Vol.57,No.8APPLIEDANDENVIRONMENTALMICROBIOLOGY,Aug.1991,p.2147-2151009 - PDF document

Vol.57,No.8APPLIEDANDENVIRONMENTALMICROBIOLOGY,Aug.1991,p.2147-21510099-2240/91/082147-05$02.00/0Copyright©1991,AmericanSocietyforMicrobiologyEcologyofBacteriophagesInfectingActivatedSludgeBacteriaJARKKOHANTULA,1ANNEKURKI,2PERTTIVUORIRANTA,2ANDDENNISH.BAMFORD'*DepartmentofGenetics,UniversityofHelsinki,Arkadiankatu7,SF-00100Helsinki,'andTampereUniversityofTechnology,WaterandEnvironmentalEngineering,SF-33101Tampere,2FinlandReceived30November1990/Accepted15May1991Littleisknownabouttheendemicbacteriophagesofactivatedsludge.Inthisinvestigation49virus-hostsystemswerestudiedbyisolatingco-occurringbacteriaandbacteriophagesfromtheaerationbasinofasewagetreatmentplantduring5successiveweeks.Thephagetiterswerehighandfluctuatedduringthetimeperiod.Theoccurrenceofphage-sensitiveand-resistanthostsdidnotdependonthepresenceorabsenceofphages.Severalphage-hostsystemsexpressedvariableplatingefficiencies.Inaddition,phageswithbroadhostrangeswereobserved.Theseresultsshowthatphagesareanactivepartofthisecosystemandthattheymayexertselectionpressureforphageresistanceontheirbacterialhostpopulations.Phagesareabundantinnaturalecosystems(2,18,19,21).However,ourunderstandingoftheecologyofbacterio-phageshasbeenlargelybasedontheoreticalconsiderationsandlaboratory-scalestudiesofseveralwell-characterizedtypestrainsandtheirphages.TheearlychemostatstudiesofEscherichiacolianditsphagesT2,T3,andT4(9,16,17)indicatedthatastableequilibriumisreached.Thetheoryemergingfromthislineofresearch(reviewedinreference11)predictsthatmostofthebacteriaareresistanttoco-occurringphages,whicharemaintainedbyasubdominantphage-sensitivepopulation.Dominantbacterialclonessen-sitivetoco-occurringphagesaretransientphenomena.Therearedifficultiesincollectingdataonbacterialivinginnaturalenvironments.(i)Itisnotpossibletocultivateandquantitateallbacteriainanyavailablemedium(5,23).(ii)Thebacterialfloraofnaturalecosystemslikeactivatedsludgeissodivergentthatonanonselectivemediumonlythemostdominantstrainscanbeisolatedmorethanonce(8).Thus,thedominancepatternsofdifferentbacterialstrainsorclonesarenoteasytoelucidate.(iii)Studiesongrowthratesorcompetitivedifferencesofbacterialclonescannotbecarriedoutinanativeenvironment.Suchstudiescanbecarriedoutinlaboratoryconditions,butitisnotclearwhethertheseresultsarerelevanttothesameisolatesinnaturalcommunities.Asfarasweareaware,thebacteriophagesofactivatedsludgehavenotbeenstudied,exceptforthosethatinfectinfluxenterobacteria(forareview,seereference4).Theactivationofsludgedrasticallydecreasesthenumbersofthesebacteriophagesandtheirenterobacterialhosts.How-ever,thereisnoinformationaboutthephage-hostrelationsinthisecosystem.Thepresentstudyisbasedonawork(8)inwhichweclassifiedactivatedsludgebacteriabyanalyzingtheirtotalcellularproteinpatternsbyusingsodiumdodecylsulfate-polyacrylamidegelelectrophoresis.Withthismethodwecanidentifyandgroupisolatesdespitemutationschangingtheircolonymorphology,growthrequirements,and/orphagesensitivity.Thegroupofbacterialisolatesthatwecallastrainismadeupofspecimensthataretaxonomicallyveryclose(7,8,10).Thisallowsustoconcludethattheobservedresistancestophageareduetosingleoratmostafewmutationsandnotjustreflectionsofgeneraldifferentiationof*Correspondingauthor.hostisolates.Byusingthebacterialcollectionofourprevi-ouswork(8),weisolatedco-occurringbacteriophagesfromtheaerationbasinofasewagetreatmentplantfor5succes-siveweeks.Wedeterminedphagedensitiesandefficienciesofplating(EOPs),theoccurrenceofphageresistanceinisolatedhostbacteria,andthehostrangesofphagesonthesebacteria.Themethodologyusedhereallowedustoobtainusefuldirectinformationaboutselectedphage-hostsystemsovertimeinactivatedsludge.MATERIALSANDMETHODSBacteriaandphages.Thebacterialstrainsusedweredescribedpreviously(8).Thefirstnumberintheisolatenumberreferstotheweeklysamplefromwhichtheisolateoriginated.Thefirstisolateofeachbacterialstrainistakenasatypeisolateafterwhichthestrainisdesignated.Thusthefoursimilarisolates3006,3010,3063,and3097belongtostrain3006.Inaddition,E.coliBandK-12(3,6)andbacteriophage46(22)wereusedinthisinvestigation.SamplesforbothbacterialandbacteriophageisolationweretakenonceaweekonWednesdaysat9:00a.m.fromtheaerationbasinofViinikanlahtimunicipaltreatmentplantinTampere,Finland,from14Septemberthrough12October1988.Aftersettlingforhalfanhour,thebacteriawereisolatedfromthedispersedphaseasdescribedpreviously(8).Forphageisolations,100-mlsamplesofthedispersephaseswerefrozenat-20°C.Whenthebacterialstrainswereavailable,thesampleswerethawedandcentrifuged8,000rpmfor20minat4°C(SorvallSS34rotor)andfilteredthroughaMinisartNMR0.2-,um-pore-size(Sartorius)filter.Analiquotwasstoredat-20°C,andtherestwasconcen-trated8.5-foldwith10%polyethyleneglycol6000asde-scribedpreviously(24).Theconcentratedsamplewasstoredat-20°C.Thephagetestswerecarriedouton10%TGYA(tryptone-glucose-yeastextract[TGY][1]containing15gofDifcoBacto-Agarperliter)plateswith10%TGYsoftagar(1mMCaCI2and7gofBacto-Agarperliterof10%TGY).Thepresenceofphageswastestedbyapplyinga10-,uldropofeachoftheconcentratedsamplesontoeachof629bacteriallawns.Theplateswereincubatedat20°C.Whenplaqueswereobserved,theunconcentratedphagesampleswereplatedonappropriatehosts.Thephageswerepurifiedfromtheseplatesbytwosuccessivesingleplaqueisolations.Then2147 2148HANTULAETAL.allbacterialisolateswithproteinpatternsidenticaltothoseoftheoriginalhostweretestedforEOP.ThenonidenticalbacteriaofthesamestrainclusterthatweresensitivetophagewerealsoincludedintheEOPtest.WhenvariationsinEOPweredetected,thephagewasplaquepurified,whenpossible,withtheotherisolatesandtheEOPwasdeter-minedoncemoreonallsensitivehostbacteria.Thenomenclatureofviruseswasderivedfromthenomen-clatureoftheirhosts.Forexample,aphageoriginallyisolatedandplaquepurifiedonhost3010wasdesignated4)3010.Phagesfromtwoseparateplaquespurifiedonthesamehostisolateweredesignated,forexample,4)5054.1and4)5054.2.Tocontroltheeffectsoffreezingduringthesamplingprocedure,wedeterminedthetitersbeforeandafterafreeze-thawcyclefor24plaque-purifiedphageisolates(datanotshown).Thephagescapableofplaqueformationaftertheinitialfreezingandthawingcycledidnotshowconsid-erablesensitivitytofreezing.Inthetesteventhemostsensitiveisolate(4)5030)wasinactivatedonlytoone-fourthofitsoriginaltiter.Phageparticleconcentrationandpurification.Softagarlayersof30identicalsemiconfluentplateswerecollectedandcentrifugedat10,000rpmfor20minat4°C(SorvallSS34rotor).Thepelletwasresuspendedin10mlof10%TGYandcentrifugedagain;thetwosupernatantswerecombined.Thephageswereconcentrated75-foldwithpolyethyleneglycol6000andresuspendedin400[lof10mMpotassiumphos-phate(pH7.2)containing1mMMgCl2(bufferA).TheremainingagaroseinthesamplewasreducedbycentrifuginginanEppendorfmicrocentrifuge(10,000rpmfor5minat4°C).Phageswerepurifiedbyratezonalcentrifugationin5to20%(wt/vol)sucrosegradientinbufferA(SW41rotor,23,000rpmat10°Cfor25or120min,dependingonthephage).Tenfractionswerecollected,andthephagetitersweredetermined.Thehighest-titerfractionsweredilutedtwofold,andthephagewerepelletedbycentrifugation(TiSOrotor,32,000rpmfor2or5hat5°C).Theresultingpelletwasresuspendedin50[LofbufferA.Whenpossible,thephageproteinpatternwasanalyzedbysodiumdodecylsulfate-gelelectrophoresisasdescribedpreviously(15).Themolecularsizesofmajorproteinswereestimatedbycomparingtheirmobilitieswiththoseofthestructuralproteinsofbacterio-phage4)6,thegenesofwhichhavebeensequenced(14).Thesemolecularsizemarkerscovertherangefrom9.5to85.0kDa.TABLE1.DetectionofphagesinfivesamplesHostPFU(103)in1mlofsamplefromweek:eI,altA300230103038306430853109312231254001404140654085408640914094410150015022503050335054506050785087509851005105600360046009601660266043604460576058607761157015702270277039705070517066708771041ND"NDNDND0.080.02NDNDjobND0.010.620.010.25ND15.000.020.020.010.880.01NDND0.13NDND0.03ND0.080.07NDNDND0.10NDNDNDNDND0.01ND0.022.802.560.76NDND2ND0.130.41ND2.500.04NDND100.030.070.130.010.02ND11.00NDND0.03NDNDNDNDNDNDNDND0.020.03NDNDNDND0.11ND0.040.010.04NDNDND8.964.321.043.84ND0.0530.151.524.80ND0.39NDND0.02107.520.020.010.52ND0.0520.00ND3.900.430.01ND0.060.01NDNDNDND0.010.18NDNDNDND0.33ND0.010.160.130.03NDND1.368.080.672.400.010.2340.070.6432.800.090.240.01ND0.021029.200.01ND0.72ND0.4140.00ND20.000.30NDND0.09NDND0.080.01ND0.060.86ND0.150.010.050.64ND2.000.052.000.03NDND1.783.440.3624.00ND0.13RESULTSGeneralobservations.Intheinitialscreening,inwhichthephagesampleswerepipettedon629bacteriallawns,wefoundplaquesonisolatesbelongingto56ofthe302strains.Theexacttiterdeterminationswithsoftagar(50°C)suc-ceededonlyfor47ofthese.Thisindicatesthat16%ofthephageswerethermosensitive.Theplaquemorphologywasusuallyclear,whichistypicalforvirulentphages.Thisinvestigationisbasedonlyonthethermoresistantphagesthatsurvivedafreeze-thawcycle.Densityofphagesinactivatedsludge.Thetotalnumberofphageparticlescapableofformingvisibleplaquesonthe629hoststestedwiththeseconditionsvariedfrom3.3x104to1.7x105PFU/mlinfivedifferentsamples(Table1).Intheplaqueassay47differentstrains,correspondingtoabout15%ofallstrainsavailable,weresensitivetothesephages(Table1).Thetitersdeterminedfordifferentphagesvariedfrom1x101to5x104PFU/mlinonesample.Thenumbers"ND,notdetected(densityofvirusparticleslowerthan10PFU/ml).Accuratetitersonstrain4001couldnotbedeterminedduetounclearplaquemorphology.ofphagesfluctuatedconsiderablyduringthe5-weekperiod(Table1).Phageresistance.Theoccurrenceofphageresistancewastestedwhenbacterialstrainswithisolatessensitivetoaparticularphagewereisolatedmorethanonceduringthe5-weekstudy(8).Thephagesinfectingthese18strainsformedthreecategoriesinrespecttotheirhostranges.Thelargestgroupcomprisedphageisolatesthatwereabletoinfectonlyafewoftheisolatesbelongingtothesamehoststrain(12cases).Sixphageisolatesinfectedalloftheindependentisolatesofthehoststrain.Inthreecasesthephagegrowthwassupportedbymorethanonestrain(4)4094,4)7066,4)3122,andsimilarphages;seebelow).5ND0.294.7650.000.010.100.02ND101.220.01ND0.07ND0.112.96ND3.400.070.022.082.36NDNDND0.95ND2.003.54ND0.04ND2.900.230.050.020.070.16NDND0.060.160.950.103.760.020.05APPL.ENVIRON.MICROBIOL. ACTIVATEDSLUDGEBACTERIOPHAGES2149TABLE2.Phagesinfectingisolatesofstrain3006PhageTestisolateEOP,)3010300603010130630309704)30633006030100.130631309714)30973006030100.13063130971Animportantquestioninunderstandingtheeffectofphagesontheirhostsistodeterminetheoccurrenceofresistanceandsensitivityinrelationtotheoccurrenceofthephage.Thiswastestedwithphage-hostsystemsinbacteriaisolatedfromsamplesthatdidnothavethemaximaldensityofphagesinfectingotherisolatesofthatstrain(55sensitiveand16resistantisolates).Theoccurrenceofresistantandsensitivehostsinthesesamplesdidnotdifferfromrandomdistribution(X2test).ObserveddifferencesinEOP.Threeofthefourisolatesofstrain3006supportedplaqueformationintheprimaryscreeningoftheactivatedsludge(isolates3010,3063,and3097).Whenaplaqueonstrain3010waspurifiedandthatphagestockwastestedfortheinfectivitytotheotherthreeisolates(Table2),isolates3063and3097inadditionto3006wereresistanttopurifiedphage4)3010.Startingfromtheoriginalsterilizedactivatedsludge,plaqueswerepurifiedon3063and3097.Bothofthesephages(4)3063and4)3097)wereabletoinfecteachother'shostsandalsotoinfectstrain3010,butataloweredEOP.Fromtheseresultswecon-cludedthattwoseparatephageswereisolated,oneinfectingonlyhost3010andtheotherinfectingallofthesensitivehostsbutinfecting3010withaloweredEOP.Thepurifica-tionofvirusparticlesdidnotsucceedforthesephages,anditwasnotpossibletoverifyourconclusionatthemolecularlevel.Phage4)3064wasinfectivetoallsixisolatesofthehoststrain.Theisolates,however,fellintwocategories:theEOPson4002,4036,and5058were6ordersofmagnitudelowerthanthoseon3064,4035,and4064.Theplaquesonlow-EOPhostscouldnotbepurifiedtotesttheirEOPsonthehigher-EOPisolates;thusthegeneticbasisforvariationinEOPremainsunknown.ThevariationsofEOPofphages4)4094and47087.1,whichinfectseveralhoststrains,arediscussedbelow.Broadhostrange.Thehostrangesofphagesweretestedbyplatingthemonnonidenticalbutrelatedhoststhathadsupportedphagegrowthduringtheinitialscreeningphaseofthestudy.Phage44094infectedtheisolate5078ofstrain3112withanEOPof10-6.When4)4094wasplaquepurifiedon5078,itgaveastockwithalowtiter(3x106)andinfected4094withalowEOP(about10-6).Thelowinfectivityofphagestocksontheseisolateswhenproducedindifferenthostssuggeststhatparticlesthatinfecteitherhostdifferbyonemutation.Mostoftheknownhostrangemutationsareactuallyextendedhostrangemutations(11),butthissystemseemstorepresentatruehostrangemutationbecausethemutantcannolongergrowontheoriginalhost.TABLE3.Phagesinfecting3122-likebacteriaPhage~PropagationTestEOPhagestrainstrainEOP4)31223122312217066170871-)70667066312207066170870.00170877066170871-)7087.1708731220.570660.5708714)7087.27087312217066170871Amorecomplicatedsituationwasobtainedforstrains3122,7066,and7087,whicharecloselyrelated(8).Phagesplatingonstrain7066werefoundinallfiveweeklysamples,whereasphagesplatingonstrains3122and7087werefoundonlyinoneandtwosamples,respectively(Table1).4)7066infectedstrain7087,althoughataloweredEOP(Table3).Thisexplainsthetiterdissimilaritiesonstrains7066and7087indifferentsamples(Table1).When4)7066wasplaquepurifiedonstrain7087,theEOPwasthesameforbothhosts.Apossibleexplanationforthismightbearestriction-modi-ficationsystemactinginstrain7087butnotin7066.4)3122infectedbothisolatesofthehoststrain(3122and4037)andstrains7066and7087.47087.1and47087.2infectedstrains3122and7066,although4)7087.1hada50%reducedEOPonthesehosts.Inaddition,therewereplaquemorphologyandEOPdifferencesbetween4)7087.1and4)7087.2onLuria-Bertanimedium(13):4)7087.1formedsmallplaqueswithanEOPthatwasseveralordersofmagnitudelowerthanthatof4)7087.2,whichformedlargeplaqueswithanalmostunalteredEOPonhost7087.Thesimilarityofthesefourphageisolatesinfectingthesamehostswastestedbyusingparticlesedimentationandproteincompositionanalysis.Thefastsedimentationof4)7066indi-catedthatitwasunrelatedto4)3122,47087.1,and4)7087.2.Theproteingelanalysisofpurified4)3122,4)7087.1,and4)7087.2confirmedthemalltobesimilarslowlysedimentingsmallphageswithmajorcapsidproteinofapproximately16kDa(datanotshown).Theproteincompositionandsedi-mentationbehaviorofthesephagessuggestthattheyaresingle-strandedRNA-typephagesliketheMS2phageofE.coli.Plaquemorphologyvariation.Whenthephage4)4101wasplaquepurifiedonallsensitiveisolatesofitshoststrain(4081,4101,4102,and4115),theEOPwasthesameonallisolatesandindependentofthehostusedtopropagatethephage.However,theplaquemorphologywasdependentonboththehostusedfortiterdeterminationandthehostusedtopropagatethephage.Returnofphage.Aphagethatformedplaquesonstrain5054wasobtainedwithsamplesfromthefirstandlastweeks(Table1).Phagesisolatedfrombothsamplesinfectedbothisolatesofthehoststrain(5054and5055).Theyalsohadsimilarplaquemorphologies.TheproteingelanalysisofpurifiedphageparticlesshowedthatthesetwophageisolatesVOL.57,1991 2150HANTULAETAL.hadidenticalproteinpatterns,identifyingthemasthesamephagewithamajorcapsidproteinofabout35kDa(datanotshown).Otherexamplesofstrainswithphagesdetectedtwicewithasamplelackingdetectablephagesinbetweenare3109,5022,5033,and7087(Table1).Inallofthesecasesthephagehadtitersclosetoourdetectionlimiteitherbeforeorafteritsdisappearance.DISCUSSIONThebacteriophagesisolatedhereareundoubtedlyrepro-ducinginthisecosystem,whichsuggeststhattheirhostsarealsoactive.Thiswasverifiedbydeterminingtitersofcoliph-ages,whichdeclineinactivatedsludgebutareknowntobefrequentincommunalsewage(4).ThephagetitersontwodifferentstrainsofE.coliwere60PFU/mlatmostandinmostsampleswerebelowourdetectionlimitandthuslowerthanthoseofmostofthephagesstudiedhere.Fifteenpercentofthebacterialstrainsisolatedweresensitivetoco-occurringphages.Thisnumbershouldbetakenasaminimumestimate,because(i)thesensitivityofthedetectionisabout12PFU/mi,(ii)twofreeze-thawcyclesareappliedtothesample,(iii)somephagesadsorbtothesterilizationfilter,(iv)allphagesdonotformplaquesintheconditionsused(thisbecameevidentastheplaquemorphol-ogyofdifferentphagesvariedfrombigclearplaquestoverytinyonesthatcouldnotbecounted),and(v),sincemoststrainswereisolatedonce,thephageswerenotdetectedwhentheonlyisolatebelongedtoaresistantclone.Ifweassumethat90%ofthephageswerenotdetected(whichstillmightbeanunderestimate),andifwetakeintoaccounttheestimatedtotalnumberofstrainsinactivatedsludge(severalthousands[8])andsupposethatthesameproportionsofallstrainsareinfectedwithcomparablenumbersofviruses,thetotalnumberofphagesinthedispersedphaseofactivatedsludgewouldbeabout106to108PFU/ml.Thisfigureisstillsmallerthantheestimateforfreshwatersbasedonelectronmicroscopy(108PFU/ml[2])andstrengthensourassump-tionthatthegivennumberisanunderestimate.Thephageinfectioncycleseemstobeveryslowinactivatedsludge.Typicallyittakesafewweekstorisefromthefirstemergencetohighestphagedensity(+3038,43064,44041,4A086,45022,+6004,P6058,4.6115,47050,and47066).TheobservationofTanandReanney(21)thatphageinfectioninenrichmentculturesofsoilismuchslowerthaninlaboratoryconditionsisinaccordancewithourresults.ThetimescaleofTanandReanneyisinhours,whereasourobservationsindicatemuchslowerdynamics.Weconcludethatinnaturalecosystemsphageinfectionsareveryslowcomparedwiththoseinlaboratoryconditions.Sincewedonotknowtheexactdensitiesorthegenerationtimesofthehosts,itisnotpossibletospeculatewhetherthisreflectsslowadsorptionkineticsorslowlifecyclesofthehost.Anotherfeatureemergingfromourdataistheslowdisap-pearanceofphages(43038,43085,P4041,44085,44091,,4101,45022,+5030,+6058,46115,47039,+7050,+7051,andP7066),butalsothereappearanceofphageswasde-tected(43109,+5022,45033,+5054,andphagesagainst7087).Inalloftheseexamplesthephagehadtitersclosetoourdetectionlimiteitherbeforeorafteritsdisappearance,whichmakestheinterpretationofthisresultdifficult.Thephagesisolatedbeforeandafterdisappearancewereintwocases(+5054and+7087.1,47087.2)confirmedtobethesamebyusingpurifiedphagepreparations.Theseobserva-tionsindicatethatthenumberofphagesoscillatesorstaysforweeksatlowdensitiesinthisecosystem.Possibleexplanationsforthesephenomenaarethecirculationofsludgeduringplantoperationandaprobableequilibriumofphagesbetweenflocculatedanddispersedphases,althoughthemixingofactivatedsludgeisrigorous.Thechemostatstudiesshowthathostresistanceevolveswithinashorttimeasaresponsetothepresenceofphage(11,12).Ourresultsareinaccordancewiththisprediction;(i)phageswerefoundinonly15%ofallstrains,and(ii)onlyabout40%ofthephage-sensitivebacterialstrainsthatwereisolatedseveraltimeswerecomposedofisolateswhichwereallsensitivetotheappropriatephages.Itshouldbepointedoutherethat,becausenotallofthephagesweredetected(seeabove),thenumberofsensitiveisolatesissomewhatunderestimated.Thevastmicrobialdiversityofthisecosystemhamperedourattempttostudymanyisolatesofeachstraininasinglesample,buttherandomizedisolationprocedureused(8)allowsustoconcludethatmost(ifnotall)oftheisolatesrepresentdominantorcodominantclonesandthatsubdom-inantcloneswerenotdetected.Oneofthepredictionsofthetheoreticalworkisthatthesensitivehostsshoulddominatebeforephageinfectionandthattheresistantformsshoulddominateafterphageinfection.However,therewasnorelationbetweentheoccurrenceofphageandresistantbacteria.Itcanbespeculatedthatphage-resistantbacteriahaveevolvedduringanearlieroscillationcycleofphageinfection.Thelackofrelationwiththeoccurrenceofphagesandsensitivehostsindicateseitherthatthesebacteriacansurviveduringtheinfectionorthattheycangrowbacktohighnumbersaftertheinfection.Thus,someofourobser-vationsagreewithandsomecontradictthetheoreticalmodel.Theobservationthatthephageisolatedfromanearliersampleforstrain7087waslessvirulenttoitshostthantoalatterisolateisinaccordancewiththechemostatresultsofSpanakisandHorne(20),whoshowedthat,asaphagebecomesmorevirulenttoitshost,thehostevolvestobecomemoreresistant.Inourcase,however,itisnotpossibletoexcludethepossibilitythatalaterphagehasnotevolvedfromtheformeronebutisaseparatebutverycloselyrelatedspecies.Mostofthephagesinfectedonlyasinglestrain.However,somephages(+4094,47066,and43122andsimilarvirusesisolatedon7087)wereabletoinfectmorethanonestrain,suggestingthatbroad-host-rangephagesarecommoninnaturalecosystems.Thusitcanbearguedthattheclassicalapproachofstudyingsinglephage-singlehostsystemsistoonarrowtoobtainacomprehensiveunderstandingofthephageecology.Weshowherethatphagesareanactivepartoftheactivatedsludge.Althoughitiswellknownthatphagesareparasites,theireffectonhostpopulationsisnotproperlyunderstood.Evenlessunderstoodistheireffectontheactivatedsludgeprocessitself.Thisinvestigationdoesnotindicatethattheyarestrongregulatorsofspeciescomposi-tion.Instead,theyseemtomodifyspecific,phageresis-tance-related,geneticpropertiesofbacteriabyexertingselectionpressureforphageresistance.ACKNOWLEDGMENTSD.H.B.isaseniorscientistattheAcademyofFinland.ThisinvestigationwassupportedbyaresearchgrantfromtheMaj&TorNesslingfoundation(toJ.H.andA.K.).APPL.ENVIRON.MICROBIOL. ACTIVATEDSLUDGEBACTERIOPHAGES2151REFERENCES1.AmericanPublicHealthAssociation.1985.Standardmethodsfortheexaminationofwaterandwastewater,16thed.AmericanPublicHealthAssociation,Washington,D.C.2.Bergh,0.,K.Y.B0rsheim,G.Bratbak,andM.Heldal.1989.Highabundanceofvirusesfoundinaquaticenvironments.Nature(London)340:467-468.3.Bertani,G.,andJ.J.Weigle.1953.Hostcontrolledvariationinbacterialviruses.J.Bacteriol.65:113-121.4.Bitton,G.1987.Fateofbacteriophagesinwaterandwastewatertreatmentplants,p.