toolswww.6633.us

LawrenceBerkeleyNationalLaboratoryLawrenceBerkeleyNationalLaboratoryTitleMetagenomics:DNAsequencingofenvironmentalsamplesPermalinkhttps://escholarship.
org/uc/item/1hg5v2pgAuthorsTringe,SusannahGreenRubin,EdwardM.
PublicationDate2005-09-01PeerreviewedeScholarship.
orgPoweredbytheCaliforniaDigitalLibraryUniversityofCalifornia1Metagenomics:DNAsequencingofenvironmentalsamplesSusannahGreenTringeandEdwardM.
Rubin*DOEJointGenomeInstitute,2800MitchellDrive,WalnutCreek,CA94598,USA*Towhomcorrespondenceshouldbeaddressed:emrubin@lbl.
gov2Whilegenomicshasclassicallyfocusedonpure,easy-to-obtainsamples,suchasmicrobesthatgrowreadilyincultureorlargeanimalsandplants,theseorganismsrepresentbutafractionofthelivingoroncelivingorganismsofinterest.
Manyspeciesaredifficulttostudyinisolation,becausetheyfailtogrowinlaboratoryculture,dependonotherorganismsforcriticalprocesses,orhavebecomeextinct.
DNAsequence-basedmethodscircumventtheseobstacles,asDNAcanbedirectlyisolatedfromliveordeadcellsinavarietyofcontexts,andhaveledtotheemergenceofanewfieldreferredtoasmetagenomics.
Completegenomesequenceshavebeenobtainedfromhundredsoforganisms.
Inthewell-studied,easilymanipulatedorganismstargetedbyearlygenomeprojects,genotypicandphenotypicdatacouldbecomparedandgenome-basedhypothesestestedbyexperiment.
Comparativegenomicsallowedexperiment-basedannotationstobetransferredtonovelgenomes,andquicklygainedprominenceasavaluabletoolforunderstandingbothgenesandgenomes1.
DNAisuniversal,andprotocolsforitspurificationarewellestablished;thoughsomeoptimizationisusuallyrequiredforDNAextractionfromnovelorganisms,theeffortinvolvedisgenerallymuchlessthanthatrequiredtodeveloptechniquesforgeneticmanipulation.
Asaresult,thefocusofsomegenomicsequencinghaschangeddramatically,suchthatDNAsequenceisusedtopredictfeaturesandbehaviorsofotherwisepoorlyunderstoodorganismsaswellastounderstandthegeneticbasisofcharacterizedtraits.
Barrierstogenomesequencingrangefromthelackofsufficientmaterialfortheconstructionofsequencinglibrariestothecostofsequencing.
Improvementsincloning3andsequencingtechnologieshaveconsistentlydecreasedtheamountofstartingmaterialneededforlibraryconstruction,makingDNAsequencingfeasibleforavarietyoforganismsthatareotherwisedifficulttostudy.
Meanwhile,theprogressivereductioninthecostofhigh-throughputsequencinghasmadefeasiblethesequencingoflibrariesconstructedfrommixturesoforganisms,eventhose"contaminated"withgenomesotherthanthatofthetargetedorganism2.
Thishasopenedthedoortosequence-basedstudiesoforganismsandenvironmentspreviouslythoughtinaccessible,includingobligatepathogensandsymbionts,whichcannotsurviveoutsidetheirhosts;environmentalmicrobes,mostofwhichcannotbegrowninpureculture;andancientorganismswhoseonlyrecordisfossilizedremains.
DNAforthesestudiesisextracteddirectlyfromtheorganismsintheirnaturalhabitat,suchashosttissueorsoil,andclonedintosequencingvectors.
Theresultinglibrariescontaingenomefragmentsfromaheterogeneousmixofspecies,strainsandsubpopulations.
Thusthesequencedatafromtheselibrariesharborawealthofinformationoncommunitydynamics,suchasspeciesinteractionsandselectiveprocesses.
ThisarticlefocusesontheinsightsthathaveemergedfromDNAsequencingofnaturallyoccurringpopulationsandcommunities.
Thefirstsectiondescribesmethodologicaladvancesthathaveenabledthesequencingofnaturalpopulations,thesecondsectiongivesexamplesofstudiesthathaveusedthesetechniquesandthethirdsectionsuggestsfuturedirectionsthesestudiesmaytake.
Environmentalnucleicacidanalysis4NaturalsamplescontainDNAinavarietyofpackages,includingfreeDNA,virusparticles,andprokaryoticandeukaryoticcells.
Thesecanbesuspendedinwater,boundtoasolidmatrixlikesoil,orencasedinabiofilmortissue.
ExtractionmethodsmustbechosencarefullybasedonthemediumandtheDNApopulationofinterest.
Aquaticsamplesmustbeconcentrated,typicallybyimpactortangentialflowfiltration,andmayalsobeprefilteredtoremovelargecellsordebris3.
Thechoiceoffiltersizesisacriticalone,ascellsthataresmallerorlargerthanthesizefractiontargetedwillbeinvisibletofurtheranalysis.
Thusfiltrationprotocolscanbechosentoenrichforeukaryoticcells,prokaryoticcells,orviralparticles4,5.
Cellsinsoilsandsedimentsarelesseasilyconcentratedthanaquaticsamplesandoftencontainenzymeinhibitors,suchashumicacids,thatmustberemovedpriortoamplificationorcloning.
Solid-matrixDNAisolationiseitherdirect,inwhichcellsarelysedwithinthesamplematerial,orindirect,inwhichcellsareseparatedfromnoncellularmaterialpriortolysis.
Ineithercase,contaminantsthattendtocopurifywithDNAfromsampleshighinorganicmattercanberemovedbymethodssuchasagarosegelelectrophoresisorcolumnchromatography6,7.
DirectisolationmayalsocaptureDNAfromvirusparticlesorfreeDNAfromdeadcells;whenthesenoncellularDNAsaretheintendedtarget,theycanbedirectlysolubilizedandconcentratedwithoutlysisofcellsinthesample8,9.
ThetechniquesusedtolysecellsmayalsoaffectthecompositionofenvironmentalDNAlibraries,astheharshlysismethodsnecessarytoextractDNAfromeveryorganismwillcausedegradationoftheDNAfromsomeorganisms7.
Hard-to-lysecells,suchasGrampositivebacteria,maythereforebeunder-orover-representedinenvironmentalDNApreparations10.
Oftenthedesireforcompletelysismustbebalancedwiththeneed5forhigh-qualityDNA,especiallywhenpreparinghighmolecularweightDNAforlarge-insertlibraries11,12.
OnceDNAhasbeenobtained,itcanbedirectlyclonedintosmall-insertvectorsforhigh-throughputsequencing(see,forexample,http://www.
jgi.
doe.
gov/sequencing/protocols/).
Alternatively,itcanbeclonedintolarge-insertlibrariesandscreenedforcloneswithactivitiesorgenesofinterest,whicharethensubclonedandsequenced.
DNAsequence-basedinsightsinto"inaccessible"organismsThefirstforaysintosequencingofnaturalsamplesaimedtocharacterizethegenomesoforganismsthatoccurintightassociationwithoneormoreadditionalspecies,andthereforecannotbeeasilystudiedinisolation.
Here,thechallengeistoextracttherelevantsequencefromamixed-specieslibrary,whichmaycontainonlyasmallfractionofclonesfromthetargetspecies.
Avarietyofpre-andpost-sequencing"sifting"techniqueshaveenabledthegenomiccharacterizationoforganismsthatcannotbecultivated,suchasobligatepathogensandsymbionts,andevenlong-extinctspecies.
16SrRNA:alaunchpadfornovelprokaryoticgenomesThegenomicstudyofnaturalcommunitieshasbeenlargelydrivenbyinterestinthe~99%ofmicrobesthatarenoteasilyisolatedinculture.
Thesespeciesareidentifiedbytheir16S/18SsmallsubunitrRNAgenes,whicharecommonlyusedasphylogeneticmarkersbecauseeverycellularorganismcontainsthegene,andvirtuallyallgenevariants6canbeamplifiedbystandardsetsofdegenerateprimers(Box1).
SeveralinvestigatorshaveusedrRNAgenesasastartingpointtoexplorethegenomesofuncultivatedmicrobesvialarge-insertclonesequencing.
