ApplMicrobiolBiotechnol(2011)91:1215–1225 DOI10.1007/s00253-011-3408-y ENVIRONMENTAL BIOTECHNOLOGY Ammonia-oxidizing archaea and ammonia-oxidizing bacteria in six full-scale wastewater treatment bioreactors Tong Zhang&Lin Ye&Amy Hin Yan Tong& Ming-Fei Shao&Si Lok Received:14March2011/Revised:20May2011/Accepted:22May2011/Publishedonline:25June2011 #TheAuthor(s)2011.ThisarticleispublishedwithopenaccessatSpringerlink.com Abstract In this study, dideoxy sequencing and 454 high- Keywords Ammonia-oxidizing archaea.Ammonia- throughput sequencing were used to analyze diversities of oxidizing bacteria.Municipal wastewater treatment the ammonia monooxygenase (amoA) genes and the 16S plants.454 Pyrosequencing rRNA genes of ammonia-oxidizing archaea (AOA) and ammonia-oxidizingbacteria(AOB) insixmunicipalwaste- watertreatmentplants.TheresultsshowedthatAOBamoA Introduction genes were quite diverse in different wastewater treatment plantswhilethe16SrRNAgeneswererelativelyconserved. Nitrification, the biological oxidation of ammonium to BasedontheobservedcomplexityofamoA and 16S rRNA nitriteandnitrate,isanessentialprocessinnitrogencycling genes,mostoftheAOBcanbeassignedtotheNitrosomonas and wastewater treatment bioreactors. Several groups of genus, with Nitrosomonas ureae, Nitrosomonas oligotropha, microorganisms are involved in the two-step process: Nitrosomonasmarina,andNitrosomonasaestuariibeingthe ammonia-oxidizing archaea (AOA), ammonia-oxidizing fourmostdominantspecies.FromthesequencesoftheAOA bacteria (AOB), and nitrite-oxidizing bacteria (NOB). In − amoAgenes,mostAOAobservedinthisstudybelongtothe the first step, AOA and/or AOB oxidize NH to NO , and 3 2 − − CGI.1bgroup,i.e.,thesoillineage.TheAOBamoAand16S in the second step NOB oxidize NO to NO . The first 2 3 rRNA genes were quantified by quantitative PCR and 454 step is rate limiting and has been relatively well studied high-throughput pyrosequencing, respectively. Although the (Limpiyakorn et al. 2006; Park and Noguera 2004; resultsfromthetwoapproachesshowsomedisconcordance, Tokutomi et al. 2010). Although some heterotrophic they both indicated that the abundance of AOB in activated bacteria (Robertson and Kuenen 1990) and anaerobic sludge was very low. ammonia-oxidizing (anammox) bacteria (Strous et al. 1999) can also oxidize ammonia to nitrite, AOA and AOB are thought to be the main contributors for environ- mental ammonia oxidation and in bioreactors (Nicol and Schleper 2006). According to the previous reports, Nitro- Electronicsupplementarymaterial Theonlineversionofthisarticle somonasandNitrosospiraarethemostimportant generaof (doi:10.1007/s00253-011-3408-y)containssupplementarymaterial, AOB in activated sludge (Park and Noguera 2004; whichisavailabletoauthorizedusers. Purkhold et al. 2000). Of the two genera, Nitrosomonas : : T.Zhang(*) L.Ye M.-F.Shao has been shown to be the dominant AOB in many EnvironmentalBiotechnologyLaboratory, bioreactors (Limpiyakorn et al. 2006; Park and Noguera TheUniversityofHongKong, 2004;Wells etal.2009).Bycontrast, Nitrosospirasp.were HongKongSAR,China e-mail:[email protected] rarely found in activated sludge (Schramm et al. 1999), : probablyduetotheirrelativelowgrowthrate(Siripongand A.H.Y.Tong S.Lok Rittmann 2007) resulting in an underrepresentation in GenomeResearchCenter,LiKaShingFacultyofMedicine, bioreactors. The limited studies of AOA species in TheUniversityofHongKong, HongKongSAR,China activatedsludgeshowedacompositiondifferentfromthose 1216 ApplMicrobiolBiotechnol(2011)91:1215–1225 existing in other environments such as soil, water column, volume of 30 μl containing 15 μl of iQTM SYBR® Green andsediment(Parketal.2006). Moreover, their abundances Supermix (Bio-Rad), 5 μl of DNA template with the seemed to be much lower (four orders or more) than AOB concentration of about 1 ng/μl, and 0.3 μM of each primer based on analysis of amoA gene copy number (Jin et al. usingthesamecyclingconditions.ThearchaealamoAgene 2010; Limpiyakorn et