Copyright(cid:211) 1999bytheGeneticsSocietyofAmerica Genetic Diversity of Archaea in Deep-Sea Hydrothermal Vent Environments Ken Takai and Koki Horikoshi Deep-SeaMicroorganismsResearchGroup,JapanMarineScienceandTechnologyCenter,Yokosuka237-0061,Japan ManuscriptreceivedJanuary26,1999 AcceptedforpublicationApril12,1999 ABSTRACT Molecularphylogeneticanalysisofnaturallyoccurringarchaealcommunitiesindeep-seahydrothermal ventenvironmentswascarriedoutbyPCR-mediatedsmallsubunitrRNAgene(SSUrDNA)sequencing. As determined through partial sequencing of rDNA clones amplified with archaea-specific primers, the archaeal populations indeep-sea hydrothermal vent environmentsshowed a great geneticdiversity, and most members of these populations appeared to be uncultivated and unidentified organisms. In the phylogenetic analysis, a number of rDNA sequences obtained from deep-sea hydrothermal vents were placedindeeplineagesofthecrenarchaeoticphylumpriortothedivergenceofcultivatedthermophilic membersofthecrenarchaeotaorbetweenthermophilicmembersoftheeuryarchaeotaandmembersof themethanogen-halophileclade.Wholecellinsituhybridizationanalysissuggestedthatsomemicroorgan- isms of novel phylotypes predicted by molecular phylogenetic analysis were likely present in deep-sea hydrothermal vent environments. These findings expand our view of the genetic diversity of archaea in deep-seahydrothermalventenvironmentsandofthephylogeneticorganizationofarchaea. RECENT molecular phylogenetic analyses based on 1997),Yellowstonehotsprings(Barnsetal.1994,1996), smallsubunitrRNAgene(SSUrDNA)sequencing shallowmarinehydrothermalventwater,andaJapanese have indicated that the microbial diversity of naturally acidic hot spring (Takai and Sako 1999). The results occurringmicrobialcommunitiesismuchgreaterthan of these studies suggested that a number of previously previously assumed based on standard cultivation and unknownanduncharacterizedmembersofthearchaea isolationmethods(Stahletal.1984;Giovannonietal. exist in such environments. 1990; Delong 1992; Fuhrman et al. 1992; Ward et al. Todate,anumberofextremethermophilesorhyper- 1992;Barnsetal.1994,1996;Delongetal.1994;Rey- thermophiles have been isolated from terrestrial solfa- senbachetal.1994;Prestonetal.1996;Bintrimetal. taricfieldsandhotsprings,andshallowordeepmarine 1997;Massanaetal.1997;Munsonetal.1997;Schleper hydrothermal vent environments (Stetter et al. 1990; etal.1997a;Dojkaetal.1998;Hugenholtzetal.1998). Adams 1993; Adams et al. 1995; Sako et al. 1996a,b). Although many studies focusing on the identification Thediversityofthermophilicmicroorganismsinterres- anddiversityofprokaryotesofthedomainBacteriahave trialhotspringsandinashallowmarinehydrothermal beenconducted,increasingattentionisnowbeingpaid venthasbeeninvestigatedbyboththecultivation-isola- to the genetic diversity and ecological significance of tion technique and the molecular phylogenetic ap- prokaryotes of the domain Archaea (Delong 1992; proach (Stetteret al.1990; Adams1993; Adamset al. Fuhrman et al. 1992; Barns et al. 1994, 1996; Delong 1995;Barnsetal.1994,1996;Sakoetal.1996a,b;Takai et al. 1994; Preston et al. 1996; Bintrim et al. 1997; andSako1999).Sincethediscoveryofdeep-seahydro- Kudo et al. 1997; Massana et al. 1997; Mcinerney et thermal vents in 1977, such environments have been al. 1997; Munson et al. 1997; Schleper et al. 1997a; regarded as one of the major habitats of thermophiles Vetrianietal.1998;TakaiandSako1999).Molecular and have been also a major source of novel deep-sea phylogenetic surveys of archaeal diversity have focused thermophilic archaea and bacteria (Gonzalez et al. onavarietyofenvironments,forinstance,oceanwater 1995, 1998; Prieur et al. 1995; Blo¨chl et al. 1997). (Fuhrman et al. 1992; Mcinerney et al. 1997), coastal However, the genetic diversity of archaea in deep-sea water(Delong1992;Massanaetal.1997),apolarsea hydrothermalventenvironmentsispoorlyunderstood, (Delongetal.1994),sedimentsfromacontinentalshelf andmolecularphylogeneticsurveysofthearchaealpop- (Vetrianietal.1998),asaltmarsh(Munsonetal.1997), ulationsinsuchenvironmentshavenotyetbeenunder- a freshwater lake (Schleper et al. 1997a), agricultural taken. soils(Bintrimetal.1997),paddyfieldsoil(Kudoetal. Inthisstudy,wesoughttodeterminethegeneticdiver- sity of archaea in deep-sea hydrothermal vent environ- ments such as effluent vent water and chimney struc- tures.Thehydrothermalventsystemsinthisstudywere Correspondingauthor:KenTakai,Deep-SeaMicroorganismsResearch located at Suiyo Sea Mount and Myojin Knoll in the Group, Japan Marine Science and Technology Center, 2-15 Natsu- shima-cho,Yokosuka237-0061,Japan. E-mail:[email protected] OgasawaraareaandatIheyaBasinintheOkinawaarea, Genetics152:1285–1297(August1999) 1286 K.TakaiandK.Horikoshi