Onesuch"phylogeneticanchoring"studyledtothediscoveryofproteorhodopsin,atypeoflight-harvestingprotein,inoceanicbacteria–asurprisenotonlybecausethesemicrobeswerepreviouslybelievedtodependonorganicmatter,notlight,asanenergysource,butbecauserhodopsin-likeproteinshadneverbeenseeninthebacterialdomain13-15.
Othershaveprovidedglimpsesofthegenomesofseveraladditionaluncultivatedprokaryotes,includingcrenarchaeota11,16-18andAcidobacteria10frommultiplehabitats.
Insomecases,thesesequenceshaveprovidedevidenceforunexpectedfunctions11orhorizontalgenetransfers19.
Host-associatedbacteria:genomicinsightsintopathogenesisandsymbiosisThoughdiscussionofuncultivatedmicrobesmostoftenbringsenvironmentalorganismstomind,theseveraluncultivatedmicrobeswhosegenomeshavealreadybeensequencedarelargelyobligatepathogensorsymbionts20-32.
Theamenabilityofthesehost-associatedmicrobestophysicalseparationmakesthemwell-suitedtothisapproach(SeeTable1),incontrasttoorganismsresidingincomplexenvironmentalcommunities.
Thefirstcompletegenomeofanunculturedbacterium,thesyphilisspirocheteTreponemapallidum,wasreleasedin1998–alandmarkingenomesequencing20.
Whilethebacterialoriginofsyphiliswasrecognizedacenturyago,theinfectiousagenthasneverbeenisolatedincontinuousculture.
DNAforsequencingtheintracellularpathogenwasobtainedfromthetestesofinfectedrabbits-some400ofthem-byaseriesoflysisandcentrifugationstepsthateventuallyresultedinanessentiallypurebacterial7preparation(Table1).
Sequenceanalysisimmediatelyidentifiedpotentialcontributorstovirulence,andaidedthedevelopmentofDNA-baseddiagnostics33.
AyearandahalfofpainstakinggrowthincoculturewithhumanfibroblastswasnecessarytoobtainsufficientDNAtosequencethegenomeoftheWhipple'sdiseasebacteriumTropherymawhipplei,whichrevealeddeficienciesthatsuggestedanexplanationforthefailuretopropagateinaxenicculture.
Basedonthesegenomicinsights,Renestoetal.
thenusedastandardtissueculturemedium,supplementedwithaminoacidsimplicatedbythesequenceanalysis,tosuccessfullycultivateT.
whippleiintheabsenceofhostcells,shorteningtheirdoublingtimebyanorderofmagnitude34.
ThisisoneofmanycasesinwhichDNAsequenceinformationhasbeenusedtoimproveculturetechniques,diagnosticsandtherapiesforfastidiousorganisms35-37.
Severalgenomesofobligateintracellularsymbionts,primarilyfrominsecthosts,thatcouldnotbegrownbyconventionalmeanshavealsobeenobtainedbyvariousseparationandpurificationmethods(Table1).
ThefirstwasBuchneraaphidicola24,anE.
colirelativethatprovidesnutrientstosupplementitsaphidhost'srestricteddietofplantsap.
Bacteriomes–specializedsymbiont-harboringorgans-from2000aphidswereisolatedbydissectionpriortocrushingandfiltration,resultinginvirtuallypureBuchneracellsforDNAisolation.
Symbiontsoftsetseflies,fruitflies,carpenterants,anematodeandtwootheraphidspecieshavesincehadtheirgenomessequenced,ashasoneunculturedplantpathogen25-29,31,32.
Ineachproject,techniquessuchasdissection,differentiallysisandpulsed-fieldgelelectrophoresis,oftenincombination,havehelpedenrichforprokaryoticmaterial(Table1);wherereported,between5%and47%ofthesequenceswerehost-derived27-29.
Anotheressentiallycompletesymbiontgenomerecentlyemergedasa8byproductofametazoangenomeproject,asthesequencinglibrarieswereconstructedfromsymbiont-harboringwholeembryos30.
PaleogenomicsEvolutionarybiologydependsheavilyonDNAsequencedatatoreconstructevolutionarypathways,butthesemoleculartreesarelimitedtothemodernspeciesattheendsofthebrancheswhoseDNAisreadilyavailable.
Phylogeneticplacementandhypothesizedphenotypesoftheorganismsatthebranchingnodes,orthebranchesthatterminatebeforethemodernera,arebasedprimarilyuponmorphologicalexaminationoffossilizedspecimens.
Theabilitytosequencegenomesfromancientorganismswouldoffera"genomictimemachine"tostudythesepoorlycharacterizedspecies.
Whenananimaldies,itstissuesarequicklyexploitedasanorganicnutrientsourcebyavarietyofcreatures,particularlysingle-celledmicrobes.
Rarely,conditionsaresuchthatthecarcassescapestotaldecompositionandparts,particularlybone,remainpreserved;however,theDNAcontainedthereinisnotonlydamagedandfragmentedbutmixedwiththegenomesoftheabundantopportunisticmicrobesthathaveinvadedthetissue.
Nonetheless,gentleandrigorouslysterileDNAisolationprocedureshaveallowedthegenerationofverifiablemitochondrialandnuclearsequencefrommaterialssuchasbones,teeth,andcoprolites(fossilizedfecalmaterial)datingtoaslongas50,000yearsago38,39.
Thesestudies,relyingonPCR-amplifiedmitochondrialsequence,havebeenusedtoresolvephylogeneticrelationshipsbetweenextinctandmodernanimals40.
Mitochondriaarepresentinmorethan1000copiespercellandarethereforerelativelyeasilyamplified;thesingle-copynucleargenome,whichcouldofferfarmorephenotypic9information,haveremainedminimallyexploredduetotechnicalhurdles41,42.
Low-costhigh-throughputsequencing,coupledwithametagenomicapproach,nowprovidesameanstoaccessthenucleargenomesofextinctorganismswithoutamplification.
Thiswasrecentlyappliedtotheanalysisofcavebear,Ursusspelaeus,arelativeofmodernbrownandblackbearsthatlivedincavesthroughoutEuropeinthelatePleistocenebutbecameextincttensofthousandsofyearsago.
TheinvestigatorsexploitedametagenomicstrategytodemonstratethepresenceofverifiablecavebearsequenceinlibrariescreatedbydirectlycloningDNAextractedfrom40,000-year-oldbones43.
Alibraryconstructionprotocolthatinvolvedneitherlysisnorshearingenabledthecloningofend-repairedancientDNAisolatedfromcavebearboneandtoothsamples.
Whilecavebearsequenceconstitutedamere1-5%ofthelibrariesdescribedbyNoonanetal.
,thesesequenceswerereadilyidentifiedbytheirhighsequenceidentitytoarelatedcarnivore,dog,whosegenomeisfullysequenced44(Figure1).
Roughly27kbofputativecavebearsequencewasobtained,andPCRamplificationoforthologoussequencesfrommodernblack,brownandpolarbearsverifiedtheiroriginandallowedthereconstructionofaphylogenetictreecongruentwiththatbasedonmitochondrialsequences.
Modernhumancontaminationfromlaboratorypersonnel,asurprisinglylow0.
05%ofclones,waseasilyidentifiedasthisproof-of-principlestudyfocusedonaspecieswhichisreadilydistinguishablefrommodernhuman.
Thesetechniquesopenupthepossibilityofgenomeprojectstargetingextinctspeciesandcouldrevolutionizepaleobiology.
Ourclosesthominidrelatives,theNeanderthals,divergedfrommodernhumansroughly500,000yearsagobutsurviveduntilthelatePleistocene,andnumerousNeanderthalremainsofagescomparabletothe10sequencedcavebearsampleshavebeenfound.
Byprovidingsequencefromanotherhominid,theNeanderthalgenomecoulddefinehuman-specificsequencesandexpandknowledgeofthebiologyofbothHomosapiensandNeanderthals.
High-throughputshotgunsequencingofenvironmentalsamplesEnvironmentalsamplesaremanytimesmorecomplexthansingleorganisms,astheymaycontaintens,hundredsoreventhousandsofdistinctspecies,andwerethereforeuntilrecentlywidelyconsideredunsuitableforhigh-throughputsequencing.
Similarconcernsonceaccompaniedtheapplicationofwhole-genomeshotgun(WGS)sequencingtolargegenomes,asitwasthoughtthatassemblyofWGSreadsintochromosomesandgenomeswouldprovetoocomputationallycomplex.
YetWGShasproventobethemostefficientandeffectiveapproachtogeneratingcompletegenomesbothlargeandsmall,largelyduetocomputationaladvances.
Inthecaseofenvironmentalgenomics,analysistoolshaveonceagainrisentothetask,enablingthesimultaneousstudyofwholeensemblesofgenomesviahigh-throughputsequencing.
Anewperspective,inwhichgenesandgenomesareviewedassubunitsofalargerwhole,ischangingthewayinwhichweunderstandevolutionandadaptation.
Thefirstlarge-scaleenvironmentalshotgunsequencingprojectinterrogatedtheorganismsmakingupanacidminebiofilm45.
Acidminedrainageisanenvironmentallydevastatingconsequenceofcommercialminingwhichresultsfromtheproductionofsulfuricacidwhenpyrite(FeS2)isexposedtoairandwaterduringminingoperations.
Microorganismshavelongbeenrecognizedasimportantplayersinthisprocess,asthe11rate-limitingstepofferric(Fe3+)ionregenerationisslowundersterileconditionsbutcanbegreatlyacceleratedbymicrobesthatderiveenergyfromthereaction(chemolithotrophs)46.
Microbialcommunitiesflourishundertheseseeminglyhostileconditions,formingextensiveunderwaterstreamersandfloatingbiofilmsanchoredinpyriticsediments,butaretypicallyofrelativelylowdiversityasfeworganismscantoleratetheextremeacidity.
Toaddressthephysiologyoftheuncultivatedmicrobesinthemine,Tysonetal.
builtashort-insertgenomiclibraryfrombiofilmDNAandgenerated76.
2millionbasepairsofsequencefromtheresidentbacteriaandarchaea45.
Fromthis,theyassemblednear-completegenomesfortwocommunitymembersandpartialgenomesforthreemore,enablingmetabolicreconstructiontoassesstheroleofeachindividualorganism.
Interestingly,oneorganism,anuncultivatedLeptospirillumgroupIII,wastheonlymemberofthiscommunitythatpossessedthegenesfornitrogenfixation.
Asthisprocessisessentialinsuchanutrient-limitedenvironment,thislow-abundancespeciesappearstobealinchpinforthewholecommunityand,theoretically,apotentialbiologicaltargetforcleanupefforts.
AnotherstudyreportedthemetagenomicsequencingofthesurfacewatermicrobialcommunityoftheSargassoSea,abodyoflow-nutrientwaterintheNorthAtlantic4.
PlanktonicmicrobeswerecollectedfrommultiplelocationsandextractedDNAwasusedtoconstructsevenindependentlibraries,fromwhichatotalofmorethan1.
6GbofDNAsequencewasgenerated.
ReflectingtheunexpectedcomplexityoftheSargassoSeaplanktoniccommunity,just3%ofthissequencewascoveredat3Xdepthormore–eventhoughthisquantityofrawsequencewouldbesufficienttocompleteas12manyas50prokaryoticgenomesat8Xdepth.
Morethan1.
2milliongeneswerefoundtohavesignificantsimilaritytodatabaseentries.
Althoughlessthanathirdcouldbeassignedtentativecellularroles,somefunctionsstoodout,includingnumerousrhodopsin-relatedgenesandgenesinvolvedinphosphorusuptakeandmetabolism,consistentwiththeneedtoefficientlyutilizetheplentifulsunlightandlimitedphosphateavailableinthisenvironment47.
MostofthepredictedgenesintheSargassoSeadatacouldnot,however,bedefinitivelylinkedtoparticularphylogeneticgroups,muchlessindividualspecies.
Thesedatahavesincebeenminedforavarietyofgenesincludingiron-sulfurproteins,chitinases,proteorhodopsins,andelectrontransportproteins48-51.
Eachofthesestudieshasidentifiedgeneshighlydivergentfromknownfamilymembers,highlightingthenoveltyofenvironmentalsequencesascomparedtogenomesequencesofculturedisolates.
WhileacidminesandtheSargassoSearepresentrelativelynutrient-poorenvironments,arecentstudybyTringeetal.
52exploredtwodifferentnutrient-richenvironments:agriculturalsoilanddeepseawhaleskeletons,a.
k.
a.
"whalefalls,"whichsustainthrivingcommunitiesofmicro-andmacro-organismsastheydecompose53.
Thecombinationoftheseenvironmentswiththepreviouslysequencedsamplesspansawiderangeofenvironmentalvariablessuchastemperature,pHandillumination,providingarichtestinggroundforcomparativeanalysis.
Justascomparativegenomicsformsthefoundationformostgenomeannotationefforts,itwasreasonedthatpatternsofgeneabundanceamongenvironmentswouldenhanceunderstandingofboththeenvironmentsandthegeneproducts.
13Genomicsequencingofcomplex,nutrient-richsamplesdidnotresultinassembledgenomes-indeed,itwasestimatedthatseveralbillionbasesofsequencewouldneedtobegeneratedfromacomplexenvironmentlikesoilbeforegenomeswouldbegintoassemble-butdididentifygenefamiliesimportantforsurvivalintheenvironmentssampled.
Inthisgene-centricapproach,eachsequenceobtainedwastermedanEnvironmentalGeneTag(EGT),becauseitcontainedasnippetofsequencepotentiallyencodingaproteinadaptiveforthatenvironment(Box2).
PredictedgenesontheEGTsfromeachsamplewerecomparedwitheachotherandwithsequencesfrompreviousenvironmentalsequencingprojects4,45.
Anumberofcharacterizedanduncharacterizedorthologousgroups,functionalmodulesorbiochemicalprocessesemergedthatwereunevenlydistributedacrossthesamples52.
ThisprovidedanEGT"fingerprint"ofeachenvironmentanddemonstratedthatsimilarenvironments,suchastwowhaleskeletons8000milesapartontheoceanfloor,havesimilargenecontent.
Analysisoffunctionsoverrepresentedinparticularnichesprovideduniqueinsightsintothedemandsplacedonorganismslivingthere.
Oneofthemostsignificantdisparitiesingenedistributiontoemergefromthisanalysiswastheoverabundanceofrhodopsin-likeproteinsintheSargassoSeaascomparedtonon-illuminatedenvironments.
Similarly,asmightbepredictedinhindsight,numeroushomologsofcellobiosephosphorylase,anenzymeinvolvedinthebreakdownofplantmaterial,werefoundinthesoilsample,takennearasilagebunker,butnotintheothersamples.
Apreponderanceofsodiumtransportandosmoregulationproteinsinallthemarinesamples,bothsurfaceanddeepsea,wasconsistentwiththehighsodiumcontentofseawater.
Thesoilsample,bycontrast,containedfarmorepotassiumtransporters,and14biochemicalanalysisrevealedthatpotassiumionsoutnumberedsodiuminthesampleseventoone.
Overall,variationsingenedistributionweremostevidentintransportersandmetabolicenzymes-thosemoleculesmostinvolvedininteractingwith,andpresumablyadaptingto,theenvironment.
Themanyuncharacterizedorthologousgroupsexhibitinghighlyskeweddistributionsacrosssamplesmayfunctioninnicheadaptationandthereforemakepromisingcandidatesforfutureinvestigations.
Withthesecomparativetoolsinhand,researcherscannowinvestigatethefactorsthatinfluencemicrobialcolonizationorthechangesthatoccurinenvironmentsunderstress,withouttheconstraintsondiversitycreatedbytheneedtoassemblegenomes.
FuturedirectionsThegoalsofmetagenomicprojectsvaryconsiderably,fromcharacterizingoneparticularspeciestounderstandingthedynamicsofawholecommunity.
Whilethe"difficulttoaccess"genomeprojectsdescribedhereinmightseemtosharelittleincommonwithenvironmentalprojectsexaminingcomplexcommunities,manyofthemethodsandchallengesoverlap.
Thesetwopreviouslyseparatefieldsarerapidlyconvergingintheseveralmetagenomicprojectsnowtargetingeitherindividualmembersoffree-livingcommunities,suchasmarineCrenarchaeota54,orentirecommunitiesofsymbioticorganisms,suchasthesyntrophicconsortiuminhabitingthemarineoligochaeteOlaviusolgarvensis55(ForinformationontheseandotherongoingprojectsattheDOEJointGenomeInstitute,see).
A"secondhumangenome15project"hasevenbeenproposedtosequencethegenomesofthehuman-associatedmicrobiota56.
Thisreviewhasdescribedmanyinnovationsthathaveimprovedourabilitytostudy"inaccessible"genomes.
However,thecurrentmethodsofDNAisolation,libraryconstruction,sequenceassemblyandbioinformaticanalysisareallstilloptimizedforsingle-genomeanalysisandwilllikelyneedtobemodifiedforapplicationtometagenomicprojects.
DNAisolationandlibraryconstructionThemethodsusedtoisolateDNAfrommixedsamplesandconstructlibrariessubstantiallyaffecttheresultsobtained,ascellsdifferintheirsensitivitytolysisandDNA"cloneability"varieswidely.
Forenvironmentalsamples,particularefforthasbeendevotedtoobtainingDNArepresentativeofallorganismspresent,tobeststudythecommunityasawhole.
Theserepresentativelibrariesareeffectivetoolsforcommunityoverviewsandforcharacterizingthedominantactivitiesinanenvironment52.
However,whencompletegenomesaredesired,representativelibrariesareaninefficientmeansofsequencingnon-dominantcommunitymembers;oneorganismintheSargassoSeastudy,forexample,wassequencedat21Xcoverage4.
Anumberoftechniqueshavebeenusedtonormalizeorenrichenvironmentallibrariesforavarietyofapplications,basedongenericpropertieslikecellsizeorDNAcomposition.
Filtrationisonethathasalreadybeenmentionedasameansofseparatingcellsbasedonsize,particularlyforseparatingprokaryotesfromeukaryotes;ithasalsobeenusedtoseparatemulticellularconsortiafromindividualcells57.
SeparationofDNA16onbisbenzimidazolegradientsallowsfractionationbasedonGCcontent,exploitingthechangeinbuoyantdensitythatoccurswhenbisbenzimidazolebindstoadenineandthymidine58.
Othertechniquesthathavebeenappliedtohost-associatedmicrobesincludedifferentialcentrifugation59,densitygradients25,57,differentiallysis20,pulsed-fieldelectrophoresis32andselectiveuseofrestrictionenzymes60.
Insomecases,aparticularorganismorgroupoforganismsinacommunityisofinterest;forexample,thosethatcarryoutaparticularmetabolicprocessoraremembersofanuncharacterizedphylogeneticgroup.
Successfultargetingoftheseorganismscouldsignificantlyreducetheamountofsequenceneededforgenomecoverageandsimplifyassembly.
Stableisotopeprobing(SIP)holdspromiseasameanstoobtainDNAfromorganismscapableofmetabolizingaparticularsubstrate,andmayserveasavaluablemethodforcommunityfractionation61.
Flowcytometryisahighlyspecificmethodtoisolateorganismsbasedonviability62,membraneproperties63,surfaceproteinexpression64,orSSUrRNAsequence65.
Finally,affinitypurificationmightholdpromiseforseparatingoutsomegroups66,67basedoncellwallcharacteristicsorextracellularmarkers.
BuildinglibrariesfromsuchenrichedDNAwillgreatlyimprovesequencingefficiencyascomparedtowhole-communitylibraries.
Wholegenomeamplificationviaisothermalstranddisplacementcoulddramaticallyopenupthepossibilitiesforsequencingunculturableorganismsbysignificantlyreducingtheamountofstartingmaterialrequiredforlibraryconstruction.
DNAfromprokaryoticandeukaryoticcellshasbeenamplifiedbythistechniqueandusedforavarietyofPCR-andhybridization-basedgenomicanalyses68.
Encouragingresultswererecentlyreportedforametagenomicsample,wherePCRresultsfrom17amplifiedandunamplifiedDNAwerecomparable69.
Short-insertshotgunsequencinglibrarieshavealsobeenconstructedfromwhole-genome-amplifiedsamples70,71;however,ahighrateofsequencingartifactshasthusfarprecludedgenomeassembliesbasedontheselibraries(P.
Richardson,personalcommunication).
Libraryconstructionisapotentiallymajorsourceofbias,assomegenomesegmentsareuncloneableand/orlethaltoE.
coli.
New,highlyparallelnon-Sangersequencingtechnologiesalreadybeingmarketed,suchaspyrosequencing,obviatetheneedforlibrariesofanysort72.
Byeliminatingthismajorsourceofbiaswhiledecreasingtime,effortandexpense,theycouldhaveamajorimpactonthefield;however,thiswillrequiresurmountingkeyobstaclessuchasshortreadlengths.
DataanalysisOneofthemostpressingissuesinmetagenomicsisgenomeassembly,whichiscriticalforsometypesofgenomicanalysis.
Themostbasicobstacletoassemblyissimplythecostofachievingsufficientsequencecoverageofasinglemicrobeinacommunitythatmaycontainhundredsofspecies;however,giventhedroppingcostofsequencing,thismaysoonbelessofaproblem.
Anotherconcernishowassemblyalgorithmswillperformwhenconfrontedwithmixeddatafrommultiplespecies.
Fortunately,experiencesuggeststhatcross-speciesassembliesarenotacommonoccurrence45,exceptinthecaseofhighlyconservedgenessuchasrRNA73.
Perhapsthemostseriouschallengeinassemblinggenomesfrommetagenomicdataispopulationheterogeneity,intheformofsequencepolymorphismsandgenomicrearrangements.
18Assemblyalgorithmsappeartobefairlyrobusttosequencepolymorphisms28,45,74,thoughveryhighpolymorphismcaninterferewithproperassemblyespeciallyincomplexgenomes75.
Genomicrearrangements,however,mayrequireseriousrethinkingofthemeaningandpurposeofgenomeassembly76.
Itisnotyetclearwhatlevelofheterogeneityis"typical":inthelimitedsetofcommunitiesthathavebeenexplored,somepopulationsarevirtuallyclonal,someexhibithighpolymorphism,andsomecontainextensiveinsertions,deletions,andtranslocations4,45,57.
Itwillbeinterestingtoseewhetherheterogeneitycorrelateswithfeatureslikegrowthrate,competitionorcommunitystability.
Oncesequenceshavebeengenerated,betheywholegenomes,largescaffolds,orindividualreads,weoftenwanttoassignthemtophylogeneticgroups.
Forclosedornearlyclosedgenomesscaffoldassignmentisstraightforward,becausefunctionalgenesaredirectlylinkedtophylogeneticmarkerslike16SrRNA.
Butevenunderoptimalconditionseachgenomemaybedividedintomultiplescaffolds,andmanysequences,particularlythosefromlow-abundancecommunitymembers,willremaininsmallcontigsorunassembledreadslackingobviousmarkergenes.
Thesimplestmethodoftaxonomicassignment,bestBLASThit,shouldbeusedwithcaution:itisonlyreliablewhencloserelativesareavailableforcomparison,andisessentiallyuselesswhennorelativeshavebeenfullysequenced77.
Otherfeaturesthathavebeenusedto"bin"scaffoldsorcontigsintotaxonomicgroupsincludeGCcontentandoligonucleotidefrequency,coveragedepth,andsimilaritytosequencedgenomes4,43,45,78.
Anotherfieldthatisinitsinfancyisgenecallinginmetagenomicdatabecausethedataisfragmented,heterogeneous,andabundant.
Homology-basedmethodsarevery19accuratebutnotverysensitive,particularlyforgenomesthatlacksequencedrelatives,andwillalwaysmissnovelgenes,whicharepotentiallythemostinteresting.
Abinitiomethodscanpredictnovelgenes,buttrainingisoptimallyperformedoncompletegenomesandfalsepositiveratesmaybehighevenforassembledgenomes79.
Onemethodforcircumventingthisproblemistouseasamplingofsequencedgenomesasatrainingset,preferablyofasimilarphylogeneticrangeasthespeciesinthesample52,butimprovementscouldalmostcertainlybemadeandthisisanimportantareforfuturework.
Furthervalidationofpotentialnovelgenescanbeobtainedthroughsequenceclustering:predictedproteinsthathavehomologswithinthedatasetarelikelytobevalid4.
Geneannotationisalsoachallengeformetagenomicprojects,astheamountofdatageneratedislikelytobelargeformanualannotation.
Fortunately,thereareseveralhigh-qualityautomatedannotationtoolsforcompletemicrobialgenomes,suchasERGO80,GenDb81andPRIAM82.
Ingeneral,thesecanbeadaptedwithminimalefforttometagenomicdatasets;accuracy,however,isalwaysaconcernasnoautomatedmethodscanfullyreplacemanualannotation.
Thegreatestimprovementsinaccuracyarelikelytoresultfromthefurtherproductionofhigh-qualitycompletegenomes,particularlyinphylogeneticgroups,suchasChloroflexiandAcidobacteria,thatarewell-representedintheenvironmentbutpoorlyrepresentedinsequencedatabases83.
Suchhigh-qualitygenomedatawillprovidebettersubstratesforhomologysearches.
20ConclusionsGenomesequencinghasmadeinvaluablecontributionstoevolutionarybiology,medicine,andagriculturalscienceandisrapidlybeingadaptedtostudiesoforganismsintheirnaturalhabitats.
Suchstudiesofferanumberofuniquebenefitsbeyondthoseoftraditionalgenomicstudiesofclonallaboratorystrains.
ThemostobviousbenefitofsequencingDNAfromnaturalsamplesistheabilitytoaccessamuchwiderrangeofgenomes.
Manyorganismsfailto"reproduceincaptivity"andthereforecannotbesubjectedtolaboratorymanipulationandgenomicstudy.
Theseincludenotonlyexoticgroups(e.
g.
Nanoarchaeota),butmanycloserelativesofcultivablemicrobes.
Othersspeciesareextinct,andthereforecannotprovidecleanmaterialforDNAisolation–mostnotably,ancienthominidssuchastheNeanderthalswhichmaysoonbethetargetoftheirown"humangenomeproject.
"Alessimmediatelyapparentadvantageofthistechniqueistheabilitytocapturethegenomicdiversitywithinanaturalpopulation.
WhileDNAsequencefromaclonalstrainiseasiertogenerateandassemble,anindividualgenomerepresentsasinglesnapshotofthepopulationfromwhichitderives.
Bothclonalstrainsequencingandenvironmentalstudiesrevealthattherecanbesubstantialvariationingenecontent,geneorderandnucleotidesequenceevenwithinpopulationsthoughtofasasinglespecies84-86.
Sequencefromnaturalsamplesreflectsthisvariationandrevealstheprevalenceofspecificsubgroups.
Byofferingaccesstogenomesofhard-to-studyorganisms,environmentalgenomicsanditsoffshootshaveadvancedourunderstandingofspeciesinterrelationships,environmentalnicheadaptation,andhumanevolutionaryhistory.
Technologiesnow21underdevelopmentwillcontinuetolowerthebarrierstogenomesequencing,allowingthestudyofeverscarcerandmorecomplexsamplesandvastlyexpandingtherangeofspeciesonthegenomicsradar.
Definitions:Metagenomics:thegenomicanalysisofassemblagesoforganisms.
Meta-isusedtoindicateacollectionofsimilaritems,asinmeta-analysis87;genomicsisthestudyofgenomes.
Grampositivebacteria:MembersoftheActinobacteriaandFirmicutesphyla,whichhaveasinglemembraneandathickcellwallmadeofcross-linkedpeptidoglycanandthereforecanbestainedwiththeGramstainingprocedure.
Phylogeneticanchoring:Atechniquethatinvolvesscreeninglarge-insertlibrariesmadefromenvironmentalDNAforclonescontainingphylogeneticmarkergenes,andsequencingthoseclonesintheirentirety.
Horizontalgenetransfer:Thetransferofgeneticmaterialfromonespeciestoanother.
Axenic:Apurecultureofasinglespeciesofmicroorganism.
Metagenomic:Atermusedtodescribetechniquesthatcharacterizethegenomesofwholecommunitiesoforganismsratherthanindividualspecies.
whole-genomeshotgun:AnapproachtogenomicsequencingthatinvolvesbreakingtheDNAupintosmallpiecesandcloningthemintovectors,thensequencingclonesatrandom.
Biofilm:Alayeredaggregateofmicroorganisms.
densitygradient:Asolutioninwhichtheconcentrationofthesoluteislowestatthetopandgraduallybecomesmoredenseasitgetsdeeper.
differentiallysis:AtechniquethatusesconditionsthatwillonlylysecertaincellssothattheDNAfromthosecellscanbeisolatedfromothercellsinacommunity.
pulsed-fieldelectrophoresis:TheuseofpulsedelectricfieldsofalternatingpolaritytoseparatelargefragmentsofDNA.
Flowcytometry:Atechniquethatmeasuresthefluorescenceofindividualcellsastheypassthroughalaserbeaminanindividualstream.
22fluorescenceinsituhybridization(FISH):AtechniquethatusesfluorescentlylabeledDNAprobesthathybridizetocellularDNAorRNAtolabelindividualcellsthatcanbeexaminedunderamicroscope.
affinitypurification:Ameansofpurifyingcellsormoleculesbasedonspecificbindingtoaproteinorothermoleculethathasbeenimmobilizedonasolidsubstratelikebeadsoracolumn.
Isothermalstranddisplacement:ADNAamplificationtechniqueusingrollingcircleamplificationwithphi29DNApolymerasetogeneratelargequantitiesofDNAwithoutthermalcycling.
Pyrosequencing:aDNAsequencingtechniquethatreliesondetectionofpyrophosphatereleaseuponnucleotideincorporationratherthanchainterminationwithdideoxynucleotides.
Contig:acontinuousstretchofDNAsequenceassembledfrommultipleindependentsequencingreads.
Methanogens:agroupofhydrogen-consumingArchaeathatgeneratemethanebyreductionofcarbondioxide.
23References1.
Boffelli,D.
,Nobrega,M.
A.
&Rubin,E.
M.
Comparativegenomicsatthevertebrateextremes.
NatRevGenet5,456-65(2004).
2.
Shendure,J.
,Mitra,R.
D.
,Varma,C.
&Church,G.
M.
Advancedsequencingtechnologies:methodsandgoals.
NatRevGenet5,335-44(2004).
3.
Somerville,C.
C.
,Knight,I.
T.
,Straube,W.
L.
&Colwell,R.
R.
Simple,rapidmethodfordirectisolationofnucleicacidsfromaquaticenvironments.
ApplEnvironMicrobiol55,548-54(1989).
4.
Venter,J.
C.
etal.
EnvironmentalgenomeshotgunsequencingoftheSargassoSea.
Science304,66-74(2004).
ThisprojecttosequencetheentiremetagenomeoftheSargassoSeasurfacewatersrevealedunexpectedcommunitycomplexityandsequencediversity.
5.
Breitbart,M.
etal.
Genomicanalysisofunculturedmarineviralcommunities.
ProcNatlAcadSciUSA99,14250-5(2002).
6.
LaMontagne,M.
G.
,Michel,F.
C.
,Jr.
,Holden,P.
A.
&Reddy,C.
A.
EvaluationofextractionandpurificationmethodsforobtainingPCR-amplifiableDNAfromcompostformicrobialcommunityanalysis.
JMicrobiolMethods49,255-64(2002).
7.
vonWintzingerode,F.
,Gobel,U.
B.
&Stackebrandt,E.
Determinationofmicrobialdiversityinenvironmentalsamples:pitfallsofPCR-basedrRNAanalysis.
FEMSMicrobiolRev21,213-29(1997).
8.
Kolman,C.
J.
&Tuross,N.
AncientDNAanalysisofhumanpopulations.
AmJPhysAnthropol111,5-23(2000).
9.
Breitbart,M.
etal.
Metagenomicanalysesofanunculturedviralcommunityfromhumanfeces.
JBacteriol185,6220-3(2003).
10.
Liles,M.
R.
,Manske,B.
F.
,Bintrim,S.
B.
,Handelsman,J.
&Goodman,R.
M.
AcensusofrRNAgenesandlinkedgenomicsequenceswithinasoilmetagenomiclibrary.
ApplEnvironMicrobiol69,2684-91(2003).
11.
Stein,J.
L.
,Marsh,T.
L.
,Wu,K.
Y.
,Shizuya,H.
&DeLong,E.
F.
Characterizationofuncultivatedprokaryotes:isolationandanalysisofa40-kilobase-pairgenomefragmentfromaplanktonicmarinearchaeon.
JBacteriol178,591-9(1996).
12.
Berry,A.
E.
,Chiocchini,C.
,Selby,T.
,Sosio,M.
&Wellington,E.
M.
IsolationofhighmolecularweightDNAfromsoilforcloningintoBACvectors.
FEMSMicrobiolLett223,15-20(2003).
13.
Suzuki,M.
T.
,Beja,O.
,Taylor,L.
T.
&Delong,E.
F.
PhylogeneticanalysisofribosomalRNAoperonsfromuncultivatedcoastalmarinebacterioplankton.
EnvironMicrobiol3,323-31(2001).
14.
Schmidt,T.
M.
,DeLong,E.
F.
&Pace,N.
R.
Analysisofamarinepicoplanktoncommunityby16SrRNAgenecloningandsequencing.
JBacteriol173,4371-8(1991).
15.
Beja,O.
etal.
Bacterialrhodopsin:evidenceforanewtypeofphototrophyinthesea.
Science289,1902-6(2000).
24Aseminalpaperinmetagenomics,thisstudyidentifiedanovelproteinonaBACfromtheuncultivatedSAR86groupofbacterioplanktonthatwaslaterrevealedtorepresentapreviouslyunknown,widespreadgroupofecologicallyimportantlight-harvestingproteins.
16.
Beja,O.
etal.
Comparativegenomicanalysisofarchaealgenotypicvariantsinasinglepopulationandintwodifferentoceanicprovinces.
ApplEnvironMicrobiol68,335-45(2002).
17.
Lopez-Garcia,P.
,Brochier,C.
,Moreira,D.
&Rodriguez-Valera,F.
Comparativeanalysisofagenomefragmentofanuncultivatedmesopelagiccrenarchaeoterevealsmultiplehorizontalgenetransfers.
EnvironMicrobiol6,19-34(2004).
18.
Quaiser,A.
etal.
Firstinsightintothegenomeofanuncultivatedcrenarchaeotefromsoil.
EnvironMicrobiol4,603-11(2002).
19.
Quaiser,A.
etal.
Acidobacteriaformacoherentbuthighlydiversegroupwithinthebacterialdomain:evidencefromenvironmentalgenomics.
MolMicrobiol50,563-75(2003).
20.
Fraser,C.
M.
etal.
CompletegenomesequenceofTreponemapallidum,thesyphilisspirochete.
Science281,375-88(1998).
Thispaperreportedthefirstgenomeofamicrobethatcouldnotbegrownincontinuouspureculture.
21.
Andersson,S.
G.
etal.
ThegenomesequenceofRickettsiaprowazekiiandtheoriginofmitochondria.
Nature396,133-40(1998).
22.
Cole,S.
T.
etal.
Massivegenedecayintheleprosybacillus.
Nature409,1007-11(2001).
23.
Bentley,S.
D.
etal.
SequencingandanalysisofthegenomeoftheWhipple'sdiseasebacteriumTropherymawhipplei.
Lancet361,637-44(2003).
24.
Shigenobu,S.
,Watanabe,H.
,Hattori,M.
,Sakaki,Y.
&Ishikawa,H.
GenomesequenceoftheendocellularbacterialsymbiontofaphidsBuchnerasp.
APS.
Nature407,81-6(2000).
Thispaperreportedthefirstcompletegenomeofanuncultivatedintracellularsymbiontandrevealedsignificantgenomereduction.
25.
Tamas,I.
etal.
50millionyearsofgenomicstasisinendosymbioticbacteria.
Science296,2376-9(2002).
26.
Akman,L.
etal.
Genomesequenceoftheendocellularobligatesymbiontoftsetseflies,Wigglesworthiaglossinidia.
NatGenet32,402-7(2002).
27.
Gil,R.
etal.
ThegenomesequenceofBlochmanniafloridanus:comparativeanalysisofreducedgenomes.
ProcNatlAcadSciUSA100,9388-93(2003).
28.
vanHam,R.
C.
etal.
ReductivegenomeevolutioninBuchneraaphidicola.
ProcNatlAcadSciUSA100,581-6(2003).
29.
Wu,M.
etal.
PhylogenomicsofthereproductiveparasiteWolbachiapipientiswMel:astreamlinedgenomeoverrunbymobilegeneticelements.
PLoSBiol2,E69(2004).
30.
Salzberg,S.
L.
etal.
SerendipitousdiscoveryofWolbachiagenomesinmultipleDrosophilaspecies.
GenomeBiol6,R23(2005).
31.
Foster,J.
etal.
TheWolbachiagenomeofBrugiamalayi:endosymbiontevolutionwithinahumanpathogenicnematode.
PLoSBiol3,e121(2005).
2532.
Oshima,K.
etal.
Reductiveevolutionsuggestedfromthecompletegenomesequenceofaplant-pathogenicphytoplasma.
NatGenet36,27-9(2004).
33.
Liu,H.
,Rodes,B.
,Chen,C.
Y.
&Steiner,B.
Newtestsforsyphilis:rationaldesignofaPCRmethodfordetectionofTreponemapalliduminclinicalspecimensusinguniqueregionsoftheDNApolymeraseIgene.
JClinMicrobiol39,1941-6(2001).
34.
Renesto,P.
etal.
Genome-baseddesignofacell-freeculturemediumforTropherymawhipplei.
Lancet362,447-9(2003).
UsinginformationonT.
whipplei'smetabolicdeficienciesrevealedbyitsgenomesequence,investigatorssuccessfullycreatedthefirstpureculturesystemforthisorganismandreduceditsinvitrogenerationtimebyafactorof15.
35.
Fenollar,F.
&Raoult,D.
Moleculargeneticmethodsforthediagnosisoffastidiousmicroorganisms.
Apmis112,785-807(2004).
36.
Ogata,H.
&Claverie,J.
M.
Metagrowth:anewresourceforthebuildingofmetabolichypothesesinmicrobiology.
NucleicAcidsRes33DatabaseIssue,D321-4(2005).
37.
Lemos,E.
G.
,Alves,L.
M.
&Campanharo,J.
C.
Genomics-baseddesignofdefinedgrowthmediafortheplantpathogenXylellafastidiosa.
FEMSMicrobiolLett219,39-45(2003).
38.
Hofreiter,M.
,Serre,D.
,Poinar,H.
N.
,Kuch,M.
&Paabo,S.
AncientDNA.
NatRevGenet2,353-9(2001).
39.
Cooper,A.
etal.
Completemitochondrialgenomesequencesoftwoextinctmoasclarifyratiteevolution.
Nature409,704-7(2001).
40.
Hofreiter,M.
etal.
Evidenceforreproductiveisolationbetweencavebearpopulations.
CurrBiol14,40-3(2004).
41.
Poinar,H.
,Kuch,M.
,McDonald,G.
,Martin,P.
&Paabo,S.
Nucleargenesequencesfromalatepleistoceneslothcoprolite.
CurrBiol13,1150-2(2003).
42.
Greenwood,A.
D.
,Capelli,C.
,Possnert,G.
&Paabo,S.
NuclearDNAsequencesfromlatePleistocenemegafauna.
MolBiolEvol16,1466-73(1999).
43.
Noonan,J.
P.
etal.
GenomicSequencingofPleistoceneCaveBears.
Science(2005).
ThefirstreportofDNAsequencefromanextinctspeciesgeneratedwithoutPCRamplification.
44.
Kirkness,E.
F.
etal.
Thedoggenome:surveysequencingandcomparativeanalysis.
Science301,1898-903(2003).
45.
Tyson,G.
W.
etal.
Communitystructureandmetabolismthroughreconstructionofmicrobialgenomesfromtheenvironment.
Nature428,37-43(2004).
Thispaperreportsthefirstassembledgenomestoemergefromshotgunsequencingofenvironmentalsamples,allowingmetabolicreconstructionofcommunitymembers.
46.
Johnson,D.
B.
&Hallberg,K.
B.
Themicrobiologyofacidicminewaters.
ResMicrobiol154,466-73(2003).
47.
Wu,J.
,Sunda,W.
,Boyle,E.
A.
&Karl,D.
M.
PhosphatedepletioninthewesternNorthAtlanticOcean.
Science289,759-62(2000).
2648.
McDonald,A.
E.
&Vanlerberghe,G.
C.
AlternativeoxidaseandplastoquinolterminaloxidaseinmarineprokaryotesoftheSargassoSea.
Gene349,15-24(2005).
49.
Sabehi,G.
,Beja,O.
,Suzuki,M.
T.
,Preston,C.
M.
&DeLong,E.
F.
DifferentSAR86subgroupsharbourdivergentproteorhodopsins.
EnvironMicrobiol6,903-10(2004).
50.
Meyer,J.
Miraculouscatchofiron-sulfurproteinsequencesintheSargassoSea.
FEBSLett570,1-6(2004).
51.
LeCleir,G.
R.
,Buchan,A.
&Hollibaugh,J.
T.
Chitinasegenesequencesretrievedfromdiverseaquatichabitatsrevealenvironment-specificdistributions.
ApplEnvironMicrobiol70,6977-83(2004).
52.
Tringe,S.
G.
etal.
ComparativeMetagenomicsofMicrobialCommunities.
Science308,554-557(2005).
Thisstudyrevealedthatdifferencesingenecontentamongcommunitiesareapparenteveninunassembledgenomicdata.
53.
Smith,C.
R.
&Baco,A.
R.
EcologyofWhaleFallsattheDeep-seaFloor.
inOceanographyandMarineBiology:anAnnualReview,Vol.
41(eds.
Gibson,R.
N.
&Atkinson,R.
J.
A.
)311-354(Taylor&Francis,2003).
54.
Karner,M.
B.
,DeLong,E.
F.
&Karl,D.
M.
ArchaealdominanceinthemesopelagiczoneofthePacificOcean.
Nature409,507-10(2001).
55.
Dubilier,N.
etal.
Endosymbioticsulphate-reducingandsulphide-oxidizingbacteriainanoligochaeteworm.
Nature411,298-302(2001).
56.
Relman,D.
A.
&Falkow,S.
Themeaningandimpactofthehumangenomesequenceformicrobiology.
TrendsMicrobiol9,206-8(2001).
57.
Hallam,S.
J.
etal.
Reversemethanogenesis:testingthehypothesiswithenvironmentalgenomics.
Science305,1457-62(2004).
AgenomicanalysisofunculturedArchaeafromdeep-seasedimentsthatprovidedevidencefora"reverse-methanogenesis"mechanismofanaerobicmethaneoxidation.
58.
Nusslein,K.
&Tiedje,J.
M.
CharacterizationofthedominantandraremembersofayoungHawaiiansoilbacterialcommunitywithsmall-subunitribosomalDNAamplifiedfromDNAfractionatedonthebasisofitsguanineandcytosinecomposition.
ApplEnvironMicrobiol64,1283-9(1998).
59.
Waters,E.
etal.
ThegenomeofNanoarchaeumequitans:insightsintoearlyarchaealevolutionandderivedparasitism.
ProcNatlAcadSciUSA100,12984-8(2003).
60.
Garcia-Chapa,M.
,Batlle,A.
,Rekab,D.
,Rosquete,M.
R.
&Firrao,G.
PCR-mediatedwholegenomeamplificationofphytoplasmas.
JMicrobiolMethods56,231-42(2004).
61.
Dumont,M.
G.
&Murrell,J.
C.
Stableisotopeprobing-linkingmicrobialidentitytofunction.
NatRevMicrobiol3,499-504(2005).
62.
Bernard,L.
etal.
Anewapproachtodeterminethegeneticdiversityofviableandactivebacteriainaquaticecosystems.
Cytometry43,314-21(2001).
63.
Park,H.
S.
,Schumacher,R.
&Kilbane,J.
J.
,2nd.
Newmethodtocharacterizemicrobialdiversityusingflowcytometry.
JIndMicrobiolBiotechnol32,94-102(2005).
2764.
Gu,F.
etal.
Insituandnon-invasivedetectionofspecificbacterialspeciesinoralbiofilmsusingfluorescentlylabeledmonoclonalantibodies.
JMicrobiolMethods62,145-60(2005).
65.
Sekar,R.
,Fuchs,B.
M.
,Amann,R.
&Pernthaler,J.
Flowsortingofmarinebacterioplanktonafterfluorescenceinsituhybridization.
ApplEnvironMicrobiol70,6210-9(2004).
66.
Lin,Y.
S.
,Tsai,P.
J.
,Weng,M.
F.
&Chen,Y.
C.
AffinityCaptureUsingVancomycin-BoundMagneticNanoparticlesfortheMALDI-MSAnalysisofBacteria.
AnalChem77,1753-60(2005).
67.
Bundy,J.
L.
&Fenselau,C.
Lectinandcarbohydrateaffinitycapturesurfacesformassspectrometricanalysisofmicroorganisms.
AnalChem73,751-7(2001).
68.
Hawkins,T.
L.
,Detter,J.
C.
&Richardson,P.
M.
Wholegenomeamplification--applicationsandadvances.
CurrOpinBiotechnol13,65-7(2002).
69.
Erwin,D.
P.
etal.
Diversityofoxygenasegenesfrommethane-andammonia-oxidizingbacteriaintheEasternSnakeRiverPlainaquifer.
ApplEnvironMicrobiol71,2016-25(2005).
70.
Detter,J.
C.
etal.
Isothermalstrand-displacementamplificationapplicationsforhigh-throughputgenomics.
Genomics80,691-8(2002).
71.
Kwon,Y.
M.
&Cox,M.
M.
Improvedefficacyofwholegenomeamplificationfrombacterialcells.
Biotechniques37,40,42,44(2004).
72.
Margulies,M.
etal.
Genomesequencinginmicrofabricatedhigh-densitypicolitrereactors.
Nature(2005).
73.
DeLong,E.
F.
Microbialcommunitygenomicsintheocean.
NatRevMicrobiol3,459-69(2005).
74.
Dehal,P.
etal.
ThedraftgenomeofCionaintestinalis:insightsintochordateandvertebrateorigins.
Science298,2157-67(2002).
75.
Holt,R.
A.
etal.
ThegenomesequenceofthemalariamosquitoAnophelesgambiae.
Science298,129-49(2002).
76.
Allen,E.
E.
&Banfield,J.
F.
Communitygenomicsinmicrobialecologyandevolution.
NatRevMicrobiol3,489-98(2005).
77.
Koski,L.
B.
&Golding,G.
B.
TheclosestBLASThitisoftennotthenearestneighbor.
JMolEvol52,540-2(2001).
78.
Teeling,H.
,Waldmann,J.
,Lombardot,T.
,Bauer,M.
&Glockner,F.
O.
TETRA:aweb-serviceandastand-aloneprogramfortheanalysisandcomparisonoftetranucleotideusagepatternsinDNAsequences.
BMCBioinformatics5,163(2004).
79.
McHardy,A.
C.
,Goesmann,A.
,Puhler,A.
&Meyer,F.
Developmentofjointapplicationstrategiesfortwomicrobialgenefinders.
Bioinformatics20,1622-31(2004).
80.
Overbeek,R.
etal.
TheERGOgenomeanalysisanddiscoverysystem.
NucleicAcidsRes31,164-71(2003).
81.
Meyer,F.
etal.
GenDB--anopensourcegenomeannotationsystemforprokaryotegenomes.
NucleicAcidsRes31,2187-95(2003).
82.
Claudel-Renard,C.
,Chevalet,C.
,Faraut,T.
&Kahn,D.
Enzyme-specificprofilesforgenomeannotation:PRIAM.
NucleicAcidsRes31,6633-9(2003).
2883.
Hugenholtz,P.
Exploringprokaryoticdiversityinthegenomicera.
GenomeBiol3,REVIEWS0003(2002).
Aprovocativediscussionoftheproblemsofculturebiasandtheneedforgenomicinvestigationofunderrepresentedbacterialandarchaealphyla.
84.
Thompson,J.
R.
etal.
Genotypicdiversitywithinanaturalcoastalbacterioplanktonpopulation.
Science307,1311-3(2005).
85.
Spencer,D.
H.
etal.
Whole-genomesequencevariationamongmultipleisolatesofPseudomonasaeruginosa.
JBacteriol185,1316-25(2003).
86.
Rocap,G.
,Distel,D.
L.
,Waterbury,J.
B.
&Chisholm,S.
W.
ResolutionofProchlorococcusandSynechococcusecotypesbyusing16S-23SribosomalDNAinternaltranscribedspacersequences.
ApplEnvironMicrobiol68,1180-91(2002).
87.
Handelsman,J.
Metagenomics:applicationofgenomicstounculturedmicroorganisms.
MicrobiolMolBiolRev68,669-85(2004).
88.
Stahl,D.
A.
,Lane,D.
J.
,Olsen,G.
J.
&Pace,N.
R.
AnalysisofHydrothermalVent--AssociatedSymbiontsbyRibosomalRNASequences.
Science224,409-411(1984).
89.
Stahl,D.
A.
,Lane,D.
J.
,Olsen,G.
J.
&Pace,N.
R.
CharacterizationofaYellowstonehotspringmicrobialcommunityby5SrRNAsequences.
ApplEnvironMicrobiol49,1379-84(1985).
90.
Giovannoni,S.
J.
,Britschgi,T.
B.
,Moyer,C.
L.
&Field,K.
G.
GeneticdiversityinSargassoSeabacterioplankton.
Nature345,60-3(1990).
91.
Weisburg,W.
G.
,Barns,S.
M.
,Pelletier,D.
A.
&Lane,D.
J.
16SribosomalDNAamplificationforphylogeneticstudy.
JBacteriol173,697-703(1991).
92.
Amann,R.
I.
,Ludwig,W.
&Schleifer,K.
H.
Phylogeneticidentificationandinsitudetectionofindividualmicrobialcellswithoutcultivation.
MicrobiolRev59,143-69(1995).
93.
Theron,J.
&Cloete,T.
E.
Moleculartechniquesfordeterminingmicrobialdiversityandcommunitystructureinnaturalenvironments.
CritRevMicrobiol26,37-57(2000).
AcknowledgementsThisworkwasperformedundertheauspicesoftheUSDepartmentofEnergy'sOfficeofScience,BiologicalandEnvironmentalResearchProgramandthebytheUniversityofCalifornia,LawrenceLivermoreNationalLaboratory,LawrenceBerkeleyNationalLaboratoryandLosAlamosNationalLaboratoryandSGTwassupportedbyanNIHNRSATrainingandFellowshipgrant(THL007279F).
WewouldliketothankPhilHugenholtzandTanjaWoykeforhelpfulcommentsonthemanuscript.
Onlinelinks:ARB:http://www.
arb-home.
de/RibosomalDatabaseProject:http://rdp.
cme.
msu.
edu/index.
jsp29Box1:16SrRNAanalysisofmicrobialcommunitiesTTTGTAAA-TCTTCAGATAA.
.
.
TTTGTCAAGTCTTTGGTGAA.
.
.
TTTGTCAAGTCTTTGGTGAA.
.
.
.
.
.
EnvironmentalsamplesGenomicDNA16SrRNAsequencesPhylogenetictreesDNAextractionPCRandsequencingSequencecomparisonInthe1980senvironmentalmicrobiologistsrealizedthatonlyasmallfractionofthemicroscopicallyobservableorganismsinasamplewerecapableofcolonyformation.
PioneeringexperimentsbyNormanPaceandcolleaguesrevealed,throughribosomalRNA(rRNA)sequencing,thatthese"unculturable"microbesrepresentednovelspeciesoftenonlydistantlyrelatedtoknown,culturedlineages14,88,89.
AnrRNAsequencecanserveasauniquemolecular"barcode"toidentifyanorganismandplaceitinanevolutionarycontext,providingafirstglimpseintothebroaddiversityinvisibletoculture-basedapproaches.
Thelabor-intensivemethodsinitiallyused,suchasdirectsequencingofisolated5SrRNAorscreeningofgenomiclibrariespriortosequencing,wereeventuallysupplantedbyPCR-basedmethods.
Thisisbecausewell-conservedsequencesthatparticipateinsecondarystructureformationcanbetargetedforamplificationby"universal"primerstogenerateclonelibraries90,91.
Inthesestudies,DNAisextracteddirectlyfromanenvironmentalsamplesuchasoceanwater,soil,orabiofilm,andthe16SgenesofthecommunitymicrobesarethenamplifiedfromthemixedgenomicDNAusingPCR(forreview,see92or93).
ThePCRproductsareclonedintovectorsandsequenced,producingrRNA"signatures"forthemicrobesthatwerepresentinthesample.
Comparisonofthesesequencesagainstdatabasesof16SribosomalRNAgenesallowsthemtobephylogeneticallyclassified.
ThefrequenciesofparticularSSUrRNAclonesequencesprovidearoughpreliminaryestimateofthecommunitystructure,assequencesfromdominantcommunitymembersshouldbemoreabundant.
Insomecases,thepresenceofSSUrRNAsequencesfromspecializedcladessuchasmethanogenscansuggestfunctionalactivitiesaswell.
Thedownside,however,isthatevenspeciesthatarecloselyrelatedbasedonSSUrRNA30sequencecanhaveverydifferentlifestyles,andthephylogeneticpositionoforganismswithnoculturedcloserelativesfrequentlyofferslittletonoinsightintotheirphenotypiccharacteristics.
16SrRNAgeneshavebeenamplified,clonedandsequencedfromthousandsofdistinctenvironmentalniches,yetthesesurveysroutinelycontinuetoidentifyuniquenewbacterialandarchaealtaxa.
Tools(suchasARB,,andEstimateS)anddatabases(suchastheRibosomalDatabaseProject,)havebeendevelopedtomanageandanalyzethisfloodofdata.
31GenomeSizeHost/HabitatSeparationtechniqueReferenceTreponemapallidum1.
1MbHuman,rabbitDissection,differentiallysis20Rickettsiaprowazekii1.
1MbHuman,chickenDifferentialcentrifugation21Mycobacteriumleprae3.
3MbHuman,armadilloGradientcentrifugation22Tropherymawhipplei0.
9MbHumanDifferentialcentrifugation23Buchneraaphidicolasp.
APS0.
6MbAphid(A.
pisum)Dissection,differentiallysis,filtration24Buchneraaphidicolasp.
Sg0.
6MbAphid(S.
graminum)Gradientcentrifugation25Wigglesworthiaglossinidiabrevipalpis0.
7MbTsetsefly(G.
brevipalpis)Dissection,differentiallysis26Blochmanniafloridanus0.
7MbCarpenterantsDifferentiallysis27Buchneraaphidicolasp.
BBp0.
6MbAphid(B.
pistaciae)Differentiallysis,filtration28WolbachiapipientiswMel1.
27MbFly(D.
melanogaster)Differentiallysis,pulsed-fieldelectrophoresis29WolbachiapipientiswAna1.
4MbFly(D.
ananassae)None30WolbachiapipientiswBm1.
1MbParasiticnematodeworm(B.
malayi)BAClibraryscreening31Phytoplasmaasteris,lineOY-M0.
9MbPlantsandleafhoppersDifferentiallysis,pulsed-fieldelectrophoresis32Nanoarchaeumequitans0.
5MbIgnicoccussp.
cocultureDifferentialcentrifugation59FerroplasmaacidarmanustypeII1.
8MbAcidminebiofilmNone45Leptospirillumsp.
GroupII2.
2MbAcidminebiofilmNone45Burkholderiasp.
~8.
8MbSargassoSeaFiltration4Shewanellasp.
~5MbSargassoSeaFiltration4Shewanellasp.
~5MbSargassoSeaFiltration4Table:Assembledgenomesofuncultivatedmicrobes32Figure1:AncientDNAsequencingCarnivoreMicrobeHumanUnknownEtc…GACTGCATCCGACGT…CTACTGATTATCGA…ATGACTNCTCGATCG…CGTTAACGTACG…BLASTE>1e-3;>30bpsequencingGenomicsequenceofextinctorganismscanbeobtainedfromtheDNAinancientremainssuchasbone.
Bonesarefirstmilledintopowder,thenimmersedinasolutiontoextracttheDNA.
ThedamagedendsoftheDNAmoleculesarethenrepairedenzymaticallyandclonedintoasequencingvector.
Theclonesarethensequencedaccordingtostandardprotocols,andprobablespeciesoforigindeterminedbyBLAST.
InthestudybyNoonanetal.
43,upto5%oftheclonesfoundtheirclosestmatchinthedoggenome,acarnivorecloselyrelatedtobears.
Onlyafew(~0.
05%)ofthereadswereofhumanorigin,while10-20%onlyhadsignificantmatchestoenvironmentalsequences.
33Box2:EnvironmentalGeneTagsAdaptivegeneforhabitatAAdaptivegeneforhabitatBEssentialgeneABEGTsEachorganisminacommunityhasauniquesetofgenesinitsgenome;thecombinedgenomesofallthecommunitymembersmakeupthemetagenome.
Essentialgenesarepresentineachindividualgenome,regardlessofenvironment,andwillthusoccurfrequentlyinthemetagenome.
Amongnonessentialgenes,thosethatareadaptiveforaparticularnichewillappearinthegenomesofmanyorganismsinthatenvironment,whilethosethatarenotadaptivemayappearatlowabundances.
EnvironmentalGeneTags(EGTs)areshortsequencesfromtheDNAofmicrobialcommunitiesthatcontainfragmentsoffunctionalgenes.
EachEGTderivesfromadifferentmemberofthecommunity,butgenesthatareimportantforsurvivalandadaptationwillbepresentinmanygenomes(possiblyinmorethanonecopy)andwillthereforeappearrepeatedlyintheEGTdata.
WhenthegeneabundancesintheEGTdataarecomparedbetweenenvironments,genesthatareadaptiveinonlyonecontextaremoreabundantinthatenvironment.