al. 2010; Wells et al. 2009). was amplified by PCR using the primer set Arch-amoAF In this study, the AOA and AOB diversity in six (5′-STAATGGTCTGGCTTAGACG-3′) and Arch-amoAR bioreactorsfromthewastewater treatmentplants (WWTPs) (5′-GCGGCCATCCATCTGTATGT-3′) (Francis et al. in four countries was investigated using amoA and 16S 2005).PCRamplificationwasperformedina50-μlvolume rRNA genes as the biomarkers. High-throughput pyrose- comprising 25 μl 2× MightyAmp Buffer (TaKaRa), 1 μl quencing of the 16S rRNA genes was used to tabulate the (1.25 U) MightyAmp DNA Polymerase (TaKaRa), 0.3 μM AOB diversity and their relative abundance in the total of each primer, and 20–50 ng of genomic DNA. PCR was bacterial community. Quantitative polymerasechain reaction first incubated at 98 °C for 1 min and was followed by 35 (qPCR) was also used to quantify AOB amoA genes in the cyclesat98°Cfor10s,60°Cfor15s,and68°Cfor60s. same activated sludge samples. Several SYBR green-based The PCR products were visualized by agarose (1%) gel qPCRsystemshavebeentriedtoquantifyAOAamoAgene. electrophoresis in the presence of suitable size markers. However, non-specific amplification and/or the formation of For high-throughput 454 pyrosequencing, the bacterial primer dimer hindered the accurate quantification of AOA DNA was amplified with a set of primers targeting the amoA using this approach. hypervariable V4 region of the 16S rRNA gene. The forward primer is 5′-AYTGGGYDTAAAGNG-3′ and the reverse primers are the mixture of four equally mixed Materials and methods primers: 5′-TACCRGGGTHTCTAATCC-3′, 5′-TACCA- GAGTATCTAATTC-3′, 5′-CTACDSRGGTMTCTAATC-3′, Activated sludge sampling and DNA extraction and5′-TACNVGGGTATCTAATCC-3′(RDP'sPyrosequenc- ing Pipeline, http://pyro.cme.msu.edu/pyro/help.jsp). Barc- In this study, activated sludge samples were taken from odes that allowed sample multiplexing during aeration tanks of eight full-scale WWTPs treating municipal pyrosequencingwereincorporatedbetweenthe454adaptor wastewater in China, Singapore, Canada, and the USA. A and the forward primer. Relevant parameters about these WWTPs are shown in Supplementary Table S1. Sludge samples from the aeration Dideoxy DNA sequencing tank were fixed with 50% ethanol (v/v) on site before transporting to the laboratory for DNA extraction. Ten PCR products were purified by using PCRquick-spinTM milliliters offixed activated sludge samples was centrifuged PCR Product Purification Kit (iNtRON Biotechnology, at 4,000 rpm for 10 min at 4 °C. Approximately 200 mg of Korea). The purified PCR products were ligated to samplepelletwasrecoveredforDNAextractioninduplicate pMD®18-T Vector (TaKaRa). Recombinant plasmid was with a FastDNA® SPIN Kit for Soil (Qiagen, CA, USA), transformedintoE.coliandwhitecoloniesthatgrewonLB which was found to be the most suitable DNA extraction plates containing ampicillin (60 μg/ml), X-Gal, and IPTG methodforthesamplesinthisstudy,asbeingcomparedwith were picked to conduct colony PCR amplification with the other commercial reagents. primer set M13F and M13R. The PCR products were purifiedandsequencedbyABI3730xlcapillarysequencers PCR and quantitative PCR (Applied Biosystems, Foster City, CA, USA) with the primer M13F. Primer set amoA-1F (5′-GGGGTTTCTACTGGTGGT-3′) and amoA-2R (5′-CCCCTCKGSAAAGCCTTCTTC-3′) OTU definition and phylogenetic analysis (Rotthauwe et al. 1997) was used to amplify bacterial amoAgeneina30-μlmixturecontaining0.2μlofTaKaRa AnAOBamoAgeneclonelibrarywasconstructedforeachof Ex TaqTM, 3 μl of 10× Ex Taq Buffer (TaKaRa), 3 μl of the eight activated sludge samples (sample IDs—CN-NJ-SJ, 10 mM dNTP mixture (TaKaRa), 0.2 μM of each primer, CN-HR-UN, SG-SG-UP, CN-QD-TD, CN-SH-TS, CN-BJ- and 20–50 ng of genomic DNA. Thermal cycling param- JX, US-CO-CO, and CA-GP-GP). Individual AOA amoA eters followed the protocol of Rotthauwe and colleagues geneclonelibrariesweresuccessfullyconstructedforsamples (1997).AOBamoAgenecopynumberswerequantifiedby