tctzbJhcsheataohayumptemdesidasohmrseipno,neiia,atddcharonaaeeneftdtledolrdtypdmthwhdd-eUseaeipaietfMlrthaatfgehaverAeetrcbhsnahTneheynieonEtdandt1steRsriec9yfoatIfrhs9oArtdtowhoe5efiLmve)emmoSre.rrtaoremsThAnostaieNehhangtrrxyleeeeDeahravamgoaneiMhrgenfndocihEonot1htpg-Tte0tawehhredH0enaaemy.0aptOvrlel–oiphrDr1(gpeioT3eceSorn5annaapml0btveluuyilterelmriaoacnetn1.ntoieso;dmTrcnIgwhogessaesehennyirsotinsiees---- aCharacteristics HightemperaturefluidcontainingrelativelyhighconcentrationsofCa,Sr,Ba,AlComposedofchalcopyrite,sphaler-ite,anhydrite,barite.HighAucontent(cid:122)(cid:56)100clearfluiddischargedfrom(cid:122)(cid:56)180black-coloredfluiddischargedfrom.Composedofbarite,pyrite,andotherminorsulfidesHightemperaturefluidcontainingquitehighconcentrationsofCO,2,andothercarbonhydrates,CH4andHS2Composedofcarbonate-richhydro-thermalprecipitateswithdissemin-atedsulfides SamplecollectionandDNAextraction:Samplesfromvari- ousdeep-seahydrothermalventenvironmentswereobtained pth 9m 8m0m fromthehydrothermalfieldsatSuiyoSeaMountandMyojin e 6 33 d 3 33 KnollintheOgasawaraareaandfromthehydrothermalfields / m 1 11 m e / // m amtaInhneeyda sBuabsmineirnsibthleeSOhikninkaaiw2a0a0r0eian,sJeavpearna,lbdyivemse(aDnisveosf9t7h7e, ratur 1369 nside nsidenside 972 1398 1007,1016,and1111).Effluentblacksmokerventwater(11 y e / i ii / / lsimteorsk;e(cid:122)rc3h0i0m(cid:56);ndeyes(i(cid:122)gn7a0t0edg;adsesSigSnMaVteWd)asaSnSdMthCe)wtiepreocfoallebclatecdk sstud Temp (cid:122)(cid:56)300 (cid:122)(cid:56)300 (cid:122)(cid:56)100(cid:122)(cid:56)180 (cid:122)(cid:56)300 (cid:122)(cid:56)100 fromahydrothermalventsiteatSuiyoSeaMount.Thetipof hi t ablacksmokerchimney((cid:122)500g;thetemperatureofthevent n i E E EE E E wsmatoekrewracsh(cid:122)im1n8e0y(cid:56);(d(cid:122)e3s0ig0nga;tetdheasteMmCpe2r)aatunrdetohfethtiepvoefntawcaletearr used (cid:57)630 (cid:57)630 (cid:57)004(cid:57)004 (cid:57)900 (cid:57)300 was(cid:122)100(cid:56);designatedasMC1)wereobtainedfromahydro- s 8. 8. 2.2. 3. 8. e 3 3 55 5 5 twhaetremr(a1l2velintetrssi;t(cid:122)ea3t00M(cid:56);ydojeisnigKnnaotelld.aEsfflIVuWen)tabnldacskedsmimoeknetrfvroenmt ampl ation (cid:56)140 (cid:56)140 (cid:56)139(cid:56)139 (cid:56)126 (cid:56)126 tgIhS;)ethuweperpteeemro1pb–tea1ria0nt-ceumdreflraooyfmetrhatehtestehhdeyimwdareotnethtreswirmamsma1le0vr0ei(cid:56)nn;gtdpseiostieignnatat(t(cid:122)Iehd3e0ya0as ABLE1 alvents Loc (cid:57)4.090N, (cid:57)4.090N, (cid:57)6.208N,(cid:57)6.208N, (cid:57)7.220N, (cid:57)2.800N, 95). Basin.Thesamplesources,locations,andpropertiesaresum- T m 3 3 00 4 3 9 wmonearriecizeiemadnmidnesdtToiaarbteeldleya1tfr.4oW(cid:56)z,eaanntedrattsha(cid:50)em8tpi5pl(cid:56)esusonfwtceihlreipmriomncemeyssesaedndiad.teselydcimooelnetds ydrother (cid:56)rc28 (cid:56)rc28 (cid:56)32(cid:56)32 (cid:56)27 (cid:56)27 Urabe(1 Nucleic acids were extracted by physical disruption (the h a a elaborategrindingoffilterwithmortarandpestle)andchemi- ea ra ra arcarc nd cSvS1(eAaae0nla-kd(cid:109)tlovyMmaws(nioa-1spttu9eeoon9crrf,9teft)T-rohs.oioreAzmkepIcyhNpoeeerl,ailyotcsJaexahxapiBmcoafiacfnalostit)terneh.drlesMyiwna1haignc0ysrddtlosioret6boeq.itt0rahuhs-l(cid:109)eeeopnrmfmmateir-apefatfllitollchyurlvoeeeefisd-nnsliwtttzoebeesrfirletafieTedclsatcketkoahsrtalmrlpieSooacuuapktingeyeehddorr Listofdeep-s Source nt,Izu-Ogasawa nt,Izu-Ogasawa Izu-OgasawaraIzu-Ogasawara MiddleOkinawa MiddleOkinawa omIshibashia tmw1fhr5eome0rmec-mdatithsmaheemenoNseefwatteCafihrslleh/tceet1erid0lpel0usdwlomoiwtfshmaectNheaEirEcmDeTstnTaabemAteyu)psffiflateewlntrseidc(ore5ssn0e(adA0mni.mdd2mv2esa-Ttn(cid:109)nortmtrsiese,c--d(cid:122)Hp)o.C3arTt0le,0h(cid:50)-speig2Hzefi0(,(cid:56)w8l.t.4ee0I7trn/)s- Suiyoseamou Suiyoseamou MyojinKnoll,MyojinKnoll, IheyaBasin,trough IheyaBasin,trough artreferredfr ofthesamplematerialwasresuspendedin500mlofMJsyn- p tthherotiucgsheaaw0a.t2e2r-(cid:109)(Sma-kpooreet-sailz.e19fi9lt6ebr),awndhicahuthoacdlavbeede,nafinldterthede W) arge satapwannomiddrcepeps-lseeeitdzosetimlmreNae.eitmTtneerthoxiveapfielarlrtwemetsiaruciscalserpncsoredubanasindah6dlee.d0dct-he(cid:109)sileanlmnmst-ofppsrcolofirermenewee-nsatispehzedsieeofictnchlertisrceuoarswuthpeigetadhhdpfeoaacrrh.1mMi10mo0-i(cid:109)crnstrmeeaocyr-- on) water(SSMV ney(SSMC) ney(MC1)ney(MC2) water(IVW) ediments(IS) wereforthel bteddtbhaniiielaaaesvllmccifrorpfoeorinalbatnlrteoecamtrtdwimicoeiclaniennenbgstsloaalwuenvwlexalteoDsspras(eeNwneTradAciaasmo.hscnlteeelHonentdreacitegrttwedaehisedtatfihamtaol(cid:50)tnlNno.eelr2E1ega.09Tca(cid:56)90Tut.8bi.lhvT2a)uei2ro,sf-fca(cid:109)cewofihmrfienletlitt-etgwcprerhkooirctlrfwweoecDaio-ratssNhnniezotwdxAeaupa,iuntsew4hsrim7eanie-mdsmgidcefernmonaoxors---- Sample(abbreviati Blacksmokervent Blacksmokerchim ClearsmokerchimBlacksmokerchim Blacksmokervent Watersimmerings aCharacteristics tracted and purified from all filters including the filter em- ArchaealDiversityinHydrothermalVents 1287 ployed as a negative control as described previously (Takai the same manner as described in this study. A total of 26 of andSako1999). 33 clones revealed the correct identical rDNA sequence of Microscopicobservation:Theventwaterandchimneysam- thisorganism.Atotalof6of33cloneshadanerrorsuchas plesfilteredthrough6.0-(cid:109)m-pore-sizefilterpaperwerefixed asinglenucleotidesubstitution,insertion,ordeletion,and1 for 12 hr in 3.7% formaldehyde, and the fixed microbial of33clones hadbothasinglenucleotide substitutionanda particles were collected on 0.22-(cid:109)m-pore-size, 13-mm-diame- single nucleotide insertion. The overall error frequency of ter polycarbonate filters (Advantec). Each filter was rinsed partial sequencing analysis was calculated to be 0.063 error twiceinMJsyntheticseawater(Sakoetal.1996b),whichhad per100nucleotides. been filtered through a 0.22-(cid:109)m-pore-size filter and auto- SequencesweremanuallyalignedtoSSUrDNAdatafrom claved,andthenstainedbytreatmentwithMJseawatercon- the RDP on the basis of primary and secondary structure taining 4(cid:57),6-diamidino-2-phenylindole (DAPI; 10 (cid:109)g/ml) or considerationsand werealso submittedtoanalysis usingthe ethidiumbromide(10(cid:109)g/ml)at4(cid:56)for20min.Thefilterwas CHECK CHIMERAprogramtodetectthepresenceofchime- brieflyrinsedinMJseawaterandexaminedunderepifluores- ric artifacts. Phylogenetic analyses were restricted to nucleo- cenceusing aLeica DMRBmicroscopewith aLeica MPS30 tide positions that were unambiguously alignable in all se- camerasystem. quences.Least-squaresdistancematrixanalysis(Olsen1988), PCRamplificationofrDNA:SmallsubunitribosomalRNA based on evolutionary distances, was performed using the genes(rDNAs)wereamplifiedbyPCRusingLATaqpolymer- correction of Kimura (1980). Neighbor-joining analysis was asewithGCbuffer(TaKaRa,Kyoto,Japan),asrecommended accomplishedusingtheODENsoftwarepackage(version1.1; by the manufacturer. Reaction mixtures were prepared in National Institute of Genetics, Mishima, Japan). Maximum- whichtheconcentrationofeacholigonucleotideprimerwas likelihoodanalyseswereperformedusingthePHYLIPpackage 0.1(cid:109)mandthatoftheDNAtemplatewas1ng/(cid:109)l.Thermal (version 3.5; obtained from J. Felsenstein, University of cyclingwasperformedusingtheGeneAmpPCRsystem9600 Washington, Seattle).Bootstrap analysis was usedto provide (Perkin-Elmer, Foster City, CA) and the conditions were as confidenceestimatesforphylogenetictreetopologies. follows: denaturation at 96(cid:56) for 20 sec, annealing at 50(cid:56) for Whole-cellhybridizationanalysis:RibosomalRNA-targeted 45sec,andextensionat72(cid:56)for120secforatotalof30cycles. oligonucleotide probes were designed for detection of ar- Theoligonucleotideprimersequencesandcombinationsused chaeal members that were potentially predominant in the were as follows: Arch21F (Fuhrman et al. 1992) and 1492R deep-seahydrothermalventenvironments.TheyweretheMGI (Lane1985)forarchaealrDNA,andUni515FandUni1408R probe (5(cid:57)-AAAYCACTCGGAYTAACCTT-3(cid:57)) and AAG probe (TakaiandSako1999)forallmicrobialrDNAs. (5(cid:57)-CCTAGCACTCGGGCTCGCGG-3(cid:57)), which corresponded Cloning and sequencing of amplified rDNA: Amplified topositions391–411and401–421inEscherichiacoli16SrDNA, rDNAsfromfiveseparatereactionswerepooled,fractionated respectively,andweredesignedfordetectionofmarinegroup byelectrophoresisona1.2%(w/v)agarosegel,thenextracted I(MGI)crenarchaeoticmembersandafewancientarchaeal from the gel slicessequentially with phenol, phenol/chloro- group(AAG)membersbasedonthealignmentinthephyloge- form/isoamyl alcohol and chloroform/isoamyl alcohol, and neticanalysis.Bothprobesequenceswereanalyzedusingthe precipitated with ethanol. After centrifugation, DNA pellets PROBE CHECKfromtheRDP(Larsenetal.1993)andthe were resuspended in sterile distilled water. The purified gapped-BLASTsearchalgorithm(Altschuletal.1997;Ben- rDNAs were cloned in the vector pCR2.1 using the original sonetal.1998)toexaminewhetheranyotherrDNAsequences TA cloning kit (Invitrogen, Carlsbad, CA), and then the ar- have similarity to these probe sequences. In addition, dot- chaeal and universal primer PCR libraries were built from hybridizationanalysiswascarriedoutwithrepresentativearch- DNAobtainedfromeachdeep-seahydrothermalventsample. aeal rDNA clones obtained from the deep-sea hydrothermal Clonescontaininginsertsoftheappropriatesizewereidenti- vent environments to check the specificity of the probes to- fied byelectrophoresis of alkaline lysisplasmid preparations wardtheMGIcrenarchaeoticandAAGmembers,respectively. (Maniatiset al.1982).Denatured, double-strandedplasmid The purified plasmid DNAs (100 ng) containing (cid:122)1.5 kb of templates were sequenced by the dideoxynucleotide chain- rDNA sequences were blotted onto positively charged nylon termination method with Taq DNA polymerase FS (Perkin- membranes (Boehringer Mannheim, Indianapolis). The Elmer)followingthemanufacturer’srecommendations.Uni- membranes were hybridized with the MGI-D probe or the versal rRNA-specific primers (Lane 1985) and M13 forward AAG-D probe (5 ng/(cid:109)l), which was labeled at the 5(cid:57)-end and reverse primers (Maniatis et al. 1982) were used in se- with digoxigenin (DIG) and purified by HPLC (Amersham quencingreactions. PharmaciaBiotech).Afterhybridization,themembraneswere Sequenceandphylogeneticanalyses: ThepartialrDNAse- washed,and theDIG-labeledoligonucleotide probeshybrid- quenceswereanalyzedusingSIMILARITY RANKandALIGN izedwiththetargetedrDNAsequencesweredetectedbyusing SEQUENCE from the Ribosomal Database Project (RDP; the DIG luminescent detection kit for nucleic acids (Boeh- Larsenetal.1993)andthegapped-BLASTsearchalgorithm ringerMannheim).Allhybridizationandwashingprocedures (Altschul et al. 1997; Benson et al. 1998) to estimate the werecarriedoutinthesameconditionsasdescribedbelow. degree of similarity to other rDNA sequences. Sequences of For whole-cell hybridization experiments, the vent water 400bpinlengthdeterminedbymeansofUni515Fprimerin andchimneysamplesfilteredthrough6.0-(cid:109)m-pore-sizefilter onestrandwereusedinthesimilarityanalysis.Thedatabases paper were fixed for 12 hr in 3.7% formaldehyde, and the usedforRDPandgapped-BLASTanalysisweretheprokaryotic fixedmicrobialparticleswerecollectedon0.22-(cid:109)m-pore-size, SSU rRNA database and the nonredundant nucleotide se- 13-mm-diameterpolycarbonatefilters(Advantec).Eachfilter quence database from GenBank, EMBL, and DDBJ, respec- wasrinsedsequentiallyinMJsyntheticseawaterthathadbeen tively.Inthisstudy,wetentativelydefinedthat(cid:46)98%similarity filtered through a 0.22-(cid:109)m-pore-size filter and autoclaved in inUni515F-dependentsequencesrepresentedthesamerDNA 1:1(v/v)MJseawater/deionizeddistilledwater(DDW)and clone type among rDNA sequences obtained from deep-sea DDW.Thefilterwasdehydratedinanethanolseries(30,60, hydrothermal vent environments. In addition, error fre- 80, 90, and 95%, v/v) and air-dried (Amann et al. 1990). quency,whichpresumably occurredthroughPCRamplifica- The hybridization was carried out in hybridization solution tion,cloning,andpartialsequencing,wasroughlyestimated. containing 20 mm Tris-HCl (pH 7.2), 0.9 m NaCl, 0.1% The16SrDNAofPyrococcushorikoshiiJCM9974(Gonzalezet (w/v)sodiumdodecylsulfate,and50%(v/v)formamidefor al. 1998) was amplified, cloned, and partially sequenced in 12 hr at 48(cid:56) in the case of the MGI-F probe (5 ng/(cid:109)l) or at 1288 K.TakaiandK.Horikoshi 55(cid:56) in the case of the AAG-F probe (5 ng/(cid:109)l), which were ducedduringtheDNAextractionandpurificationpro- each labeled at the 5(cid:57)-end with fluorescein and purified by cedures,asdescribedbyTanneretal.(1998).However, HPLC (Amersham Pharmacia Biotech). After hybridization, norDNAwasamplifiedfromthenegativecontrolsam- the filter was washed in hybridization solution lacking the probes,atthesametemperatureasthatemployedforhybrid- ples using either the archaea-specific primers or the ization,for20mintwiceandstainedwitha10(cid:109)g/mlsolution universal primers under the conditions of this study. ofDAPIinDDWat4(cid:56)for20min.Thefilterwasbrieflyrinsed The first step in our strategy was to examine the ar- in DDW and was examined under epifluorescence using a chaeal populations in naturally occurring microbial LeicaDMRBmicroscopewithaLeicaMPS30camerasystem. communitiesinvariousdeep-seahydrothermalventen- Twotypesofnegativecontrolswereemployedineachexperi- ment,usingidenticalcellpreparationsandhybridizationcon- vironmentsonthebasisofpartialnucleotidesequence ditions.Onetypeofnegativecontrolconsistedofsamplesin information obtained through analysis of a number of whichtheunlabeledMGI-ForAAG-Fprobewasaddedin50- insert-containingplasmids.First,a400-ntsequence(cor- fold excess (250 ng/(cid:109)l) of the fluorescein-labeled MGI-F or responding to positions 536–935 of E. coli rDNA num- AAG-F probe added at the standard concentration (5 ng/ (cid:109)l). Theother type ofnegative control consistedof samples bering) determined in one strand from each plasmid employed to examine hybridization with cultivated thermo- was obtainedfrom universaland archaealprimed PCR philic microorganisms, including Aeropyrum pernix JCM9820 libraries of each sample. In this study, 132, 116, 110, (Sakoetal.1996a),ThermococcuspeptonophilusJCM9653(Gon- 102,and 124clones werepartially sequencedfrom the zalez et al. 1995), P. horikoshii JCM9974 (Gonzalez et al. universal primer PCR libraries established from the 1998), Methanococcus jannashii DSM2661 (Jones et al. 1983), deep-sea hydrothermal vent samples of SSMC, MC1, RhodothermusobamensisJCM9785(Sakoetal.1996b),andTher- maerobacter marianensis (Takai et al. 1999). Cells were grown MC2,IVW,andIS,respectively(Table2).Furthermore, as described in the references given above, and cells in the 104, 164, 139, 145, and 188 clones were partially se- stationaryphaseofgrowthwereusedforthesehybridization quenced from the archaeal primer PCR libraries of experiments. SSMCA(SSMC-Archaeallibrary),MC1A,MC2A,IVWA, Nucleotide sequence accession numbers: The sequences reported here have been deposited in the DDBJ database and ISA (Table 3). Gapped-BLAST analysis of these undertheaccessionnos.AB019714–AB019759. sequences provided a profile of the PCR-dependent rDNA composition of the various environmental sam- ples (Tables 2 and 3). As described in materials and RESULTS methods,wetentativelydefinedthat(cid:46)98%similarityin Characterization of microbial rDNA recovered from Uni515F-dependentsequencesrepresentedanidentical deep-seahydrothermalventenvironments:DNAwasiso- rDNAclonetypeamongrDNAsequencesobtainedfrom lateddirectlyfromsamplescollectedfromvariousdeep- deep-sea hydrothermal vent environments. The error sea hydrothermal vent environments except for the ef- frequency during the partial sequencing analysis was fluent water from the black smoker vent (SSMVW) at estimated to be (cid:44)0.1%. thehydrothermal ventsite atSuiyo SeaMount. Inthat From the black smoker vent chimney of Suiyo Sea sample, low DNA recovery was expected due to the Mount(SSMC),128ofthe132clones(97%)withinserts very small number of microorganisms ((cid:44)10 cells/ml) amplifiedbymeansofuniversalprimerswerefoundto observed in the sample by epifluorescence microscopy be bacterial SSU rDNA sequences (Table 2). Archaeal after staining with DAPI or ethidium bromide (Table rDNAsequencesweredetectedinalibraryofsequences 2).MicrobialribosomalRNAgeneswereselectivelyam- derivedfromSSMCpreparedbyPCRusingtheuniversal plified by PCR and cloned. PCR amplification was also primer.All4ofthesecloneswerefoundtohavethesame carriedoutusingextractspreparedfromnegativecon- rDNAsequence,andthissequencewasthepredominant trol samples to check for possible contaminants intro- rDNAsequence(pSSMCA1)foundinthearchaealpopu- TABLE2 RibosomalDNAcompositionofsamplesfromvariousdeep-seahydrothermalventenvironments Bacterial Archaeal Eucaryal Sample(no.ofclonessequenced) (%) (%) (%) Microbialcellsa SSMVW ND ND ND ND SSMC(132) 97.0 3.0 0 104/gwet MC1(116) 82.8 17.2 0 107/gwet MC2(110) 95.5 4.5 0 107/gwet IVW(102) 73.5 26.5 0 104/ml IS(124) 96.8 3.4 0 107/gwet Ribosomal DNA composition was determined by partial sequencing analysis of rDNA clones of universal primerPCRlibraries.ND,notdetected. aDirectcountsofmicrobialcellsstainedwithDAPIandethidiumbromide. ArchaealDiversityinHydrothermalVents 1289 lation of this sample. The archaeal rDNA clones derived foundtoconstitutetwolargephylogeneticassemblages; fromSSMCconsistedofonlytwoclonetypes,represented one consisted of the very deep lineages of rDNA se- by pSSMCA1 (99 of 104 clones) and pSSMCA108 (5 of quences within the Crenarchaeota and the other was 104clones),bothofwhichshowedrelativelylowsimilarity placedinanintermediatepositionbetweenthethermo- ((cid:44)80%) to any other rDNA sequences from cultivated philic Euryarchaeota and the methanogens-halophiles or uncultivated archaea (Table 3). (Figure 1). Most of the rDNA clones obtained from The proportions of archaeal rDNA sequences were deep-seahydrothermalventenvironmentswereplaced similar in the universal primer PCR libraries prepared in novel phylogenetic branches distinct from those of fromtheblacksmokerchimneysamplecollectedfrom cultivated members of the archaea and other environ- Myojin Knoll (MC2) and the clear water simmering mental clones (Figure 1). sediment sample from Iheya Basin (IS; Table 2). The Of the many novel rDNA sequences found, several archaeal primer PCR library prepared from MC2 con- rDNAsequencesrepresentedbypMC2A256,pMC2A249, sisted of a variety of rDNA sequences, and the same and pMC2A14/pSSMCA1 (AAG) were placed in the clonetypeasthatrecoveredfromSSMCwasfoundalso deepest branches within the crenarchaeotic phylum. in this library (Table 3). The archaeal rDNA composi- In this phylogenetic tree, the crenarchaeotic phylum tion of IS differed significantly from that of the black accommodated korarchaeotic rDNA sequences and smokerchimneysamplesofSSMCandMC2.Inthecase clonepOWA133,whichwerepreviouslydenotedaspos- of the archaeal primer PCR library prepared from IS, sible outgroups prior to the divergence of Crenar- a high proportion of Euryarchaeota was found (Table chaeota and Euryarchaeota, respectively (Barns et al. 3). These euryarchaeotic rDNA clones were unidenti- 1994;TakaiandSako1999).Thepreferenceinplace- fiedsequencesandconsistedofavarietyofclonetypes ment of these deep lineages of rDNA sequences into (Table 3). the crenarchaeotic phylum was strongly influenced by Arelativelyhighproportionofarchaeawasfoundin therDNAsequenceofMethanopyruskandleri.Whenthe the universal primer PCR libraries prepared from the M.kandlerirDNAsequencewasomittedfromthephylo- clear smoker chimney sample obtained from Myojin geneticanalysis, thephylogenetictrees obtainedbased Knoll (MC1)and the effluent blacksmoker vent water on the neighbor-joining and maximum-likelihood samplefromIheyaBasin(IVW).Intheuniversalprimer methodsshowedfivepossibleseparatearchaealphyletic PCR library prepared from MC1 and IVW, 20 of 116 groupsatthekingdomlevel;thedeepestoutgroupwas clones (17.2%) and 30 of 102 clones (26.5%) were ar- represented by the AAG rDNA clones, the second out- chaeal rDNA sequences, respectively (Table 2). All of groupby thepOWA133clone,and thethirdoutgroup thesearchaealrDNAcloneswerecloselyrelatedtoma- bythekorarchaeoticclones,priortothedivergenceof rine planktonic clones, designated as MGI (Delong Crenarchaeota and Euryarchaeota (data not shown). 1992;Fuhrmanetal.1992;Delongetal.1994;Massana TheseverydeepbranchesofrDNAsequencesobtained et al. 1997; Mcinerney et al. 1997); however, none of fromvarioushotwaterenvironmentshadrelativelylow these sequences showed (cid:46)98% similarity to the se- bootstrap significance and were sensitive to the se- quences of MGI clones described previously. In addi- quences used; however, base composition disparities tion, the archaeal primer PCR library prepared from had little influence on the topology, which was sup- MC1containedonlytwoclonetypes,bothofwhichwere portedbytransversiondistance-matrixanalysis(Woese categorized within the MGI (39 clones for pMC1A103 etal.1991).TheseresultsindicatedthattheAAGrDNA type or 71 clones for pMC1A11 type of 164 clones), clonesrecoveredfromdeep-seahydrothermalventenvi- and another major unidentified clone type (pMC1A4) ronments undoubtedly represent one of the deepest categorized within the Euryarchaeota (54 of 164 lineagesofanyarchaealrDNAsequencesderivedfrom clones),whilethearchaealprimerPCRlibraryprepared cultivated members or unidentified environmental from IVW mainly consisted of a variety of rDNA clone clones,althoughwhetherthephylogeneticclassification types categorized within the MGI (Table 3). oftheserDNAclonesintothekingdomCrenarchaeota Phylogeneticanalyses ofrDNA sequencesrepresent- is appropriate remains uncertain. ingunidentifiedmicroorganisms:Toinferphylogenetic Inaddition,agreatdiversityofrDNAsequenceswere affiliationsofarchaealrDNAclonesobtainedfromvari- alliedinpositionsclosetotheThermoplasma-Picrophi- ous deep-sea hydrothermal vent environments, nearly luscladeandinintermediatepositionsbetweenthermo- full-length sequences ((cid:122)1400 nt) of representative philic Euryarchaeota branches and methanogen-halo- rDNAclonetypesweredeterminedandusedforappro- phile branches (DHV euryarchaeotic groups I and II; priate phylogenetic analysis. Phylogenetic analysis by Figure1).Thesesequencesweremainlyobtainedfrom the neighbor-joining and maximum-likelihood meth- the black smoker chimney at Myojin Knoll (MC2) and odsresultedintreeswithsimilartopologies(Figure1). theclearwatersimmeringsedimentsatIS,andmostof Inanoverviewofthetree,thearchaealrDNAsequences themshowednosignificantsimilaritytoanyotherrDNA obtainednotonlyfromdeep-seahydrothermalventen- sequences known to date. When the phylogenetic tree vironmentsbutalsootherhotwaterenvironmentswere was reconstructed by neighbor-joining methods to in- 1290 K.TakaiandK.Horikoshi TABLE3 DistributionofrepresentativearchaealrDNAclonetypesinvariousdeep-seahydrothermalenvironments Clonetype Phylogeneticgroup (positiononblot) SSMC MC1 MC2 IVW IS Crenarchaeota 39/99a 90/110 46/63 103/141 5/6 Ancientarchaealgroupb 39/99 — 14/22 — — DHVA1b pMC2A256(1e) — — 1/2 — — pMC2A249(1d) — — 1/1 — — pMC2A14(1c)/pSSMCA1(1a/1b) 39/99 — 12/19 — — DHVcrenarchaeoticgroupb — — 12/14 — — DHVC1b pMC2A15(1f) — — 1/1 — — pMC2A209(1g) — — 4/5 — — pMC2A36(1h) — — 4/4 — — pMC2A308(1i) — — 3/4 — — MarinegroupI(MGI) — 90/110 20/27 103/141 — pMC1A103(2a) — 31/39 — — — pMC1A11(2b) — 59/71 — — — pMC2A1(2c) — — 20/27 — — pIVWA1(2d) — — — 21/23 — pIVWA2(2e) — — — 28/39 — pIVWA3(2f) — — — 11/16 — pIVWA5(2g) — — — 29/47 — pIVWA11(2h) — — — 3/4 — pIVWA101(2i) — — — 8/9 — pIVWA108(2j) — — — 3/3 — THScrenarchaeoticgroupb — — — — 5/6 THSC1b pISA9(1j) — — — — 4/5 pISA7(3a) — — — — 1/1 Euryarcheaota 3/5 36/54 66/76 3/4 144/182 Methanococcales — — 1/1 — — pMC2A384(3b) — — 1/1 — — Archaeglobales — — 3/3 — — pMC2A228(3c) — — 3/3 — — DHVeuryarchaeotic groupIb 3/5 — 39/46 — 52/56 DHVE1b pMC2A24(3d) — — 17/18 — — pMC2A203(3e) — — 3/3 — — pMC2A33(3f) — — 2/2 — — DHVE2b pMC2A10/pSSMCA108(3g) 3/5 — 17/23 — — pISA12(3h) — — — — 23/24 pISA42(3i) — — — — 29/32 DHVeuryarchaeotic groupIIb — 36/54 23/26 3/4 59/78 DHVE3b pMC2A25(3j) — — 2/3 — — pMC2A21(4a) — — 4/4 — — pMC2A232(4b) — — 8/8 — — pMC2A211(4c) — — 6/8 — — DHVE4b pMC2A17(4d) — — 1/1 — — pISA35(4e) — — — — 1/1 DHVE5b pISA18(4f) — — — — 6/7 pISA38(4g) — — — — 4/4 pISA1(4h) — — — — 4/4 (continued) ArchaealDiversityinHydrothermalVents 1291 TABLE3 (Continued) Clonetype Phylogeneticgroup (positiononblot) SSMC MC1 MC2 IVW IS DHVE6b pISA3(4i) — — — — 3/5 pMC2A35(4j) — — 2/2 — — pMC1A4(5a) — 36/54 — — — pISA48(5b) — — — — 31/41 pISA13/pIVWA104(5c) — — — 3/4 10/16 DHVeuryarchaeotic — 33/48 groupIIIb — — — DHVE7b pISA16(5d) — — — — 26/37 pISA14(5e) — — — — 7/11 Total 42/104 126/164 112/139 106/145 149/188 ArchaealrDNAcompositionwasdeterminedbypartialsequencinganalysisofrDNAclonesofarchaealprimerPCRlibraries. aThis indicates the number of sequence variations per total number of clones designated into each phylogenetic group or clonetypeonthebasisofpartialsequencinganalysis. bNovel uncultivated phylogenetic groups shown in Figure 1 and abbreviations are as follows: DHVA, deep-sea hydrothermal vent ancient archaea; DHVC, deep-sea hydrothermal vent crenarchaeota; THSC, terrestrial hot spring crenarchaeota; DHVE, deep-seahydrothermalventeuryarchaeota. cludeotherenvironmental,unidentifiedeuryarchaeotic obtainedfromtheblacksmokerchimney(pMC2A)and clones based on the partial sequences (corresponding the clear smoker chimney (pMC1A) at Myojin Knoll, to positions 438–887 inE. coli 16S rDNA), some phylo- andtheblacksmokerventwater(pIVWA)atIheyaBasin types of sequences (DHVE1) within the DHV euryar- (Figure 3). When the phylogenetic tree was recon- chaeoticgroupIwerecloselyrelatedwithsomeuniden- structedincludingthesenewMGIclones,twoevolution- tified euryarchaeotic clones obtained from microbial ary arrays were found in the MGI cluster. The finding communitiesincoastalsaltmarshandcontinentalshelf of further genetic diversity among the members of the sediments (salt marsh and marine benthic euryarchae- MGIindeep-seahydrothermalventenvironmentsserves ota; Munson et al. 1997; Vetriani et al. 1998; Figure asanimportantcluetounderstandingtheubiquityand 2).However,nosignificantphylogeneticrelatednesswas originoftheseuncultivatedmembersofthearchaeain observed between a variety of euryarchaeotic rDNA se- marine environments. quencesfromthedeep-seahydrothermalventenviron- Whole-cellhybridizationanalysis:RibosomalRNA-tar- ments and the marine group II obtained from plank- geted oligonucleotide probes were employed in visual- tonic microbial communities in ocean water, coastal izationofformalin-fixedmicrobialcellsfromthedeep- water, and polar sea water (MGII; Delong 1992; seahydrothermalventenvironmentstofacilitateidenti- Delongetal.1994).Itseemslikely,therefore,thatun- fication of members of MGI or the AAG represented cultivated and unidentified members of the euryar- bypMC2A249,pMC2A14,andpSSMCA1.Thedesigned chaeotadisplayinggreatgeneticdiversityarepresentin probes were confirmed to have no specificity to any themicrobialcommunitiesofvariouscoastalanddeep other rDNA sequences by analyses using the PROBE sea sedimentary environments, as in the case of mem- CHECK from the RDP (Larsen et al. 1993) and the bers of MGI in planktonic microbial communities gapped-BLASTsearchalgorithm(Altschuletal.1997; (Delong1992;Fuhrmanetal.1992;Delongetal.1994; Bensonetal.1998),andwerealsoconfirmedtohybrid- Massana et al. 1997; Mcinerney et al. 1997). ize specifically with the targeted rDNA sequences of Figure3showsaphylogenetictreepreparedfocusing MGI and AAG members by dot-hybridization analysis ontheMGIrDNAclones.TheseMGIrDNAclonesare (Figure4).Usingtheseprobes,whole-cellhybridization the most frequently occurring rDNA clones in marine analysiswascarriedoutwiththesamplesofSSMC,MC1, planktonic microbial communities (Delong 1992; MC2,andIVW(Figure5).HybridizationwiththeMGI- Fuhrman et al. 1992; Delong et al. 1994; Massana et specificfluorescentoligonucleotideprobewasobserved al. 1997; Mcinerney et al. 1997) and have often been inthesamplesofMC1andIVW,andhybridizationwith recoveredfrommarinesedimentsamplesandevensedi- the AAG-specific probe was observed in the case of mentadjacenttoahydrothermalvent(TakaiandSako SSMCandMC2.Thearchaealcellsinthecaseofmem- 1999). A variety of archaeal rDNA sequences closely bers of the AAG were irregular cocci, on average 1.0 related to the sequences of the MGI clones were also (cid:109)m in diameter (Figure 5, C and D, G and H), and 1292 K.TakaiandK.Horikoshi ArchaealDiversityinHydrothermalVents 1293 Figure 2.—Phylogenetic tree of rDNA clones focused on a variety of environmental, unidentified euryarchaeotic rDNA sequences. The tree was inferred by neighbor-joining analysis of 436 homologous positions of the rDNA sequence in the case of each organism or clone. Abbreviations indicate rDNA clones corresponding to uncultivated organisms derived from the following environments: SBAR16, OARB, and WHARN, from coastal water in North America (Delong 1992); ANTARCTIC5, fromantarcticmarinepicoplankton(Delongetal.1994);2MT,fromsaltmarshsediment(Munsonetal.1997);andBBA,from continentalshelfsediments(Vetrianietal.1998).BoldlettersindicaterDNAclonesobtainedfromdeep-seahydrothermalvent environments.Thescalebarrepresents0.025nucleotidesubstitutionspersequenceposition.Thepercentageof1000bootstrap resamplingsisindicated.Thefilledcirclesindicatethebranchpointsnotsupportedbymaximum-likelihoodanalysis. thoseinthecaseoftheMGIwereirregularcocci,(cid:122)1.0 or in hybridization experiments in which MGI-specific (cid:109)m in diameter and rod shaped, on average 2.0 (cid:109)m orAAG-specificprobeslackingthefluorwereincluded long and 0.6 (cid:109)m wide (Figure 5, A and B, E and F). simultaneously with fluor-labeled probes, but at a 50- Preston et al. (1996) reported that marine sponge- fold higher concentration. associatedMGIcellswererodshaped,0.8(cid:109)mlong,and OfthetotalprokaryoticcellscountedafterDAPIstain- 0.5 (cid:109)m wide, and the fluorescence signals of probes ing, 0.4 and 12.5% were identified as members of the labeled with Texas Red were detected at both cellular MGI in the case of MC1 (7 of 1753) and IVW (59 of poles.Inthisstudy,nofluorescentsignalsweredetected 474), respectively, and 2.2 and 0.8% were identified as at the poles of the short rod-shaped cells. No fluores- membersoftheAAGinthecaseofMMSC(29of1319) cencewas observedwhenhybridization wasperformed and MC2 (11 of 1375), respectively. The percentages using the MGI-specific and AAG-specific probes in the of certain rDNA clone types among the total rDNA case of cultivated bacterial and archaeal thermophiles sequences, based on PCR-mediated partial sequencing Figure 1.—Phylogenetic tree of rDNA clones derived from various deep-sea hydrothermal vent environments. The tree was inferredbyneighbor-joininganalysisof1211homologouspositionsoftherDNAsequenceinthecaseofeachorganismorclone. AbbreviationsindicaterDNAclonescorrespondingtouncultivatedorganismsderivedfromthefollowingenvironments:pOWA andpUWA,fromshallowmarinehydrothermalventwaterandterrestrialacidichotspringwater,respectively(TakaiandSako 1999);pJP,fromsedimentinaYellowstoneNationalParkhotspring(Barnsetal.1994);andsoilcloneSCA,fromagricultural soil(Bintrimetal.1997).BoldlettersindicaterDNAclonesobtainedfromdeep-seahydrothermalventenvironments.Thescale bar represents 0.1 nucleotide substitutions per sequence position. The percentage of 1000 bootstrap resamplings is indicated. Thefilledcirclesindicatethebranchpointsnotsupportedbymaximum-likelihoodanalysis. 1294 K.TakaiandK.Horikoshi Figure3.—Phylogenetictree ofrDNAcloneswithinmarine group I (MGI) derived from variousdeep-seahydrothermal ventenvironments.Thetreewas inferred by neighbor-joining analysis of 1211 homologous positionsoftherDNAsequence inthecaseofeachorganismor clone. Abbreviations indicate rDNAclonescorrespondingto uncultivated organisms derived from the following environ- ments:pJP,fromsedimentina YellowstoneNationalParkhot spring(Barnsetal.1994);soil clone SCA, from agricultural soil(Bintrimetal.1997);ma- rineclonesCandPM,fromwa- ter of the northeast Atlantic ocean at 500 m depth (Mcin- erneyetal.1997);marinefos- mid clone 4B7, from Pacific oceanwater(Steinetal.1996); andmarinecloneSB95-57,from coastal water of the Santa Bar- bara channel (Massana et al. 1997). Bold letters indicate rDNA clones obtained from deep-seahydrothermalventen- vironments.The scale bar rep- resents 0.05 nucleotide substi- tutionspersequenceposition. The percentage of 1000 boot- strapresamplingsisindicated. The topology of this tree was supported by maximum-likeli- hoodanalysis. analysis, were 11.5% MGI cells, 25.8% MGI cells, 2.9% and some were closely related to previously discovered AAG cells, and 0.7% AAG cells in the whole microbial environmental archaeal clones. Whole-cell hybridiza- communities of MC1, IVW, SSMC, and MC2, respec- tionanalysistofacilitateidentificationofrepresentative tively. Although there were differences found between uncultivated environmental clones revealed the exis- therDNApopulationandthefluorescence-labeledcell tence of organisms of such uncultivated phylotypes in population, both experiments strongly indicated that the naturally occurring microbial communities of the uncultivated and unidentified archaeal members are deep-sea hydrothermal vent environments. This is an presentasacertainproportionofthemicrobialcommu- outstanding example showing that a great variety of nitiesindeep-seahydrothermalventenvironments,and archaea remain undiscovered and uncharacterized in the rDNA from theseorganisms was successfully recov- various natural microbial habitats on the earth. ered in certain proportions from the bulk of DNA di- Samples have been collected from various environ- rectly extracted from these environments. ments with different geologic, physical, and chemical properties(Table1;IshibashiandUrabe1995).Envi- ronmentalcharacteristicshaveanimpactonthemicro- DISCUSSION bial communities in a given environment and might AgreatgeneticdiversityofarchaealrDNAcloneswas be important factors in determining the magnitude, evident upon analysis of DNA recovered from various content, and diversity of the microbial communities. deep-seahydrothermalventenvironments.Mostofthe For instance, in the case of the samples of MC2 and archaealrDNA sequencesobtainedfrom blacksmoker IS, the samples examined contained almost the same ventwaterandchimneys,aclearsmokerchimney,and number of microbial cells, and the universal primer clear water simmering sediments represented the se- PCRlibrariesshowedsimilarrDNAcomposition(Table2). quences of as-yet-uncultivated phylotypes distinct from Thesesamplesalsodisplayedagreaterdiversityofarch- anyothercultivatedarchaeaandenvironmentalclones, aealrDNAclonesthantheotherdeep-seahydrothermal
Description: