Copyright2003bytheGeneticsSocietyofAmerica Mode of Selection and Experimental Evolution of Antifungal Drug Resistance in Saccharomyces cerevisiae JamesB.Anderson,*,1CarolineSirjusingh,*AinslieB.Parsons,†CharlesBoone,†ClaireWickens,* Leah E. Cowen* and Linda M. Kohn* *Department of Botany, University of Toronto, Mississauga, Ontario L5L 1C6, Canada and †University of Toronto, Banting and Best DepartmentofMedicalResearch,Toronto,OntarioM5G1L6,Canada ManuscriptreceivedSeptember25,2002 AcceptedforpublicationJanuary7,2003 ABSTRACT Weshowthatmodeofselection,degreeofdominanceofmutations,andploidyaredeterminingfactors intheevolutionofresistancetotheantifungaldrugfluconazoleinyeast.Inexperiment1,yeastpopulations were subjected to a stepwise increase in fluconazole concentration over 400 generations. Under this regimen,twomutationsinthesametwochromosomalregionsrosetohighfrequencyinparallelinthree replicatepopulations.Thesemutationsweresemidominantandadditiveintheireffectonresistance.The firstofthesemutationsmappedtoPDR1andresultedintheoverexpressionoftheABCtransportergenes PDR5 and SNQ2. These mutations had an unexpected pleiotropic effect of reducing the residual ability ofthewildtypetoreproduceatthehighestconcentrationsoffluconazole.Inexperiment2,yeastpopula- tionsweresubjectedtoasinglehighconcentrationoffluconazole.Underthisregimen,asinglerecessive mutation appeared in each of three replicate populations. In a genome-wide screen of (cid:1)4700 viable deletionstrains,13wereclassifiedasresistanttofluconazole(ERG3,ERG6,YMR102C,YMR099C,YPL056C, ERG28,OSH1,SCS2,CKA2,SML1,YBR147W,YGR283C,andYLR407W).Themutationsinexperiment2 allmappedtoERG3andresultedintheoverexpressionofthegeneencodingthedrugtargetERG11,but not PDR5 and SNQ2. Diploid hybrids from experiments 1 and 2 were less fit than the parents in the presenceoffluconazole.Inavariationofexperiment2,haploidsshowedahigherfrequencyofresistance thandiploids,suggestingthatdegreeofdominanceandploidyareimportantfactorsintheevolutionof antifungaldrugresistance. RESISTANCE to antimicrobial agents enters micro- sistancetoazoledrugscanoccurthroughdiversemech- bial populations through mutation or immigra- anisms, including: (i) alteration in sterol biosynthesis tion. Resistant genotypes then increase in frequency thatresultsinthesubstitutionofothersterolsforergos- in response to the natural selection imposed by the terol, (ii) overexpression of the target protein so that presenceofthedrug.Thekeydeterminantofwhether sufficientenzymeactivityremainseveninthepresence resistance spreads and persists in a population is the of the drug, (iii) overexpression of various membrane fitnessoftheseresistantgenotypesinthepresenceand effluxpumpsthatreduceintracellulardrugconcentra- in the absence of an antimicrobial agent (Andersson tion, and (iv) alteration in the aminoacid sequence of and Levin 1999; Levin et al. 2000). the target protein that reduces its binding affinity for For controlling pathogenic fungi, only a small num- azoles(Sanglardetal.1998;Lupettietal.2002).Addi- ber of different kinds of antifungal drugs are available tionalmechanismsofresistance,notyetdescribed,may (Georgopapadakou and Walsh 1994; Cowen et al. exist. Because of the different possible mechanisms of 2002b). Of the antifungal drugs that are commonly azole resistance, each with different possible fitness ef- used, most target the ergosterol biosynthesis pathway. fects,wehypothesizedthattheparticularmodeofselec- Ergosterolisthemajorsterolinfungalcellmembranes tioncoulddeterminethemechanismsofresistancethat andisnotpresentinanimalcellmembranes.Inaddition ultimately become established in a fungal population. toitseffectonfungalmembranefluidityandpermeabil- For example, selection with different drug concentra- ity,ergosterolhasaroleinregulatingcelldivision.The tions may favor different mechanisms of resistance to widely used azole drugs target the cytochrome p450 the same agent. enzyme, lanosterol demethylase, which is encoded by Much of what is known about antifungal drug resis- ERG11,ageneessentialforergosterolbiosynthesis.Re- tance comesfrom studies of theyeast Candida albicans, a widespread commensal and important pathogen of humans. In C. albicans, Cowen et al. (2000) provided evidencefor theevolutionofdivergent mechanismsof 1Correspondingauthor:DepartmentofBotany,UniversityofToronto, resistance to the antifungal drug fluconazole (FLC) in 3359MississaugaRd.,NorthMississauga,ONL5L1C6,Canada. E-mail:[email protected] experimental populations. The divergent response of Genetics163:1287–1298(April2003) 1288 J.B.Andersonetal. the C. albicans populations in that study may be attrib- TABLE1 uted in part to population size, which was reduced to PCRprimersfortaggedreplacementsatURA3 (cid:1)106 cells per daily batch transfer, and to the random nature of mutation availability. In smaller populations, FirstPCRfrompFA6KanMX4 the random nature of the various possible mutations Forward for resistance is expected to lend more of an element GCAGGAAACGAAGATAAATC-tag-CGTACG ofchancetotheoutcomethanitdoesinlargerpopula- CTGCAGGTCGAC tions in which mutations are more available and selec- Reverse TTTACTTATAATACAGTTTT-tag-ATCGAT tion is more efficient, resulting in more frequent fixa- GAATTCGAGCTCG tion of the fittest genotypes. Tags Inrecentstudies,wecharacterizedthetwodivergent 1ATCTTACAAAATTTGGTTTA programs of adaptation to the presence of FLC in 2TCTGACATGACTAAGTTCAC C.albicans(Cowenetal.2000,2002a).Thefirstprogram 3GTCTTAGGTATCGACGGCAT ofadaptationwasobservedonlyonceamongsixexperi- 4CGAACATAGTTGCTAATGCT mental populations and included constitutive overex- 5CATCTGGAAGTGAAATCCAT InvariantProbe pressionofCDR1andCDR2,whichencodeeffluxpumps GCCATCAAAATGTATGGATGC (ABC transporters) known to play a role in azole resis- SecondPCR tance,andalteredexpressionofeightadditionalgenes. Forward Theotherprogramofadaptationincludedconstitutive TCTTAACCCAACTGCACAGAACAAAAACCTGCAG overexpression of MDR1 (Cowen et al. 2000, 2002a), GAAACGAAGATAAATC which encodes another kind of efflux pump (major Reverse facilitator)alsoknowntoplayaroleinFLCresistance. GCTCTAATTTGTGAGTTTAGTATACATGCATTTAC TTATAATACAGTTTT Thissecondprogramofadaptationaroseinparallelin three different populations and was accompanied by altered expression of (cid:1)100 other genes. This second program was further characterized by a succession of to a stepwise increase in FLC concentration from low early and late patterns of gene expression. Although to high over 400 generations; in experiment 2, yeast theearlyandlateexpressionprofileswerehighlysimilar populationsweresubjectedattheoutsettoahighcon- amongpopulations,thesepatterns,exceptforthecon- centration of FLC. sistentoverexpressionofMDR1,werealmostcompletely dissimilar from one another. The patterns of gene ex- pression associated with thefirst and second programs MATERIALS AND METHODS of adaptation are not unique to experimental popula- Strains:The progenitorof allexperimentswas aprototro- tions. Essentially the same patterns of gene expression phicMATastrainderivedfromacrossoftwostrains(“FY69,” alsooccurinclinicalisolatesofC.albicansthatareresis- MATa leu2(cid:2) GAL2, ATCC 90842 and “S288C-ura3,” MAT(cid:3) tant to FLC (Cowen et al. 2002a). ura3,ATCC90842)closelyrelatedtothelaboratorystandard Inadditiontotheirdifferentpatternsofgeneexpres- S288C. The entire URA3 open reading frame (ORF) of this sion,thetwoprogramsofadaptationinC.albicanshad strainwasreplacedwiththeKanMX4cassetteflankedbytwo unique20-nucleotidebarcodestocreatefiveuniquelytagged markedly different fitness profiles. The first program, strains. The PCR primers used to prepare the transforming whichincludedoverexpressionofCDR1andCRD2,was DNA from plasmid pFA6a-KanMX4 (Wach et al. 1994) are associatedwithextremelyhighfitnessbothinthepres- listedinTable1;transformationwasbyamulti-wellprocedure enceandintheabsenceofFLC(Cowenetal.2001).The (http://sequence-www.stanford.edu/group/yeast_deletion_ second program of adaptation, which included over- project/protocols.html; Winzeler et al. 1999). Each trans- formantwas auxotrophicforuraciland resistantto5-fluoro- expression of MDR1, was associated with various levels orotic acid (5-FOA). One of the two tags from each of the of fitness, all lower than that associated with the first five strains was used throughout this study; these tags are program of adaptation (Cowen et al. 2001). designated1–5. Althoughthephenotypicattributesofdrugresistance TestsofminimuminhibitoryconcentrationofFLC:These in C. albicans are well studied, genetic analysis is ham- testswereconstructedandinterpretedaccordingtostandard protocol(NationalCommitteeforClinicalLaboratory pered by the inability to make meiotic crosses. In this Standards 1997) except that 0.5(cid:4) yeast peptone dextrose study, we turned to Saccharomyces cerevisiae, a yeast rela- (YPD;Adamsetal.1997)wasthetestmediumusedthroughout tively closely related to C. albicans, but with well-estab- thisstudy.FLCwasobtainedasagiftfromPfizerCanada.All lished methods for molecular genetics and functional taggedstrainshadaninitialminimuninhibitoryconcentration genomics. Our goal was to test the hypothesis that the (MIC)of16(cid:5)g/mlofFLC.Althoughthemajorityofstrains hadidenticalMICsamongreplicatetests,theresistantstrains mode of selection to which populations are exposed fromexperiment1occasionallyvariedtwofoldinMICofFLC. determinestheevolutionofresistanceinexperimental This level of variation is within the range of variation com- populations. Two modes of selection were used in this monly observed in MIC tests (National Committee for study:inexperiment1,yeastpopulationsweresubjected ClinicalLaboratoryStandards1997). EvolutionofAntifungalDrugResistance 1289 Contaminationchecksandfitnessassays:Withthepossibility method (Sherman and Hicks 1991). Hybrid diploids were ofuniformevolutionarytrajectoriesof adaptation,itwasim- constructed by mixing overnight cultures of haploid strains portanttoruleoutcross-contaminationamongthereplicate with compatible mating types in liquid 0.5(cid:4) YPD medium populations. The unique tags were used to confirm identity overnight.Zygoteswereisolatedbymicromanipulation. throughouteachexperiment;nocontaminationwasobserved Quantification of mRNA transcripts: Northern hybridiza- and each tag appeared where expected without exception. tions were used to quantify mRNA levels corresponding to Thetagswerealsousedasameansforquantifyingfitness,or PDR5, SNQ2, FLR1, and ERG11 relative to a standard, YEF3, reproductiveoutput,bymeasuringchangeintheproportion and were performed exactly as described by Cowen et al. of strainsin mixed cultures overtime. To measurefitness, a (2000),exceptthatthereferenceforbasalexpressionwasthe segment of the URA3 region in the replacement strains was progenitor,P1. amplifiedfromgenomicDNAfrommixedcultureswithprim- Genome-wide screen for fluconazole resistance: Approxi- ersAGAAGGTTAATGTGGCTGTGG andGCCATCAAAATG mately4700MATahaploiddeletionstrainsfromtheS.cerevis- TATGGATGC.Ampliconsof525bp,whichincluded279bp iaedeletionconsortiumweremaintainedinanorderedarray oftheremainingURA3regionand246bpofKanMX4,were on16single-wellagarplatesof86(cid:4)128mmatadensityof transferredtonylonmembranesbycapillaryblottingandwere 768 strains per plate (384 individual strains in duplicate). then probed in succession with 32P-end-labeled oligonucleo- Strains were robotically pinned onto 0.5(cid:4) YPD (cid:6) adenine tides complementary to the tagged regions, as well as with mediumcontaining 64(cid:5)g/ml FLCand incubatedat 30(cid:7)for one oligonucleotide complementary to an invariant region 2days.Strainsviableatthisdrugconcentrationwereidentified within the KanMX4 replacement. There was no detectable as putative suppressors of fluconazole sensitivity. Resistant cross-hybridizationamongthetagsequences.Signalintensity strainswereindividuallyconfirmedwith spotassays;four10- wasmeasuredwithaphosphorimagerwithappropriateback- fold serial dilutions starting at OD (cid:8) 1 were spotted onto groundsubtraction(backgroundwasusually(cid:9)1%ofthesig- media containing 64 (cid:5)g/ml FLC and growth was compared nal).Exposuresweretimedtoyieldsignals(cid:9)5%ofthesatura- to a wild-type control after 2 days at 30(cid:7). Strains that grew tioncapacityofthephosphorscreen. on theFLC platesat all dilutionswere classifiedas “strongly Foreachstraininamixedculture,thenumberofdoublings resistant,”strainsthatgrewonlywhenOD(cid:1)0.01wereclassi- was calculatedas log(R (cid:4) D/R (cid:4)D), whereR isthe ratio fied as “moderately resistant,” and strains that grew only at 2 f f i i f ofthesignalforthetagtothesignalfortheinvariantprobe OD(cid:8)1wereclassifiedas“mildlyresistant.”Strainsnotgrow- attheendoftheincubationandR isthecorrespondingratio ingatanydilutionswereclassifiedas“notresistant.” i at the beginning of a batch culture. D is the optical density Construction of isogenic MATa/a and MAT(cid:1)/(cid:1) diploids: f (530 nm) at the end of the incubation and D is the optical Recovery of diploids homozygous for mating type followed i densityat thebeginning. Forpairs oftagged strains,calibra- the procedure of Reynolds and Fink (2001). A MATa/(cid:3) tionswereconstructedbydilutingonestationary-phaseculture diploid (P1) was transformed with pGAL1-HO and trans- insixtwofoldincrementsintoanothercultureandviceversa. formants were grown overnight in rich medium containing For each tagged strain, the ratio of the signal for the tag to galactose as a carbon source. Colonies that had lost pGAL1- theinvariantprobewasplottedasafunctionoftheproportion HO were identified on medium with 5-FOA. Colonies that ofthestraininthemixedculture.ThemeanR2forallcalibra- secretedeitherpheromonewerelocatedbyreplicaplatingto tionswas0.99. mediumcontainingcellshighlysensitivetooneortheother At each time point of experiment 1, cells from the (cid:10)80(cid:7) pheromone(strainsSY2014,MAT(cid:3)ste3(cid:2)306::LEU2sst2(cid:2)and archive were streaked out on YPD agar and a single colony SY2625, MATa bar1(cid:2)). Colonies secreting pheromone were wasusedtoestablisha10-mlovernightculture.Therationale identified by the presence of a halo of growth inhibition in for choosing single colonies, rather than mass cultures, for thebackgroundcells.Theputativehomozygousdiploidsdid fitnessassayswasthatresistanceevolvedrapidlyinexperiment notsporulateandhadcellvolumesgreaterthanthoseofthe 1,mostlikelyresultinginnearfixationofthefittestgenotype. haploids. Thefitnessofeachstrainwasmeasuredrelativetotheprogeni- torat0,16,32,64,and128(cid:5)g/mlFLCbymixingstationary cultures of P1 (“P” means progenitor; the number indicates RESULTS thetag),D2,D3,andD4(“D”meanspropagatedinincreasing concentrationsofthedrug;thenumberindicatesthetag)in Experiment 1: Mutation and fitness: Three haploid theproportions0.50,0.17,0.17,and0.17,respectively.These MATa-taggedstrains,D2,D3,andD4,werepropagated mixtureswerepropagatedintwosuccessivebatchculturesof separately in 10-ml batch cultures of 0.5(cid:4) YPD, with 0.5(cid:4)YPD with100-fold dilutionatthe timeof transfer.The daily transfer of 0.1 ml of the stationary-phase culture number of cell doublings over each daily increment was to9.9mloffreshmediumforanaverageof6.6genera- summedforthetotalperiod.Theotherfitnesstestsweredone tionsperday.TheconcentrationofFLCwas16(cid:5)g/ml in exactly the same way except that strains were mixed in equalproportionsinpairs.Insomeofthetests,theprogenitor for the first 100 generations, 32 (cid:5)g/ml for the second strain was P5 (“P” means progenitor; the number indicates 100 generations, 64 (cid:5)g/ml for the third 100 genera- thetag). tions, and 128 (cid:5)g/ml for the fourth 100 generations, Construction of hybrid diploids and isolation of tetrads: for a total of 400 generations. Samples of populations TheprogenitorsP1andP5andtheevolvedlinesD2,D3,and were archived at (cid:10)80(cid:7) in 15% glycerol at 0, 100, 200, D4 at generation 400 were diploidized by transforming with URA3-based plasmid pCY709, which contains the HO gene 300, and 400 generations. under its own promoter. Transformants were cultured over- The entire fitness surface for experiment 1 is shown nightinmediumwithouturacilandthencoloniesthathadlost inFigure1.Atgeneration0,noneofthetaggedstrains theplasmidwereidentifiedonmediumwith5-FOA(Adamset hadbeensubjectedtoFLCandallrespondedsimilarly al.1997).MATa/(cid:3) diploidywasconfirmedby thecriteriaof to the various concentrations of FLC. All populations lack of ability to mate, ability to sporulate, and increased cell size relative to haploids. Diploid strains were allowed to underwent (cid:1)13 doublings in the absence of FLC and sporulateandtetradsofsporeswereisolatedbythestandard 7–8doublingsevenat128(cid:5)g/mlofFLC.Bygeneration 1290 J.B.Andersonetal. ml, in which the fitness of the evolved and progenitor populations was approximately equal. By generations 300and400,populationsD2,D3,andD4showedhigher fitnessthantheprogenitorinallconcentrationsofFLC. AtnotimeintheexperimentwasthefitnessofD2,D3, andD4lessthanthatoftheprogenitorP1intheabsence of FLC; no fitness cost of resistance was evident. Genetic analysis: The following observations showed that D2, D3, and D4 each accumulated