181-195.InS.M.Goyal,C.P.Gerba,andG.Bitton,(ed.),Phageecology.JohnWiley&Sons,Inc.,NewYork.5.Giovannoni,S.J.,T.B.Britschgi,C.L.Moyer,andK.Field.1990.GeneticdiversityinSargassoSeabacterioplankton.Na-ture(London)345:60-63.6.Gray,C.H.,andE.L.Tatum.1944.X-rayinducedgrowthfactorrequirementsinbacteria.Proc.Natl.Acad.Sci.USA30:404-410.7.Hantula,J.I.,T.K.Korhonen,andD.H;Bamford.1990.Determinationoftaxonomicresolutioncapacityofconventialone-dimensionalSDS-polyacrylamidegelelectrophoresisofwhole-cellproteinsusingEnterobacteriaceae.FEMSMicro-biol.Lett.70:325-330.8.Hantula,J.I.,A.Kurki,P.Vuoriranta,andD.H.Bamford.1990.RapidclassificationofbacterialstrainsbySDS-polyacryl-amidegelelectrophoresis:populationdynamicsofthedominantdispersedphasebacteriaofactivatedsludge.Appl.Microbiol.Biotechnol.34:551-555.9.Horne,M.T.1970.CoevolutionofEscherichiacoliandbacte-riophagesinchemostatculture.Science168:992-993.10.Jackman,P.J.H.1985.Bacterialtaxonomybasedonelectro-phoreticwhole-cellproteinpatterns,p.115-129.InM.Good-fellowandD.E.Minnikin(ed.),Chemicalmethodsinbacterialsystematics.AcademicPress,Inc.(London),Ltd.,London.11.Lenski,R.E.1987.Dynamicsofinteractionsbetweenbacteriaandvirulentbacteriophage.Adv.Microb.Ecol.10:1-44.12.Lenski,R.E.,andB.R.Levin.1985.Constraintsofthecoevolutionofbacteriaandvirulentphage:amodel,someexperiments,andpredictionsfornaturalcommunities.Am.Nat.125:585-602.13.Maniatis,T.,E.F.Fritch,andJ.Sambrook.1982.Molecularcloning:alaboratorymanual.ColdSpringHarborLaboratory,ColdSpringHarbor,N.Y.14.Mindich,L.,andD.H.Bamford.1988.Lipidcontainingbacte-riophages,p.475-520.InR.Calendar(ed.),Thebacterio-phages,vol.2.PlenumPublishingCorp.,NewYork.15.Olkkonen,V.M.,andD.H.Bamford.1989.Quantitationoftheadsorptionandpenetrationstagesofbacteriophage46infection.Virology171:229-238.16.Paynter,M.J.B.,andH.R.BungayIII.1969.Dynamicsofcoliphageinfections,p.323-335.InD.Perlman(ed.),Fermen-tationadvances.AcademicPress,Inc.,NewYork.17.Paynter,M.J.B.,andH.R.BungayIII.1971.CharacterizationofvirulentbacteriophageinfectionsofEscherichiacoliincon-tinuousculture.Science172:405.18.Proctor,L.M.,andJ.A.Fuhrman.1990.Viralmortalityofmarinebacteriaandcyanobacteria.Nature(London)343:60-62.19.Reanney,D.C.,andS.C.N.Marsh.1973.TheecologyofvirusesattackingBacillusstearothermophilusinsoil.SoilBiol.Biochem.5:399-408.20.Spanakis,E.,andM.T.Horne.1987.Co-adaptationofEsche-richiacoliandcoliphageXvirincontinuousculture.J.Gen.Microbiol.133:353-360.21.Tan,J.S.A.,andD.C.Reanney.1976.Interactionsbetweenbacteriophagesandbacteriainsoil.SoilBiol.Biochem.8:145-150.22.Vidaver,A.K.,R.K.Koski,andJ.L.VanEtten.1973.Bacteriophage+6:alipid-containingvirusofPseudomonasphaseolicola.J.Virol.11:799-805.23.Ward,D.M.,R.Weller,andM.M.Bateson.1990.16SrRNAsequencesrevealnumerousunculturedmicroorganismsinanaturalcommunity.Nature(London)345:63-65.24.Yamamoto;K.R.,B.M.Alberts,R.M.Behzinger,L.Lawhorne,andG.Trieber.1970.Rapidbacteriophagesedimen-tationinthepresenceofpolyethyleneglycolanditsapplicationtolarge-scaleviruspurification.Virology40:734-744.VOL.57,1991