ebook img

Candida tropicalis Antifungal Cross-Resistance Is Related to Different Azole Target PDF

13 Pages·2013·1.4 MB·English
by  
Save to my drive
Quick download
Download
Most books are stored in the elastic cloud where traffic is expensive. For this reason, we have a limit on daily download.

Preview Candida tropicalis Antifungal Cross-Resistance Is Related to Different Azole Target

Candida tropicalis Antifungal Cross-Resistance Is Related to Different Azole Target (Erg11p) Modifications A.Forastiero,a*A.C.Mesa-Arango,a,bA.Alastruey-Izquierdo,aL.Alcazar-Fuoli,aL.Bernal-Martinez,aT.Pelaez,cJ.F.Lopez,d J.O.Grimalt,dA.Gomez-Lopez,aI.Cuesta,eO.Zaragoza,aE.Melladoa MycologyReferenceLaboratory,CentroNacionaldeMicrobiologia,InstitutodeSaludCarlosIII,Majadahonda,Madrid,Spaina;GroupofInvestigativeDermatology, UniversidaddeAntioquia,Medellín,Colombiab;MicrobiologyandInfectiousDiseasesDepartment,HospitalGeneralUniversitarioGregorioMarañón,andDepartmentof Medicine,FacultyofMedicine,UniversidadComplutense,Madrid,Spainc;DepartamentodeQuímicaAmbiental,InstituteofEnvironmentalAssessmentandWater Research(IDÆA),ConsejoSuperiordeInvestigacionesCientíficas,Barcelona,Spaind;BioinformaticUnit,CentroNacionaldeMicrobiologia,InstitutodeSaludCarlosIII, Majadahonda,Madrid,Spaine D CandidatropicalisranksbetweenthirdandfourthamongCandidaspeciesmostcommonlyisolatedfromclinicalspecimens. o w InvasivecandidiasisandcandidemiaaretreatedwithamphotericinBorechinocandinsasfirst-linetherapy,withextended-spec- n trumtriazolesasacceptablealternatives.Candidatropicalisisusuallysusceptibletoallantifungalagents,althoughseveralazole lo drug-resistantclinicalisolatesarebeingreported.However,C.tropicalisresistanttoamphotericinBisuncommon,andonlya a d fewstrainshavereliablydemonstratedahighlevelofresistancetothisagent.TheresistancemechanismsoperatinginC.tropi- e d calisstrainsisolatedfromclinicalsamplesshowingresistancetoazoledrugsaloneorwithamphotericinBcross-resistancewere f elucidated.Antifungaldrugresistancewasrelatedtomutationsoftheazoletarget(Erg11p)withorwithoutalterationsofthe ro m ergosterolbiosynthesispathway.Theantifungaldrugresistanceshowninvitrocorrelatedverywellwiththeresultsobtainedin vivousingthemodelhostGalleriamellonella.Usingthispanelofstrains,theG.mellonellamodelsystemwasvalidatedasasim- h t ple,nonmammalianminihostmodelthatcanbeusedtostudyinvitro-invivocorrelationofantifungalsinC.tropicalis.The tp : developmentinC.tropicalisofantifungaldrugresistancewithdifferentmechanismsduringantifungaltreatmenthaspotential // a clinicalimpactanddeservesspecificprospectivestudies. a c . a s m Inrecentyears,anepidemiologicalchangeamongfungalinfec- expressionseemstoplayaroleinresistance.(ii)Theupregulation . o tionshasbeenobserved.AlthoughCandidaalbicansisstillthe of multidrug efflux transporters (ABC [ATP-binding cassette]/ r g mostcommonspeciesisolatedfrompatientsinfectedwithpatho- MFS[majorfacilitatorsuperfamily])leadstodecreaseddrugcon- / o genicyeasts,thefrequencyofisolationofnon-albicansCandida centrationswithinthefungalcell.(iii)Abypassintheergosterol n species has increased (1, 2). The frequency of isolates of non- biosynthesispathwaymaydevelopasaresultofmutationsinste- J albicansCandidaspeciesdifferswithgeographicallocation.Can- rol(cid:2)5,6-desaturase(ERG3)(18–20).InC.albicans,themostfre- an didatropicalisranksbetweenthirdandfourthamongthespecies quentazoleresistancemechanismsreportedareincreasedefflux u a mostcommonlyisolated(3,4),althoughitisconsideredthemost ofthedrug(85%)andmutationsintheERG11sequence(65%) r y frequentlyisolatedspeciesintheAsia-Pacificregion(5–7)andthe (21).Infact,numerousErg11pmutationshavebeendescribed, 3 secondmostfrequentlyisolatedspeciesinBrazilandLatinAmer- butonlyafewhavebeendirectlyassociatedwithazoleresistance 1 , ica(20.9%and13.2%,respectively)(1,8).Moreover,C.tropicalis (22).Also,theroleofaminoacidsubstitutionshasbeenelucidated 2 isalmostalwaysassociatedwiththedevelopmentoffungalinfec- 0 byheterologousexpressionoftheERG11gene(23,24).Inaddi- 1 tions,anditisoftenrelatedtohighermortalitythanothernon- 9 tion,increasingevidencepointstoarelationshipbetweenamino albicansCandidaspeciesandC.albicans,particularlyinneutro- b acidsubstitutionsandchangesinproteinconformationinvolved y penicandoncologypatients(2,9,10). inazoleresistance(25,26).ForC.tropicalis,overexpressionofC. g Invasivecandidiasisandcandidemiaaretreatedwithampho- u tropicalisERG11(CtERG11)associatedwithmissensemutations e tericinBoranechinocandinasfirst-linetherapy,withextended- s (Y132ForS154F)hasbeendescribedastheazoleresistancemech- t spectrum triazoles as acceptable alternatives (11, 12). Candida anisminclinicalisolates(14,27).Additionally,invitrodevelop- tropicalisisusuallysusceptibletoallantifungalagents,although severalcasesoffluconazoleandazolecross-resistanceinclinical isolateshavebeenreported(5,13,14),especiallyintheAsia-Pa- Received8March2013 Returnedformodification16April2013 cificregion(1,15).Ontheotherhand,C.tropicalisresistanceto Accepted12July2013 amphotericinBisrare,andveryfewstrainshavereliablydemon- Publishedaheadofprint22July2013 stratedhigh-levelresistancetothisagent(16,17). AddresscorrespondencetoEmiliaMellado,[email protected]. Limitedspecificinformationisavailableaboutthemolecular *Presentaddress:A.Forastiero,LaboratoriodeMicología,HospitalBritánico, mechanismsofresistancetoazolesinC.tropicalis,althoughthey CiudadAutonomadeBuenosAires,Argentina. arelikelytobesimilartothoseidentifiedinotherCandidaspecies. Supplementalmaterialforthisarticlemaybefoundathttp://dx.doi.org/10.1128 Todate,Candidasp.azoleresistancehasbeenlinkedtodifferent /AAC.00477-13. molecular mechanisms. (i) Mutations in the azole target, 14(cid:1)- Copyright©2013,AmericanSocietyforMicrobiology.AllRightsReserved. steroldemethylase(Erg11p),resultinareductionorlossofaffin- doi:10.1128/AAC.00477-13 itywithazolesoranincapacitytobindazoles.Also,ERG11over- October2013 Volume57 Number10 AntimicrobialAgentsandChemotherapy p.4769–4781 aac.asm.org 4769 Forastieroetal. ment of fluconazole resistance in C. tropicalis linked to the up- Testing(AFST-EUCAST)(39).ForamphotericinB,thebreakpointswere regulation of two multidrug efflux transporter genes, CtMDR1 definedbasedonthedistributionofMICsforwild-typestrains,deter- andCtCDR1,hasbeenreported(28). mined by a EUCAST method (susceptibility breakpoint, (cid:1)1.0 (cid:3)g/ml) Ingeneral,adequatemeasurementofinvitroantifungalactiv- (40).Forfluconazole,voriconazole,andanidulafungin,breakpointspro- ityintermsofMICsisconsideredofclinicalrelevanceandoften posedbytheAFST-EUCASTwereusedtointerpretsusceptibilityresults (41). givesusefultherapeuticinformationforpatientmanagement.The Metabolicinhibitorsusceptibilityassays.Thesusceptibilitiesofthe correlationbetweeninvitroandinvivodatacanbeexploredusing strainstodifferentclassesofmetabolicinhibitorsweredeterminedbya differentexperimentalmodels.Amousemodelofsystemiccandi- visualspottingassayonYNBmediumwithdifferentdrugconcentrations diasis has been used to evaluate the effects of azole resistance (42).StrainsweregrowninSabouraudliquidmediumasdescribedabove, mechanismsontheefficacyofazoletreatment(29,30).Murine andcellconcentrationswereadjustedforastartingsolutioncontaining modelsarestillconsideredthegoldstandardforthemeasurement 105cellsperml.Then10-foldserialdilutionswerecarriedout,and5(cid:3)lof ofinvivodrugefficacybuthaveasamaindisadvantagethatalarge eachdilutionwasspottedontoagarplatescontainingeither2(cid:3)g/mlam- number of animals are needed to determine the dose-response photericinB,10(cid:3)g/mlbrefeldinA(Sigma-AldrichQuimicaSA,Madrid, relationshipsoftheantifungal.Increasingconcernabouttheuse Spain), 200 (cid:3)g/ml hygromycin B (Sigma-Aldrich Quimica SA), or 2 D of mammals, i.e., the cost, the number used, and ethical issues (cid:3)g/mlterbinafine(Novartis,Basel,Switzerland).Drug-freeYNBmedium o relatedtothepainoftheanimalsduringtheprocedures,haspro- wasincludedasacontrol.Theplateswereincubatedfor48hat37°C w n beforereading. motedtheuseofalternative,nonmammalianinvertebratemodels lo for testing microbial virulence and for screening the efficacy of Amplificationandsequencingofergosterolbiosynthesisgenes.The a ERG11andERG3geneswerePCRamplifiedandsequencedusingapanel d antimicrobial agents. Among these, Galleria mellonella presents e of oligonucleotide primers (synthesized by Sigma Genosys, Madrid, d certainbenefitsincomparisonwithothernonmammalianmodels Spain)(seeTableS1inthesupplementalmaterial).GenomicDNAwas f ofinfection(31).PreviousstudieshavedocumentedthatG.mel- extracted using the DNeasy Plant minikit (Qiagen Inc., Valencia, CA, ro lonellaisavalidmodelfortheevaluationofmicrobialvirulence USA)andwasusedasthetemplateforPCRamplification.Amplification m and/ortheefficaciesofdifferentantimicrobialagentsagainstin- conditionsconsistedof5minofdenaturationat94°C,followedby30 h t fectionscausedbydifferentbacteriaandfungi(32–38). cyclesof30sat94°Cfordenaturation,40sat50°Cforannealing,and50 tp Inthiswork,weinvestigatedtheresistancemechanismsoper- sat72°Cforelongation,andbyafinalelongationstepof10minat72°C. :/ / atinginC.tropicalisstrainsshowingazolecross-resistancealone PCRproductswerepurifiedwiththeHighPurePCRproductpurification a a orcombinedwithamphotericinBresistance.Wealsotestedthein kit(RocheDiagnostics,Madrid,Spain)accordingtothemanufacturer’s c vivoantifungalresponsesofC.tropicalisstrains,bothsusceptible recommendations. PCR products were semiquantified by agarose gel .a electrophoresisandwereusedasthetemplatesforsequencing,whichwas s andresistanttoamphotericinB,fluconazole,andvoriconazole,in m performed with the Dye Terminator cycle sequencing quick-start kit themodelhostG.mellonellainordertocorrelatethesusceptibility .o (BeckmanCoulterInc.,Fullerton,CA).ThesequencingPCRproducts phenotypesofthosestrainswiththoseshowninvitro. werethenpurifiedthroughG50Sepharosecolumns(AmershamBiosci- rg / ences)and,finally,weresequencedonaCEQ8000DNAanalysissystem o MATERIALSANDMETHODS (BeckmanCoulter)withtheforwardandreverseprimersusedforPCR n J Strains,media,andgrowthconditions.FourCandidatropicalisstrains amplification(seeTableS1inthesupplementalmaterial). a Theaminoacidsequencesofthegenesencodingputative14(cid:1)-sterol n were used in this study: (i) two ATCC strains, ATCC 750 and ATCC 200956,and(ii)twoclinicalisolates(CL-6835andTP-13650)recovered demethylase(ERG11)andsterol(cid:2)5,6-desaturase(ERG3)werededuced ua fromtwopatientswithcandidemiawhohadbeentreatedwithantifungal fromnucleotidesequencesandwereanalyzedusingtheMegAlignsoft- ry drugspreviously(seeTable1).Theseisolateswereidentifiedbymorpho- warepackage(DNAStar,Inc.,Lasergene,Madison,WI,USA).Theamino 3 logicalfeaturesandwereconfirmedbyDNAinternaltranscribedspacer acidalignmentswerederivedbyCLUSTALanalysis(43). 1 , (ITS)sequencing.YeastcellsweregrownovernightinSabouraudliquid Sterolanalysis.SterolswereextractedfromC.tropicalisafter18hof 2 medium(Becton,DickinsonandCompany,MD,USA)at30°Cwithcon- growthinSabouraudliquidmediumat37°Cwithshaking(150rpm), 0 1 stantshaking(150rpm).Cellswereharvestedbycentrifugationandwere usingtheprotocoldescribedpreviously(44,45).Briefly,cellswerehar- 9 washedtwicewithsteriledistilledwater.Inoculumsizeswereadjusted vested by centrifugation, washed with sterile water, lyophilized, and b usingahemocytometerchamber.Yeastnitrogenbase(YNB)solidme- weighedtodeterminethedryweight(ingrams)ofthefungus.About150 y diumwasusedinthemetabolicinhibitorassays. to200mgwasusedfortotal-sterolextraction.A3-mlvolumeofa25% gu Antifungaldrugs.AmphotericinB(AMB)(Sigma-AldrichQuimica, alcoholic potassium hydroxide solution (methanol-ethanol, 3:2) was e Madrid,Spain),fluconazole(FLC)(PfizerSA,Madrid,Spain),voricona- addedtothedriedpelletsandwasmixedbyvortexingfor1min.The st zole(VRC)(PfizerSA),andanidulafungin(ANF)(Merck&Company, mixturewasincubatedat85°Cinawaterbathfor1h.Neutrallipidswere Inc.,NJ,USA)wereusedforthedeterminationofMICsandforassess- extractedtwicebyadditionofamixtureof1mlofwaterand1.5mlof mentoftheirefficaciesagainstC.tropicalisinfectionintheGalleriamel- hexane,followedbyvigorousvortexmixingfor3min.Theupperhexane lonellamodel.Theantifungaldoseswerecalculatedbytakingintoaccount layer was transferred to a clean glass tube, evaporated, and stored at thetherapeuticdoseusedforhumansandtheMICvaluesforthestrains (cid:4)20°Cuntilanalysis.Onehundredmicrogramsofandrostanol(Sigma) studied(11).Stocksolutionsofvoriconazole,fluconazole,amphotericin wasaddedtothesamplesasaninternalstandardforsterolrecoveryand B,andanidulafunginwerefirstpreparedindimethylsulfoxide(DMSO), quantification.Sterolswereconvertedintotheirtrimethylsilylethersby andthendilutionsinwaterforinjectionweremadetoobtainthefollow- reactionwitha1:1solutionofN,O-bis-(trimethylsilyl)trifluoroacetamide ingconcentrations:400and80(cid:3)g/mlofvoriconazole;800,360,and80 (85°C,60min)intoluene.NeutrallipidfractionswereanalyzedonaTrace (cid:3)g/ml of fluconazole; 120 and 80 (cid:3)g/ml of amphotericin B; and 400 gas chromatograph coupled to a quadrupole mass analyzer (Thermo (cid:3)g/mlofanidulafungin. FisherScientific,Manchester,UnitedKingdom).TheGCprogramand Antifungalsusceptibilitytesting.Antifungalsusceptibilitywasdeter- MSoperationconditionswerethesameasthosedescribedpreviously(45, minedbybrothmicrodilution(BMD).MICsweredeterminedbyfollow- 46).Sterolidentificationwasbasedoninterpretationandcomparisonof ingtherecommendationsoftheAntifungalSusceptibilityTestingSub- massspectraandretentiontimedatawithpreviousdata(45,47).The committee of the European Committee on Antibiotic Susceptibility amountofeachsterol(inmicrograms)relativetothedryweight(in 4770 aac.asm.org AntimicrobialAgentsandChemotherapy ResistanceofCandidatropicalistoAntifungalDrugs grams)ofthefunguswascalculatedfromthreebiologicalreplicates referencesequence(GenBankaccessionnumberM23673).Thefullnu- analyzed in triplicate byGC-MS.Therelativesterolcompositionwas cleotidesequencesoftheERG11genesofATCC750,ATCC200956,and expressed as the ratios of individual sterol concentrations to the total thetwoC.tropicalisclinicalisolates,CL-6835andTP-13650,andtheir sterolcontent.Thesignificanceofdifferencesinsterolcomposition(P(cid:5) deducedaminoacidsequenceshavebeendepositedintheGenBankda- 0.0001)wasdeterminedbyanonparametricMann-Whitneytest. tabaseunderaccessionnumbersKC542323,KC542324,KC542325,and HomologymodelingofErg11p.Homologymodelsofwild-typeC. KC542326,respectively.ThefullnucleotidesequenceoftheERG3gene tropicalisErg11pandmutatedsequencescontainingaminoacidsubstitu- (encoding sterol (cid:2)5,6-desaturase) of C. tropicalis MYA-3404 was used tions G464D, G464S, Y132H, and Y132F were constructed using the as the reference sequence for ERG3 (GenBank accession number SWISS-MODEL platform (http://swissmodel.expasy.org ) 4(8–50). The XM002550136).ThefullnucleotidesequencesoftheERG3genesofthe crystalstructureofhumanlanosterol14(cid:1)-demethylase(HS-CYP51)in fourstrainsusedinthisstudy(ATCC750,ATCC200956,CL-6835,and complexwithketoconazole,depositedintheProteinDataBank(http: TP-13650)havebeendepositedintheGenBankdatabaseunderaccession //www.rcsb.org) with code 3LD6 (51), was used as the template. The numbersKC542319,KC542320,KC542321,andKC542322,respectively. amino acid sequence identity between this template and C. tropicalis Erg11pwas39%.TheN-terminalregionofC.tropicalisErg11p(residues RESULTS 1 to 50), defined to comprise the membrane-spanning domains, and Antifungalsusceptibilitytesting.TheMICvaluesoffluconazole D thereforewithoutcounterpartsinthetemplateprotein,wasdeletedprior (FLC),voriconazole(VRC),amphotericinB(AMB),andanidu- o toproteinmodeling. lafungin (ANF) are shown in Table 1. Candida tropicalis strain w n AnalysisofR6Guptakeandefflux.Theactivityofeffluxpumpsin ATCC 750 was susceptible to all the antifungals tested. Isolate lo azole- and/or amphotericin B-susceptible and -resistant C. tropicalis TP-13650wasresistanttofluconazoleandvoriconazolebutsus- a strains(ATCC200956,ATCC750,CL-6835,andTP-13650)wasevalu- d ceptible to amphotericin B and anidulafungin. Resistance to all e ated using rhodamine 6G (R6G) according to the protocol previously antifungal drugs except anidulafungin was observed for strains d describedbyVandeputteetal.in2005(14)andbyMaesakietal.in1999 f ATCC200956andCL-6835. r (52).AccumulationofR6GwasmeasuredbyflowcytometryintheFL2 o channel(collectingtheemissionsignalat585(cid:6)42nm).Atotalof10,000 Metabolicinhibitorsusceptibilityassays.Thesusceptibility m cells were recorded using a FACSCalibur cytometer (FL2 channel; ofC.tropicalisstrainstodifferentmetabolicinhibitorswasusedas h CellQuest;BDBiosciences),andthefluorescenceintensityofR6G-stained a screening method to identify those yeasts with alterations in ttp cellswasdetermined.Unstainedcellswereincludedasanegativecontrol membrane fluidity. Candida tropicalis strains with cross-resis- :/ / inallcases.DatawereprocessedusingFlowJosoftware(version7.1.6; tancetoazoledrugsandamphotericinBweremoresusceptibleto a a TreeStarInc.,Ashland,OR,USA).Experimentswereperformedatleast metabolicinhibitors(brefeldinAorhygromycinB)andtoterbi- c twicewitheachstrainondifferentdays.Resultswereexpressedasmean nafinethanthewild-typestrainorthestrainwithazoledrugre- .a fluorescenceintensities(MIF)andstandarddeviations(SD). sistanceonly.Changesinmembranefluiditywereclearlydetect- sm Evaluation of antifungal efficacy in Galleria mellonella infected ablewhenbrefeldinAwasused(seeFig.S2inthesupplemental . o withC.tropicalis.Galleriamellonellacaterpillarsinthefinalinstar(stage oflarvaldevelopment)(R.J.MousLiveBait,theNetherlands)wereused. material).TheseresultssuggestthatamphotericinB-resistantiso- rg Larvaewerestoredinthedarkat25°Cpriortouse.Tencaterpillars(body lates may have an alteration of the membrane that increases its o/ weight, 400 (cid:6) 100 mg) were employed per group. Candida tropicalis permeability,andtherefore,thesecompoundscouldpassthrough n strainsweregrowninliquidSabouraudmediumasdescribedabove.Pre- themembranerelativelyeasily,resultinginthehypersusceptible Ja liminary experiments were performed with all the strains to define an phenotypeobservedinvitro. n appropriate range of inoculum concentrations that would cause 80 to ERG11 and ERG3 gene sequence analyses. (i) ERG11 gene ua 100%mortalityat72to96hpostinfection.Theinoculaweredetermined sequence. No changes were observed between the ERG11 se- r y bycountinginahemocytometerchamberandwereadjustedinphos- quenceofstrainATCC750determinedhereandthatdepositedin 3 phate-bufferedsaline(PBS)to2(cid:7)108cellspermlforATCC750,CL- GenBank(accessionnumberM23673).ComparisonoftheERG11 1 6855,andTP-13650andto4(cid:7)108cellspermlforATCC200956.Viable gene sequence of strain ATCC 200956 with the available corre- , 2 countswereperformedtoconfirmthecorrectsizeoftheinocula. 0 A10-(cid:3)lHamiltonsyringe(FisherScientific,Madrid,Spain)wasused sbppotnhdaitncgosrerqesupeonncdesfotorstthreaianbAseTnCceCo7f5404raemveianleodaaciddeslewtiitohninofth1e32I 19 toinfectthelarvaebyinjecting10(cid:3)loftheinoculumsuspensionintothe helix((cid:2)276–319)oftheprotein(Table1;seealsoFig.S1Binthe b hemocoelofeachG.mellonellathroughthelastleftproleg.Within2hof y the infection, 10 (cid:3)l of an antifungal solution (either 400 or 80 (cid:3)g/ml supplementalmaterial)andanaminoacidsubstitution(D275V) g voriconazole,800,360,or80(cid:3)g/mloffluconazole,120or80(cid:3)g/mlof with restoration of the open reading frame. PCR amplification ue amphotericinB,or400(cid:3)g/mlofanidulafungin)wasinjectedintoadif- usingtwoprimercombinationswithoneoftheoligonucleotides s t ferentprolegusingthesametechnique.Afterantifungaltreatment,cater- designedwithinthedeletedsequenceclearlyshowedthatthede- pillarswereincubatedat37°Cfor7days.Larvasurvivalwasmonitored letionhadoccurredinthetwoalleles(seeFig.S4inthesupple- daily, and larvae were considered dead when they did not respond to mentalmaterial).Therefore,strainATCC200956washomozy- physical pressure. The following control groups were included: un- gous for the ERG11 gene deletion. Analysis of the ERG11 gene touchedlarvae,piercedlarvae,larvaereceivingPBS,andlarvaegiventoxic sequencesofstrainsCL-6835andTP-13650revealeduniquemis- dosesofantifungals(400(cid:3)g/mlofvoriconazole,800(cid:3)g/mlofflucona- sensemutationsforeachstraincorrespondingtoaminoacidsub- zole,120(cid:3)g/mlofamphotericinB,or400(cid:3)g/mlofanidulafungin).Each stitutionsG464DandY132F,respectively(seeFig.S1CandAin experiment was performed at least three times. and the results are re- thesupplementalmaterial).Silentmutationswerealsoobserved portedasmeanvalues.Tosimplifysomeofthefigures,thePBScontrol inthesestrains(Table1). groupresultsarenotpresented. (ii)ERG3genesequence.Thefullnucleotidesequenceofthe Thesurvivalrateofcaterpillarswasplottedagainsttime,andPvalues ERG3 gene (encoding sterol (cid:2)5,6-desaturase) of C. tropicalis werecalculatedbythelogrank(Mantel-Cox)testusingGraphPadPrism, version5.Pvaluesof(cid:5)0.01wereconsideredsignificant. MYA-3404wasusedasareferencesequenceforERG3(GenBank Nucleotide sequence accession numbers. The full nucleotide se- accessionnumberXM002550136).SequencingoftheERG3gene quence of the ERG11 gene of C. tropicalis ATCC 750 was used as the showed single missense mutations, C773T and A334G, in the October2013 Volume57 Number10 aac.asm.org 4771 Forastieroetal. TABLE1MutationsinERG11andERG3genes,correspondingaminoacidsubstitutions,andMICsofantifungaldrugsforCandidatropicalis isolates Genesequenceanalysisb 14(cid:1)-Steroldemethylase Sterol(cid:2)5,6-desaturase MIC((cid:3)g/ml)c Isolate Previoustreatmenta ERG11(bp) Erg11p(aa) ERG3(bp) Erg3p(aa) AMB FLC VRC ANF ATCC750 Unknown G366A R122R 0.25 0.5 0.03 0.03 ATCC200956 Unknown (cid:2)132bp (cid:1)44aa C773T S258F 2 (cid:9)64 (cid:9)8 0.03 —d D275V G1362A K454K C1534G P511A T1554C I517I CL-6835(clinical) FLC,AMB,andCAS T906C I302I A334G S113G 2 (cid:9)64 (cid:9)8 0.03 G1391A G464D G366A R122R T637C L212L D TP-13650(clinical) FLC,VRC,andTRB T225C C75C 0.25 (cid:9)64 (cid:9)8 0.03 o w G264A L88L n A395T Y132F lo T783C I261I a d G1362A K454K e T1554C I517I d f aCAS,caspofungin;TRB,terbinafine. ro bSubstitutionsinaminoacidsequencesareshowninboldface. m cDeterminedbybrothmicrodilutionaccordingtotheestablishedmethodofAFST-EUCAST.AMB,amphotericinB;FLC,fluconazole;VRC,voriconazole;ANF,anidulafungin. EUCASTantifungalclinicalbreakpointsforC.tropicalisareasfollows:forAmB,(cid:1)1(cid:3)g/mlforsusceptibilityand(cid:9)1(cid:3)g/mlforresistance;forFLC,(cid:1)2(cid:3)g/mlforsusceptibilityand ht (cid:9)4(cid:3)g/mlforresistance;forVRC,(cid:1)0.12(cid:3)g/mlforsusceptibilityand(cid:9)0.12(cid:3)g/mlforresistance;andforANF,(cid:1)0.06(cid:3)g/mlforsusceptibilityand(cid:9)0.06(cid:3)g/mlforresistance. tp d—,132-bpdeletion,betweennucleotides824and945,resultsina44-amino-aciddeletion,betweenaa276and319.Thechangeintheopenreadingframeisthecauseofthe :/ / D275Vaminoacidsubstitution. a a c . a s ERG3sequencesofATCC200956andCL-6835,respectively,re- cellsbythesametypesoftransportersasthoseusedforazolesin m sultingintheS258FandS113Gaminoacidsubstitutions.Inad- yeasts(52).TherewerenoremarkabledifferencesamongtheR6G . o dition,CL-6835presentedtwopointmutationsintheERG3gene uptake/efflux ratios of the different C. tropicalis strains, despite rg (T637CandG366A)thatdidnotaffecttheaminoacidsequence. theirdifferentantifungaldrugsusceptibilityprofiles(seeFig.S5in / o NeitherATCC750norisolateTP-13650presentedanymissense the supplemental material). Levels of retention of R6G, after n mutationsintheentireERG3genesequence(Table1). washingwithPBSandincubationinfreshYPDmedium(15min), J a Sterol analysis. To investigate the contribution of mem- were similar for all strains. The percentages of retention of the n branesterolcompositiontoantifungaldrugresistance,theste- totalR6Gquantifiedinitiallyrangedfrom58to83%.Somediffer- u a rolprofilesofsusceptibleandresistantstrainswereanalyzedby encesintheinitialmeanfluorescenceintensities(MIF)wereob- r y gas chromatography (GC)–mass spectrometry (MS). Sterols served.ForstrainsATCC750andTP-13650,bothsusceptibleto 3 wereclassifiedaccordingtotheirGCretentiontimesandmass amphotericinB,theMIFwere239(cid:6)28and365(cid:6)196,respec- 1 , spectra,andatotalof22sterolsweredetected(Table2).The tively.Incontrast,fortheamphotericinB-resistantstrainsATCC 2 0 relative sterol composition for each strain is given in Table 2 200956andCL-6835,theintensitiesoffluorescencewerehigher 1 based on the retention times of the sterols. Ergosterol was (1,881(cid:6)28and949(cid:6)4,respectively). 9 b found to be the major sterol in strains ATCC 750 and TP- HomologymodelingofErg11p.Thecrystalstructureofhu- y 13650.Incontrast,noergosterolwasdetectedineitherATCC manlanosterol14(cid:1)-demethylase(HS-CYP51)wasusedtogener- g 200956orCL-6835.Thisabsenceofergosterolsuggestedpos- atea3-dimensional(3D)homologymodeloftheaminoacidse- u e siblealterationsinthesterolcontentsofthesetwostrains.We quenceofwild-typeC.tropicalisErg11p.Thehemecofactorwas s t then grouped these metabolites according to the presence of extractedfromtheHS-CYP51structureandwasmergedintothe themethylgrouponcarbonatom14.FigureS3inthesupple- modelsobtained.Alignmentofthetwosequencesrevealedthat mentalmaterialshowsstatisticallysignificantaccumulationsof there were no insertions or deletions in any of the secondary- C-14-methylated sterols in ATCC 200956 and CL-6835 that structureregionsoftheprotein.Thefungus-specificinsertionin werenotobservedinstrainsATCC750andTP-13650.C-14- theC-terminalpartofthesequencewasmodeledasaloopandwas methylated sterols in ATCC 200956 and CL-6835 comprised supportedbytheANOLEAprediction(53),whichreportedneg- 14-methyl fecosterol, 14-methylcholesta-8,24-dien-3(cid:8)-ol, ativeenergyvaluesformostoftheresiduesofthissection.More- 4,14-dimethylcholesta-8,24-dien-3(cid:8)-ol,lanosterol,andeburi- over,noneoftheresiduesstudiedinthispaperwerelocatedwithin col. thisregionofthemodel;rather,theywerelocatedinconserved AnalysisofR6Guptakeandefflux.Rhodamine6G(R6G)isa regions that have been intensively studied, as discussed below. fluorescent substrate of the multidrug transporter superfamily Therefore,modelsforErg11pwiththeG464SorY132Fsubstitu- (conferring multidrug resistance [MDR]) that has already been tion were generated using the same approach. Modeling of the usedtodemonstratetheactivityoftheseproteinsinazole-resis- partiallydeletedErg11pfromstrainATCC200956wasnotpossi- tantisolatesofCandidaalbicans.Infact,thisdyeisexportedfrom ble because of the lack of sufficient identity with the available 4772 aac.asm.org AntimicrobialAgentsandChemotherapy ResistanceofCandidatropicalistoAntifungalDrugs TABLE2SterolcompositionsofCandidatropicalisstrains Sterol %oftotalsterolsainstrain: Compound no. Systematicnameb Commonname ATCC750 ATCC200956 CL-6835 TP-13650 1 Squalene 8.54 0.17 1.45 ND 2 24-Methylcholesta-5,7,9(11),22-tetraen-3(cid:8)-ol 0.15 ND ND ND 3 S28:3 0.08 ND ND ND 4 Cholesta-8,24-dien-3(cid:8)-ol Zymosterol 8.97 ND ND 20.60 5 14-Methylcholesta-8,24-dien-3(cid:8)-ol ND 0.11 0.07 ND 6 24-Methylcholesta-5,7,22-trien-3(cid:2)-ol Ergosterol 64.96 ND ND 60.84 7 24-Methylcholesta-7,22-dien-3(cid:8)-ol 0.20 ND ND 0.51 8 24-Methylcholesta-5,7,22,24(28)-tetraen-3(cid:8)-ol 0.81 ND ND 5.20 9 14,24-Dimethylcholesta-8,24(28)-dien-3(cid:8)-ol 14-Methylfecosterol ND 61.08 52.67 ND 10 24-Methylcholesta-8,24(28)-dien-3(cid:8)-ol Fecosterol 1.91 ND ND 2.31 11 4,14-Dimethylcholesta-8,24-dien-3(cid:8)-ol 0.12 1.64 0.63 ND D 12 4(cid:1)-S28:2 ND ND ND 0.21 o 13 24-Methylcholesta-5,7,24(28)-trien-3(cid:8)-ol 2.56 ND ND 3.43 w 14 24-Methylcholesta-5,7-dien-3(cid:8)-ol 4.11 ND ND 2.23 n 15 24-Methylcholesta-7,24(28)-dien-3(cid:8)-ol Episterol 4.45 ND ND 2.06 lo a 16 24-Methylcholest-7-en-3(cid:8)-ol ND ND ND 0.03 d 17 4(cid:1),4(cid:8),14-Trimethylcholesta-8,24-dien-3(cid:8)-ol Lanosterol 2.32 25.80 27.75 1.83 e d 18 4(cid:1),24-Dimethylcholesta-8,24(28)-dien-3(cid:8)-ol 0.11 ND ND ND f 19 4(cid:1),4(cid:8)-Dimethylcholesta-8,24(28)-dien-3(cid:8)-ol 0.53 ND ND 0.65 ro 20 4(cid:1),4(cid:8),14,24-Tetramethylcholesta-8,24(28)-dien-3(cid:8)-ol Eburicol 0.15 8.68 17.42 0.10 m 21 4(cid:1),4(cid:8),24-Trimethylcholesta-8,24(28)-dien-3(cid:8)-ol 0.03 ND ND ND h 22 Unidentified ND 2.51 ND ND tt p aND,notdetected. :/ / bS28:3and4(cid:1)-S28:2refertosterolswith28carbonatomsand3or2doublebonds,respectively. a a c . a crystal structures of proteins that could be useful as templates. acid substitutions affected the heme environment in different s m Therefore,thismodelisnotshownbutisdiscussedbelow.Figure ways.Thewild-typestrainshowsatyrosineatposition132(Fig. . 1Ashowsthepositionsofthemutationsanalyzed,bothcloseto 1B),whichinteractsdirectlywiththehemethroughthehydroxyl o r the heme group. The different models revealed that the amino group. Two mutations were analyzed for this site, Y132F and g / o n J a n u a r y 3 1 , 2 0 1 9 b y g u e s t FIG1PositionsofthepointmutationsofCandidatropicalisErg11panalyzedaccordingtothehomologymodelsgeneratedinthisstudy.(A,B,andD)Wild-type CandidatropicalisErg11p(Y132andG464).(C,EandF)Positionsofthemutationsanalyzedintheprotein:Y132F,G464D,andG464S,respectively.Theresidues studiedareshowninstickrepresentation(green,carbonatoms;red,oxygenatoms;blue,nitrogenatoms).Hemeisrepresentedinred.Thedistancebetweenthe hemeandtheclosestatomoftheaminoacidstudiedisindicatedbyadashedgreenlineandanumbergivingthedistanceindaltons.Thecrystalstructureof human lanosterol 14(cid:1)-demethylase (HS-CYP51) in complex with ketoconazole (Protein Data Bank code 3LD6) was used as a template. Y, tyrosine; F, phenylalanine;G,glycine;D,asparticacid;S,serine. October2013 Volume57 Number10 aac.asm.org 4773 Forastieroetal. Y132H,butonlytheformer(theonefoundinthiswork)isshown. nisms responsible for azole drug resistance in Candida species Interactionoftheproteinwiththehemegroupwouldbeavoided havebeenextensivelystudiedandreviewed(18–20,55).InCan- because of the absence in the F residue of the hydroxyl group didaalbicans,numerousERG11mutationshavebeendescribed, presentintheYresidue,positionedinthe3-Åenvironmentofthe but only a few of them have been associated with amino acid hemegroup(Fig.1C).Anothertwomutationswereanalyzedfor substitutionsandazoleresistance(22).Themostfrequentlyre- theaminoacidatposition464.IncontrastwithY132,glycineat portedmodificationsarethesubstitutionsY132H,D278E,S405F, position464(Fig.1D)isclosetothehemegroupbutnotwithin G464S,andR467K(14,22).However,littleinformationisavail- the3-Åenvironmentandprobablydoesnotinteractwithit.In ableconcerningthemolecularmechanismsleadingtoazoleresis- contrast,thesubstitutionG464SorG464Dmightaffecttheenvi- tanceinthepathogenicyeastC.tropicalis.Acquiredresistanceto ronment of the heme group, changing the interaction, due not azoledrugsinclinicalisolatesofC.tropicalisduetooverexpres- onlytothenucleophiliccharacterofserine(Fig.1F)ortheacidic sion of a CtERG11 gene associated with missense mutations characterofasparticacid(Fig.1E)butalsotothefactthattheside (Y132FandS145F)hasbeenreported(13,26). chainsofthesetwoaminoacidsarelargerthanthatofglycine. Inthisstudy,weintensivelyinvestigatedthemolecularbases Galleria mellonella in vivo infection model and antifungal forC.tropicalisstrainspresentingapatternofresistancetoazole D response.Galleriamellonellalarvaewereinfectedwithaninocu- drugs,aloneorincombinationwithresistancetoamphotericinB. o lumof2(cid:7)106CFU/larvawhenstrainATCC750,CL-6835,or Cross-resistancetoazolesandamphotericinBdrugswasobserved w n TP-13650wasused.Aninoculumof4(cid:7)106CFU/larvawasused in two of the C. tropicalis strains analyzed (ATCC 200956 and lo forATCC200956,sincethisstrainshowsreducedvirulenceinthis CL-6835),whilestrainTP-13650wasresistantonlytoazoledrugs. a d model(37).Inallcases,80%ofinfectedanduntreatedlarvaedied We also confirmed this in vitro phenotype in vivo by using an e betweenthe3rdand4thdaypostinfection.Foreachstraininocu- alternativeminihostmodelofcandidiasis. d f lated,thesurvivalofthelarvaewasdependentonboththeanti- Enhanced drug efflux activation is one of the major mecha- r o fungaldoseadministeredandtheinvitroMIC(Fig.2to4). nismscausingazoleresistanceinyeast,althoughitcouldbeac- m Theresultsshowedthatattherapeuticdosesofazoledrugs(10 companiedbyotherresistancemechanisms.Thismechanismisa h mg of voriconazole/kg of body weight/day, equivalent to 4 (cid:3)g/ consequenceoftranscriptionalactivationofgenesencodingdrug tt p larva,and9mgoffluconazole/kgofbodyweight/day,equivalent effluxpumpproteinsbelongingtotheABC(encodedbyCandida :/ to 3.6 (cid:3)g/larva), larvae infected with azole-susceptible strain drugresistance[CDR]genes)andmajorfacilitator(encodedby /a a ATCC750showedsignificantlyimprovedsurvivalrelativetothat MDRgenes)superfamiliesoftransporters(20).Inordertosearch c . of their untreated control groups (P, 0.0033 for both VRC and forconcomitantresistancemechanismsrelatedtoincreasedefflux a s FLC)(Fig.2).Incontrast,whenG.mellonellalarvaewereinfected pump activity, we tested this possibility using rhodamine 6G m withanazole-resistantstrain(ATCC200956,TP-13650,orCL- (R6G).Thiscompoundisafluorescentsubstrateofthemultidrug . o 6835)andtreatedwithatherapeutic(10mgofvoriconazole/kg/ transporter superfamily and has been used to demonstrate the rg day, equivalent to 4 (cid:3)g/larva, or 9 mg of fluconazole/kg/day, activityoftheseproteinsinazole-resistantisolatesofC.albicans, / o equivalent to 3.6 (cid:3)g/larva) or subtherapeutic (2 mg/kg/day, becausethisdyeusesthesametransportersasazolesinyeasts(52). n equivalent to 0.8 (cid:3)g/larva) dose of an azole, the percentages of OurresultsshowedthatallstrainscapturedR6G,althougham- J a survivalofthesegroupsdidnotdiffersignificantlyfromthoseof photericinB-resistantstrainsshowedhigheruptake/effluxratios n theirrespectiveuntreatedcontrolgroups(Fig.2).Inaddition,an than amphotericin B-susceptible strains at time zero. This phe- u a increaseinthefluconazoledosageto20mg/kg/day(equivalentto nomenoncouldbeexplainedbyhighermembranepermeability r y 8(cid:3)g/larva)failedtoimprovesurvivalwhenlarvaewereinfected duetochangesinthesterolcompositionofthemembrane.How- 3 withthesestrains(Fig.2). ever,analysisoftheuptakeandeffluxofR6Gforallazole-resistant 1 , Amphotericin B treatment (1.2 (cid:3)g/larva, equivalent to a orwild-typeC.tropicalisstrainsshowsthattheeffluxpumpsdo 2 0 therapeuticdoseof3mg/kg/day)improvedthesurvivaloflar- notseemtoplayafundamentalroleinazoleresistanceinC.tropi- 1 vaeinfectedwithstrainATCC750(P,0.0015)orTP-13650(P, calis,inagreementwithpreviousresults(14,27). 9 b 0.0002) but not that of those infected with a strain (ATCC The antifungal susceptibility phenotype of strains ATCC y 200956orCL-6835)cross-resistanttoazoleandamphotericin 200956andCL-6835resembledthatofC.albicansERG11/ERG11 g B(Fig.3). doublemutantstrainsshowingresistancetoazolesaswellasto u e In contrast, anidulafungin was the most active antifungal amphotericinBoneitherthewild-typeortheERG3/ERG3mutant s t drugagainstallstrainsregardlessoftheirazole/amphotericinB background(42).Also,ourdatawereconsistentwiththoseforthe susceptibilityprofiles,withalarvalsurvivalrangebetween30 C. albicans ERG11-null mutation when the strains were tested and 50% at the end of the experiment (Fig. 4). None of the againstdifferentmetabolicinhibitors(seeFig.S2inthesupple- drugsweretoxictothelarvaeatanyoftheconcentrationsused mental material). These data suggest that the susceptibilities of (datanotshown). bothantifungalcross-resistantC.tropicalisstrainstothesemeta- boliccompoundsmightberelatedtochangesinthecomposition DISCUSSION andpermeabilityofthefungalmembraneandarelikelyassociated Candida tropicalis is usually susceptible to all antifungal agents. withthefunctionalityofErg11pandalterationsinergosterolbio- However,antifungalshavebeenusedextensivelyfortheprophy- synthesis(42,56).Sterolanalysisconfirmedtheabsenceofergos- laxisortreatmentofcandidiasis,andthereareincreasingreports terol(Table2)andtheaccumulationof14-methylsterolsinC. of azole resistance for this species (13, 14, 27, 54). In fact, the tropicalisstrainsATCC200956andCL-6835(Fig.5;seealsoFig. azole-resistantclinicalisolatesofC.tropicalisstudiedinthiswork S3 in the supplemental material). In fact, this sterol profile was wererecoveredfromtwopatientssufferingfromcandidemiaand similar to those of different Candida species with ERG11 gene treatedwithdifferentantifungals(Table1).Theprecisemecha- deletions(42,56).Sequencingandstructuralanalysisofthepro- 4774 aac.asm.org AntimicrobialAgentsandChemotherapy ResistanceofCandidatropicalistoAntifungalDrugs D o w n lo a d e d f r o m h t t p : / / a a c . a s m . o r g / o n J a n u a r y 3 1 , 2 0 1 9 b y FIG2Efficaciesoffluconazole(FLC)andvoriconazole(VRC)forGalleriamellonellainfectedwithdifferentstrainsofCandidatropicalis.Tenlarvaepergroup g wereinfectedwith2(cid:7)106cellsofC.tropicalisATCC750,CL-6835,orTP-13650.ForC.tropicalisATCC200956,aninoculumof4(cid:7)106cellswasused.After u e inoculation,larvaeweretreatedwith0.8or4(cid:3)gofVRC/larva(subtherapeuticandtherapeuticdoses,respectively)orwith0.8,3.6,or8(cid:3)gofFLC/larva s (subtherapeuticandtherapeuticdosesandtwicethetherapeuticdose,respectively).Anuntreatedcontrolgroupwasincluded.Larvasurvivalwasmonitored t daily;thelarvaeareconsidereddeadwhentheydonotrespondtotouch. teinsshowedthatstrainATCC200956hasadeletionof44amino proteinthatislacking(helixI)islikelytobethecauseofthelackof acidsinErg11pthatcorrespondstotheexcisionofthe“I”helixin interaction either with the substrate (lanosterol) or with azole theazoletarget.Thatdeletionwouldpresumablyrenderthepro- drugs (26, 57). In fact, in Candida glabrata and Saccharomyces teinnonfunctional,whichwouldexplainboththelackofergos- cerevisiae,asinglemutationatthatlevelisenoughtonullifythe terolandtheaccumulationof14(cid:1)-methylatedsterols,andthere- functionofnativeErg11p(56,58). foretheresistancetoamphotericinB.Infact,duringtherevision The effects of Erg11p mutations at G464 and Y132 on azole ofthiswork,anidenticalazole/amphotericinBresistancemecha- susceptibilityhavebeenreportedpreviouslyindifferentyeastsand nismwasdescribedinaclinicalisolateofC.tropicalisisolatedin filamentousfungi(59–63).TheY132HandG464Smodifications Tunisia(54). conferredazoleresistanceonC.albicans(59).Also,theeffectsof ExaminationoftheC.albicansErg11p3Dproteinmodelsin differentERG11mutationshavebeenexploredbyheterologous combination with azole drugs suggests that the portion of the expression in S. cerevisiae (24, 64). In addition, several studies October2013 Volume57 Number10 aac.asm.org 4775 Forastieroetal. D o w n lo a d e d FIG3EfficacyofamphotericinB(AMB)inGalleriamellonellainfectedwithdifferentstrainsofCandidatropicalis.Tenlarvaepergroupwereinfectedwith2(cid:7) fr 106cellsofC.tropicalisATCC750,CL-6835,orTP-13650.ForC.tropicalisATCC200956,aninoculumof4(cid:7)106cellswasused.Afterinoculation,larvaewere o treatedwith0.8or1.2(cid:3)gofAMB/larva(subtherapeuticandtherapeuticdoses,respectively).Anuntreatedcontrolgroupwasincluded.Larvasurvivalwas m monitoreddaily;thelarvaeareconsidereddeadwhentheydonotrespondtotouch. h t t p : / performedwithmicrosomalfractionsandpurifiedenzymeprep- ThatisthecaseinstrainTP-13650;thepresenceofphenylala- /a arations have been used to analyze the relationship between nineatposition132producesthelossofabridgeofhydrogenthat a c Erg11p amino acid substitutions and reductions in the level of isnormallyformedbetweenhemeandtyrosine.Thismodification . a azole binding (22, 65). Those authors concluded that the azole in the environment of heme would alter the binding to azoles s m nitrogencouldnotinteractwiththeironofthehemeofC.albicans withoutmodifyingtheactivityoftheenzyme.Thisdifferenceis . o Erg11p(CaErg11p)withY132HorG464S,asnormallyoccursto markedbyawild-typesterolprofile,resistancetometabolicin- r g produceinhibition.Incontrast,thealteredproteinretainedthe hibitors(duetoanormalcytoplasmicmembrane),andsuscepti- / abilitytometabolizethesubstrate,comparablytotheERG11wild- bilitytoamphotericinB.Themodelingofwild-typeproteinanda o n type strain, allowing resistant mutants to produce ergosterol variantwithequivalentresiduesmutatedinfungalplantpatho- J (23,64). gens,suchasY137FinMycosphaerellagraminicola,hasprovided a n u a r y 3 1 , 2 0 1 9 b y g u e s t FIG4 Efficacyofanidulafungin(ANF)inGalleriamellonellainfectedwithdifferentstrainsofCandidatropicalis.Tenlarvaepergroupwereinfectedwith2(cid:7)106 cellsofC.tropicalisATCC750,CL-6835,orTP-13650.ForC.tropicalisATCC200956,aninoculumof4(cid:7)106cellswasused.Afterinoculation,larvaewere treatedwith4(cid:3)gofANF/larva.Anuntreatedcontrolgroupwasincluded.Larvasurvivalwasmonitoreddaily;thelarvaeareconsidereddeadwhentheydonot respondtotouch. 4776 aac.asm.org AntimicrobialAgentsandChemotherapy ResistanceofCandidatropicalistoAntifungalDrugs D o w n lo a d e d f r o m h t t p : / / a a c . a s m . o r g / o n J a n u a r y 3 1 , 2 0 1 9 b y g u e s t FIG5ErgosterolbiosyntheticpathwayinCandidatropicalis.MajorsterolintermediatesthataccumulateinC.tropicalisstrainswithnonfunctionalErg11pare boxed.Brokenarrowsindicatemultipleenzymaticsteps;solidarrowsindicatesingleenzymaticsteps.Thenumbersinparenthesesrefertothesterolsdescribed inTable2. similarresults(65),whichareconsistentwiththeresistancere- interdomainconformationalchangesuponinhibitororsubstrate cordedforotherplantpathogens(Y136FinMycosphaerellafijien- binding(57).InstrainCL-6835,theabsenceofergosterolanda sis, Y136F in Erysiphe graminis, and Y136F in Uncinula necator patternofsterolmetabolitessimilartothatofstrainATCC200956 [66–68]). would suggest that the Erg11p amino acid substitution G464D On the other hand, although G464 cannot interact with flu- (Fig.1E),locatedatthecatalytichemebindingdomain,willaffect conazoledirectly,thispositionmightbeimportantforthecon- equallytheaffinityoftheenzymefortheazoleanditsaffinityfor formationofthehemeenvironment.Changingglycinetoother thesubstrate.Thiscouldbeduetothenegativechargeorthelarger residueswouldbeexpectedtodecreasetheflexibilityrequiredfor sizeofasparticacid,producingagreatdisturbanceintheenviron- October2013 Volume57 Number10 aac.asm.org 4777 Forastieroetal. ment of the heme and preventing the binding of the substrate/ gillusfumigatus(36).Recently,Mesa-Arangoetal.(37)demon- inhibitor.Infact,inC.albicansErg11pwiththeG464Ssubstitu- stratedthatC.tropicalisisabletoinfectandkillG.mellonella.The tion, although no difference in the affinity of the substrate was lethaleffectofC.tropicalisdependsonboththeyeastdoseandthe observedbetweenthemutantandthewild-typeprotein,reduced incubation temperature. In larvae infected with antifungal-sus- steroldemethylasecatalyticactivitywasobserved(63).Thesterol ceptibleC.tropicalis,thefungalburdensincreasedinatime-de- profileconfirmedthatpossibility,sincethisstrainhasacomplete pendent manner, and treatment with either amphotericin B or lackofergosterolandanaccumulationof14(cid:1)-methylatedsterols, fluconazolereducedthenumberofCFUintheworm.Also,anti- thusshowingcross-resistancetoazoleandamphotericinB,aswas fungaltreatmentwasassociatedwiththeformationofcellaggre- thecasewithstrainATCC200956.Interestingly,strainCL-6835is gatesaroundinfectedareas.Takingthisintoaccount,alongwith thefirstreportedisolatethatcarriesanasparticacidsubstitutionat otherpublishedworks(34,35),weusedtheG.mellonellamodelto positionG464. evaluate the efficacies of different antifungal drugs against C. Ithasbeenreportedthatdisruptionofergosterolbiosynthesis tropicalisinfection.Theinvivoresponsecorrelatesverywellwith attheErg11plevelresultsintoxicmetabolites.Cellgrowtharrest theantifungalsusceptibilityphenotypeshowninvitro.Flucona- correlates with the accumulation of 14(cid:1)-methyl-ergosta- zoleorvoriconazoletreatmentiscapableofimprovingthesur- D 8,24(28)-dien-3(cid:8),6(cid:1)-diolinyeastswithasterol14(cid:1)-demethylase vivaloflarvaeinfectedwithazole-susceptibleC.tropicalis(ATCC o genedisruption.Inthissituation,14(cid:1)-methylatedsterolsarecon- 750).Thesedrugsimprovesurvivalinadose-dependentmanner. w n verted to toxic 3,6-diol derivatives by (cid:2)5,6 desaturation per- Infact,voriconazoleandfluconazoleareeffectivewhenthedoses lo formed by the enzyme Erg3p (Fig. 5). However, cells can over- recommendedforhumanstotreatinvasivecandidiasisorcandi- a d cometheeffectofsuchablockbyasuppressormutationinsterol demiaareused(11).Ontheotherhand,theseantifungaldrugsare e C5,6desaturationandcanthusacquireazoleresistance(69).In ineffectiveforlarvaeinfectedwithazole-resistantstrains(ATCC d f agreement with this observation, azole-resistant fungal isolates 200956, CL-6835, and TP-13650). However, amphotericin B r o withnonfunctionalERG3alleleshavebeendescribedforC.albi- treatmentattherapeuticdoses(1.2(cid:3)g/larva)improveslarvalsur- m cansandCandidadubliniensis(70–72).Bothstrainswithaputa- vivalwhenstrainsATCC750andTP-13650(amphotericinBsus- h tiveblockattheErg11plevel(CL-6835andATCC200956)carried ceptible)areusedforinfectionbutfailtoimprovethesurvivalof tt p ERG3 mutations (Table 1) and showed an accumulation of 24- larvaeinfectedwithstrainCL-6835orATCC200956(amphoter- :/ / methylcholesta-5,7,24(28)-trien-3-ol (Table 2), suggesting that icin B resistant). The response to anidulafungin was used as a a a thesetwostrainsmighthaveevolvedtoavoidtheproductionof controlfordrugefficacyandtoevaluateifthemodelwasableto c toxic sterols like 14,24-dimethylcholesta-8,24(28)-3(cid:8),6(cid:1)-diol reproducetheinvitrophenotypeofsusceptibilitytoechinocandin .a s (Fig.5).Ontheotherhand,itispossiblethatchangesinoverall drugsforallthestrains.Anidulafungintreatmentshowedade- m membrane composition could compensate for perturbations in creaseinthemortalityoflarvae,independentlyofthestrain.We . o sterolcontent,ascenarioconsistentwithreportsofotherergos- confirmedthattheG.mellonellahostisveryusefulforevaluation rg terol-deficientyeastswithdysfunctionalErg11pthatremainaer- of the in vivo efficacies of antifungal agents in an experimental / o obicallyviable(56).Althoughourresultsagreewiththoseofother modelofcandidiasis. n studies regarding activity alterations of Erg11p and Erg3p (54), Insummary,wereportthestudyofdifferentmechanismsre- J a future work would include functional characterization of these sponsibleforsingleandcombinedazoleandamphotericinBre- n enzymes by gene deletion or other genetic approaches, such as sistanceinC.tropicalisclinicalisolates.Althoughincreasedefflux u a heterologousexpressionandsite-directedmutagenesis,inorder or missense mutations in ERG11 are the most frequent mecha- r y to fully understand the enzymatic activities of these proteins in nismsinvolvedinazoleresistance,thedecreasedsusceptibilityof 3 theseparticularclinicalisolates. theseisolatestoazolesseemedtobeduetoanERG11genedele- 1 , Sincetwoofthestrains,CL-6835andTP-13652,wereisolated tion/aminoacidsubstitutioninanareaoftheproteinaffectingthe 2 0 frompatientswithcandidiasisrefractorytoantifungaltreatment, siteforbothsubstratebindingandazolebinding.Theinterrup- 1 weattemptedtodeterminethetherapeuticeffectsofazolesinanin tion of the ergosterol biosynthesis pathway would, in turn, be 9 b vivomodel.Duetothestrongstructuralandfunctionalsimilari- responsibleforamphotericinBresistanceinstrainslackingfunc- y tiesbetweentheinnateimmuneresponsesofinsectsandmam- tionalErg11p.Theresultsofourstudyelucidateimportantchar- g mals (73), several minihosts are now being employed to study acteristicsandpotentialmechanismsoffluconazoleandampho- u e alterationsinmicrobialvirulenceandtheefficacyofantimicrobial tericin B resistance in C. tropicalis. In addition, we validated G. s t agents.Amongthem,Galleriamellonellapresentscertainbenefits mellonellaasasimplenonmammalianmodelsystemthatcanbe in comparison with other nonmammalian models of infection usedtostudyinvitro-invivocorrelationofantifungals.Galleria (31).Forexample,thelarvaecanbemaintainedatatemperature mellonellalarvaeinfectedwithCandidatropicalisstrainsshowing rangeof25°Cto37°C,thusreproducingbothatthetemperature invitroresistancetoamphotericinB,fluconazole,andvoricona- atwhichfungiexistintheenvironmentandatthetemperatureat zolefailedtreatmentwiththeseantifungalsbutrespondedtoechi- which they are capable of producing invasive infections in hu- nocandins. mans.Inaddition,anaccurateinoculumofthepathogencanbe delivereddirectlyintothehost’sbody(34).Finally,thenumberof ACKNOWLEDGMENTS larvae used and the ease of handling also represent important E.M.wassupportedbytheEuropeanScienceFoundation(Fuminomics advantages.PreviousstudieshavedocumentedthatG.mellonella 06-RNP-132),theResearchProjectsfromtheSpanishMinistryofScience isavalidmodelfortheevaluationofmicrobialvirulenceand/or andInnovation(ERA-NETPathogenomics[7thFP],BFU2008-04709-E/ theefficaciesofdifferentantimicrobialagentsininfectionscaused BMC),andtheFondodeInvestigacionSanitaria(PI12_02376,MPY1003/ bydifferentbacteria(32–38)andfungi,suchasCryptococcusneo- 13).L.A.-F.isfundedbytheSpanishFondodeInvestigaciónSanitaria formans(34),Candidaspp.(38),Fusariumspp.(35),andAsper- withaMiguelServetfellowship(FISref.CP11/00026).A.F.isfundedbya 4778 aac.asm.org AntimicrobialAgentsandChemotherapy

Description:
TABLE 2 Sterol compositions of Candida tropicalis strains. Compound no. daily; the larvae are considered dead when they do not respond to touch. Resistance of .. Desalermos A, Fuchs BB, Mylonakis E. 2012. Selecting an
See more

The list of books you might like

Most books are stored in the elastic cloud where traffic is expensive. For this reason, we have a limit on daily download.