two semidomi- nantmutationsinthesametwounlinkedchromosomal regionsinthesameorderduringexperiment1andthat thetwomutationswereapproximatelyadditiveintheir effect on MIC of FLC. Tomeasurethedegreeofdominance,sixhybriddip- loidswereconstructedbymatingMATaandMAT(cid:3)mei- otic offspring of diploid D2, D3, and D4, which had MICsof256(cid:5)g/ml,withthoseoftheprogenitordiploid P1, which had a MIC of 16 (cid:5)g/ml. The six hybrid dip- loids had intermediateMIC values of 64or 128 (cid:5)g/ml (Table 2), indicating semidominance. These same six hybrid diploids had levels of fitness intermediate be- tween the P1 and the D2, D3, and D4 homozygous, nonhybriddiploidsin128(cid:5)g/mlFLC,againindicating semidominance. All diploids showed approximately equal fitness in the absence of FLC (Table 2). Toexaminesegregationofmutations,thesixhybrid diploids representing crosses between P1 and D2, D3, Figure 1.—Fitness of the progenitor P1 (*) and evolved and D4 were allowed to sporulate and meiotic tetrads populations D2 ((cid:1)), D3 ((cid:2)), and D4 ((cid:4)) from experiment 1in0–128(cid:5)g/mlFLC. were analyzed. A total of 8 tetrads were interpreted as parental ditypes, 7 as nonparental ditypes, and 30 as tetratypes (see Figure 3 for examples of each). This 100, populationsD2, D3,and D4 allshowed enhanced ratio closely approximates the expected 1:1:4 ratio of fitnessat16and32(cid:5)g/mlFLC,butnotathighercon- parental ditypes, nonparental ditypes, and tetratypes centrations of the drug. Unexpectedly, the fitness of expected with segregation of alleles at two unlinked theprogenitorwassignificantlyhigherthanthatofthe lociwithalargecombinedgene-to-centromeredistance. evolved populations D2, D3, and D4 at generation 100 Ourinterpretationofthesesegregationpatternsisthat when measured in 64 and 128 (cid:5)g/ml FLC. This effect onemutantgenedeterminesagreaterlevelofresistance wasreproducibleonagarmediumwith128(cid:5)g/mlFLC (MIC 64 (cid:5)g/ml) than the other (MIC 32 (cid:5)g/ml) and (Figure2);theprogenitorproducedsmallcoloniesthat thatbothmutantgenestogetherconferanevenhigher stoppedgrowingafter1day,buttheevolvedpopulations levelofresistance.Therewasnoambiguityaboutwhich produced even smaller colonies indicating fewer cell mutationscamefirstinexperiment1:thelargerjumps divisions, a result consistent with the fitness assays. At in MIC, which were accompanied by point mutations generation200,populationsD2,D3,andD4hadhigher in PDR1 (see below) and elevated expression of PDR5 fitness at all concentrations of FLC, except at 128 (cid:5)g/ and SNQ2 (see below), all occurred in the first 100 Figure 2.—Unexpected fitness deficit of D2 fromgeneration100on0.5(cid:4)YPDagarwith128 (cid:5)g/mlFLC.TheprogenitorP1(left)underwent more cell divisions, resulting in larger colonies, than did the evolved, “resistant” (MIC, 64 (cid:5)g/ ml)strainD2(right)during2daysofincubation. EvolutionofAntifungalDrugResistance 1291 TABLE2 FitnessofdiploidsconstructedfromtheprogenitorhaploidandFLC-resistanthaploidsfromexperiment1 No.ofdoublings(cid:11)SD(n(cid:8)3replicates) Diploids/ MIC progenitora FLC TagXb TagYb Tag5(allP5) Parentdiploidsc P1/P5 16 6.2(cid:11)0.1 6.2(cid:11)0.2 D2/P5 256 14.7(cid:11)0.0 5.5(cid:11)0.4 D3/P5 256 14.1(cid:11)0.1 6.6(cid:11)0.8 D4/P5 256 14.4(cid:11)0.0 7.2(cid:11)1.0 Reconstructeddiploids P1(cid:4)P1/P5 16 7.2(cid:11)0.1 6.9(cid:11)0.2 D2(cid:4)D2/P5 256 14.5(cid:11)0.2 4.3(cid:11)2.8 D3(cid:4)D3/P5 256 14.4(cid:11)0.2 7.3(cid:11)1.5 D4(cid:4)D4/P5 256 14.0(cid:11)0.1 7.1(cid:11)1.9 Hybriddiploids(D(cid:4)D) D2(cid:4)D3/P5 256 14.5(cid:11)0.1 14.4(cid:11)0.1 8.8e D2(cid:4)D4/P5 256 14.4(cid:11)0.1 14.0(cid:11)0.2 6.4(cid:11)1.5 D3(cid:4)D2/P5 256 14.4(cid:11)0.2 14.5(cid:11)0.2 6.7(cid:11)1.8 D3(cid:4)D4/P5 256 14.2(cid:11)0.1 14.2(cid:11)0.1 7.1(cid:11)1.0 D4(cid:4)D2/P5 256 14.7(cid:11)0.1 14.5(cid:11)0.2 6.5(cid:11)1.0 D4(cid:4)D3/P5 256 14.4(cid:11)0.1 14.4(cid:11)0.0 7.2(cid:11)1.1 Hybriddiploids(P(cid:4)D)d P1(cid:4)D2/P5 64 8.4(cid:11)0.1 8.5(cid:11)0.1 7.1(cid:11)0.1 P1(cid:4)D3/P5 128 7.4(cid:11)0.2 7.6(cid:11)0.2 7.3(cid:11)0.2 P1(cid:4)D4/P5 64 9.2(cid:11)0.4 9.2(cid:11)0.3 7.5(cid:11)0.3 D2(cid:4)P1/P5 64 8.4(cid:11)0.3 8.3(cid:11)0.2 7.0(cid:11)0.6 D3(cid:4)P1/P5 128 11.8(cid:11)0.2 11.3(cid:11)0.2 6.8(cid:11)0.5 D4(cid:4)P1/P5 64 8.8(cid:11)0.2 8.9(cid:11)0.0 7.1(cid:11)0.2 The mean number of doublings for all diploid strains in 0.5(cid:4) YPD with no FLC was 13.5 (cid:11) 0.2 (data not shown). aDiploids are listed with MAT(cid:3) strain first and the MATa strain second. Progenitor diploid P5 after the slash(/)wasthereferencestrainineachfitnessassay. bFitnessof eachdiploidcarryingtwo tags(e.g.,D2 (cid:4)D3)was measuredwithbothtags (e.g.,tag2 andtag 3). cParent diploids were constructed from the haploid strains by transient transformation with a plasmid carryingtheHOgene. dBecause these assays were done at a concentration of FLC that was higher than the measured MIC, the overalldensityattheendofthesecompetitionassaysremainedlowandthesemeasuresoffitnessweretherefore morevariablethanthosedoneunderconditionsinwhichthemixedculturesreachedhighdensityattheend oftheassayperiod. eNostandarddeviationavailable;onlyonemeasurementwasmade. generations.The mutationsoflessereffect onMICbe- determinants of resistance in D2, D3, and D4 were lo- came apparent only in subsequent generations. cated in the same chromosomal regions. In addition to the six hybrid diploids of P1 with D2, As a control for the genetic analyses above, homozy- D3,andD4atgeneration400,sixhybriddiploidswere gous,nonhybriddiploidversionsofP1andD2,D3,and also constructed among the fluconazole-resistant D2, D4atgeneration400werealsoallowedtosporulateand D3, and D4 at generation 400. Each of these hybrids 5–10 tetrads were analyzed from each. No segregation had the same MIC of 256 (cid:5)g/ml FLC as the parent inthelevelsofMICwasobservedinanyofthesetetrads. diploid versions of D2, D3, and D4 (Table 2) and had The spore progeny of diploid D2, D3, and D4 were all the same high levels of fitness in 128 (cid:5)g/ml FLC as highlyresistant(MIC256(cid:5)g/ml)andthoseofdiploid thenonhybriddiploidsandtheirpredecessorhaploids. P1 had the basal level of resistance (MIC 16 (cid:5)g/ml). Meiotic offspring of hybrids among D2, D3, and D4 TofurthercharacterizetheFLC-resistantphenotypes, showednosegregationforresistancetofluconazole;in the expression of four genes known to play a role in a total of 34 tetrads from these hybrids, all spores had resistance to FLC was measured. From generation 100 MICs of 256 (cid:5)g/ml. This result showed that the two on, populations D2, D3, and D4 all overexpressed the 1292 J.B.Andersonetal. for expression of the same set of four genes (Figure 3). In all cases, the putative genotype interpreted as containingthefirstmutationexpressedPDR5andSNQ2 at a high level approximately equal to the evolved par- ents and all genotypes interpreted as lacking the first mutation expressed PDR5 and SNQ2 at basal levels. Sporesofthesetetradswerealsoexaminedforability toreproducein128(cid:5)g/mlFLConagarmedium.Geno- types from these tetrads containing only the first resis- tance mutation showed the same fitness deficit in 128 (cid:5)g/ml FLC as did D2, D3, and D4 at generation 100 (see Figure 2), at which time only the first mutations had become established. Mapping of mutations: The semidominant nature of thefirstmutationsfromexperiment1andtheireffects on the expression of PDR5 and SNQ2 were consistent Figure3.—MICofFLCandexpressionofPDR5andSNQ2 inthreemeiotictetradsfromthecrossP1(cid:4)D2(generation with the action of known mutations in PDR1 or PDR3 400, experiment 1). Genotypes are for the first and second (KolaczkowskaandGoffeau1999;DeRisietal.2000). mutations in experiment 1: r, the mutant allele conferring For these crosses, the first mutation from the D2 line resistance;s,thewild-typeallele;T,tetratype;NPD,nonparen- in experiment 1 was placed in a MAT(cid:3) background talditype;PD,parentalditype.ForMICs,thebluebarsindicate lacking G418 resistance. This strain was then crossed the greater contribution of the first (PDR1) mutations (64 (cid:5)g/ml) and the green bars indicate the contribution of the withthePDR1andPDR3knockoutstrains,bothofwhich second mutations (32 (cid:5)g/ml). For PDR5 and SNQ2, black carried the KanMX4 cassette with G418 resistance at indicatesthe basallevelof expressionofthe progenitorand the deletion site and had MICs of 16 (cid:5)g/ml FLC or red indicates an increase in expression comparable to those lower. In the cross with the PDR1 knockout strain, all inTable3. 30 tetrads were parental ditype with two FLC-resistant sporesandtwoFLC-sensitive,G418-resistantspores.The unknown resistance mutation was therefore tightly ABCtransportergenesPDR5andSNQ2,three-tofour- linked to PDR1. In the cross with the PDR3 knockout fold with respect to P1 and all of its derivatives (Table strain, all tetrads also segregated 1:1 for resistant vs. 3).Noneofthesesamestrainsoverexpressedthemajor sensitive,buttheG418resistance(encodedatthePDR3 facilitatorgeneFLR1,theexpressionofwhichwasbarely knockoutsite)didnotcosegregatewithlowresistance. detectable in any of the strains assayed in this study Thisindicatesthatthefirstmutationsfromexperiment (data not shown). The expression of ERG11 was more 1 were not in the same chromosomal region as PDR3. variable in D2, D3, and D4, with no consistent trend ThefirstresistancemutationsinD2,D3,andD4were (Table3).ThehybriddiploidsbetweenP1andD2,D3, further pinpointed to PDR1 by sequencing both DNA and D4 expressed PDR5 and SNQ2 at a level between stands of the entire PDR1 ORF plus the flanking in- thatoftheprogenitorandparentsevolvedfor400gen- tergenic regions in the progenitor P1 and in D2, D3, erations in FLC. Three tetrads representing parental and D4 at generations 100 and 400. A single mutation ditypes,nonparentalditypes,andtetratypesforthetwo wasobservedineachofthethreeevolvedlinesatgenera- segregatinggenesfromexperiment1werealsoassayed tions100and400.Eachmutationwasnearthecarboxy TABLE3 Expressionofthreegenes(mean(cid:11)SD)relativetoYEF3 MIC((cid:5)g/ml) Strains(n(cid:8)no.measured) FLC PDR5 SNQ2 ERG11 Haploidprogenitors(n(cid:8)3) 16 1.2(cid:11)0.3 1.1(cid:11)0.2 1.2(cid:11)0.2 HaploidD2,D3,andD4,gen.100(n(cid:8)3) 64 4.8(cid:11)0.8 2.7(cid:11)0.5 1.2(cid:11)0.1 HaploidD2,D3,andD4,gen.400(n(cid:8)3) 256 3.5(cid:11)0.3 2.3(cid:11)0.5 2.5(cid:11)0.9 DiploidD2,D3,andD4,gen.400(n(cid:8)3) 256 3.2(cid:11)0.2 2.9(cid:11)0.6 1.3(cid:11)0.3 HybriddiploidsD2,D3,andD4(cid:4)P1(n(cid:8)6) 64or128 1.8(cid:11)0.3 1.8(cid:11)0.2 1.3(cid:11)0.3 HaploidO1,O2,andO3(n(cid:8)3) 256 0.8(cid:11)0.5 1.3(cid:11)0.4 3.5(cid:11)0.3 HybriddiploidO1,O2,O3(cid:4)D2,D3,D4(n(cid:8)9) 64or128 3.3(cid:11)0.8 2.0(cid:11)0.5 1.2(cid:11)0.2 HybriddiploidO1,O2,O3(cid:4)P1(n(cid:8)3) 16 1.5(cid:11)0.5 1.5(cid:11)0.2 1.0(cid:11)0.2 Gen.,generation. EvolutionofAntifungalDrugResistance 1293 TABLE4 FitnessofprogenitorhaploidandFLL-resistanthaploidsfromexperiments1and2 No.ofdoublings(cid:11)SD(n(cid:8)3replicates) Haploid Tag1 Tag2 Tag3 Tag4 Tag5 O1/O2/O3/P5 13.9(cid:11)0.4 14.1(cid:11)0.3 13.7(cid:11)0.4 — 8.0(cid:11)0.2 D2/D3/D4/O1/P5 11.3(cid:11)0.3 14.7(cid:11)0.1 14.6(cid:11)0.2 14.3(cid:11)0.1 9.0(cid:11)0.1 O2/D3/D4/P5 — 11.8(cid:11)0.2 14.6(cid:11)0.3 14.4(cid:11)0.4 8.7(cid:11)0.2 O3/D2/D4/P5 — 14.7(cid:11)0.2 11.5(cid:11)0.4 14.1(cid:11)0.4 8.3(cid:11)0.3 Mean number of doublings for all strains in 0.5(cid:4) YPD in the absence of FLC was 13.3 (cid:11) 0.3 (data not shown).Boldfacetypeindicatesone-stepmutantstrain.UnderliningindicatesDstrain.Nounderliningindicates P5.—,notagpresent.Slashes(/)separatethedesignationsofthestrainscompeted. terminusofthepredictedpolypeptide,neartheactiva- region. The MICs of the heterozygous diploid hybrids tion domain (Kolaczkowska et al. 2002) of this tran- ofO1,O2,andO3withP1wereall16(cid:5)g/ml,indicating scriptional regulator: D2, T817K; D3, C862W; and D4, that the resistance determinant was recessive with re- L722P. specttoMIC.Thefitnessofthesesamehybriddiploids, Thesecondmutationsinexperiment1:Althoughthesec- however, was somewhat higher (9.7 (cid:11) 0.5 doublings) ond mutations in experiment 1 remain unidentified, thanthatofthecompetitorprogenitordiploidP5(7.4(cid:11) certaingenescanbeexcludedascandidates.Sincethe 0.3 doublings) in 128 (cid:5)g/ml FLC, indicating that the second mutations were semidominant, they are not resistance determinant was not completely recessive likelytobetheresultofsimplelossoffunctionandare withrespecttothiscriterion.Thesesameheterozygous therefore not likely to correspond to any of the FLC- hybrid diploids were allowed to sporulate and all 30 resistant, gene-deletion strains described below under tetradsdissectedshoweda1:1ratioforMICsof16and experiment 2. Also, because the second mutations did 256 (cid:5)g/ml; this indicated the segregation of a single notaffecttheexpressionofPDR5andSNQ2,theregula- mutation.WhenO1andO3werecrossedwithaMAT(cid:3), tor PDR3 is not a likely candidate. Further transcrip- FLC-resistant (MIC 256 (cid:5)g/ml) segregant from the tional profiling may provide clues about the nature of cross of O2 with P1, the hybrid diploids had MICs of the second mutations. 256 (cid:5)g/ml and none of 29 meiotic tetrads examined Experiment 2: Mutation and fitness: Three haploid showedanysegregation;allsporesshowedMICsof256 MATa-taggedstrains,O1,O2,andO3(“O”meansselec- (cid:5)g/ml.ThisshowedthatthemutationsinO1,O2,and tioninasinglehighconcentrationofFLC;thenumber O3wereeachlocatedinthesamechromosomalregion. indicatesthetag),werespreaddirectlyonmediumcon- O1,O2,andO3consistentlyoverexpressedERG11rela- taining 128 (cid:5)g/ml FLC. On each plate, 104 cells from tivetoP1,butPDR5,SNQ2,andFLR1wereallexpressed an overnight culture in 0.5(cid:4) YPD were distributed as at levels similar to that of P1 (Table 3). evenlyaspossible.Becausewild-typecellsundergoseven Mappingofthemutation:Toidentifytherecessivemuta- toninedoublingsinthepresenceofFLC,cellnumbers tionfromexperiment2,wefirsttookacomprehensive on each plate reached 1–5 (cid:4) 106 within 2 days (see and unbiased approach to identify FLC-resistant yeast below). Large, continuously growing colonies were mutants.Approximately4700viablehaploiddeletionmu- picked after 4–5 days of incubation. Three mutants of tants were screened for FLC resistance. Because these O1, O2, andO3, respectively, all had MICsof 256 (cid:5)g/ strains are viable, this set should be highly enriched for ml of FLC. loss-of-function mutations with minimal fitness defects. Thefitnessofthestrainsfromexperiment2wascom- In total, 13 deletion mutants displayed a FLC-resistant paredtothatoftheprogenitorandtheresistantstrains phenotype (Figure 4). This set of FLC-resistant strains is from experiment 1 (Table 4) in competitive growth significantly enriched for genes classified within the Mu- assays containing 128 (cid:5)g/ml FLC. The fitness of O1, nichInformationCenterforProteinSequencesdatabase O2, and O3 was substantially higher than that of P5, (http://mips.gsf.de)asfunctioninginlipid,fatty-acid,and butlessthanthatofD2,D3,andD4.ThefitnessofO1, isoprenoidbiosynthesis(P(cid:8)7.7(cid:4)10(cid:10)5;Robinsonetal. O2,andO3intheabsenceofFLCwasthesameasthat 2002; http://funspec.med.utoronto.ca). The strength of ofalloftheotherstrains,includingtheprogenitor;no theFLC-resistantphenotypewasdistinguishedbycolony fitness cost of resistance was detected. size on FLC medium (Figure 4). Consistent with pre- Geneticanalysis:ThefollowingresultsshowedthatO1, viously published observations (Sanglard et al. 1998; O2,andO3eachcontainedasinglerecessivemutation Lupetti et al. 2002), loss of function at ERG3 resulted for resistance that mapped to the same chromosomal inarelativelystrongFLC-resistantphenotype.Deletion 1294 J.B.Andersonetal. Figure 4.—Haploid deletion strains resistant to 64 (cid:5)g/ml FLC. Cellular roles are as defined by the Yeast Proteome Database (http://www. incyte.com/proteome). mutations of three other genes implicatedinergosterol doublings) in 128 (cid:5)g/ml FLC. Hybrids between O1, biosynthesis,ERG6,ERG28,andOSH1(Paltaufetal.1992; O2, and O3 and all other strains expressed ERG11 at Beh et al. 2001; Gachotte et al. 2001); a gene associated basallevels,aresultconsistentwiththerecessivenature with inositol metabolism, SCS2 (Kagiwada et al. 1998); a of these mutations with respect to the MIC of FLC. In geneassociatedwithcellcyclecontrol,CKA2;ageneassoci- the hybrids between O1, O2, and O3 and D2 and D3, atedwithnucleotidemetabolism,SML1;andsixuncharac- expression levels of PDR5 and SNQ2 were about equal terized genes, YMR099c, YMR102c, YPL056c, YBR147w, to those of the hybrids of D2, D3, and D4 with P1, a YGR283c,andYLR407w,resultedinaFLC-resistantphe- result consistent with the semidominant nature of the notype.Allofthesemutationsprovidecluesaboutpossi- first mutation in the evolved populations established ble mechanisms for the evolution of a FLC-resistant above for experiment 1. In contrast, in the hybrids be- phenotype. The genes YMR099c and YMR102c flank tween O1,O2, and O3 andD2 and D3,the PDR5 mes- the gene that encodes Srt1p, a protein involved in the sage was present at levels about equal to those of D2, synthesisofdolichol,afamilyoflong-chainpolyprenols D3, and D4, a result more consistent with full domi- (Sato et al. 2001). Because gene deletions can cause nance.Thereasonforthedifferenceinexpressionlevels overexpression of neighboring genes (Hughes et al. ofSNQ2andPDR5inthehybridsbetweenexperiments 2000),itispossiblethatoverexpressionofSRT1causes 1 and 2 is not known. theFLC-resistantphenotypeobservedfortheYMR099c Differential response of haploids and diploids to and YMR102c deletion strains. strong selection: Because the mutations for resistance Onthebasisofthesescreendata,theresistancemuta- in experiment 1 were semidominant, while those from tionsinO1,O2,andO3fromexperiment2wereplaced experiment2wererecessive,weexaminedtheeffectof inaMAT(cid:3)backgroundandthencrossedwiththeERG3 ploidyonthefrequencyofresistanceat128(cid:5)g/mlFLC. knockoutstrain.Fromthesecrosses,allofthe35tetrads Inanotherversionofexperiment2,equivalentnumbers tested showed no segregation for FLC resistance; all of isogenic haploid (MATa and MAT(cid:3)) and diploid spores had MICs of 256 (cid:5)g/ml FLC. The mutations in (MATa/(cid:3),MATa/a,andMAT(cid:3)/(cid:3))strainswerespread experiment2arethereforelocatedinthesamechromo- on0.5(cid:4)YPDcontaining128(cid:5)g/mlFLC.Morecolonies somalregionasERG3.LiketheO1,O2,andO3mutants, appearedamonghaploidthanamongdiploidcellssub- the ERG3 deletion strain has aMIC of 256 (cid:5)g/ml FLC jected to this kind of selection (Figure 5, Table 6). andoverexpressesERG11atgreaterthanthreefold.This Thosecoloniesthatdidappearamongdiploidcellswere effectoftheERG3knockoutontheexpressionofERG11 observed after only 2 days. Among the haploid cells a wasreportedpreviously(Hughesetal.2000).Thecom- few resistant colonies were evident by day 2, but the bined genetic linkage and phenocopy of the O1, O2, majority of colonies appeared later. We conclude that and O3 mutants to the ERG3 deletion suggest that the thehaploidshadagreaterfrequencyofmutantpheno- mutations in O1, O2, and O3 likely reside in ERG3. typesthanthediploids.Thenatureofthemutationsin Hybrids from experiments 1 and 2: Hybrid diploids these experiments has not yet been investigated. wereconstructedbetweenO1,O2,andO3andP1,D2, D3,andD4(Table5).TheMICofthehybridsofstrains DISCUSSION O1, O2, and O3 with D2, D3, and D4 was either 64 or 128 (cid:5)g/ml, similar to hybrids of D2, D3, and D4 with Ourresultsshowthatthemodeofselectionisastrong P1(Table1).Thefitnessofthesesamehybridswashigher determinant of the mechanism of drug resistance that (mean 11.7 (cid:11) 0.9) than that of diploid P5 (7.9 (cid:11) 0.6 is favored in a fungal population. The two different EvolutionofAntifungalDrugResistance 1295 TABLE5 FitnessofdiploidsconstructedfromtheprogenitorhaploidandFLC-resistanthaploids fromexperiments1and2 No.ofdoublings(cid:11)SD(n(cid:8)3replicates) MIC Diploidsa FLC TagXb TagYc Tag5(allP5) O2(cid:4)O1/P5 256 13.8(cid:11)0.4 13.8(cid:11)0.4 9.5(cid:11)0.5 O2(cid:4)O3/P5 256 13.2(cid:11)0.2 13.3(cid:11)0.2 8.7(cid:11)0.6 O1(cid:4)P1/P5 16 9.0(cid:11)0.9 (cid:10) 7.2(cid:11)0.7 O2(cid:4)P1/P5 16 10.3(cid:11)1.2 10.1(cid:11)1.1 7.2(cid:11)0.7 O3(cid:4)P1/P5 16 9.4(cid:11)0.7 9.7(cid:11)0.5 7.8(cid:11)0.9 O1(cid:4)D2/P5 64 11.2(cid:11)0.5 11.2(cid:11)0.5 7.9(cid:11)0.4 O1(cid:4)D3/P5 128 12.5(cid:11)0.8 12.2(cid:11)0.6 8.6(cid:11)0.4 O1(cid:4)D4/P5 64 10.9(cid:11)0.1 11.3(cid:11)0.1 8.0(cid:11)0.1 O2(cid:4)D2/P5 64 11.4(cid:11)0.2 (cid:10) 7.8(cid:11)0.4 O2(cid:4)D3/P5 128 12.7(cid:11)0.3 12.3(cid:11)0.1 8.6(cid:11)0.1 O2(cid:4)D4/P5 128 11.3(cid:11)0.3 11.1(cid:11)0.4 7.9(cid:11)0.4 O3(cid:4)D2/P5 128 11.2(cid:11)0.2 11.0(cid:11)0.2 7.7(cid:11)0.3 O3(cid:4)D3/P5 128 13.2(cid:11)0.1 13.2(cid:11)0.1 9.0(cid:11)0.2 O3(cid:4)D4/P5 128 10.8(cid:11)1.2 10.6(cid:11)1.2 7.3(cid:11)1.1 Mean number of doublings for all strains in 0.5(cid:4) YPD in the absence of FLC was 13.6 (cid:11) 0.2 (data not shown). aDiploids are listed with MATa strain first and the MAT(cid:3) strain second. Progenitor diploid P5 after the slash(/)wasthereferencestrainineachfitnessassay. bFitnessof eachdiploidcarryingtwo tags(e.g.,D2 (cid:4)D3)was measuredwithbothtags (e.g.,tag2 andtag 3). cMinus((cid:10))meansthatonlyonetagwaspresentinthehybriddiploid. selection regimens used in this study resulted in the that the nature of the selection applied is the main appearanceofcompletelydifferentmechanismsofresis- determinant.Whenprogenitorcellsweregrowninliq- tance.Inbothexperiments1and2,contaminationwas uidculturesathighFLCconcentrations(64or128(cid:5)g/ ruledoutbythepresenceofthemarkertagsandparallel ml),thelossoffunctioninERG3wasalwaysthemutation evolution was the rule. Under the stepwise selection favored (data not shown). Although additional loss- regimenofexperiment1,twosuccessivemutationsthat of-function mutations equivalent to the 12 other gene weresemidominantandapproximatelyadditiveintheir deletions in Figure 4 probably occurred in the experi- effect on resistance appeared independently in three mentalpopulationsunderstrongselection,thesewould different populations.In contrast,under thesingle ex- notbeexpectedtorisetohighfrequencybecausetheir posuretohighconcentrationsofFLCinexperiment2, fitnessismuchlessthanthatoftheERG3mutationsin one recessive mutation appeared independently in high FLC concentrations. three different populations. Although the population Therecruitmentoftwodivergentkindsofresistance sizes and transfer regimens differed between experi- to FLCin S. cerevisiaepopulations is reminiscentof the ment 1 and 2, additional evidence suggests that the twodivergentkindsofresistancefoundinexperimental differentoutcomesarenotsensitivetothesefactorsand populations of C. albicans (Cowen et al. 2000, 2002a), Figure 5.—Progenitor haploid (left) and its corresponding isogenic MATa/a diploid (right) after5daysofincubationon0.5(cid:4)YPDwith128 (cid:5)g/ml FLC. Inoculum for each plate consisted of 104 cells. The plate with haploid cells shows numerousresistantcolonieswhiletheplatewith diploidcellsdoesnot. 1296 J.B.Andersonetal. TABLE6 of resistance in both experiments 1 and 2, where a fitness cost might have been expected, none was de- Numberofcoloniesarisingfrom104haploidanddiploidcells spreadonmediumwith128(cid:2)g/mlFLC tected in any of the strains. If such a cost exists, it was too small to be detected under the conditions used Plates with 128 here.Thisresultissimilartothatfoundinexperimental (cid:5)g/mlFLC populationsofC.albicans,inwhichthemajorityofresis- Standard tant strains showed no significant fitness cost, and for Strains 1 2 3 Average deviation thefewthatdid,thecostwasslightandwaseliminated MATahaploid 31 49 20 33.3 14.6 with further evolution (Cowen et al. 2001). MAT(cid:3)haploid 57 60 63 60.0 3.0 Althoughnocostofresistancewasfoundwhereitwas MATa/(cid:3)diploid 8 8 1 5.7 4.0 expected, there was a strong and unexpected fitness MATa/(cid:3)diploid 1 0 0 0.3 0.6 deficit in D2, D3, and D4 at generation 100 of experi- MATa/adiploid 5 3 5 4.3 1.2 ment 1 at the two highest concentrations of FLC. This fitness cost was statistically significant and repeatable even when the progenitor and generation 100 strains buttheactualresistancemechanismsdocumentedhere were grown separately (Figure 2). In effect, the first showedbothsimilaritiesanddifferencestothosefound mutations(PDR1),whichconferresistancetothelower earlier in C. albicans. The resistance in S. cerevisiae re- concentrations of FLC, actually reduce the ability of sultinginoverexpressionoftheABCtransportersPDR5 cellstoreproduceathigherconcentrationsofthedrug, and SNQ2 in experiment 1 is very similar to the resis- relative to the progenitor (Figure 2). The presence of tance in C. albicans resulting in the overexpression of these PDR1 mutations can therefore render additional the homologous ABC transporters, CDR1 and CDR2. resistancemutations,includingtheERG3mutationsfa- In contrast, the resistance in S. cerevisiae resulting in vored in experiment 2, unavailable in high concentra- apparent loss of function in ERG3 and overexpression tionsoffluconazoleduetoinsufficientpopulationsize. of ERG11 was not accompanied by any equivalent in For example, when populations from generation 100 the experimental populations of C. albicans. Also, the ofexperiment1wereplatedonmediumwith128(cid:5)g/ml resistanceinC.albicansaccompaniedbyoverexpression FLC(exactlyasinexperiment2),populationexpansion of the major facilitator gene MDR1 had no equivalent was severely limited and accumulation of further resis- in this study, as none of the resistant mutants overex- tance mutations did not occur (data not shown). pressedthehomologFLR1.Interestingly,inS.cerevisiae, This discrepancy between MIC and fitness similar to itwastheresistancepatterninexperiment1thatshowed that observed at generation 100 of experiment 1 has a temporal succession of changes, while in C. albicans, been noted before (Cowen et al. 2001) and may be itwastheMDR1patternthatshowedatemporalsucces- due to differences in what capabilities the two assays sion. Additional mechanisms of resistance may well be measure.MICmeasurestheconcentrationatwhichthe possibleinS.cerevisiae,butdifferentselectionregimens finalcelldensityisreducedbyhalfrelativetothesame and/or smaller populations with greater replication mediumwithoutthedrug,whilethefitnessassaysmea- from those used here may be necessary to find them. surethenumberofcelldoublingsoveradefinedperiod Thetwodivergentpathwaysofresistancerecruitedin of time.The discordance in thesemeasures (i.e., when populations under different kinds of selection showed MIC is low/intermediate) suggests two ways of coping no immediate advantage when combined in hybrids. with the presence of FLC, one of which results in an TheF hybridscontainingallthreeresistancemutations increaseinMICandtheotherofwhichallowsaresidual 1 as heterozygotes had lower MIC and fitness in high numberofcelldivisionsevenatthehighestconcentra- concentrations of FLC than did either parent alone. tions of FLC. This residual growth at high concentra- Further evidence from meiotic offspring of these hy- tionsofthedrugiswellknownasthe“trailing”pheno- brids (data not shown) suggests that haploids con- typeshownbycertainstrainsofC.albicansinMICtests taining all three mutations merely show the maximum (Cowenet al.2001,2002b).Although thetrailingphe- MIC of the parents. That none of the mutations in notypeofC.albicansinFLCiseliminatedbycyclosporine experiments1and2arefullydominantwithrespectto (Marchettietal.2000)andmaybemitigatedbyalter- MIC or fitness and that different modes of selection ing pH conditions (Marr et al. 1999), the reduction favor different kinds of resistance make it unlikely for of the trailing phenotype here is due instead to the both mechanisms to predominate together in diploid pleiotropicnatureofthePDR1mutationsinexperiment populationsevolvinginthepresenceofFLCandinthe 1.Thetrailingphenotypewasnotapparentatthelater absence of genetic exchange between cell lineages. timepointsofexperiment1because,withthepresence In both experiments 1 and 2, fitness was measured ofbothmutationsconferringhighfitnessinallconcen- in addition to MIC to detect any reduction of fitness trations of FLC, this phenotype is overwhelmed. If this in the absence of the drug that might accompany the kindofdiscrepancy,thatis,low/mediumMICwithmod- evolution of resistance. Despite the rapid appearance erately high fitness in high concentrations of FLC, is
Description: