyl::LEU2].enzymestomonitortheorientationoftheinsert(thereisasingleBamHIsitepositionedasymmetricallywithintheGTEPfragment).Finally,thesingleSstIfragmentfromoneofthederivativescontainingtheGTEPinsertinthecorrectorienta-tion(aswellasaclonecontainingthesamefragmentintheoppositeorientation)wasinsertedintotheuniqueSstIsiteoftheoriginalvector(YEpADE8).ImmunofluorescenceofSokl-HA.Strainswerestainedforindirectimmunofluorescenceessentiallyasdescribedprevi-ously(27).Theprimaryantibodywasanti-HAmonoclonalantibody12CA5(thekindgiftofKenFergusonandMikeWigler)andwasdiluted1:1,000forstaining.Thesecondaryantibodywasfluoresceinisothiocyanate(FITC)-conjugatedgoatanti-mouseimmunoglobulinGfromBoehringerMannheim.RESULTSIsolationandcharacterizationofAkinasetemperature-sensitivemutations.Toisolatetemperature-sensitiveAkinasemutations,wetookadvantageofthestress-sensitivephenotypeofstrainswithelevatedAkinaseactivity(4).StrainslackingtheregulatorysubunitoftheAkinaseandcontainingoneormoreofthethree,redundantcatalyticsubunitgenesfailtosurviveregimenssuchasbriefexposuretohightemperatures(10min,55°C)andnutrientdeprivation.Thisstresssensitivitywaspreviouslyexploitedtoisolatepartial-loss-of-functionmuta-tionsintheAkinasecatalyticsubunitgenes(3).AlthoughthemutationsisolatedbyCameronetal.(3)werenotanalyzedforconditionalactivity,wereasonedthatasubsetofthemmightresultinthecompletelossofAkinasefunctionatanelevatedtemperature.Strainscontainingsuchmutationswouldberesistanttoheatshockandviableatthepermissivetempera-42S76S42S76S230C34.5'CFIG.2.High-copy-numbersuppressorsofconditionalAkinasemutants.PatchesofstrainMWY63[tpk2-63(Ts)ade8]containingtheindicatedhigh-copy-numberplasmidswerereplicatedtominimalmedium(lackingadenine)andincubatedat23and34.5'Cforsev-eraldays.PlasmidswereYEp(YEpADE8),35S(YEpADE8-SOKI)(pMW25),42S(YEpADE8-SOKI)(pMW42),and76S(YEpADE8-SOK2)(pMW43).ture(i.e.,23°C)andwouldarrestinG1onashifttothenonpermissivetemperature.AsimilarschemewasusedbyotherstoidentifyconditionalallelesofRAS2(26).Stress-resistantrevertantsofatpklTPK2tpk3bcylstrainwereisolatedbyexposingcellsto55°Cfor10minandthenincubatingthecellsat23°Cfor3days.Toensurethatalloftherevertantswereindependent,onlyoneheatshock-resistantcolonyfromeachpatchwassaved(seeMaterialsandMeth-ods).Stress-resistantrevertantswerethenretestedfortheirabilitytosurviveexposuretoextremeheataswellasfortheirviabilityat36°C(Fig.1).Ofatotalof150heatshock-resistantsurvivors,30(20%)exhibitedatemperature-sensitivegrowthdefect.Atleastthreeofthestress-resistantrevertantscontaintemperature-sensitivemutationsinTPK2asjudgedbyseveralcriteria(Table2).First,thetemperature-sensitivegrowthdefectofeachstrainwasrecessiveandwascomplementedbyalow-copy-numberplasmidcarryingTPK1orTPK3.Second,thetemperature-sensitivemutationswereshowntobetightlylinkedtoTPK2bymatingeachoftherevertantstoaTPKJtpk2::HIS3tpk3::TRP1BCY1strainandscoringtetrads.OnlyHis-Ura+strainsweretemperaturesensitiveforgrowth(of50tetrads),suggestingthattheconditionalmutationwastightlylinkedtoTPK2andwasmaskedbythepresenceofawild-typeTPK1allele.Third,thetemperature-sensitivegrowthdefectwassuppressedbythedisruptionofagene,YAK1,identifiedpreviouslyasarecessivesuppressorofconditionalAkinaseactivity(13).Finally,allofthemutantsaccumulatedatleastmoderatelevelsofglycogen,incontrasttothestress-sensitiveTPK2bcylparent,whichwasunabletoaccumulatemeasurableglycogenunderanycondition.InS.cerevisiae,glycogensynthesisanddegradationhavebeenshowntobeTABLE2.Characterizationoftpk2(Ts)mutantsaGrowthat:ComplementationTPK2StrainRelevantgenotypeHSbIodine'lika23°C34.5°C36°CTPK1TPK2linkageSGY356TPK2+++/--YNDNDNDMWY63tpk2-63(Ts)+--+B++YesMWY65tpk2-65(Ts)++/--+B++YesMWY123tpk2-63(Ts)yakl::HIS3+++/-NDNDNDNDNDMWY131tpk2-65(Ts)yakl::HIS3+++/-NDNDNDNDNDa+,growth;+/-,slowgrowth;-,nogrowth;ND,notdone;Y,yellow(noglycogenaccumulation);B,brown(glycogenaccumulation).bGrowthat23'Cafterexposureto55°Cfor10min.cColorofcoloniesafterexposuretoiodinevapors.VOL.14,19945622WARDANDGARRETMABCDEF230CBg11stNcoNheNheSstIII1IIABCDEF34.5tCBgi11-ilkbC+D+EFHLS3FIG.3.CloningandrestrictionmapofSOKI.PatchesofstrainMWY63(tpk2-63ade8)containingtheindicatedplasmidswerereplicatedtominimalmedium(lackingadenine)andincubatedforseveraldaysat23and34.5°C.TheyeastDNAfragmentsofsomeoftheplasmidsareshownatthebottomalongwiththeresultsofthetestofsuppression.PlasmidsnotshownareA(YEpADE8)andB(pMW25,theoriginalSOKIclone).regulatedbyAkinasephosphorylation,withglycogenlevelsexhibitinganinverserelationtocellularAkinaseactivity(4,12).Bythesecriteria,theconditionalgrowthdefectsofatleastthreeoftheheatshock-resistantrevertants

wereduetoatemperature-sensitivemutationi
wereduetoatemperature-sensitivemutationinTPK2(Table2).Eightotherconditionalmutantscontainedrecessive,temperature-sensitivemutationsinTPK2asjudgedbytheirinabilitytocomplementoneofthetpk2(Ts)mutants(tpk2-63mutant)oftheoppositematingtype.IsolationofgenedosagesuppressorsoftheconditionalAkinasemutations.Thetechnicaldifficultiesofcharacterizingandcloningdominantsuppressorshavebeenpartiallycircum-ventedbytherecentexploitationofgenedosage,orhigh-copy-number,suppressors.Genedosagesuppressorsofthecondi-tionalAkinasemutationswereisolatedbytransformingtwoofthetemperature-sensitiveAkinasemutants(tpk2-63andtpk2-65mutants)withahigh-copy-numberplasmidlibraryandincubatingthetransformantsatthenonpermissivetempera-tureforseveraldays.SinceAkinasephosphorylationmayaffectmanyessentialprocesses,wedeterminedthelowesttemperatureatwhicheachofthestrainscouldbeincubatedwithoutexhibitinganygrowth.Bydoingso,wehopedtomakethephosphorylationofasingleessentialsubstratelimiting.Inaddition,theplasmidlibraryfromwhichthehigh-copy-numbersuppressorswereisolatedwasgeneratedfromatpkltpk2tpk3yaklstrain(seeMaterialsandMethods).TheuseofthislibraryeliminatedthereisolationofthethreeTPKgenes,anyoneofwhichwasabletocomplementtheconditionalAkinasemutationandalleviatethegrowthdefect.Toconfirmthatthetemperature-resistantgrowthofthetransformantswasplasmiddependent,coloniesthatgrewatthenonpermissivetemperaturewereallowedtolosetheplasmid(bygrowthinnonselectivemediumat23°Cforseveraldays)andthenretestedforconditionalgrowth.Bythiscrite-rion,30colonies(ofatotalof55,000Ade+transformants)exhibitedplasmid-dependent,temperature-resistantgrowth.PlasmidDNAfromeachofthe30colonieswasrescuedinE.coliandusedtoretransformtheoriginaltemperature-sensitiveyeaststrain(MWY63orMWY65)fromwhicheachwasderived.Inthissecondscreen,plasmidDNAfrom8ofthe30temperature-resistantcolonieswasabletoretransformtheconditionalAkinasemutantstotemperatureresistance(Fig.2).Atleast6oftheremaining22coloniesmayhavecontainedamixtureofhigh-copy-numberplasmids,sincetemperatureresistancewasrestoredtogreaterthanone-eighthoftheyeasttransformantswhenplasmidDNAfromapooledbacterialtransformation,ratherthanasinglecolony,wasused.Thegenedosagesuppressorsdefinetwonewgenes,SOKIandSOK2.SincetheplasmidlibrarywasconstructedfromastrainlackingthethreeAkinasecatalyticgenes,wecouldruleoutthepossibilitythatanyoftheeightgenedosagesuppres-sorscontainedafunctionalTPKgene.Todeterminetheidentitiesoftheinsertscontainedwithintheeighthigh-copy-numberplasmids,eachplasmidwassubjectedtorestrictionfragmentanalysis.Plasmidswithnonidenticalbutoverlappinginsertsweregroupedintotwoclasses:classIandclassIIweredefinedbyfiveandthreeplasmids,respectively.Consistentwiththisphysicalassignment,classIIplasmidssuppressedthetemperature-sensitivegrowthdefectmoreweaklythanthefiveclassIsuppressors(seebelow).DespitethisquantitativeMOL.CELL.BIOL.YEASTAKINASESUPPRESSOR5623TCTTCAAATAAAATAGGTCCGATTCCTGACCCACTATTTTGCATTCTTCTTTAAGAAAATTATCATCCCACAATCAATACTATCGTAATCAAAATATTCTTTTAAATCATCCAACTCTTCCGGCATATCAAAGAAAAAATTTTCTAGAAACCATCATCAACCATATTTGCATTCTAATAATCCTCTCAGTTCAAACCCTCTTTCATTAAAGCGCGCAATTTTTTTAAATCAACAAATATCAGGTAACGCAAGCACTAACGCTAATAATGACAACATTAATAATTCCACTGCCAATTCAATGACAAATCAAAGCTTCCTATCAAGCTCAAACTTTGACTTAACTTTGGAAGATAGAATANTNQSFLSSSNFDLTLEDRIAACTACATAAAGGCTACTCCAACGCCTGTTCCATTTCCTCCTATAAATTTGCAGGGCTTANYIKATPTPVPFPPINLQGLKEIDLQEILKNPQLRHDIIFDPLLQFRPNLDGERGNKKRQTTGGCGAATATCTATTGGAATGATGTTCAAAATGAAATTTATGTTTACTCTAAGAGGCCTLANIYWNDVQNEIYVYSKRPEIFQYNRSRLVPLFDTLRDVTTGTTAACGATAGTCCCACAAAAAGAGTCTCCGATGATAAATAATGTACTGGACACAGAALLTIVPQKESPNINNVLDTETTGAACATTCAAGAACTATTGAAAGGTTCTCTGATAATGTCTAACTTGTCAGGCTGGTTGLNIQELLKGSLIMSNLSGWLGCTGATTTATTTAAACATCATTGTGCCCCCATGAGGGACCCATGGGTGGATAAAATGAGCADLFKHHCAPMRDPWVDKMSNKFKEAERDSSLTRLIEGLRTTGGTTTTTCAAATTTTGGAAACAATGAAATTGGATATTGCCAATCATCAAATAAGAATALVFQILETMKLDIANHQIRICTAAGGCCAGCTCTGTTAAGTAATGCTGTAGAATTTGAGAAACAGTATTTCAACACTCTTLRPALLSNAVEFEKQYFNTLATAGCCTCTAAAAGGGTGAATTTAAATACTTCCCTACTTTGGTTTGATAAAAAATTCAACIASKRVNLNTSLLWFDKKFNGAAAATGTTACCGCTGGCCTTGTTAGAAATCCAAGTTCTATCACTATCCCTGATGTCTACENVTAGLVRNPSSITIPDVYAATATTTGCATTAGAAGTATAATTAACCTATTGTCATGTAGGAAGATGGTGAGAGAGTACNICIRSIINLLSCRKMVREYCCAACTCCGCTTTCTTTTGATCATCGAAGATTGATCCTTTTGCGTGCCGATATACGTCAAPTPLSFDHRRLILLRADIRQATTGTTTGTATTTTAGTTTGCCGTTTACTTTTCCAACAATTGGTGGCCAACGATCCTTCAIVCILVCRLLFQQLVANDPSATGGATAAAGCTACAAAGGAATATGTTATTCATACATACTCAACTAAAAGGCTGAAGAATMDKATKEYVIHTYSTKRLKNGAAATTATCAGCATTATAACAGATGAACACGGCAATTGTAGGTGGACTAAAAACACAATGEIISIITDEHGNCRWTKNTM1441TCTATTGCTGTTCATCTTTGCAAAGTTATTGACGATCTTCACAGGGAGTACGACAACAAT381SIAVHLCKVIDDLHREYDNNGGT!AGTTGTG%'AACAAGCCtAGiAAGGCCTCAATTGCCS'lTTCTAG=ACAWACTC'AAAAATAACGSCEQARRPQLPSLDNSKITTTTGCTAAATCTTGGTTATCTAAGCAAACTCAACCCCTCAGTGAAGTTTATGGTGTCCTAFAKSWLSKQTQPLSEVYGVLGAAAATAGAGTATTCAAATCACTAGAGGACGCTATCTTCAACAGGTCCGAGTGCACAATTENRVFKSLEDAIFNRSECTIGATGGACGCGTTAAACAAGACTTTGTGTACCTCTACAACACAAACAATGGCAACGTAGGTDGRVKQDFVYLYNTNNGNVGAGCACTAACACTTTGAGTACTACTACAGATACTGCTAGCGTTAAAATCAGCCCGTCTTTGSTNTLSTTTDTASVKISPSLATGTCTCCTTCTAAAACCTCCACCACCACACCTACTGGCAATGCGATTGCATCCAGAGGTMSPSKTSTTTPTGNAIASRGTTATTCGCAGCAACAGAGCTGGAGGAATTCGAGAATGTTTATCGCCACTTATATGCATTALFAATELEEFENVYRHLYALATCAACCTTCATTGGTCCGTATTCGGTCCTCATTATATCGAAATGTTAGGAGATAAAGTT

INLHWSVFGPHYIENLGDKVNKKGI*CTTTATGTAACTTC
INLHWSVFGPHYIENLGDKVNKKGI*CTTTATGTAACTTCATATATCATTTTTAAAGCGTACATTTCTTATAATTTCACTTCTTAAATGCAGTCACATATTTACTGCTATCAGACAAAGTAAAGCGTCGCTAATTACTTGCCTCCACGAGTTGCTTGTCAATCAGGGGCGCTTCATCTCGCCAGCGTAAGCGCGTTGTCCATCTTATTTAACAAGAGAAAAAAAAGATAATTCAACTTAAAAAGAACGTGCTATCAGCTATATAAAATGTAGACACATTATGTATCAGTGTTTCTTTAGCAATGCGAAAAATATAGGCTTAGTGTCATTTACTTAAAGTTTAGAAACTCTGCACACTTCACCAAGAACAAGTTTAACCGCTCATTACTCCAAACGGATTTTTTTGCCTAAAGAATCACGACAATGAAAGTATGTTATCACTCTAAAACTGCCATGCTCTTTATTCGAAATTACTAACATTGTACCT2500difference,plasmidsfromeachclasswereisolatedassuppres-sorsofbothconditionalAkinasestrains.Thus,bybothphysicalandphenotypicanalyses,theeightplasmidsdefinedtwogeneswhosepresenceinelevatedcopynumbersuppressedthegrowthdefectoftwoindependentAkinaseconditionalmutants.TheSCH9genewaspreviouslyidentifiedasadosagesuppressorofaconditionaldefect[cdc25(Ts)]intheAkinasepathway(30).Toconfirmthatnoneofthesuppressingplas-midscontainedincompleteclonesofSCH9,alabelledprobeofSCH9DNAwashybridizedtoclassIandclassIIplasmidDNAsisolatedfromE.coli.Aspredictedbytherestrictionmaps,theSCH9probefailedtohybridizetotheDNAofeithersuppressorclassbutelicitedastrongsignalfromacontrollanecontaininganidenticalamountofSCH9DNA(datanotshown).Wehaveconfirmedthatthetemperature-sensitiveAkinasedefectcanbesuppressedbyanSCH9high-copy-numberplasmidbutthatsuppressionisweakrelativetoSOK1overex-pression(datanotshown).FragmentsfromonerepresentativeplasmidoftheclassIsuppressorsweresubclonedbackintotheYEpADE8vectorandscreenedfortheirabilitytosuppresstheconditionalAkinasedefect.AsshowninFig.3,theclassIsuppressorwasdefinedbyaDNAinsertof3kb,andinsertionofHIS3intothesingleNcoIsiteofthesamefragmentcompletelydestroyedsuppressoractivity.ThelocusdefinedbytheHIS3insertionwasdesignatedSOKI,forsuppressorofAkinase.TheclassIIsuppressorshavebeendesignatedSOK2.TheSOKIgeneencodesapolypeptideexhibitingsequencesimilaritytoamousetestis-specificprotein.ThenucleotidesequenceoftheSOKIgeneandflankingDNAwasdeter-mined.ThesequencerevealedonelongopenreadingframedisruptedbytheNcoIsite(Fig.4).TheopenreadingframecorrespondingtotheSOK1genewouldencodeapredictedproteinof565aminoacidsifthefirstATGencodedtheinitiatingmethionine.Asecondreadingframe2kbdown-streamfromandtranscribedintheorientationoppositetothatofSOKIwasidenticaltoRAD57(17).ThephysicalproximityofthesetwogeneswasinagreementwithgeneticmappingdatathatrevealedtightlinkagebetweenbothgenesandtheCEN-linkedTRP1geneofchromosomeIV(datanotshown).ThepredictedSOK1geneproductwasusedinahomologysearchwithpreviouslyidentifiedproteins.TheonlyproteintoexhibitsignificanthomologywithSOKIwasadevelopmentallyregulatedmouseprotein(thepredictedproductofatranscriptdesignatedpBs13)ofunknownfunction(21).Overall,thetwoproteinshave25%identityand49%similarity.Thesequenceidentityincreasesinseveralstretchesinthemiddleofthetwoproteinsandincludesastretchof146residuesexhibiting33%identityand59%similarity(Fig.5).AlthoughouranalysisoftheSoklsequencehasnotbeenextensive,thesequencedoesnotexhibitsimilaritytoanyofthecanonicaltranscriptionfactors,inagreementwithconclusionsreachedforthetestis-specificmousegene(21).SincetheSOKIgenewasisolatedasahigh-copy-numbersuppressorofaconditionalAkinasemutant,thepredictedpolypeptidewasscannedforpotentialAkinasephosphoryla-tionsitesaswellasforotherconsensusmotifswithapossibleroleinthefunctionofSokl.WhileseveralputativeproteinkinaseCphosphorylationsiteswerefound,Akinaserecogni-FIG.4.NucleotideandpredictedaminoacidsequencesofSOKI.Thenucleotidesequenceextends2,500bpfromtheleftmostSstIsite(thesequencedoesnotincludetheSstIsite)toabouthalfwaybetweenthetwoNheIsitesshowninFig.3.TheNcoIandHindlllrestrictionsites(thesitesofinsertionoftheHIS3marker[Fig.3]andtheHAepitopetag,respectively)areindicatedbylinesabovetheDNAsequence,andthechain-terminatingcodonisdenotedbyanasterisk.VOL.14,19941611211812413011361214214148161541816011016611217211417811618411819012019612211021241108126111412811201301126132113213411381361150140115614211621441168146117414811801501186152119215411981561204121012161222122812341240124615624WARDANDGARREMT111VPLFDTLRDVLLTIVPQKESPMINNV...LDTELNIQELLKGSLIMSNLS157-1--:I11::.::I.:1:11::1-1:111175LELLKEIKEILLSLLLPRQSRLKNEIEEALDMEFLQQQADRGDLNVSYLS224158GWLADLFKHHCAPMRDPWVDKMSNKFKEAERDSSLTRLIEGLRLVFQILE207225KYILNMMVLLCAPIRDEAVQRLEN..........ISDPVRLLRGIFQVLG264208TMKLDIANHQIRILRPALLSNAVEFEKQYF237265QMXMDMVNYTIQSLQPQLQEHSVQFERAQF294FIG.5.AminoacidsequencehomologybetweentheSOK1productandatestis-specificmouseprotein.SequencesimilaritybetweenSokl(residues111to237)andtheproteinencodedbypBsl3isdisplayed.Aminoacididentityisdenotedbybars,andconservativechangesaredenotedbycolons.tionsiteswereconspicuouslymissingfromthepredictedprotein.Anepitope-taggedSoklproteinislocalizedtothenucleus.SincetheprimarysequenceoftheSoklproteindidnotsuggestanobviousfunctionoractivity,wedetermineditscellularlocalizationbyindirectimmunofluorescence(27).TheSoklproteinwasepitopetaggedbyinsertinga115-bpoligonucleo-tidespecifyingthreerepeatsofthe9-amino-acidHAepitopeintothesingleHindlllsitewithinthe5'endoftheSOK]codingregion(Fig.4;seeMaterialsandMethods).Insertionoftheoligonucleotideinframeandinthecorrectorientationdidnotperturbthesuppressoractivityoftheal

teredSOK1gene(datanotshown),sowereasoned
teredSOK1gene(datanotshown),sowereasonedthatthetaggedpolypeptidewouldbeproperlylocalized.Theepitope-taggedSoklproteinwaslocalizedtothenucleiofcellscontainingthehigh-copy-numberYEpADE8-SOKl-GTEPplasmidbyusingthean-ti-HAmonoclonalantibody12CA5andafluoresceinisothio-cyanate-conjugatedgoatantimouseantibody(Fig.6).Incontrast,nonuclearstainingwasobservedwithstrainsbearingvariouscontrolplasmids,includingtheYEpADE8vectorandaabFIG.6.CellularlocalizationoftheSoklprotein.Thelocalizationofanepitope-taggedderivativeoftheSoklproteinwasvisualizedbyindirectimmunofluorescencewithfluoresceinisothiocyanate-conju-gatedsecondaryantibody(a)andcontrastedwiththelocalizationofcellularDNAasjudgedbyDAPI(4',6-diamidino-2-phenylindole)staining(b).Cellscontainedanepitope-taggedderivativeoftheoriginalhigh-copy-numberSOKIsuppressor(seeMaterialsandMeth-ods).TABLE3.SuppressionofAkinasedefectsbyoverexpressionofSOKIorSOK2Growth"withthefollowingsuppressor:Mutantgenotypeyak]HCb-SOKJHC-SOK2rasiras2-34(Ts)++cdc25-5(Ts)++tpkltpk2-63(Ts)tpk3+++1-tpkltpk2-65(Ts)tpk3+++/-tpkltpk2tpk3±1-+1-aSeeTable2,footnotea.bHC,highcopynumber.SOKIderivativecontainingtheepitopeintheinverseorienta-tion(datanotshown).Consistentwiththisobservation,itispossibletoidentifyatleastonepotentialnuclearlocalizationsignalwithintheSoklcodingregion(Fig.4).Themousetestis-specificproteinalsocontainsasequencethatmightserveasanuclearlocalizationsignal,althoughneithersequencefallswithintheregionssharedbythetwoproteins.Finally,wehaveconfirmedbyWesternblot(immunoblot)analysis(datanotshown)thatstrainsbearingtheSOKJ-GTEPplasmid,butnotthosebearingtheSOK1-PETGconstruct,specifyaproteinthat,at65kDa,isconsistentwiththesizepredictedfortheepitope-taggedSOK1geneproduct(565plus27aminoacids).SOKJoverexpressionsuppressestotallossofAkinasefunction.TodeterminetherelationoftheSOKIandSOK2productstotheRas/Akinasepathway,wetestedtheabilitiesofthecorrespondingplasmidstosuppressothermutationsofthepathway.Arepresentativehigh-copy-numberplasmidfromeachclasswastransformedintotemperature-sensitivecdc25-5(Ts)andrasiras2(Ts)strains,andAde+transformantsweretestedforgrowthatseveraltemperatures.AsshowninTable3,theSOK]plasmidwasabletopartiallysuppressthegrowthdefectscausedbybothmutations.Bycontrast,thehigh-copy-numberSOK2plasmidsuppressedneithermutation,consistentwiththeobservationthatSOK2wasarelativelyweaksuppres-sorofthetpk2(Ts)defect.Disruptionofyak]oroverexpressionofSCH9restoredgrowthtoastrainlackingallthreeAkinasecatalyticgenes,aswellastostrainslackingcdc35andbothrasgenes(13,30).Accordingly,wetestedwhethereitherhigh-copy-numbersup-pressorisolatedinthesestudiescoulddothesame.TheheterozygoustpkdiploidS7-7AxS7-5AwastransformedtoAde+withaplasmidbearingeitherSOK]orSOK2,sporu-lated,andpickedtorichmedium.HaploidsporescontainingdisruptionsofTPKJ,TPK2,andTPK3germinatedonlywhentheycontainedthehigh-copy-numberplasmidcontainingSOK](Table3).Interestingly,thepatternofsuppressionbySOK]overexpressionwassimilartothatconferredbyinacti-vationofYAK1,asjudgedbytheslowgrowth(Table3)(13)andhyperaccumulationofglycogen.Thus,neitherSOK1over-expressionnorYaklinactivationiscapableofcompletelyrelievingtheAkinasegrowthdependenceofacell.Incontrast,colonieslackingallthreecatalyticsubunitgeneswereneverrecoveredfromthediploidcontainingtheSOK2plasmid,despitethehightransmissionfrequencyoftheSOK2plasmidduringsporulation.OverexpressionofSOKJhasnoeffectonglycogenaccumu-lationorsensitivitytostress.SeveraldistinctphenotypeshavebeenassociatedwithstrainscontainingdiminishedorelevatedlevelsofAkinase.Forexample,activationoftheAkinasepathwayresultsincellsthatareexquisitelysensitivetovariousformsofstressaswellasinafailuretoaccumulatestorageMOL.CELL.BIOL.VOL.14,1994SOK1YAK1soklYAK1soklyaklEuSOK1yakl240C36°CFIG.7.EpistaticrelationbetweenSoklandYakl.Fourpatchesoftheindicatedstrainswerereplicatedtorichmedium(yeastextract-peptone-dextrose)agarandincubatedat24and36°Cforseveraldays.Strains:SOKIYAK],MWY63[tpk2-63(Ts)SOKIYAKI];soklYAK1,MWY285[tpk2-63(Ts)sokl::HIS3YAKI];soklyakl,MWY313[tpk2-63(Ts)sokl::HIS3yakl::ADE8];SOKIyakl,MWY273[tpk2-63(Ts)SOKIyakl::ADE8].carbohydratessuchasglycogen.StrainsthatarecompromisedforAkinaseactivityareabnormallyresistanttostressandaccumulateelevatedlevelsofglycogen.ToexaminetheeffectofSOK1overexpressiononthesephenotypes,wedeterminedtheheatshockandstarvationsensitivitiesofawild-typestraincontainingthehigh-copy-numberSOK1plasmidortheYEpADE8vector.Thesameplasmid-containingstrainswereinvertedoveriodinetogaugetheeffectofSOK1overexpres-siononglycogenaccumulation.BythesecriteriatheSOK1-overexpressingstrainwasidenticaltoitsisogenicYEpADE8controlandwasinstarkcontrasttoacongenicbcylstrain,whichwasexquisitelysensitivetoallformsofstressandfailedtoaccumulateglycogenevenonprolongedincubation(datanotshown).DisruptionoftheSOK]genealsohadnoappre-ciableeffectonstresssensitivityorthecapacitytoaccumulateglycogen.DisruptionofSOKIpreventssuppressionoftheAkinasedefectbylossofYaklfunction.TodeterminetheroleoftheSOK]geneproductincellgrowthanddivision,asokl::HIS3disruptionwasplacedinthechromosome.The4.7-kbfragmentofplasmidpMW26(YEpADE8-sok1::HIS3)wasusedtotrans-formdiploidstrain1029(SOKJ/SOK1his3lhis3)toHis',andtwotransfo

rmantsweresubjectedtotetradanalysis.Both
rmantsweresubjectedtotetradanalysis.Bothtransformantscontainedasingledisruptedcopyandawild-typecopyofSOK1asdeterminedbyDNA-DNAhybridization(datanotshown).Alltetradsofbothstrains(17of17)hadfourequal-sizedcolonies,withtheHis'markersegregating2:2(datanotshown).Thus,theSOK1geneisnotessentialforgrowth.AlthoughwecannotruleoutthepossibilitythatSOK1isamemberofafamilyofgeneswithrelatedfunctions,wehavebeenunabletodetectanotheryeastgeneexhibitingsignificantstructuralsimilarityasjudgedbylow-stringencyhybridization(datanotshown).TestsofepistasisplacedtheSOK1suppressordownstreamfrom,oronapathwayparallelwiththatof,theAkinasegene(Table3).Twoothergenesthoughttoencodegrowthregula-torsrelatedtotheAkinasepathwayincludeSCH9andYAKI,identifiedasgenedosageandloss-of-functionsuppressorsofmutationsintheAkinase,respectively.WehavepreviouslyshownthatsuppressionbySoklisindependentoftheSch9kinase(15).TodeterminetherelationbetweenSoklandYakl,thesokl::HIS3disruptionwasintroducedintoisogenictpk2(Ts)YAK1andtpk2(Ts)yak]strains,whichwerethentestedforgrowthatthepermissiveandnonpermissivetemper-atures.WhilethelossofSoklfunctionhadnoapparenteffectonthegrowthoftheYakl-proficientstrain[Fig.7;comparethegrowthofthetpk2(Ts)YAK]SOK]strainwiththatofthetpk2(Ts)YAK1soklmutant],itsinactivationtotallyblockedYEASTAKINASESUPPRESSOR5625YaklKinaseSoklAKinaseGROWTHFIG.8.ModeloftheAkinase-Yakl-Soklpathway.growthofthetpk2(Ts)yaklstrainatelevatedtemperatures.Thus,suppressionoftheAkinasedefectbytheinactivationofYaklrequiredafunctionalSOK]geneproduct.DISCUSSIONWehaveidentifiedanewgene,SOKI,whoseoverexpressionalleviatesthegrowthdefectofyeaststrainslackingAkinaseactivity.Althoughthemechanismofthissuppressionisnotknown,ourresultsareconsistentwithamodelinwhichSoklidentifiesadownstreamcomponentoftheYaklkinasepath-way(Fig.8).Inthatscenario,theYakl-SoklpathwaywouldstimulateasetofessentialcellularprocessesunderAkinasecontrol.ActivationofSokl,bySOK]overexpressionorbyinactivationofthenegativeregulatorYakl,wouldrendereachoftheprocessesindependentofAkinaseactivity.GiventhenuclearlocalizationpatternofSoklonoverexpression,itistemptingtospeculatethatatleastoneprocessmadeindepen-dentofAkinaseactivitymightincludethegeneralorspecificactivationoftranscription.Theseresultsareparticularlyin-triguinginlightofthestructuralsimilaritybetweenthepre-dictedproductofSOK]andtheproductofarecentlyde-scribed,developmentallyregulatedmousegene(21).Themostimportantresultofthesestudiesisthatoverex-pressionofSOK1cansuppressthegrowthdefectofastrainlackingallthreeTPKgenes.ThismarksSOK1asonlythesecondgenewhoseoverexpressioniscapableofbypassingtheneedforAkinaseforgrowth.Thefirstsuchgene,SCH9,encodesaproteinkinaseexhibitingsignificanthomologytotheyeastAkinasecatalyticsubunits(30).ThestructuralsimilaritybetweenSch9andtheAkinase,alongwithphysiologicalandgeneticstudies(15),isatleastconsistentwiththenotionthatsuppressionofatpkstrainbySCH9overexpressionoccursasaresultoftheoverlappingspecificitiesofthetwokinases.TheSOK]gene,incontrast,correspondstoaproteinthatbearsnorelationtoknownproteinkinases(Fig.4).Thus,theSOK]high-copy-numbersuppressorappearstoactbyamechanismthatisdifferentfromthatofSCH9.OneintriguingpossibilityisthatSOK]encodesanAkinasesubstrateinvolvedincellgrowthanddivision.ItsdependenceonAkinasephosphorylationmight,therefore,beabrogatedbyanincreaseinSOK1abundance.Arguingagainstthisproposal,however,isthefactthatdeletionoftheSOK1genedoesnotresultinanoticeablegrowthdefect,contrarytotheexpecta-tionforaneffectorproteininanessentialpathway.ThisresultcouldbereconciledifSOK1wasamemberofaduplicated5626WARDANDGARRETMLgenefamily.However,SOKIandSOK2haveeachbeenisolatedmultipletimes,andgeneticandphysicalanalysessuggestthatthetwogenesarenotfunctionallyrelated(forexample,asokisok2TPKstrainexhibitsnoobviousgrowthdefects[33a]).AnalternatehypothesisisthattheSoklproteinparticipatesinapathwaythatispartiallyredundantwiththatoftheAkinase(Fig.8).Akinase-dependentprocesseswouldberegu-latedindependentlybySokl,suchthatadecreaseinAkinaseactivity(withaconcurrentdiminutioninsomemetabolicprocess)couldbecompensatedforbyanincreaseinstimula-tionbySokl.Insuchamodel,thephenotypiceffectoflosingeitherpathwaywoulddependontherelativecontributionofeachtotheoverallfunctionoftheprocess.Judgingbytheapparentwild-typegrowthofthesoklnullmutant(Fig.7),itseemslikelythatundernormalconditions,thecontributionbytheSoklpathwayissmall.Onactivation,thecontributionbySoklpresumablyincreasestolevelssufficienttorelievetheAkinaserequirement.OurmodelalsopositsthatSoklactivityisnormallyre-pressedbytheYaklkinase.Suchamodelexplainsourearlierobservation(13)thatYaklactivityisantagonistictogrowthofanAkinase-deficientmutant,anditpredictsthatthegrowthofatpk(Ts)yaklstrainwouldbeabrogatedbythelossofSoklfunction.Inotherwords,relieffromYaklrepressionwouldresultinAkinaseindependenceonlyifafunctionalSoklproteinwaspresent.Aspredicted,thegrowthofatpk(Ts)yak]strainismadeconditionalbythedisruptionofSOK].Thus,ourresultsareconsistentwithamodelinwhichthedefectinyeastAkinasecanbealleviatedbyactivationofSoklfunction,eitherbyanincreaseinSOK1expressionorbyadecreaseinYaklkinaseactivity.ThemodelshowninFig.8isconsistentwiththepossibility,butdoesnotrequ

ire,thattheinteractionbetweenYaklandSokl
ire,thattheinteractionbetweenYaklandSoklisdirect,suchthatSoklisinactivatedbyaYakl-specificphosphorylationevent.ThelackofidentitybetweenSoklandotherproteinkinasesdoesnoteliminatethepossibilitythatSOK]overexpressionactivatesakinasethatsharesoverlappingspecificitywiththeAkinase.Forexample,ifSoklstimulatedSCH9transcription,overexpressionofeitherSOK]orSCH9mightresultinsup-pressionoftheAkinasedefect.However,severalresultsargueagainstSoklregulationoftheSch9kinase.First,thesyntheticlethalityofatpk(Ts)sch9straincanbeovercomebySOK1overexpressionordisruptionofyakl(15).Thus,theSOK]andyak]suppressorsmustalleviatetheAkinasedefectbyanSCH9-independentmechanism,placingSoklfunctiondistaltoSch9function.Second,cellslackingSch9activitygrowex-tremelyslowly(15,30),whereassok]deletionmutantsareunalteredingrowth.Thus,thetwofunctionsappearunrelated.ItremainspossiblethatSoklmightalleviatetheAkinasedefectthroughtheactivationofanotherAkinasehomolog;however,suchakinasehasbeennotablyabsentfromthesuppressorsidentifiedtodate(5,9,13,30,33a).ThestructuralsimilaritybetweenSoklandatestis-specificmousetranscriptthoughttoplayaroleinspermdevelopmentimplicatesashareddeterminantorfunctionaldomainofthesetwoproteinsintheregulationofavarietyofimportantgrowthanddevelopmentalprocesses.Unfortunately,theprimarystructuresofthetwoproteinsprovidefewcluestotheirspecificbiochemicalfunctionsorthenatureoftheprocessestheyregulate.PotentialsitesofNglycosylationandmembraneattachment(21)arenotconservedbetweenthetwoproteins,andseveralputativesitesofproteinkinaseCphosphorylationthatarepresentinthemousepBs13productaredisplacedinSokl.Moreover,theregionmostconservedbetweenthetwoproteins,acentralcoreof143aminoacids,doesnotappeartorevealaconsensussequencethatissharedwithanyotherproteinsintheavailabledatabanks.Inthiscontext,itwillbeinterestingtodetermineifthemouseprotein,likeSokl,residesinthenucleus.Localizationofthemouseproteintothenucleuswouldbeconsistentwiththetwoproteinssharingsomefunction,suchastheregulationoftranscription.Finally,SoklappearstobedevoidofanypotentialAkinasephosphor-ylationsites.ThiswouldseemtoconformwelltothenotionthatSoklactivatesanAkinase-regulatedprocessbyamech-anismthatisindependentofAkinasefunction.SinceAkinaseplaysacriticalroleincomplexdevelopmentalpathwaysofotherorganisms(10,23,29),itisreasonabletoimaginethatAkinase-regulatedcircuitsinS.cerevisiaemightbeequallycomplex.IncontrastwiththebroadpatternofsuppressionexhibitedbyoverexpressionofSOK],theSOK2high-copy-numbersup-pressorswereabletoalleviatethegrowthdefectscausedbyonlyarestrictedsetofconditionaldefectsintheAkinasepathway.WhileSOK2reversedthetemperature-sensitivede-fectcausedbymutationsintheAkinasecatalyticsubunitgeneTPK2,severalconditionallesionsinupstreamelementsoftheRas/Akinasepathway[ras2(Ts)andcdc25(Ts)]wereunaf-fected.Althoughthenumberofallelestestedistoofewtomakeastrongconclusion,itistemptingtospeculatethatthisresultmaypointtothemechanismbywhichSOK2overexpres-sionalleviatestheconditionalgrowthofthetpk2(Ts)strains.Onecanimagine,forexample,thatoverproductionofSok2mightalleviateaconformationaldefectofthefree,butaltered,tpk2(Ts)productbuthaveanegligibleeffectontheactivityofawild-typeAkinasecatalyticsubunitsequesteredbyBcyl[aswouldbethecaseinaras2(Ts)orcdc25(Ts)background].Sincepreliminarysequenceandgeneticanalyseshavedeter-minedthatSOK]andSOK2donothaveoverlappingfunctions,itwillbeinterestingtodeterminetheroleofSOK2inAkinase-dependentgrowthcontrolanddivision.Finally,itseemslikelythatthetemperature-sensitiveAkinasemutantsisolatedinthisstudywillcontinuetocontributetoourunderstandingofthegrowthandcellcyclecontrolprocessesregulatedbytheyeastAkinase.Thetwohigh-copy-numbersuppressorsdescribedheremayreflectonlyasubsetofthegenesthatcanbealteredtoalleviatetheAkinasedefect.Forexample,werecentlyhaveshownthatatleastonedomi-nantsuppressorofthetpk2-63(Ts)alleleisunlinkedtoSOK]andmaydefineyetanotherfunctionoftheAkinasepathway(33a).Inaddition,analysesofthephysiologicalandmorpho-logicalpropertiesofconditionalmutantswithdefectsineachstepofthepathway[i.e.,cdc25(Ts),raslras2(Ts),cyr](Ts),andnowtpkltpk2(Ts)tpk3mutants]arecertaintocontributetoourunderstandingofAkinaseregulation(1,25)andmayhelpdefinetheAkinase-independentfunctionoftheyeastRasprotein(1,22,33,34).ACKNOWLEDGMENTSThisworkwassupportedbygrantGM44666fromtheNationalInstitutesofHealth.M.P.W.wassupportedbyNIHtraininggrant5P32GM07184,andS.G.isaJuniorFacultyScholaroftheAmericanCancerSociety(JFRA395).REFERENCES1.Broach,J.R.1991.RASgenesinSaccharomycescerevisiae:signaltransductioninsearchofapathway.TrendsGenet.7:28-33.2.Broach,J.R.,andR.J.Deschenes.1990.ThefunctionofRASgenesinSaccharomycescerevisiae.Adv.CancerRes.54:79-138.3.Cameron,S.,L.Levin,M.Zoller,andM.Wigler.1988.cAMP-independentcontrolofsporulation,glycogenmetabolism,andheatshockresistanceinS.cerevisiae.Cell53:555-566.MOL.CELL.BIOL.YEASTAKINASESUPPRESSOR56274.Cannon,J.,andKTatchell.1987.CharacterizationofSaccharo-mycescerevisiaegenesencodingsubunitsofcyclicAMP-dependentproteinkinase.Mol.Cell.Biol.7:2653-2663.5.Cannon,J.F.,J.B.Gibbs,andK.Tatchell.1986.SuppressorsoftheRAS2mutationofSaccharomycescerevisiae.Genetics113:247-264.6.Casadaban,M.J.,A.Martinez-Arias,S.K.Shap

iro,andJ.Chou.1983.,B-Galactosidasegenef
iro,andJ.Chou.1983.,B-GalactosidasegenefusionsforanalyzinggeneexpressioninEscherichiacoliandyeast.MethodsEnzymol.100:293-308.7.Deschenes,R.J.,andJ.R.Broach.1987.FattyacylationisimportantbutnotessentialforSaccharomycescerevisiaeRASfunction.Mol.Cell.Biol.7:2344-2351.8.DeVendittis,E.,A.Vitelli,R.Zahn,and0.Fasano.1986.SuppressionofdefectiveRAS]andRAS2functionsinyeastbyanadenylatecyclaseactivatedbyasingleaminoacidchange.EMBOJ.5:3657-3663.9.DiBlasi,F.,E.Carra,E.DeVendittis,P.Masturzo,E.Burderi,I.Lambrinoudaki,M.G.Mirisola,G.Seidita,and0.Fasano.1993.TheSCH9proteinkinasemRNAcontainsalong5'leaderwithasmallopenreadingframe.Yeast9:21-32.10.Dowd,D.R.,andR.L.Miesfeld.1992.Evidencethatglucocorti-coid-andcyclicAMP-inducedapoptoticpathwaysinlymphocytessharedistalevents.Mol.Cell.Biol.12:3600-3608.11.Field,J.,J.Nikawa,D.Broek,B.MacDonald,L.Rodgers,I.A.Wilson,R.A.Lerner,andM.Wigler.1988.PurificationofaRAS-responsiveadenylylcyclasecomplexfromSaccharomycescerevisiaebyuseofanepitopeadditionmethod.Mol.Cell.Biol.8:2159-2165.12.Francois,J.,M.E.Villanueva,andH.G.Hers.1988.Thecontrolofglycogenmetabolisminyeast.1.Interconversioninvivoofglycogensynthaseandglycogenphosphorylaseinducedbyglucose,anitrogensourceoruncouplers.Eur.J.Biochem.174:551-559.13.Garrett,S.,andJ.Broach.1989.LossofRasactivityinSaccha-romycescerevisiaeissuppressedbydisruptionsofanewkinasegene,YAK1,whoseproductmayactdownstreamofthecAMP-dependentproteinkinase.GenesDev.3:1336-1348.14.Garrett,S.,M.M.Menold,andJ.R.Broach.1991.TheSaccha-romycescerevisiaeYAK1geneencodesaproteinkinasethatisinducedbyarrestearlyinthecellcycle.Mol.Cell.Biol.11:4045-4052.15.Hartley,A.D.,M.P.Ward,andS.Garrett.1994.TheYaklproteinkinaseofSaccharomycescerevisiaemoderatesthermotol-eranceandinhibitsgrowthbyanSch9proteinkinase-independentmechanism.Genetics136:465-474.16.Johnston,G.C.,J.R.Pringle,andL.H.Hartwell.1977.Coordi-nationofgrowthwithcelldivisionintheyeastSaccharomycescerevisiae.Exp.CellRes.105:79-98.17.Kans,J.A.,andR.K.Mortimer.1991.NucleotidesequenceoftheRAD57geneofSaccharomycescerevisiae.Gene105:139-140.18.Kataoka,T.,D.Broek,andM.Wigler.1985.DNAsequenceandcharacterizationoftheS.cerevisiaegeneencodingadenylatecyclase.Cell43:493-505.19.Maniatis,T.,E.F.Fritsch,andJ.Sambroolk1982.Molecularcloning:alaboratorymanual.ColdSpringHarborLaboratory,ColdSpringHarbor,N.Y.20.Matsumoto,K.,I.Uno,Y.Oshima,andT.Ishikawa.1982.Isolationandcharacterizationofyeastmutantsdeficientinadeny-latecyclaseandcAMP-dependentproteinkinase.Proc.Nati.Acad.Sci.USA79:2355-2359.21.Mazarakis,N.D.,D.Nelki,M.F.Lyon,S.Ruddy,E.P.Evans,P.Freemont,andK.Dudley.1991.Isolationandcharacterisationofatestis-expresseddevelopmentallyregulatedgenefromthedistalinversionofthemouset-complex.Development111:561-571.22.Michaeli,T.,J.Field,R.Ballester,K.O'Neill,andM.Wigler.1989.MutantsofH-rasthatinterferewithRASeffectorfunctioninSaccharomycescerevisiae.EMBOJ.8:3039-3044.23.Mochizuki,N.,andM.Yamamoto.1992.Reductionintheintra-cellularcAMPleveltriggersinitiationofsexualdevelopmentinfissionyeast.Mol.Gen.Genet.233:17-24.24.Nehlin,J.O.,M.Carlberg,andH.Ronne.1992.YeastSKO1geneencodesabZIPproteinthatbindstotheCREmotifandactsasarepressoroftranscription.NucleicAcidsRes.20:5271-5278.25.Nikawa,J.,S.Cameron,T.Toda,K.M.Ferguson,andM.Wigler.1987.RigorousfeedbackcontrolofcAMPlevelsinSaccharomycescerevisiae.GenesDev.1:931-937.26.Powers,S.,K.O'Neill,andM.Wigler.1989.DominantyeastandmammalianRASmutantsthatinterferewiththeCDC25-depen-dentactivationofwild-typeRASinSaccharomycescerevisiae.Mol.Cell.Biol.9:390-395.26a.Rose,M.Personalcommunication.27.Rose,M.D.,F.Winston,andP.Hieter.1989.Methodsinyeastgenetics.ColdSpringHarborLaboratory,ColdSpringHarbor,N.Y.28.Sanger,F.,S.Nicklen,andA.R.Coulson.1977.DNAsequencingwithchain-terminatinginhibitors.Proc.Natl.Acad.Sci.USA74:5463-5467.29.Simon,M.-N.,0.Pelegrini,M.Vernon,andR.R.Kay.1992.MutationofproteinkinaseAcausesheterochronicdevelopmentofDictyostelium.Nature(London)356:171-172.30.Toda,T.,S.Cameron,P.Sass,andM.Wigler.1988.SCH9,ageneofSaccharomycescerevisiaethatencodesaproteindistinctfrom,butfunctionallyrelatedto,cAMP-dependentproteinkinasecat-alyticsubunits.GenesDev.2:517-527.31.Toda,T.,S.Cameron,P.Sass,M.Zoller,andM.Wigler.1987.ThreedifferentgenesinS.cerevisiaeencodethecatalyticsubunitsofthecAMP-dependentproteinkinase.Cell50:277-287.32.Toda,T.,I.Uno,T.Ishikawa,S.Powers,T.Kataoka,D.Broek,S.Cameron,J.Broach,K.Matsumoto,andM.Wigler.1985.Inyeast,RASproteinsarecontrollingelementsofadenylatecyclase.Cell40:27-36.33.Wang,J.,N.Suzuki,andT.Kataoka.1992.The70-kilodaltonadenylylcyclase-associatedproteinisnotessentialforinteractionofSaccharomycescerevisiaeadenylylcyclasewithRASproteins.Mol.Cell.Biol.12:4937-4945.33a.Ward,M.P.Unpublishedresults.34.Wigler,M.,J.Field,S.Powers,D.Broek,T.Toda,S.Cameron,J.Nikawa,T.Michaeli,J.Colicelli,andK.Ferguson.1988.StudiesofRASfunctionintheyeastSaccharomycescerevisiae.ColdSpringHarborSymp.Quant.Biol.53:649-655.35.Woodcock,D.M.,P.J.Crowther,J.Doherty,S.Jefferson,E.DeCruz,M.Noyer-Weidner,S.S.Smith,M.Z.Michael,andM.W.Graham.1989.QuantitativeevaluationofEscherichiacolihoststrainsfortolerancetocytosinemethylationinplasmidandphagerecombinants.NucleicAcidsRes.17:3469-3478.VOL.14,199