CN-NJ-SJ, CN-HR-UN, and SG-SG-UP, while the other usinganiCyclerIQSystem(Bio-Rad,Hercules,CA,USA) samples failed to generate library products. Approximately in triplicate with primer set amoA-1F/amoA-2R. Quantita- 20 clones in each clone library were selected randomly and tive real-time PCR amplification was performed in a total sequenced. The resulting AOB amoA gene sequences were ApplMicrobiolBiotechnol(2011)91:1215–1225 1217 aligned and the Jukes–Cantor distances between subsequent JF271927–JF271985. The pyrosequencing results are de- pairs of sequences were calculated with DNADIST from posited into the NCBI short reads archive database the PHYLIP package (http://www.phylip.com/), and were (accession number SRA026842.2). grouped into 43 OTUs with a distance cut-off of 3%. In order to construct the phylogenetic tree, one sequence in each OTU was selected, merged, and aligned with the Results reference sequences from NCBI Entrez Database. The neighbor-joining phylogenetic tree of AOB amoA gene Diversity of AOB amoA gene and 16S rRNA gene sequences was created by MEGA software (Kumar et al. in different activated sludge samples 2008). The translated protein sequences were assigned to different OTUs with a 3% distance cut-off. The phyloge- Table S2 suggested that the coverage of the AOB amoA netic tree of AOB AmoA protein sequences was also geneclonelibrarieswasover70%exceptsampleCN-SH-TS created by MEGA software. AOA amoA gene and AOA witharelativelylowcoverage(65%).AsshowninFig. 1, a AmoA protein sequences were analyzed in the same total of 43 OTUs were generated based on 163 AOB amoA manner as that of AOB. The AOB 16S rRNA gene gene sequences in eight clone libraries with a 3% distance sequences were also classified into different OTUs with a cut-off. Figure 1 reveals an interesting phenomenon that 3% distance cut-off. In addition, Good’s estimator of relativelyfewamoAOTUsweresharedamongtheactivated coverage (Good 1953) was calculated for each AOA and sludge samples, indicating that the amoA genes were quite AOB amoA gene clone library under 3% distance cut-off diverse and not widely disseminated. (Table S2). Figure 2 illustrated that most of these bacterial amoA OTUs were affiliated to Nitrosomonas genus with Nitro- High-throughput pyrosequencing somonas ureae, Nitrosomonas oligotropha, Nitrosomonas marina, and Nitrosomonas aestuarii being the four most PCR amplicon libraries were prepared using a minimal dominant species. However, only two OTUs (OTU-17 and number of amplification cycles (25 cycles) to minimize the OTU-32) were in the Nitrosospira lineage and these two accumulation of PCR artifacts. Amplicons were purified OTUs represented only eight sequences in the total 163 usingPCRquick-spinTMcolumns(iNtRONBiotechnology). sequences, indicating the low abundance of Nitrosospira in Equal amounts of amplicon products bearing individual various types of activated sludge. sequence barcode for each sludge samples were combined Figure S1 shows the relative abundance and distribution for multiplex pyrosequencing on the Roche 454 FLX of AOB AmoA protein OTUs in different activated sludge Titanium platform (Roche, Nutley, NJ, USA). samples. Compared with AOB amoA gene OTUs (shown in Fig. 1), the diversity of AmoA protein (a total of 26 Sequence analysis and phylogenetic assignment OTUs) was less than that of amoA gene (43 OTUs) due to codon wobble positions. Twelve OTUs of AOB AmoA Following pyrosequencing, Python scripts were written to protein were shared by at least two activated sludge (1)removesequencescontainingmorethanoneambiguous samples. Our findings indicate that a number of different base(‘N’),(2)checkthesequenceintegrityofthebarcodes amoA genotypes encode essentially the identical or very and partitioned the multiplex reads to the individual similar proteins possibly reflecting the functional con- samples, and (3) remove sequence reads shorter than 150 straints in efficient ammonia oxidation. As shown in the bases.Theresultingfilteredreadswerethencomparedwith phylogenetic tree of AOB AmoA protein in Fig. S2, most the Greengenes 16S rRNA gene database (DeSantis et al. OTUs were affiliated with N. ureae, N. oligotropha, N. 2006) using NCBI’s BLASTN tool (Altschul et al. 1990) marina, and N. aestuarii lineages, exactly the same as the with default parameters set to a maximum hit number of amoAgenephylogenetictree.TherewereonlyafewOTUs 100 (Claesson et al. 2009). Sequences were then assigned inNitrosospiralineageandN.europaealineage,whichwas to NCBI taxonomies with MEGAN (Huson et al. 2007) also identical with the above amoA gene phylogenetic using the Lowest Common Ancestor (LCA) algorithm and analysis. default parameters (absolute cut-off—BLAST bitscore 35; In addition to the amoA gene, the 16S rRNA gene was relative cut-off—10% of the top hits). alsousedtoinvestigatetheAOBcommunityinsixsamples (i.e., CN-NJ-SJ, CN-QD-TD, CN-HR-UN, US-CO-CO, Accession numbers SG-SG-UP, and CA-GP-GP) by high-throughput pyrose- quencing. As shown in Table 1, more than 20,000 16S The sequences obtained from clone library in this study rRNA gene fragment sequences were obtained for each were deposited in GenBank under accession numbers sample site. These sequences were assigned to NCBI 1218 ApplMicrobiolBiotechnol(2011)91:1215–1225 90% 80% 70% 60% e g 50% nta 40% ce r e 30% P OTOU-T4OU-3T4OU2-T4OU1-T4OU0-T3OU9-T3OU8-T3OU7-T3OU6-T3OU5-T3OU4-T3UO3-T3OU2-T3UO1-T3OU0-TA2UO9-TO2OU8-T2BUO7-T2 OU6-aT2UO5m-T2U4O-oT2UO3A-T2U2O -Tg2U1O-eT2U0On-T1Ue9O-T1 U8OO-T17UO-TT16UOU-T15UOs-T1U4O-1T3UO-T12UO-1T1U-O1T0UO-9TUO-8TUO-7TU-O6TUO-5TU-O4TU-O3TU-O2TU-1 CNC-NNCJ-Q-NSC-DJNHC--TRNSUD--HUSBS-N-JTGCC-SJ-OA0SX1%--G20CG0%-OPU%-PGP Fig.1 RelativeabundancesanddistributionofAOBamoAgeneOTUsindifferentactivatedsludgesamples taxonomies using BLAST and MEGAN software. A totalsequencesobtainedfrompyrosequencingandtheAOB- tabulation of AOB-like sequences is shown in Table 1. like sequences in each of the six samples. AOB sequences The filtered reads were merged, aligned, and classified into weresuccessfullyobtainedinfiveoutofthesixsamples,and 18 OTUs with a Jukes–Cantor distance cut-off of 3%. the percentages range from 0.29% to 0.64%. For US-CO- Figure 3 shows that most (about 61%) AOB 16S rRNA CO, a sample from Columbia, USA, no AOB sequences OTUs wereshared bymore than twosamplesites, and two were obtained, probably due to the low abundance of AOB majorOTUs,i.e.,OTU-14andOTU-15,accountedfor49% in this wastewater treatment plant, in agreement with amoA of total sequences. These results implied the AOB 16S gene quantification results. rRNA genes are much more conservative than amoA gene and AmoA protein. Diversity of AOA amoA gene in different activated sludge samples Abundance of AOB in activated sludge In the present study, multiple experiments were attempted Along with direct tabulation of AOB 16S rRNA genes by to amplify amoA gene from the eight samples using four pyrosequencing, we also quantify AOB abundances in the PCR systems (including TaKaRa Ex Taq, TaKaRa Mighty activated sludge samples by qPCR, normalizing the AOB Amp DNA Polymerase, Sigma Taq DNA Polymerase, and amoA gene copy number against per nanogram of the Bio-Rad SYBR® Green Supermix) at different thermal genomicDNAextractedfromactivatedsludge.Table1shows conditions (annealing temperature from 50 to 60 °C). the abundances of AOB amoA genes in different activated Amplification was successful from only five samples as sludgesamplesquantifiedbyqPCR.SampleCN-SH-TS has shown in Fig. S3 and only three samples were eventually the most abundant AOB amoA gene copy numbers among successfully cloned and sequenced. Additionally, the the eight activated sludge samples tested. Table 1 lists the presence of primer dimers was very serious in all samples ApplMicrobiolBiotechnol(2011)91:1215–1225 1219 Fig. 2 Neighbor-joining phylo- 99 Nitrosomonas oligotropha (AF272406) genetictreebasedonbacterial 50 OTU-18 amoAgenesequences.The 83 Nitrosomonas