Aalborg Universitet

The role of inoculum and reactor configuration for microbial community compositionand dynamics in mainstream partial nitritation anammox reactors

Agrawal, Shelesh; Karst, Søren M.; Gilbert, Eva M.; Horn, Harald; Nielsen, Per H.; Lackner,SusannePublished in:MicrobiologyOpen

DOI (link to publication from Publisher):10.1002/mbo3.456

Creative Commons LicenseCC BY 4.0

Publication date:2017

Document VersionPublisher's PDF, also known as Version of record

Link to publication from Aalborg University

Citation for published version (APA):Agrawal, S., Karst, S. M., Gilbert, E. M., Horn, H., Nielsen, P. H., & Lackner, S. (2017). The role of inoculum andreactor configuration for microbial community composition and dynamics in mainstream partial nitritationanammox reactors. MicrobiologyOpen, 6(4), [e00456]. https://doi.org/10.1002/mbo3.456

General rightsCopyright and moral rights for the publications made accessible in the public portal are retained by the authors and/or other copyright ownersand it is a condition of accessing publications that users recognise and abide by the legal requirements associated with these rights.

? Users may download and print one copy of any publication from the public portal for the purpose of private study or research. ? You may not further distribute the material or use it for any profit-making activity or commercial gain ? You may freely distribute the URL identifying the publication in the public portal ?

Take down policyIf you believe that this document breaches copyright please contact us at vbn@aub.aau.dk providing details, and we will remove access tothe work immediately and investigate your claim.

MicrobiologyOpen. 2017;6:e456.  | 1 of 11https://doi.org/10.1002/mbo3.456

www.MicrobiologyOpen.com

Received:28November2016  |  Revised:9January2017  |  Accepted:25January2017DOI: 10.1002/mbo3.456

O R I G I N A L R E S E A R C H

The role of inoculum and reactor configuration for microbial community composition and dynamics in mainstream partial nitritation anammox reactors

Shelesh Agrawal1,2  | Søren M. Karst3 | Eva M. Gilbert2,4 | Harald Horn2 |  Per H. Nielsen3 | Susanne Lackner1,2

ThisisanopenaccessarticleunderthetermsoftheCreativeCommonsAttributionLicense,whichpermitsuse,distributionandreproductioninanymedium,providedtheoriginalworkisproperlycited.©2017TheAuthors.MicrobiologyOpenpublishedbyJohnWiley&SonsLtd.

1TechnischeUniversitätDarmstadt,InstituteIWAR,ChairofWastewaterEngineering,Darmstadt,Germany2KarlsruheInstituteofTechnology,Engler-Bunte-Institut,ChairforWaterChemistryandWaterTechnology,Karlsruhe,Germany3CenterforMicrobialCommunities,DepartmentofChemistryandBioscience,AalborgUniversity,Aalborg,Denmark4EnviroChemieGmbH,Rossdorf,Germany

CorrespondenceSheleshAgrawal,TechnischeUniversitätDarmstadt,InstituteIWAR,ChairofWastewaterEngineering,Darmstadt,Germany.Email:s.agrawal@iwar.tu-darmstadt.de

Funding informationGermanResearchFoundation(DFG),Grant/AwardNumber:LA-3087/1-1;GermanEnvironmentalFoundation(DBU);DanishStrategicResearchCouncil(Ecodesign)

AbstractImplementation of partial nitritation anammox (PNA) in themainstream (municipalwastewatertreatment)isstillunderinvestigation.MicrobialcommunitystructureandreactortypecaninfluencetheperformanceofPNAreactor;yet,littleisknownabouttheroleofthecommunitycompositionoftheinoculumandthereactorconfigurationunder mainstream conditions. Therefore, this study investigated the communitystructureofinoculaofdifferentoriginandtheirconsecutivecommunitydynamicsinfourdifferentlab-scalePNAreactorswith16SrRNAgeneampliconsequencing.Thesereactors were operated for almost 1 year and subjected to realistic seasonaltemperaturefluctuationsasinmoderateclimateregions,thatis,from20°Cinsummerto10°Cinwinter.Thesequencinganalysisrevealedthatthebacterialcommunityinthereactorscomprised:(1)anitrifyingcommunity(consistingofanaerobicammonium-oxidizingbacteria (AnAOB), ammonia-oxidizingbacteria (AOB), andnitrite-oxidizingbacteria (NOB)); (2)differentheterotrophicdenitrifyingbacteria andotherputativeheterotrophicbacteria(HB).Thenitrifyingcommunitywasthesameinallfourreactorsat the genus level, although the biomasses were of different origin. Communitydynamicsrevealedastablecommunityinthemovingbedbiofilmreactors(MBBR)incontrast to the sequencing batch reactors (SBR) at the genus level.Moreover, thereactordesignseemedtoinfluencethecommunitydynamics,andreactoroperationsignificantlyinfluencedtheoverallcommunitycomposition.TheMBBRseemstobethereactortypeofchoiceformainstreamwastewatertreatment.

K E Y W O R D S

biofilm,bioreactors,Molecularmicrobialecology,Nitrogenmetabolism

1  | INTRODUCTION

Partial nitritation anammox (PNA) is the combination of anaer-obic ammonium oxidation (Anammox) with partial nitrification.ImplementationofPNAformainstream(municipal)wastewatertreat-ment(WWT)canresultinaparadigmshiftformunicipalWWT,asit

holds promise to lower energy demands, enables better use of theorganiccarbon,andsavescostsforexcesssludgemanagement(Lackneretal., 2015; Siegrist, Salzgeber, Eugster,&Joss, 2008).However, itsapplicabilitytothemainstreamisstillunclearandunderinvestigation.PNAisacomplexbiotechnologicalapplicationwherecontrolledpartialnitrification (nitritation) through aerobic ammonia-oxidizing bacteria

2 of 11  |     AGRAWAL et AL.

(AOB)isrequiredtoprovideanidealratioof1.32gNO-2-N/gNH

+4-N,

for anammox activity under anoxic conditions (Strous, Kuenen, &Jetten, 1999). Maintaining these conditions is challenging due toan intrinsic competitive microbial environment. To improve processunderstandingforbetterreactorperformancesandprocessstability,alinktomicrobialecologyisessential(Rittmann,2006).

SeveralstudieshaveinvestigatedthemicrobialcommunityinPNAsystems.ThekeymembersofthismicrobialcommunityareAOBandanaerobicammonium-oxidizingbacteria (AnAOB),andnitrite-oxidiz-ingbacteria(NOB)(Park,Rosenthal,Ramalingam,Fillos,&Chandran,2010;Perssonetal.,2014;Tsushima,Kindaichi,&Okabe,2007).Somestudiesalsoreportedputativeheterotrophicdenitrifyingbacteria(HB)inPNAsystemsbasedonclonelibraries(Gilbertetal.,2014;Langoneetal.,2014;Park,Rosenthal,Ramalingam,etal.2010).

Despite the importance of HB for the PNA process—positive(Desloover etal., 2011;Wang etal., 2010) or negative (Bürgmann,Jenni,Vazquez,&Udert,2011;Lackner,Terada,&Smets,2008)—char-acterizationoftheheterotrophicbacteriahasbeenverylimited,untilrecently.Only a fewstudieshaveprovideddetailed insight into thecommunity composition of the PNA systems using next-generationsequencing platforms (Chu etal., 2015; Costa etal., 2014; Pellicer-Nácheretal.,2010;Pereiraetal.,2014;Speth,in‘tZandt,Guerrero-Cruz,Dutilh,&Jetten,2016).Thesestudiesreportedtwoimportantaspects:(1)avastlydiversecommunitycompositioninPNAsystems,(2)theneedtofurtherinvestigatethewholecommunitycompositionandstructureinPNAsystems,toshedmorelightontheroleofeachcommunitymemberforremovingnitrogeninPNAsystems,directlyorindirectly.However,thesestudieswereperformedonsystemswhichwerenotoperatedundermainstreamconditions,ratherasanammoxenrichment reactors or sidestream PNA reactors. Thus, knowledgeaboutwholemicrobial community structure of PNA systems undermainstreamconditionsremainstobeelucidated.

Forstart-upofsidestreamPNAsystems,itiscommonpracticetoinoculatewithbiomassfromexistingPNAsystemsduetothelowspe-cificgrowth ratesofAnAOB.Sucha start-up strategy isevenmoreimportantformainstreamPNAsystemsduetolowtemperaturesandlow nitrogen concentrations and thus even lower specific activitiesandgrowthratesofbothAnAOBandAOB (Hendrickxetal.,2012).However,previousstudieshavereportedthatthePNAreactorscon-tain only one dominant AnAOB (Hu etal., 2010; Park, Rosenthal,Jezek,etal.2010), indicating that specificenvironmental conditionsmight support one AnAOB over another. For example, anammoxreactorswith continuously high nitrite concentrations and COD: NratiosreportedapredominanceofthegenusCa.BrocadiaoverotherAnAOBphenotypes (Jenni,Vlaeminck,Morgenroth,&Udert, 2014;Laurenietal.,2015).Thereis,however,noclearconsensusinpreviousstudiesonwhatdrivesthedevelopmentofoneparticularanammoxstrainoveranother.Suchvariationsalsoextendtotheoccurrenceofdifferent genera ofAOB.Therefore, selection of the right inoculumforamainstreamPNAsystemmightbecrucialduetovariationsinthedominantAnAOBandAOB.Thechoiceof theright inoculummightextendtoothercommunitymemberstoo,astheyarealsosignificant(Spethetal.,2016).

AllknownAnAOBgenerahavebeen identified indifferentPNAsystemconfigurations (Eglietal.,2001;Park,Rosenthal,Jezek,etal.2010; Park, Rosenthal, Ramalingam, etal. 2010; van der Star etal.,2007;Wangetal.,2010).A recentstudyhighlighted theneed foradeep taxonomic resolution, that is, down to genus or species level.TheyreportedthatthebiomasscontainingCa.Brocadiafulgidaseemedtobelessinfluencedbyadecreaseintemperaturethanbiomasscon-taining Ca. Brocadia sinica (Lotti, Kleerebezem, & van Loosdrecht,2015).OtherstudiesreportedbiofilmPNAsystemswithCa.Brocadiaasapreferredchoiceatlowtemperature(Gilbert,Agrawal,Schwartz,Horn,&Lackner,2015;Lackner&Horn,2013;Laurenietal.,2016).Faststart-upandstableperformanceofmainstreamPNAsystemsthusreliesontheinoculumcompositionandthereactorconfiguration.16SrRNAgeneampliconsequencingprovidessucharesolutionforcom-prehensivecommunitydetermination,althoughitalsohasitspitfalls(Albertsen,Karst,Ziegler,Kirkegaard,&Nielsen,2015).

Thequestion remainshowdifferent thebiomasscomposition indifferent full-scale PNA systems actually is, and thus, how import-ant it is to choose the “right” inoculum considering that themicro-bial community structuremightbedistinct fordifferentwastewatercompositions,plantconfigurations,andoperatingconditions.Also,thesubsequentimpactofatemperaturedecreaseandtheadjustedimple-mentationofPNAinthemainstreamsarestillunclear.

Inourpreviousstudy(Gilbertetal.,2015),wecomparedprocessperformanceoffourdifferentPNAlab-scalereactorsoperatedundermainstream conditions. The biomass originated from four differentfull-scale PNA reactors at three different locations.We also moni-toredtheabundancesofAnAOB,AOB,andNOBbasedontheqPCRanalysis.TheqPCRresultssuggestedthepresenceofotherabundantmicrobial members. Also, the process performance data revealedsignificantdifferences inperformanceamong the four lab-reactors .Thesefindingsraisedfurtherconcernson:(1)theanalysisofthewholemicrobialcommunityanditscomposition,(2)differencesinthecom-munitycompositionamongtheinoculaandtheresponsibilityfordif-ferencesinreactorperformances.

Therefore, inthisstudy,wefocusedonansweringthequestionsbasedonfindingsfromotherstudiesaswellasourpreviouswork.Theoverallaimofthisstudywastwofold: (1)acomprehensivecompari-sonof themicrobialcommunitycompositionof thebiomasses fromfourfull-scalesidestreamPNAsystemswhichwereusedasinoculainthefourlab-scalereactorsoperatedundermainstreamconditions;(2)in-depthmonitoringofthespeciesabundancedistributionandtheirdynamicsintheselab-scalereactors,subjectedtoagradualtempera-turedecreaseandlownitrogenconcentrations,anddeterminethecor-relationbetweenthepreviouslyreportedreactorperformanceswiththeresponseofthewholecommunitytothechangeinconditions.

2  | MATERIALS AND METHODS

2.1 | Reactor setup and operation

Ourmodelsystems,togaincomprehensiveinsightintothecompositionanddynamicsofthetotalmicrobialcommunityofthefourdifferentPNA

     |  3 of 11AGRAWAL et AL.

biomasses,werefour10llab-scalereactors(Gilbertetal.,2015)—twosequencingbatchreactors(SBRs),onewithsuspendedbiomass(SBR1)andonewithgranularbiomass (SBR2); twomovingbedbiofilmreac-tors(MBBRs),onewithBiofilmChip™Mcarriermaterial(MBBR1)andone with K3®carriermaterial(MBBR2),(bothAnoxKaldnesAB,Lund,Sweden) which were operated as described in Gilbert etal. (2015).Thesereactorswereofparticularinterest,asthebiomassinthesereac-tors originated from four different full-scale sidestream PNA whichalso differ in their biomass enrichment methods (Table1), and theirbiomassisusedtoinoculateotherPNAreactorsinEurope.Allfourlab-reactorswereoperated inparallelandfedwithsyntheticwastewaterwith50mg-Nl−1ammonium(seealsoSuppl.Information).Thereactorswereequippedwithonlinesensorsfordissolvedoxygen(DO),tempera-ture, pH, conductivity, and ammonium and nitrate (Endress+Hauser,Germany).All reactorswere controlledbasedonammoniumeffluentconcentrationswhichweresetto6–8mg-Nl−1byadjustingtheinfluentflowrate.ThepHwasmaintainedat7.3±0.3,DOandbiomassconcen-trationsareprovided inTable1.Reactorperformancesandmicrobialpopulationdynamicsweremonitoredoveraperiodof45weeks.TheappliedtemperatureprofileisshowninFig.S1andwasasfollows:reac-toroperationstartedatweek0andduringphaseI,thereactorswereoperatedat20°C(15weeks);inphaseII,thetemperaturewasgraduallydecreasedby0.5°Cevery7daysdowntoatemperatureof10°C(weeks16–35);phaseIIIcoveredoperationatconstant10°C(week36–45).

2.2 | Microbial community analysis

2.2.1 | Sampling and amplicon library preparation

Biomasssampleswerecollectedfromthereactorsduringoperationandstoredat−80°Cforfurtheranalysis(seealsoSuppl.Information).Extracted genomic DNA was used for library preparation employ-ingaprocedureadapted fromCaporasoetal. (2011).Using5ng/μl of DNA, template PCR amplification (see also Suppl. Information)was performed in duplicate using hypervariable region V4 primers

(targeting253bppartial16S rRNAgene sequence) fusedwithbar-codes and adapters for the IlluminaMiSeq platform. The duplicatelibraryreactionswerepooledtogether,andthelibrarieswerecleanedusingAgencourtAMPureXPbeads(BeckmanCoulter)withthestand-ardprotocol.ThequalityofthelibrarywascheckedusingTapeStationandD1KScreenTape(AgilentTechnologiesInc.,USA).

2.2.2 | Sequencing and sequence analysis

ThemultiplexedampliconlibrarywassequencedonaMiSeqSystem(Illumina) usingMiSeq Reagent Kit v3 (Illumina). Samples from thelibrarieswereanalyzedbyaquickquality check toevaluateoveralldata quality (check the quality score, the occurrence of ambiguousbases and k-mer abundances). Afterward, the libraries were pro-cessedto(1)truncateread1to250bplengthanddiscardread2;(2)screen forPhiXcontamination; (3) formatmetadata forQIIME;and(4)removeuniquereadsandnormalizelibrariestoanevendepthof10,000readspersample.

OTU (operational taxonomical unit) picking and classificationwasdoneusingQIIMEv.1.7(quantitativeinsightintomicrobialecol-ogy) (Caporasoetal.,2010)with itsdefaultsettings.Overall, thedenovoclusteringofOTUswasdonewith97%identity,correspondingto species level.The sequenceswere then classifiedusing theRDP(Ribosomal Database Project) classifier (80% confidence thresh-old)basedon the taxonomy in theGreengenesdatabase (97%con-fidence threshold,version13.5May2013) (McDonald etal., 2012).Additionally,NCBIBLASTdatabasewasusedtodeterminetheclosestbacterialrelatives(having100%sequencehomology)ofthereportedOTUs. OTU representative sequences have been submitted to theGenBankundertheaccessionnumbersKY226724-KY228337.

2.2.3 | Data analysis

The sequencing datawas analyzed using R to gain insight into theoverallcommunitystructuredynamicsover time inall four reactors

TABLE  1 Characterizationoftheinoculaandreactoroperationdata(averagesovertheentire45weeks)

SBR 1 SBR 2 MBBR 1 MBBR 2

Initialbiomass

Biomassorigin SidestreamSBR,Germany SidestreamICreactortheNetherlandsa

SidestreamMBBRSweden SidestreamMBBRSweden

Inoculation fromsidestreamSBR none Swedenfrompilotplant/effluentfromsidestreamMBBR

Biomasstype Suspended(x90,3 = 640 μm) Granulated(x90,3 = 2070 μm)

Biofilm(max.thickness2mm)

Biofilm(max.thickens10mm)

lab-reactors

TSS[gl−1] 0.8±0.5 2.0±0.5 5.7±0.5 9.3±0.8

VSS[%] 60±5 56±0.3 68±2 70±1.5

AverageDO[mgl−1]

0.09±0.15 0.11±0.08 0.20±0.17 0.25±0.13

HRT[d] 6.3±4.8 2.6±1.3 1.4±0.8 1.5±0.9

x90,3,particlesizefromFeretdiameter;TSS,totalsuspendedsolid;VSS,volatilesuspendedsolids;HRT,hydraulicretentiontime;DO,dissolvedoxygen.aICreactor,Internalcirculationreactor.

4 of 11  |     AGRAWAL et AL.

individually. To analyze the core representatives of the microbialcommunity,OTUcountdatafromQIIMEwasimportedintoRinbiomformat alongwith amapping file containing samplemetadata. KeyR-packagesused foranalysiswerephyloseq (v1.7.12) (McMurdie&Holmes,2013),ggplot2(v0.9.3.1),andVEGAN(v2.0.9).ThePhyloseqpackagewasusedconsideringitsfunctionasawrapperforfunctionsfrom other ecological packages including VEGAN. The Alpha andBetadiversity indicesof thesampleswerecalculated fromthenor-malized data. Community dynamics comparisonwithin the reactorscorrespondingtospecifictaxawasperformedwithconstrainedcor-respondenceanalysis (CCA). Square-root transformedand centeredOTUcountswereusedforconstrainedCCAanalysis.Datawerecon-strainedbytemperature,asloweringoftemperaturewasthevariableparameterintheexperiment.

3  | RESULTS AND DISCUSSION

3.1 | Characterization of the inocula—microbial community composition

Thebiomassesthatwereusedtoinoculatethefourlab-scalereactors,withoutanycross-inoculationduringthe45weeksofoperation,origi-natedfromdifferentfull-scalesidestreamfacilitiestreatingcentratesfromsludgedewateringunits.Theoperatingconditionsatthesesitesdifferedwithinfluentammoniumconcentrationsof500–1,000mg-Nl−1,effluentammoniumconcentrationsaround50–100mg-Nl−1,andoperatingtemperaturesbetween22and34°C,causedbytheinstalla-tioninthesidestream(afteranaerobicdigestion)(Lackneretal.,2014).

The characteristics of the biomasses are summarized inTable1.Thebiomassesvariedintheirphysicalstructure(type)asaresultofthereactortype:twoinoculawerefromsequencingbatchreactors,sus-pended (SBR1)andgranulated (SBR2)biomasses, respectively,char-acterizedbydifferentparticlesizedistributions.Theothertwoinoc-ulacameoncarriermaterialswithbiofilmsofmaximumthicknessesof 2mm (MBBR1) and 10mm (MBBR2), respectively. For clarity, inthis study, the totalmicrobial communitywasdivided intoanitrify-ingmicrobial community (NMC, typically studied community) and aheterotrophicmicrobialcommunity(HMC).TheNMCincludedAOB,AnAOB,andNOB;theHMCincludedputativeheterotrophicbacte-ria,(partial)denitrifiers(HB),anddenitrificationintermediatereducers(onlyreducingnitrogenoxides).

All four inocula had similar NMC diversity, constituting similarfamily members of the phyla Planctomycetes (AnAOB), Nitrospirae(NOB),andProteobacteria(AOB).Inallinocula,theAnAOBbelongedtothegenusCa.Brocadia,theAOBstounclassifiedmembersofthefamilyNitrosomonadaceae,andtheNOBtothegenusNitrospira; only the relative readabundancediffered (Figure1).TheclassificationofAOBwas limitedtothefamilyNitrosomonadaceae.Therefore,man-ual searchwithotherdatabases (RDPclassifier andNCBI)wasper-formed. The closest relative identity belonged to theNitrosomonas europaea-eutrophagroup.IntheMBBRs,AnAOBaccountedfornearly35%–40%,AOBfor~1%,andNOB<1%;intheSBRs,thefractionofAnAOBwas12%–20%,AOB1%–3%,andNOB2%–5%.Alsowithin

theHMC,thesamedominantphylawerepresentinalltheinocula(Fig.S2),dominatedbysimilarfamilymembersofthephylaProteobacteria,Chloroflexi,andChlorobi.Therelativereadabundancesofthisgroup(HMC)variedmainlybetweenSBRsandMBBRs,withanoverallper-centageof55% to75%of the total relativeabundances (Figure1a,Fig.S2).

SimilaritiesbetweentheinoculawerealsofoundontheOTUspe-cies level,althoughtheinoculahadsuchdifferentorigins.EspeciallywithintheNMC,alltheinoculasharedasingledominantOTUforeachNMCmember (Figure1b).Only the inoculumofSBR2hadanaddi-tionaldominantOTUwithintheAnAOB.TheclosestrelativetoOTU877foundindatabaseswasCa.Brocadiasp.40andCa.BrocadiasinicaJPN1toOTU797.AsidefromtheNMC,thepresence/occurrenceofthecommondominantOTUswithintheHMCwasalsoevaluated.Thetop10dominantOTUswereselectedfromeachreactor,andcommonOTUswerecomparedandthendistinguishedbasedonthefollowingthreecategories:(1)sharedbyallreactors,(2)sharedamongSBRsandMBBRs,respectively,and(3)commoninanythreereactors(Figure1).Similartrendswereobservedevenforthewholecommunitycomposi-tionatOTUlevel:allfourinoculahadthelargestoverlapfollowedbytheoverlapwithin theMBBRsandSBRs, respectively (Fig.S3,S.Fig.S4A).ThedominantOTUscommoninallfourinoculawereaffiliatedtothegeneraRubrivivax,Rhodoplanes,Dok59,unclassifiedmembersofthefamiliesA4b,andIgnavibacteriaceae(Figure1b).ThedominantOTUsaffiliated toSHA-31,Xanthomonadaceae, and someunclassi-fiedfamilieswerepresentonlyintheSBRs;someOTUsaffiliatedtounclassifiedfamilieswerespecifictotheMBBRsonly.ThisobserveddifferentiationbetweentheSBRsandMBBRswasfurtherconfirmedbytherichnessdiversityindexes(TableS2)andPCAanalysis(SeeFig.S3fordetails).

TheresultsclearlyshowedthatthepredominantNMCmembersandseveralHMCmemberswerepresentinalltheinocula,whichwasnotexpectedconsideringthedifferentsettings(reactortype,waste-watercomposition)ateachsite.Similartoourresultcharacterizingtheinocula,otherstudiesalsoreportedlowfractionsofAOBandNOBinPNAreactors(Almstrandetal.,2014;Kindaichietal.,2007).Therel-ativeabundanceofCa.Brocadiawasrelativelyhighinallfourinocula,but itwas theHMC that constituted themajorbiomass fractionofeachinoculum(Figure1a)althoughtheCOD/Nratioswere<2inthesystemsoforigin(Lackneretal.,2014).OurresultsareincongruencewithpreviousstudiesaboutthecoexistenceofHBinanammoxsys-tems (Costa etal., 2014;GarciaCostas etal., 2012;Kindaichi etal.,2007; Langone etal., 2014; Laureni etal., 2015; Ni, Ruscalleda, &Smets,2012;Spethetal.,2016).Inastudy(Nietal.,2012),thepres-enceof(residual)organiccontentincludingsolublemicrobialproducts(SMP)intheside-andmainstreamwastewaterswasreportedtosup-porttheHMCinPNAsystems.

Inourstudy,Ca.Brocadiawasthesingle identicalAnAOBdomi-nantinallreactors,whichdeviatesfromresearchthatdetectedeightunique dominant AnAOB phylotypes in eight different nitrogen-removalreactors(Huetal.,2010).ThegenusCa.BrocadiahadasingledominantOTUinalltheinoculaexceptSBR2,whichhadoneadditionalOTU. Ca.Brocadiasp.40andCa.Brocadiasinica,closest relatedto

     |  5 of 11AGRAWAL et AL.

F IGURE  1  (a)Microbialcommunitycompositionofthebiomassesfromfourfull-scalePNAplants(usedasinoculumforeachlab-reactor)atfamilylevel.Forheterotrophicmicrobialcommunity(HMC),familiescontributing<2%tothetotalrelativereadabundancearemergedasothers.(b)MicrobialcommunitycompositionanddynamicsatOTUlevel.Thisheatmapshowstherelativeabundanceofthemostdominantmicrobialpopulationineachreactor.Thetop10OTUsfromeachreactorwereselected(atotalof25OTUsfromallthesamples).Theassociatedphyla,families,andgeneraareontherightpaneloftheheatmap.UpperpanelshowsOTUsassociatedwithputativeheterotrophsandlowerpanelforthenitrifyingcommunity.ColorKey(naturallogscaleofrelativepercentagereadabundance)showstherelatedabundanceineachsample.Pointscale(naturallogscaleofrelativepercentagereadabundance)presentstheOTUswith=<1%ofthereadabundance.SBR1(suspendedbiomass),SBR2(granularbiomass),MBBR1(BiofilmChip™M),andMBBR2(K3®)

6 of 11  |     AGRAWAL et AL.

OTU877andOTU797,respectively,havebeenidentifiedtothriveinNH4

+andNO2-richenvironments(Laurenietal.,2015;Oshiki,Satoh,

&Okabe,2016;Oshiki,Shimokawa,Fujii,Satoh,&Okabe,2011;Park,Rosenthal,Ramalingam,etal.2010).The reason fordetecting theseOTUs in our reactorswas that the biomasses originated fromPNAsystemstreatingwastewatershighinnitrogenconcentration.

Interestingly, all biomasses fostered the same AOB and NOB,sharingasinglecommonOTUrelatedtoNitrosomonas europaea-eutro-phaandNitrospira,respectively.Theseobservationscanbeexplainedbasedon theaerationstrategiesemployed inPNAsystems—limitedoxygen,andtherefore, lowertransientnitriteavailability.Thesecon-ditions are reported to supportNitrosomonas europaea-eutropha (Lietal.,2009)andNitrospira(Park,Rosenthal,Jezek,etal.2010).

These results suggest thatvery similarNMCcompositionswereobtained in PNA systems if operated under generally similar con-ditions, and irrespective of reactor configuration.Our results are inpartial agreementwith another study (Park, Rosenthal, Jezek, etal.2010),inwhichdifferentreactortypes(i.e.,SBRandMBBRinoculatedwithdifferentbiomasses)constitutedCa.Brocadiasp.40butdiffer-ent Nitrosomonas species.A recent study (Laureni etal., 2015) alsoemphasizedthepresenceofCa.BrocadiainmainstreamPNAsystemsexposedtocomplexsubstratesoverotheranammoxspecies.

Furthermore, our results disclosed the coexistence of domi-nant OTUs of the families Comamonadaceae, Hyphomicrobiaceae,Rhodocyclaceae, A4b, Sinobacteraceae, Xanthomonadaceae, andIgnavibacteriaceae in all the inocula, although with diverse abun-dances, as also reported in previous studies (Kindaichi,Yuri, Ozaki,&Ohashi,2012).Presumably,theseHBcancoexistindifferentPNAsystems due to the existence ofvariablemacro- andmicroenviron-ments (Spethetal.,2016).ThehighabundanceandsimilarityoftheHMChighlightthatnotonlydoesNMCbelongtothecorecommunityinPNAsystemsbutalsotheHMCmembers.Thedifferentoxicandanoxicphases,andthepresenceofsubstrategradientsduetobiomassstructureprovideconditionsforsuchadiversecorecommunity.

Basedon theHMCmembers detected, the biomasses aremet-abolically very versatile, comprising members which can carry out(partial)denitrification,metabolizeorganiccarbon(an)aerobically,andassistinformingmicrobialaggregates.ThefamiliesComamonadaceae,Hyphomicrobiaceae, Rhodocyclaceae, Sinobacteraceae, Ignavibacte-riaceae,andSJA-28containknown(partial)denitrifiersand/ordenitri-ficationintermediatereducers(Liuetal.,2012;Nietal.,2011;Sadaieetal.,2007).Chloroflexiwasthedominantfilamentousspecies,knowntosupportaggregateformation(Nietal.,2011),andreportedtoscav-engesecondarymetabolitesderivedfromAnAOBunderanoxiccondi-tions(Kindaichietal.,2012).Someofthemcanalsodenitrify(McIlroyetal.,2016).Suchhighdiversityandfunctional redundancy indicatetheirimportanceinthePNAprocess.

3.2 | Reactor performance

Detailedperformanceanalysesofallfourreactorshavebeenreportedpreviously (Gilbert etal., 2015). Here, we summarize the opera-tionalperformanceofthesereactorsanddeterminereasonsforthe

observed performance change in conjunctionwith the response ofthewholemicrobialcommunitytomainstreamconditions.Withthestartofreactoroperation(week0),thetemperaturewassetto20°Candonlyammoniumandnutrientswerefedtothelab-reactors,henceremovinganyexternalcarbonsource,comparedtotheconditionsatthefull-scalefacilities(afewhundredmg/LorganiccarbonmeasuredasCOD).Thesereactorswereoperatedwithorganiccarbon-freeinflu-enttoreducethepotentialcontributionofdenitrificationtoPNAandtobeabletoattributechangesmainlytothetemperaturegradient.Allfourlab-reactorswereexposedtothesametemperatureprofileoverthreedefinedphases(Fig.S1)—phaseI(weeks0–15),start-upat20°C;phase II (weeks 16–35), temperature gradient (20°C to 10°C); andphaseIII(weeks36–45)operationat10°C.Table1providesthemainoperating parameters.Over the course of this study, biomass con-centrationswerequitestable.However,theconcentrationsdifferedsignificantlybetweenthereactors,with0.8and2.0g-TSSl−1fortheSBRs,and5.7and9.3g-TSSl−1fortheMBBRs(Table1).Deviationsinoperationalparameters(HRTandDO)werebasedonthebiomasscharacteristics,thatis,differentthicknessesandtotalamounts.

Reactor performances are summarized in Figure2 with addi-tional details in S. Table1. Initial nitrogen loadings (week 0) werehighest in theMBBRs (around200g-Nm−3 d−1), followedby SBR2(approx.80g-Nm−3 d−1)andSBR1(40g-Nm−3 d−1),whichwasmainlyattributed to the different biomass concentrations. The ammoniumremovalwasabove90%and similar in all four reactors.By theendofphaseI,thenitrogenloadingwasreducedtomaintaintheeffluentconcentration.Thebiomass-specific conversion rateswerebetween5 and 8g-N kg-TSS−1 d−1.During the temperature decrease (phaseII),theturnoverratereducedbyafactorof2,aswasexpectedforatemperaturedecreasedownto10°C.InphaseIII,ammoniumremovalremainedratherstable.

Nitrite andnitrateproduction significantlydifferedbetween thereactors.DuringphaseII,asignificantnitriteaccumulationoccurredinSBR1andSBR2(fromalmost0%(week15)to62%inSBR1and54%inSBR2(week35)),withasimultaneousdecreaseinnitrateproduc-tiontoalmostnothing(3%–6%inweek35)(Figure2,TableS1).NitriteaccumulationinMBBR1alsoincreasedsignificantlyfrom0%to30%inphase III.OnlyMBBR2didnotaccumulateanynitriteduring thetemperaturedecrease.InphaseIII,nitriteproductionremainedathighlevelsexceptforMBBR2(<5%oftheconvertedammonium).Nitrateproductiondecreasedinallreactorsduringthetemperaturegradientfrom 20°C to 10°C from 25%–40% to less than 15%. The reactorperformancescorrelatedwellwithbiomassstructureandaggregate/biofilmthickness.MBBR2withabiofilmofupto10mmthicknessper-formedbest(leastnitriteandnitrateaccumulation),whereasthereac-tors with thinner biofilms/smaller aggregates struggled with nitriteaccumulation.

3.3 | Microbial community dynamics

Theresultsofthereactoroperationsuggestthatthesystemswereinfluenced differently by the temperature gradient, although influ-entcompositionandoperatingconditionswerethesameinallfour

     |  7 of 11AGRAWAL et AL.

reactors. The questionwas nowwhether themicrobial communitycompositioncouldexplain thedifferences inbehavior.Therefore,aconstrainedcorrespondenceanalysis(CCA)wasperformedasafirststeptoascertainwhetherthetemperaturechangeaffectedthecom-munitycompositionovertime.Thisanalysisrevealedthatthemicro-bial community composition in the MBBRs was relatively stable,whereasitwashighlydynamicintheSBRs(Figure3).Theresponseof the community composition showedahigh correlation (ANOVA(analysis of variance), p<.001) to the temperature decrease andnitrite accumulation (Figure2, Figure3). The influence of the tem-peraturechangeoncommunitystructurewasmoresignificantfortheSBRscomparedto theMBBRs,especially inSBR1,whichexhibiteda continuous instability during the entire operation, correspondingto lower efficiency (Figure2, and Supplementary information fordetails).

3.3.1 | Nitrifying microbial community

Thecorrelationof thecommunitydynamicswiththenitritebuildupsuggestedasignificantinfluenceofthetemperaturedecreaseontheNitrifyingmicrobialcommunity (NMC).However,nitritebuildupcanoriginatefromlossofAnAOBactivityaswellasfromlossofNOBactiv-ityorarelativeincreaseinAOBactivity;thus,causingoverproduction

ofnitrite.AcloserlookattheindividualmembersoftheNMCmighthelptobetterunderstandtheexactcircumstances.Therelativeabun-dances,especiallyofAOBandCa.Brocadia(AnAOB),decreasedinall

F IGURE  2 Reactorperformance(averageovercompletedatasetduringeachphase)duringthethreephases:I–20°C,II–20°C–10°C,III10°C,expressedasnitrite(leftcolumn)andnitrate(rightcolumn)produced,basedontheammoniumconverted,inallfourreactors.Boxplotsshowmedianvalues(horizontalline),aswellas25thand75thpercentiles(box)and10thand90thpercentiles(errorsbars)

F IGURE  3 Constrainedcorrespondenceanalysisplotforallfourreactorsovertime,exhibitingtheinfluenceofthechangeinoperatingtemperature(i.e.,20°Cto10°C).SBR1(suspendedbiomass),SBR2(granularbiomass),MBBR1(BiofilmChipTMM),andMBBR2(K3®).The x-axis(horizontal)referstodifferencesbetweenreactortypes.The y-axis(vertical)relatestothechangeintemperature/time

8 of 11  |     AGRAWAL et AL.

fourreactorsinweek10(phaseI)comparedtotherespectiveinocula(Figure1B,Figure4,Fig.S6).Afterward,Ca.Brocadia intheMBBRs(representedbyOTU877)wasquitestablewithasequencereadfrac-tion significantly higher than in the SBRs (Figure1b, Figure4). TherelativereadabundancesintheSBRswerelessstableanddecreasedovertime. InSBR1,therelativeabundanceofCa.Brocadiareducedto<5%inweek36(phaseIII)andfurtherto1.2%attheendofphaseIII(week45,10°C);inSBR2,itreducedto~3%inweek36(phaseIII),whichcoincidedwiththeabsenceofanotherOTU(Ca.Brocadia,OTU797).NochangeintherelativeabundanceofAOBwasobservedintheMBBRs;minor changes appeared in theSBRs.AOBabundancewaslowandrepresentedbythesingleOTU1573inallthereactorsover the entire time. TheNOBs, also represented by a singleOTU(Nitrospira,OTU475)(Figure1b,Figure4),exhibitedachangeintherelativeabundancealsointheMBBRs,thoughitwasonlyminor;OTU475 increased to1.5%duringphase II in all reactors and remainedstableexceptinMBBR2(reducedagaininphaseIII).Overall,thelossin Ca.Brocadiaabundance,especiallyintheSBRs,seemedtobethemostinfluentialpartforthebuildupofnitrite.

3.3.2 | Heterotrophic microbial community

Another reason for the differences in reactor performances mightstemfromthelargeHeterotrophicmicrobialcommunity(HMC)inallreactors.DiversityandrichnesswithintheHMCattheendofphaseIincreasedinallreactorsexceptSBR1comparedtotheinoculabio-masses(Fig.S6,TableS2).Afterward,thedifferentiationincommunitydynamicsamongtheSBRsandMBBRswasanalogoustotheNMC(Fig.S2,Fig.S6).ThecommunitydiversityintheMBBRswasrelativelysta-blecomparedtotheSBRs.TherelativeabundanceprofilesofthemostdominatingOTUsassociatedwithdifferentphylawithintheHMCdif-feredinthefourreactors(Figure1b,TableS3).Theybelongedtothegenera Rubrivivax, Rhodoplanes, Dok59, Dokdonella, Thermomonas,and some unclassified members of the phyla Chloroflexi, Chlorobi,Actinobacteria,andAcidobacteria.FewOTUsintheMBBRsandSBRsbehavedsimilarlyinthedifferentphases(TableS3).DuringphaseII,the relativeabundanceof someOTUs increasedsignificantly in the

SBRs (Figure1b, Fig. S7). For instance, Thermomonas (OTU 1663)increased,especiallyinSBR1,andthisOTUaccountedforupto60%ofthetotalrelativereadabundance.SeveralotherOTUsalsoshowedspecific responses, for example, the relative abundance within thefamilies Comamonadaceae (OTU 726) and Chromatiaceae (OTU1364) increased(upto10%) inSBR1;OTU920andOTU227fromgenus Dok59 increased (up to 9%) in SBR2. No significant changeoccurred in theMBBRsduringphase II. Inphase III, theabundanceprofilesofvariousOTUsassociatedtoChloroflexi(Anaerolineaeclass)reduced inall the reactors.Additionally, theabundanceofChlorobi(OTU1072)decreasedtoafourthinSBR1;however,itincreasedinMBBRs.

3.4 | Interrelationship between microbial community, reactor type, and reactor performance

This study clearly showed that the temporal community dynamicswerelinkedtothereactortypeanddrivenbyatemperaturedecrease.Thecommunities shifted in all reactors from the initial inocula (Fig.S5),particularlytheabundanceofthenitrifyingcommunityreduced,attributedtotheselectivepressureimposedbychangesintheopera-tional conditions. Although the inocula of all four lab-reactors har-bored Ca.Brocadia(asthesameAnAOB)andthesameAOBatOTUspecieslevel,theresponsetothetemperaturechangevariedsignifi-cantly.Theimposedtemperaturedecreaseledtoasignificantreduc-tionintherelativeabundanceofCa.BrocadiaintheSBRscomparedtotheMBBRs,whichcorrespondedtothedifferentvolumetricammo-niumremovalratesat10°C(approx.5–8g-Nm−3 d−1intheSBRsand12–16g-Nm−3 d−1intheMBBRs).Theseratesweresimilartootherstudies(Dostaetal.,2008;Huetal.,2013).MBBR2performedbetterthantheotherreactors,especiallyintermsofstability(lowestnitriteaccumulation). This suggests that the reactor configuration is animportantfactor.Inthisstudy,biofilmthicknessdifferedsignificantlywithbetterperformancesforthickerbiofilms/aggregates.

Despite great interest in mainstream PNA and availability ofmany different biomasses and reactor types for bio-augmentation,efforts to evaluate the temperature effect on various biomasses is

F IGURE  4 Relativeabundanceofnitrifyingmicroorganismsinallthereactors.SBR1(suspendedbiomass),SBR2(granularbiomass),MBBR1(2mmbiofilm),andMBBR2(10mmbiofilm)

     |  9 of 11AGRAWAL et AL.

stillvery limited.Recently, a studyevaluated thebehaviorofdiffer-entbiomassesatalowertemperaturebybatchexperiments,however,restrictedtoAnAOBandAOB(Lottietal.,2015).TheyreportedthattheAnAOBactivityinbiofilmswaslessinfluencedthaninsuspendedbiomass.We found a strong correlationbetween theNMCcompo-sition and the change in temperature, the nitrite accumulation (i.e.,performance and stability), and the reactor type in our long-termstudy. Interestingly, the observed NMC dynamics occurred onlywithintheframeworkoftheinitiallydetectedOTUs.Themagnitudeoftheeffectofthetemperaturedecreasewaslinkedtothethicknessofthemicrobialaggregates.Inourstudy,SBR1(withsuspendedbio-mass) responded firstbyadecrease inAnAOBpopulationand sud-den nitrite accumulation, followed by SBR2 (granular biomass).ThebiofilmsystemofMBBR1(thinbiofilm)showedasimilarresponsetonitriteaccumulationalthoughtheAnAOBabundancewasstable.OnlyMBBR2maintainedastableperformance.Thissuggeststhattempera-ture influenced abundance and activity. Considering these reactorswere fed onlywith ammonium, the source for nitrite can be eitherammoniumoxidationbyAOBand/orreductionofavailablenitratetonitritebypartialdenitrifiers,whereasthenitritesinkcanbeanaerobicammoniumoxidationbyAnAOB,nitriteoxidationbyNOB,ornitritereduction by nitrite reducers.The continuous presence of nitrite inPhaseIIandIIIsuggestsanimbalancebetweennitritesourceandsink.AOBwereproducingnitrite,butitsconsumptionwasnotcounterbal-ancedbyeitherofthemicrobialmembersresponsibleforthenitritesink.AnAOBwereinfluencedbythetemperaturedecreaseratherthanthe competition for nitrite imposed byNOB and denitrifiers,whichis often attributed to the failure of PNA systems (Joss etal., 2011;Lackneretal., 2015).Moreover,NOBalso seemed tobenegativelyinfluencedby the temperaturedecrease.Althoughnitritewas avail-able,noincreaseintheirpopulationwasobserved.Huang,Gedalanga,Asvapathanagul, andOlson (2010)also reportednegative impactoflowtemperatureonNitrospira(alsodetectedinourstudy).Theaccu-mulation of NO2

- (Figure2) clearly indicated thatAOB activity andcompetitionfornitritewerenotresponsibleforthepoorreactorper-formancebelow13°C.

Theabsenceoforganiccarbonintheinfluentofthelab-reactorshadnosignificantinfluenceonthecompositionoftheHMC.Thus,inconsensuswithprevious studies (Ni etal., 2011,2012),most likely,solublemicrobialproductsfromtheNMCprovidedthecarbonsourcefortheHMCmembers.Temperatureandnitriteconcentrationsinflu-enced thedynamicsof theHMC.For instance, nitrite accumulationduringphase II ledtoan increaseofOTUsassociatedwithChlorobi(Ignavibacteriaceae), Xanthomonadaceae, Comamonadaceae,Rhodocyclaceae,andChromatiaceae,SJA-28(Figure1b,Fig.S7,TableS4), respectively. Such a response might have been caused by (1)highnitriteavailabilitythattriggereddirectnitritereductionwithoutrelyingonnitrateavailability.Forexample,Xanthomonadaceaemem-bersonlyreducenitritetonitrousoxide(N2O)(Finkmann,Altendorf,Stackebrandt,&Lipski,2000).(2)Increaseingrowthofonecommu-nity member support growth of others, as most members in PNAbiomass rely on exchange and transfer of nitrogen cycle intermedi-ates among each other (Speth etal., 2016). At 10°C, temperature

played an important role in the composition of the HMC. ExceptXanthomonadaceae (OTU1663, a closest relativewas identified asThermomonas fusca,whichcan thriveat10°C (Mergaert,Cnockaert,&Swings,2003;Langoneetal.,2014))othermembersof theHMCwerenegativelyaffected.Thisobservation is supportedbypreviousstudies,thatlowertemperaturehasanegativeimpactondenitrifiers(Lu,Chandran,&Stensel,2014).Additionally, the reactorconfigura-tionwasalsorelevantforthedynamicsoftheHMC.TheMBBRshadaratherstabletotalcommunitycompositioncomparedtotheSBRs.SimilarresultswereobtainedinastudybyLaurenietal.(2016),whoreportedhighercommunitydynamicsinsuspendedbiomasscomparedtobiofilmcarriersinPNAreactorsoperatedat15°C.SuchcommunitybehaviorseemedtobecausedbythelackofstructuralpreservationwhichwasavailableintheMBBRs.

4  | CONCLUSIONS

The detailed microbial community study performed on biomassesoriginating from different full-scale PNA sidestream systems (aspotential inocula formainstream application) and their response tomainstreamconditionsinthefourlab-scalereactorsledtothefollow-ingkeyconclusions:(1)DifferentlyconfiguredPNAsystemsoperatedunder generally similar conditions revealed a similarmicrobial com-munitycompositionofthefunctionalmicrobialcommunity(i.e.,AOB,AnAOB), including the NOB and putative heterotrophs (including(partial)denitrifiersa.o.).Thus,morestudiesarerequiredtodeterminethePNAcorecommunity.(2)UbiquitouspresenceoftheHMCinalllab-reactorstreatingsyntheticfeedfreeoforganiccarbonunderlinedtheir significance for nitrogen removal. Further research to under-standthisintricatecommunityanditscorrelationwiththemainstreamPNAsystems inthepresenceofexternalorganiccarbonalongwithsolublemicrobial products, that is, the influence of organic carbonontheHMCcompositionanditsroleinPNAsystems,shouldbethefocusoffutureresearch.(3)TheimpactoftemperaturedecreaseontheperformanceandcomprehensivecommunitycompositioninPNAsystemsundermainstreamconditionscorrelatedwiththereactorcon-figuration.TheMBBRsshowedbetterperformanceandamorestableoperationatlowertemperaturesasaresultofamorestablemicrobialcommunity,emphasizingtheadvantageofcarrier-basedbiofilms,withthepotentialsafeguardingofthecommunitytotransientunfavorableenvironmentalchangesandtherebypreventingwashout.

ACKNOWLEDGMENTS

This studywas funded by theGermanResearch Foundation (DFG)underGrant number LA-3087/1-1. Studywas also supported fromtheDanishStrategicResearchCouncil (Ecodesign). E.M.Gwas sup-ported by a doctoral scholarship from the German EnvironmentalFoundation (DBU). The authors also thankMads Albertsen for hissuggestions regardingdata analysis and the teamat theCentre forMicrobialCommunities,AalborgUniversityfortheirkindsupport.TheauthorsthankAdrianoJossfromEawag,Switzerlandforhishelpwith

10 of 11  |     AGRAWAL et AL.

theprogrammingandcodingforthereactorcontrolandtheEngler-Bunte-Institutssocietyfortheirinfrastructuralsupport.

CONFLICT OF INTEREST

Noconflictofinterestisdeclared.

REFERENCES

Albertsen,M.,Karst,S.M.,Ziegler,A.S.,Kirkegaard,R.H.,&Nielsen,P.H.(2015).Backtobasics-The influenceofDNAextractionandprimerchoiceonphylogeneticanalysisofactivatedsludgecommunities.PLoS ONE,10,1–15.

Almstrand, R., Persson, F.,Daims,H., Ekenberg,M., Christensson,M.,&Wilén,B.-M.,…Hermansson,M.(2014).Three-dimensionalstratifica-tionofbacterial biofilmpopulations in amovingbedbiofilm reactorfornitritation-anammox.International Journal of Molecular Sciences,15,2191–2206.

Bürgmann,H.,Jenni,S.,Vazquez,F.,&Udert,K.M.(2011).Regimeshiftandmicrobialdynamicsinasequencingbatchreactorfornitrificationandanammoxtreatmentofurine.Applied and Environmental Microbiology,77,5897–5907.

Caporaso, J. G., Kuczynski, J., Stombaugh, J., Bittinger, K., Bushman, F.D.,Costello,E.K.,…Knight,R.(2010).QIIMEallowsanalysisofhigh-throughputcommunitysequencingdata.Nature Methods,7,335–336.

Caporaso,J.G.,Lauber,C.L.,Walters,W.A.,Berg-lyons,D.,Lozupone,C.A.,Turnbaugh,P.J.,…Knight,R.(2011).Globalpatternsof16SrRNAdiversityatadepthofmillionsofsequencespersample.Proceedings of the National Academy of Sciences of the United States of America,108,4516–4522.

Chu,Z.R.,Wang,K.,Li,X.K.,Zhu,M.T.,Yang,L.,&Zhang,J.(2015).Microbialcharacterizationofaggregateswithinaone-stagenitritation-anammoxsystem using high-throughput amplicon sequencing. Chemical Engineering Journal,262,41–48.

Costa,M.C.M.S.,Carvalho,L.,Leal,C.D.,Dias,M.F.,Martins,K.L.,Garcia,G.B.,…CalabriaAraujo,J.(2014).Impactofinoculaandoperatingcon-ditionsonthemicrobialcommunitystructureof twoanammoxreac-tors. Environmental Technology,35,1811–1822.

Desloover, J., De Clippeleir, H., Boeckx, P., Du Laing, G., Colsen, J.,Verstraete,W.,&Vlaeminck,S.E.(2011).Floc-basedsequentialpartialnitritationandanammoxatfullscalewithcontrastingN2Oemissions.Water Research,45,2811–2821.

Dosta,J.,Fernández,I.,Vázquez-Padín,J.R.,Mosquera-Corral,A.,Campos,J.L.,Mata-Alvarez,J.,&Méndez,R.(2008).Short-andlong-termef-fects of temperature on the anammox process. Journal of Hazardous Materials,154,688–693.

Egli, K., Fanger,U.,Alvarez, P. J. J. J., Siegrist,H.,vanderMeer, J. R.,&Zehnder, A. J. B. B. (2001). Enrichment and characterization of ananammox bacterium from a rotating biological contactor treatingammonium-richleachate.Archives of Microbiology,175,198–207.

Finkmann, W., Altendorf, K., Stackebrandt, E., & Lipski, A. (2000).Characterization of N2O-producing Xanthomonas-like isolates frombiofilters as Stenotrophomonas nitritireducens sp. nov., Luteimonasmephitisgen.nov.,sp.nov.andPseudoxanthomonasbroegbernensisgen.nov., sp.nov. International Journal of Systematic and Evolutionary Microbiology,50,273–282.

Garcia Costas, A. M., Liu, Z., Tomsho, L. P., Schuster, S. C., Ward, D.M., & Bryant, D. A. (2012). Complete genome of CandidatusChloracidobacterium thermophilum, a chlorophyll-based photo-heterotroph belonging to the phylum Acidobacteria. Environmental Microbiology,14,177–190.

Gilbert,E.M.,Agrawal,S.,Karst,S.M.,Horn,H.,Nielsen,P.H.,&Lackner,S.(2014).Lowtemperaturepartialnitritation/anammoxinamovingbed

biofilmreactortreatinglowstrengthwastewater.Environmental Science and Technology,48,8784–8792.

Gilbert, E.M.,Agrawal, S., Schwartz, T., Horn, H., & Lackner, S. (2015).Comparingdifferentreactorconfigurationsforpartialnitritation/ana-mmoxatlowtemperatures.Water Research,81,92–100.

Hendrickx,T. L.G.,Wang,Y.,Kampman,C.,Zeeman,G.,Temmink,H.,&Buisman,C.J.N.(2012).Autotrophicnitrogenremovalfromlowstrengthwastewateratlowtemperature.Water Research,46,2187–2193.

Hu,Z.,Lotti,T.,deKreuk,M.,Kleerebezem,R.,vanLoosdrecht,M.C.M.,Kruit,J.,…Kartal,B.(2013).Nitrogenremovalbyanitritation-anammoxbioreactoratlowtempertaure.Applied and Environmental Microbiology,79,2807–2812.

Hu,B.,Zheng,P.,Tang,C.,Chen,J.,VanDerBiezen,E.,&Zhang,L.,…Kartal,B. (2010). Identification and quantification of anammox bacteria ineightnitrogenremovalreactors.Water Research,44,5014–5020.

Huang, Z., Gedalanga, P. B., Asvapathanagul, P., & Olson, B. H. (2010).Influence of physicochemical and operational parameters onNitrobacterandNitrospiracommunitiesinanaerobicactivatedsludgebioreactor.Water Research,44,4351–4358.

Jenni,S.,Vlaeminck,S.E.,Morgenroth,E.,&Udert,K.M.(2014).Successfulapplication of nitritation/anammox towastewaterwith elevated or-ganiccarbontoammoniaratios.Water Research,49,316–326.

Joss, A., Derlon, N., Cyprien, C., Burger, S., Szivak, I., Traber, J., …Morgenroth, E. (2011). Combined nitritation-anammox:Advances inunderstandingprocessstability.Environmental Science and Technology,45,9735–9742.

Kindaichi,T.,Tsushima,I.,Ogasawara,Y.,Shimokawa,M.,Ozaki,N.,Satoh,H.,&Okabe,S.(2007).Insituactivityandspatialorganizationofanaer-obicammonium-oxidizing(anammox)bacteriainbiofilms.Applied and Environmental Microbiology,73,4931–4939.

Kindaichi,T.,Yuri,S.,Ozaki,N.,&Ohashi,A.(2012).EcophysiologicalroleandfunctionofunculturedChloroflexiinananammoxreactor.Water Science and Technology,66,2556–2561.

Lackner, S., Gilbert, E. M., Vlaeminck, S. E., Joss, A., Horn, H., & vanLoosdrecht,M.C.M.C.M.(2014).Full-scalepartialnitritation/anam-moxexperiences–anapplicationsurvey.Water Research,55,292–303.

Lackner,S.,&Horn,H.(2013).Comparingtheperformanceandoperationstability of an SBR andMBBR for single-stage nitritation-anammoxtreatingwastewaterwithhighorganicload.Environmental Technology,34,1319–1328.

Lackner,S.,Terada,A.,&Smets,B.F.(2008).Heterotrophicactivitycom-promisesautotrophicnitrogenremovalinmembrane-aeratedbiofilms:Resultsofamodelingstudy.Water Research,42,1102–1112.

Lackner,S.,Thoma,K.,Gilbert,E.M.,Gander,W.,Schreff,D.,&Horn,H.(2015).Start-upofafull-scaledeammonificationSBR-treatingeffluentfrom digested sludge dewatering.Water Science and Technology, 71,553–559.

Langone,M.,Yan,J.,Haaijer,S.C.M.,O.DenCamp,H.J.M.,Mike,S.M.,&Andreottola,G.(2014).Coexistenceofnitrifying,anammoxanddenitri-fyingbacteriainasequencingbatchreactor.Frontiers in Microbiology,5,1–12.

Laureni,M.,Falås,P.,Robin,O.,Wick,A.,Weissbrodt,D.G.,Nielsen,J.L.,…Joss,A.(2016).Mainstreampartialnitritationandanammox:Long-termprocessstabilityandeffluentqualityatlowtemperatures.Water Research,101,628–639.

Laureni, M., Weissbrodt, D. G., Szivák, I., Robin, O., Nielsen, J. L.,Morgenroth,E.,&Joss,A.(2015).Activityandgrowthofanammoxbio-massonaerobicallypre-treatedmunicipalwastewater.Water Research,80,325–336.

Li,X.-R.,Du,B.,Fu,H.-X.,Wang,R.-F.,Shi,J.-H.,Wang,Y.,…Quan,Z.-X.(2009). The bacterial diversity in an anaerobic ammonium-oxidizing(anammox)reactorcommunity.Systematic and Applied Microbiology,32, 278–289.

Liu, Z., Frigaard, N.-U., Vogl, K., Iino, T., Ohkuma, M., Overmann, J., &Bryant,D.A. (2012).Complete genomeof Ignavibacteriumalbum, a

     |  11 of 11AGRAWAL et AL.

metabolicallyversatile,flagellated,facultativeanaerobefromthephy-lumChlorobi.Frontiers in Microbiology,3,1–15.

Lotti,T.,Kleerebezem,R.,&vanLoosdrecht,M.C.M.(2015).Effectoftem-peraturechangeonanammoxactivity.Biotechnology and Bioengineering,112,98–103.

Lu,H.,Chandran,K.,&Stensel,D.(2014).Microbialecologyofdenitrifica-tioninbiologicalwastewatertreatment.Water Research,64,237–254.

McDonald,D.,Price,M.N.,Goodrich,J.,Nawrocki,E.P.,DeSantis,T.Z.,Probst,A.,…Hugenholtz,P. (2012).An improvedGreengenestaxon-omywithexplicitranksforecologicalandevolutionaryanalysesofbac-teriaandarchaea.The ISME Journal,6,610–618.

McIlroy,S.J.,Karst,S.M.,Nierychlo,M.,Dueholm,M.S.,Albertsen,M.,Kirkegaard,R.H.,…Nielsen,P.H.(2016).Genomicandinsituinvesti-gationsofthenovelunculturedChloroflexiassociatedwith0092mor-photypefilamentousbulkinginactivatedsludge.The ISME Journal,10,2223–2234.

McMurdie,P.J.,&Holmes,S.(2013).Phyloseq:AnRpackageforreproduc-ibleinteractiveanalysisandgraphicsofmicrobiomecensusdata.PLoS ONE,8,1–11.

Mergaert,J.,Cnockaert,M.C.,&Swings,J.(2003).Thermomonasfuscasp.nov.andThermomonasbrevissp.nov.,twomesophilicspeciesisolatedfromadenitrificationreactorwithpoly(e-caprolactone)plasticgranulesasfixedbed,andemendeddescriptionofthegenusThermomonas.International Journal of Systematic and Evolutionary Microbiology,53,1961–1966.

Ni,B.J.,Ruscalleda,M.,&Smets,B.F. (2012).Evaluationon themicro-bialinteractionsofanaerobicammoniumoxidizersandheterotrophsinAnammoxbiofilm.Water Research,46,4645–4652.

Ni,B.J.,Zeng,R.J.,Fang,F.,Xie,W.M.,Xu,J.,Sheng,G.P.,…Yu,H.Q.(2011). Evaluation on factors influencing the heterotrophic growthon the soluble microbial products of autotrophs. Biotechnology and Bioengineering,108,804–812.

Oshiki,M.,Satoh,H.,&Okabe,S. (2016).Ecologyandphysiologyofan-aerobicammoniumoxidizingbacteria.Environmental Microbiology,18,2784–2796.

Oshiki, M., Shimokawa, M., Fujii, N., Satoh, H., & Okabe, S. (2011).Physiological characteristics of the anaerobic ammonium-oxidizingbacterium“CandidatusBrocadiasinica”.Microbiology,157,1706–1713.

Park,H.,Rosenthal,A.,Jezek,R.,Ramalingam,K.,Fillos,J.,&Chandran,K.(2010).Impactofinoculaandgrowthmodeonthemolecularmicrobialecologyofanaerobicammoniaoxidation (anammox)bioreactorcom-munities.Water Research,44,5005–5013.

Park,H.,Rosenthal,A.,Ramalingam,K.,Fillos,J.,&Chandran,K. (2010).Linkingcommunityprofiles,geneexpressionandN-removalinanam-mox bioreactors treatingmunicipal anaerobic digestion rejectwater.Environmental Science and Technology,44,6110–6116.

Pellicer-Nácher,C.,Sun,S.,Lackner,S.,Terada,A.,Schreiber,F.,Zhou,Q.,&Smets,B.F.(2010).Sequentialaerationofmembrane-aeratedbiofilmreactors for high-rate autotrophic nitrogen removal : Experimentaldemonstration. Environmental Science and Technology, 44, 7628– 7634.

Pereira,A.D.,Leal,C.D.,Dias,M.F.,Etchebehere,C.,Chernicharo,C.A.L.,&deAraújo,J.C.(2014).Effectofphenolonthenitrogenremoval

performanceandmicrobialcommunitystructureandcompositionofananammoxreactor.Bioresource Technology,166,103–111.

Persson,F.,Sultana,R.,Suarez,M.,Hermansson,M.,Plaza,E.,&Wilén,B.-M.(2014).Structureandcompositionofbiofilmcommunitiesinamov-ingbedbiofilmreactor fornitritation-anammoxat lowtemperatures.Bioresource Technology,154,267–273.

Rittmann,B.E.(2006).Microbialecologytomanageprocessesinenviron-mentalbiotechnology.Trends in Biotechnology,24,261–266.

Sadaie, T., Sadaie, A., Takada, M., Hamano, K., Ohnishi, J., Ohta, N., …Sadaie, Y. (2007). Reducing sludge production and the dominationof Comamonadaceae by reducing the oxygen supply in the waste-water treatment procedure of a food-processing factory.Bioscience, Biotechnology, and Biochemistry,71,791–799.

Siegrist,H.,Salzgeber,D.,Eugster,J.,&Joss,A. (2008).AnammoxbringsWWTP closer to energy autarky due to increased biogas produc-tion and reduced aeration energy for N-removal.Water Science and Technology,57,383–388.

Speth, D. R., in ‘t Zandt, M. H., Guerrero-Cruz, S., Dutilh, B. E., &Jetten,M. S.M. (2016). Genome-basedmicrobial ecology of anam-mox granules in a full-scale wastewater treatment system. Nature Communications,7,1–10.

Strous,M.,Kuenen,J.G.,&Jetten,M.S.M.(1999).Keyphysiologyofan-aerobicammoniumoxidation.Applied and Environmental Microbiology,65,3248–3250.

Tsushima,I.,Kindaichi,T.,&Okabe,S.(2007).Quantificationofanaerobicammonium-oxidizingbacteriainenrichmentculturesbyreal-timePCR.Water Research,41,785–794.

vanderStar,W.R.L.,Abma,W.R.,Blommers,D.,Mulder,J.-W.,Tokutomi,T.,Strous,M.,…vanLoosdrecht,M.C.M.(2007).Startupofreactorsforanoxicammoniumoxidation:Experiencesfromthefirstfull-scaleana-mmoxreactorinRotterdam.Water Research,41,4149–4163.

Wang,C.-C.,Lee,P.-H.,Kumar,M.,Huang,Y.-T.,Sung,S.,&Lin,J.-G.(2010).Simultaneouspartialnitrification,anaerobicammoniumoxidationanddenitrification(SNAD)inafull-scalelandfill-leachatetreatmentplant.Journal of Hazardous Materials,175,622–628.

SUPPORTING INFORMATION

Additional Supporting Informationmay be found online in the sup-portinginformationtabforthisarticle.

How to cite this article:AgrawalS,KarstSM,GilbertEM,HornH,NielsenPH,LacknerS.Theroleofinoculumandreactorconfigurationformicrobialcommunitycompositionanddynamicsinmainstreampartialnitritationanammoxreactors.MicrobiologyOpen. 2017;6:e456. https://doi.org/10.1002/mbo3.456