sp. NL7 (AY958704) evolutionarydistanceswere 77 Nitrosomonas sp. Nm47 (AY123830) computedusingtheJukes– OTU-35 Cantormethodandareinthe 1938 OTU-16 unitsofthenumberofbase 97 OTU-21 substitutionspersite.Bootstrap OTU-2 valuesareindicatedonbranch 3178 OTU-12 nodes.Sequencesobtainedin 69 OTU-38 thisstudyareshownwith 58 OTU-36 “OTU-”inthenames.Other 40 OTU-40 sequenceswereobtainedfrom 15 OTU-24 GenBank.Thetreewas 55 OTU-39 N. oligotropha lineage out-groupedwithNitrosococcus OTU-23 13 halophilus(gammaproteobacte- 96 OTU-8 rialAOB)amoAsequence 99 OTU-14 7 Nitrosomonas sp. Nm59 (AY123831) Nitrosomonas sp. Nm84 (AY123818) OTU-43 12 29 36 OTU-13 99 Nitrosomonas sp. Nm86 (AY123819) 98 OTU-9 Nitrosomonas sp. JL21 (AF327919) 36 OTU-5 99 OTU-4 50 53 OTU-34 86 OTU-15 OTU-32 52 53 OTU-28 N. ureae lineage 49 OTU-31 OTU-42 98 Nitrosomonas ureae (AF272403) 54 Nitrosomonas sp. AL212 (AF327918) 20 75 Nitrosomonas aestuarii (AF272400) Nitrosomonas marina (AF272405) 82 OTU-37 20 20 77 OTU-41 OTU-6 OTU- 48 11 N. marina and N. aestuarii 47 OTU-7 32 OTU-25 69 OTU-10 33 OTU-29 98 OTU-27 65 55 OTU-33 OTU-32 Nitrosospira multiformis (AY177933) 99 100 Nitrosovibrio tenuis (AY123824) 71 Nitrosospira tenuis (AJ298720) Nitrosospira lineage 64 Nitrosospira briensis (AY123821) 49 Nitrosospira sp. CT2F (AY189143) 39 OTU-17 Nitrosomonas cryotolerans (AF272402) 99 OTU-1 100 OTU-3 Nitrosomonas-like OTU-22 78 OTU-26 57 Nitrosomonas communis (AF272399) 40 32 Nitrosomonas sp. Nm58 (AY123820) Nitrosomonas nitrosa (AF272404) 55 Nitrosomonas halophila (AF272398) OTU-19 55 OTU-30 99 N.europaea lineage 98 Nitrosococcus mobilis (AF037108) 100 Nitrosomonas europaea (AF037107) Nitrosococcus halophilus (AF272521) 0.05 1220 ApplMicrobiolBiotechnol(2011)91:1215–1225 A) and it was not possible to quantify amoA using SYBR® N eD Green qPCR approach applied in this study (Supporting ng ges/n informationFig.S3). amoA(copie 34.5 0.538 2.81 142 5.55 1.29 74.3 1.72 assHesoswmeevnetr,ofitthisesdtiilvlerosfityintoefreAstOtAo haamvoeAagperneelimininathrye activated sludge samples by exploring the sequences ge obtained from the clone libraries of the samples CN-NJ- a ent SJ, CN-HR-UN, and SG-SG-UP, all of which had Good’s c OB pers estimator of coverage over 80% (Table S2). The 56 AbundanceofA 16SrRNAgeneintotalsequence 0.42%(20,955) 0.64%(24,437) 0.46%(22,593) ND ND 0%(25,499) 0.34%(23,928) 0.29%(22,593) wrsOfereoeTsqumruUeletssAnschiuOneassBritnehfdgiarsomaasmotmuAddotihysnOe,tgaiTntthUmcwreseuaelsc(tFiufspoitagl-uemo.nf1pfds)lae,otmhsfsaopwt3mlet%ehser.eeA(AJFcOulOidagAsgA.siian4afmg)ime.odofDArAoiinfmOfgteoeTrnetU1hen5esst arenotasdiverseasAOBamoAgeneindifferentactivated ce sludge samples. While it should be mentioned that the n Aue sample amount in this study was kind of limited, thus this Nq Rse issue needs more studies to be confirmed in the future. As 16Srgene 87 156 105 ND ND 0 81 66 shown in the phylogenetic tree of AOA amoA gene sequences in Fig. 6, most of the sequences obtained in this ofUs study were distinctly different from the previously reported No.OT 8 8 7 9 12 12 5 3 sequences, especially the sequences recovered from marine andsediment.OnlytwoOTUs(OTU-10andOTU-11)were grouped into the marine and sediment lineages. e c AOB amoAsequen 21 21 20 21 20 20 20 21 Discussion s U T Diversity of AOB amoA gene and 16S rRNA gene s O WTP of in different activated sludge samples W No. 6 ND 5 ND ND ND 10 ND ht A previous study reported that the amoA gene similarities g ei among different AOB species ranged from 65% to 100% AOBin Diversity AOA amoAsequence 20 ND 16 ND ND ND 20 ND o(Pbutariknheodldinetthails. s2t0u0d0y).wTehree 6s2im%ilatroitie1s00o%f.thTeheOhTiUghs d diversity of the amoA gene found in this study was likely n a due to the different geographic locations of these sludge undanceofAOA Cityandcountry Nanjing,China Qingdao,China Harbin,China Beijing,China Shanghai,China Columbia,USA Singapore Guelph,Canada strheapemTsopehrtlesWesrtWehasanuTtdlPtsssht.hoinewetdhdiifsfNesrietturnodtsyoompweoernraeatisocnoin/nsssetiewsataedgnetofwcNoitnhitdrioptsrioeovnsipsoiruians b being the dominant AOB in nitrification bioreactors in a d WWTP (Park and Noguera 2004; Wells et al. 2009). n a al y n Notably,sevenOTU-17-associatedsequenceswereallfrom ersit egio the sample CN-HR-UN taken from Harbin, a city in aryofthediv WWTP Suojincun Tuandao Harbin Jiuxianqiao Tianshan ColumbiaR UluPandan Guelph nacpmoorenroftsehniresgtrlenontwhtCowtsheeiimnthasp.aetmhrTaepthuleperdreelsavitoiniotfuu4tdsh–ei1rse0opsfo°tuCrttdhy(itsA.havOctriautNhyraitmorisobissoteehtsrepvaialrh.taii2og0nsh0pe3psist;. m Table1Sum Sample CN-NJ-SJ CN-QD-TD CN-HR-UN CN-BJ-JX CN-HR-UN US-CO-CO SG-SG-UP CA-GP-GP ND:NoData SnlOiuniTrmeiUpabog-e2ner6ga)onagfdnrodOsueTpvReUeidrtstamlfabaOrenTalnowUn2asgy0in(0Ofg7roT).mtUoW-t1heN,eiOtakrloTsnsoUoow-mo3nbo,snAOearOTsvBUee-usa2pr2oes,pcmaaienaeslad.l ApplMicrobiolBiotechnol(2011)91:1215–1225 1221 Fig.3 Relativeabundancesand distributionofAOB16SrRNA geneOTUsindifferent activatedsludgesamples 60% 50% 40% ge a nt 30% ce er 20% P OTU-O18TU-O17TU-O16TU-1O5TU-1O4TU-A1O3OTU-B1O2T1U6-1SO1T Ur-R10ONTAU-9 OgTeU-n8OeT UO-7TOTUU-s6OTU-5OTU-4OTU-3 OTU-2 OTU-1 CNC-NNJC--QNSUD-JHS-STR-CGD-CUO-SA0N-1CG-%G0O-UP%-PGP ItwasreportedthattheAOB16SrRNAgenesimilarities obtainedinthisstudyweregroupedinNitrosomonasgenus. generally ranged from 90% to 100%, much higher than There were 12 OTUs in Nitrosomonas oligotropha group, those of amoA gene (about 65–100%) and AmoA protein four OTUs in the group of Nitrosomonas ureae, and one (about 72.5–100%) (Purkhold et al. 2000). The phyloge- OTU in the N. marinaand N. aestuarii lineages, indicating netic tree in Fig. 5 shows that all of these sequences the dominance of these species in the activated sludge Fig.4 Relativeabundancesand distributionofAOAamoAgene OTUsindifferentactivated sludgesamples 70% 60% 50% e g a 40% nt e 30% erc P 20% OTU-15OTU-14OTU-13OTU-12OTAU-O11OATU-a10mOToUA-9 OgTeU-n8eOT OU-7TOUTUs-6 OTU-5 OTU-4 OTU-3 OTU-2 OTU-1 CN-NCJN-S-HJSRG0-1%-US0%NG-UP 1222 ApplMicrobiolBiotechnol(2011)91:1215–1225 OTU-1 25 17 OTU-5 N. ureae lineage 12 OTU-8 Nitrosomonas ureae(AF272414.1) 1 67 OTU-6 Nitrosomonas cryotolerans(AF272423.1) OTU-10 21 Nitrosomonas aestuarii(AF272420.1) 71 N. marina and N.aestuarii lineage 8 63 Nitrosomonas marina(AF272418.1) Nitrosomonas oligotropha(EF016119.1) 51 87 OTU-3 78 OTU-4 OTU-2 9 8 OTU-13 N.oligotropha Lineage OTU-14 51 OTU-12 35 OTU-15 15 35 OTU-16 16 OTU-9 OTU-7 Nitrosomonas-like lineage 64 97 OTU-11 Nitrosomonas communis(AF272417.1) 93 14 Nitrosomonas nitrosa(AF272425.1) 45 OTU-17 Nitrosomonas-like lineage 100 OTU-18 Nitrosomonas halophila(AF272413.1) Nitrosomonas europaea(HM446362.1) 73 86 Nitrosomonas eutropha(HM446363.1) Nitrosospira briensis(M96396.1) Nitrosococcus mobilis(M96403.1) 30 Nitrosospira tenuis(AJ298746.1) 46 Nitrosovibrio tenuis(M96397.1) 51 74Nitrosolobus multiformis(M96401.1) Nitrosococcus halophilus(AF287298.1) 0.02 Fig.5 Neighbor-joiningphylogenetictreebasedonAOB16SrRNA substitutionspersite.Bootstrapvaluesareindicatedonbranchnodes. genesequences.Theevolutionarydistanceswerecomputedusingthe Sequences obtained in this study are shown with “OTU-” in the Jukes–Cantor method and are in the units of the number of base names.OthersequenceswereobtainedfromGenBank samples.Thisalmostperfectlymatchestheaboveresultsof the betaproteobacterial AOB, no reads can be assigned to this study using amoA gene as the biomarker. A minor thegammaproteobacterialAOB16SrRNAgenes,indicating exception is that Nitrospira 16S rRNA gene was not found thedominanceofbetaproteobacterialAOBinsamplesofthis by pyrosequencing, probably due to the low abundance of study. these AOB species in activated sludge and that the sequencing depth (>20,000 reads per sample) of the Abundance of AOB in activated sludge pyrosequencing applied in this study was not of sufficient depth to represent the complexity of the rare AOB TheabundancesofAOBinactivatedsludgewerequantified sequences. Although we identified 495 sequence reads to usingqPCRandpyrosequencing.However,thetwodatasets ApplMicrobiolBiotechnol(2011)91:1215–1225 1223 showacertaindegreeofdiscordancethatmightbeduetothe oneofthesesamples;(3)theqPCRnormalizationstepmade followingreasons:(1)primerbiasinqPCRamplificationmay use of total activated sludge DNA, which contains nucleic introducemanyuncertainties,especiallywhenthetargetgene acids from eukaryotes, bacteria, archaea, and viruses to copynumbersareatlowlevels(Herrmannetal.2008; Smith which any variations in the relative distribution of these et al. 2006); (2) the 454 pyrosequencing conducted in this components between samples would distort the qPCR study may not be at a sufficient read depth to reflect the normalization process; and (4) 16S rRNA gene copy AOB complexity since AOB sequences was not detected in numbers in prokaryotic microorganisms (Klappenbach et al. OTU-8 100 59 R60-70 242 (DQ534853) [Soil] 91 OTU-4 89 OTU-5 92 OTU-14 OTU-3 41 95 OTU-6 OKR-C-4 (DQ148868) [Soil] OTU-12 100 89 OTU-1 99 100 DI-1 (DQ278513) [WWTP] 64 OTU-2 34 100 HM-1 (DQ278530) [WWTP] OTU-9 66 CB-23(DQ278512) [WWTP] OTU-7 100 OTU-15 21 OTU-13 47 Alpine2 (DQ534698) [Soil] AOAB sC01(EU339396) [Soil] HF770 18E07(EF106900) [Marine] 100 95 OTU-10 DI-21(DQ278528) [WWTP] 62 48 BS2-130mA8(EF414239) [Marine] 100 MB C130m 3(DQ148821) [Marine] 88 SL-57(EU860282) [WWTP] 96 SouthBay-C1-4 (DQ278571) [Sediment] 100 SF06E-BA10-C02(EU650806)[Sediment] 87 Nitrosopumilus maritimus SCM1(EU239959) [Sediment] MX 6 14v (DQ148614) [Marine] 96 99 MM-15 (DQ278564) [WWTP] 50 SF04-SU002S-H03(EU651171) [Sediment] SF NB1 10(DQ148642) [Sediment] 85 OTU-11 50 CBSamoA1-2-5(EF450782) [Soil] 87 95 CBSamoA1-1-4(EF450780) [Soil] Nitrosocaldus yellowstonii strain HL72. 0.05 Fig.6 Neighbor-joiningphylogenetictreebasedonAOAamoAgene Sequences recovered in this study are shown with “OTU-” in the sequences. The evolutionary distances were computed using the names.OthersequenceswereobtainedfromGenBank.Thewordsin Jukes–Cantor method and are in the units of the number of base squarebracketsindicatethesourcethesequencesoriginatedfrom substitutionspersite.Bootstrapvaluesareindicatedonbranchnodes. 1224 ApplMicrobiolBiotechnol(2011)91:1215–1225 2001) are different and the amoA gene copy numbers in Open Access This article is distributed under the terms of the Creative Commons Attribution Noncommercial License which per- different AOB are also different (Norton et al. 2002). mits any noncommercial use, distribution, and reproduction in any medium,providedtheoriginalauthor(s)andsourcearecredited. Diversity of AOA amoA gene in different activated sludge samples Although many reports described the diversity and abun- References dance of AOAs in the natural environments, only limited information is available describing AOA diversity in the activatedsludge(Limpiyakornetal.2010;Parketal.2006; AltschulS,GishW,MillerW,MyersE,LipmanD(1990)Basiclocal alignmentsearchtool.JMolBiol215(3):403–410 Wells et al. 2009; Zhang et al. 2009). One possible reason AvrahamiS,LiesackW,ConradR(2003)Effectsoftemperatureand for this situation could be due to the difficulty in PCR fertilizer on activity and community structure of soil ammonia amplificationofAOAamoAfromDNAofsludgesamples. oxidizers.EnvironMicrobiol5(8):691–705 In this study, only five out of eight samples were ClaessonM,O'SullivanO,WangQ,NikkilaJ,MarchesiJ,SmidtH, De Vos W, Ross R, O'Toole P (2009) Comparative analysis of successfully amplified, and only three samples were pyrosequencing and a phylogenetic microarray for exploring eventually successfully cloned and sequenced. microbial community structures in the human distal intestine. Previous studies revealed that most of AOA should be PloSOne4(8):e6669 assignedtotheclustersofCGI.1a(themarineandsediment DeSantis T, Hugenholtz P, Larsen N, Rojas M, Brodie E, Keller K, Huber T, Dalevi D, Hu P, Andersen G (2006) Greengenes, a lineage) and CGI.1b (the soil lineage) in the phylum chimera-checked 16S rRNA gene database and workbench Crenarchaeota (Hatzenpichler et al. 2008). In this study, it compatible with ARB. Appl Environ Microbiol 72(7):5069– was found that most of the AOA in the activated sludge 5072 samples have a closer distance to those AOA found in soil FrancisCA,RobertsKJ,BemanJM,SantoroAE,OakleyBB(2005) Ubiquity and diversity of ammonia-oxidizing archaea in water and thus belonged to the soil lineage. This result was in columns and sediments of the ocean. Proc Natl Acad Sci USA agreement with the previous studies on AOA communities 102(41):14683–14688 in activated sludge (Limpiyakorn et al. 2010; Park et al. GoodI(1953)Thepopulationfrequenciesofspeciesandtheestimation 2006), except for a study on the activated sludge in ofpopulationparameters.Biometrika40(3–4):237–264 Hatzenpichler R, Lebedeva EV, Spieck E, Stoecker K, Richter A, bioreactors treating saline sewage (Jin, et al. 2010) which Daims H, Wagner M (2008) A moderately thermophilic had dominant AOA communities belonging to the marine ammonia-oxidizing crenarchaeote from a hot spring. Proc Natl and sediment lineage. AOA AmoA protein sequences were AcadSciUSA105(6):2134–2139 classifiedintonineOTUsusing3%distancecut-off.FigureS4 Herrmann M, Saunders AM, Schramm A (2008) Archaea dominate the ammonia-oxidizing community in the rhizosphere of the shows the relative abundance and distribution of AOA freshwatermacrophyteLittorellauniflora.ApplEnvironMicrobiol AmoA protein OTUs in different activated sludge samples, 74(10):3279–3283 indicating lower diversity compared with AOA amoA gene Huson D, Auch A, Qi J, Schuster S (2007) MEGAN analysis of andthatmostoftheAOAAmoAproteinOTUswereshared metagenomicdata.GenomeRes17(3):377–386 Jin T, Zhang T, Yan Q (2010) Characterization and quantification of amongthethreesamples.Theneighbor-joiningphylogenetic ammonia-oxidizing archaea (AOA) and bacteria (AOB) in a tree based on AOA AmoA protein sequences in Fig. S5 nitrogen-removing reactor using T-RFLP and qPCR. Appl demonstrates the samepattern as shown by the AOA amoA MicrobiolBiotechnol87(3):1167–1176 gene tree (Fig. 6), that is, most of the OTUs in this study Klappenbach J, Saxman P, Cole J, Schmidt T (2001) rrndb: the ribosomal RNA operon copy number database. Nucleic Acids were grouped together and had a far distance from those Res29(1):181–184 sequences of marine and sediment. Interestingly, there are Kumar S, Nei M, Dudley J, Tamura K (2008) MEGA: a biologist- twoOTUs(OTU-6andOTU-7inFig.S5)closelyrelatedto centric software for evolutionary analysis of DNA and protein themarineandsedimentAOAlineage.ThetwoOTUswere sequences.BriefBioinform9(4):299–306 Limpiyakorn T, Kurisu F, Yagi O (2006) Quantification of ammonia- only detected in the sludge SG-SG-UP, a sample from oxidizing bacteria populations in full-scale sewage activated Singapore, implying that the occurrence of these OTUs sludgesystemsandassessmentofsystemvariablesaffectingtheir mightbeduetothegeographiclocationofSingapore,which performance.WaterSciTechnol54(1):91–99 is near the ocean. The other two samples were taken from Limpiyakorn T, Sonthiphand P, Rongsayamanont C, Polprasert C (2010)AbundanceofamoAgenesofammonia-oxidizingarchaea cities far from the ocean and thus their AOA populations andbacteriainactivatedsludgeoffull-scalewastewatertreatment mainly come from soil. plants.BioresourTechnol102(4):3694–3701 Nicol G, Schleper C (2006) Ammonia-oxidising Crenarchaeota: Acknowledgments The authors wish to thank the Hong Kong important players in the nitrogen cycle? Trends Microbiol 14 General Research Fund (HKU7197/08E) for the financial support of (5):207–212 this study. Lin Ye wishes to thank HKU for the postgraduate NortonJ,AlzerrecaJ,SuwaY,KlotzM(2002)Diversityofammonia studentship. We would also like to thank W. Chan and C. K. Wong monooxygenaseoperoninautotrophicammonia-oxidizingbacteria. fortechnicalhelpinpyrosequencing. ArchMicrobiol177(2):139–149
Description: