h tt p : / / w w w . b j m i c r o b i o l . c o m . b r /
Environmental
Microbiology
Illumina-based
analysis
of
endophytic
bacterial
diversity
of
tree
peony
(Paeonia
Sect.
Moutan)
roots
and
leaves
Ruixian
Yang
a,∗,
Ping
Liu
a,
Wenyu
Ye
baDepartmentofEnvironmentalEngineeringandChemistry,LuoyangInstituteofScienceandTechnology,Luoyang,Henan,China bCollegeofLifeScience,FujianAgricultureandForestryUniversity,Fuzhou,Fujian,China
a
r
t
i
c
l
e
i
n
f
o
Articlehistory:
Received25October2016 Accepted1February2017 Availableonline2June2017 AssociateEditor:Fernando Andreotte Keywords: PaeoniaSect Moutan Endophyticbacteria Biodiversity IlluminaMiseq
a
b
s
t
r
a
c
t
Diversecommunitiesofbacteriainhabitplanttissuesandthosebacteriaplayacrucialrole forplanthealthandgrowth.Treepeony(PaeoniaSect.Moutan)isknownforitsexcellent ornamentalandmedicinalvaluesasChinesetraditionalplant,butlittleisknownabout itsassociatedbacterialcommunityundernaturalconditions.Toexaminehowendophytic bacteriaintreepeonyvaryacrosstissuesandcultivars,PCR-basedIlluminawasappliedto revealthediversityofendophyticbacteriaintreepeony.Atotalof149,842sequencesand 21,463operationaltaxonomicunits(OTUs)wereobtained.TheOTUabundanceofrootswas higherthanleavesacrossotherthreecultivarsexceptfor‘Kinkaku’and‘Luoyanghong’.The communitywascomposedoffivedominantgroups(Proteobacteria,Firmicutes,Bacteroidetes, AcidobacteriaandActinobacteria)inallsamples.Endophyticbacteriacommunitystructures hadchangedinleavesandroots.SequencesofPseudomonasandEnterobacteriaceaewere prevalentinrootsamples,whereasSuccinivibrioandAcinetobacterwerethedominantgenus inleafsamples.Otherwise,thedistributionofeachdominantgenusamongthe5cultivars waseithervaried.Thesefindingssuggestedthatbothplantgenotypeandtissuescontribute totheshapingofthebacterialcommunitiesassociatedwithtreepeony.
©2017SociedadeBrasileiradeMicrobiologia.PublishedbyElsevierEditoraLtda.Thisis anopenaccessarticleundertheCCBY-NC-NDlicense(http://creativecommons.org/
licenses/by-nc-nd/4.0/).
Introduction
Endophytic bacteria are a classof microbes which resides withinthe interiortissues ofplantswithout causingharm
∗ Correspondingauthorat:LuoyangInstituteofScienceandTechnology,Luoyang,Henan471023,China.
E-mail:fairy19790805@163.com(R.Yang).
tohostplants.1Theyhavebeenisolatedfromabroadrange
of plants.2 Endophytic bacteria hasmany kinds of
biologi-caleffects,suchasplantgrowthpromoting,plantpathogens biocontrol,phytoremediationimprovingand manymore.3–5
http://dx.doi.org/10.1016/j.bjm.2017.02.009
1517-8382/©2017SociedadeBrasileiradeMicrobiologia.PublishedbyElsevierEditoraLtda.ThisisanopenaccessarticleundertheCC BY-NC-NDlicense(http://creativecommons.org/licenses/by-nc-nd/4.0/).
Endophytic bacteria have gained more and more research attentionfortheirpotentialbiotechnologicalapplicationsin recentyears.6,7Thus,itiscrucialtounderstandthe
commu-nitystructureand diversityofendophyticbacteria inplant tissues.Itcouldfurthercontributetounderstandingthe sig-nificantrolesofplantmicrobiotainsupportingtheirmultiple bioactivities.
Inrecentyears,manyresearchershavefocusedon endo-phytesisolatedfrommedicinalplantsbecausethesemicrobes have huge potential to synthesis of numerous novel sec-ondarymetabolicproductsincludingantibiotics,anticancer compounds,volatileorganic compounds,antifungal, antivi-ral,insecticidalandimmunosuppressantagents.8Therefore,
abetterunderstandingoftheendophytediversityand com-munitystructurefrommedicinalplantsmayhelptoelucidate theirfunctionandpotentialrole.9Chinahasover5000plants
listedinitspharmacopeiaatpresent.Theendophytic commu-nitiesshouldbestudiedintheimportantmedicinallyplants tissuesandthesemicroorganismmayrepresentarelatively untappedsourceofnovelnaturalproducts.
Tree peony(Paeonia Sect.Moutan), belongingto the sec-tion Moutan in the genus Paeonia, Paeoniaceae, have been referredto as‘theking offlowers’ ofChina. Treepeony is believedtohavebeen cultivatedasornamentaland medic-inal plants for over 1600 years.10 Moutan Cortex, the root
bark of tree peony, is widely used in traditional medicine totreatvariousdiseasesincludingatherosclerosis,infection, andinflammation.11Althoughinteresthasincreasedinthe bacterialbiodiversityassociatedwith medicinal plants,the diversityandtaxonomiccompositionofendophyticbacteria associatedwithtreepeonyremainunclear.
Most studies about endophytic bacterial diversity had madeuseofculture-dependentapproaches inrecentyears. However,bacteriaspeciesisolatedwithcultivation-dependent methods can be influenced by cultivation media, growth conditionsand planttissuemanipulation. Therefore, endo-phytic bacteria communities were analyzed based on 16S rRNAgene-basedtechniquessuchasdenaturinggradientgel electrophoresisand16SrRNAgenecloningandsequencing. Culture-independenttechniquescanprovideamorespecific, replicableand detailed description ofmicrobial diversity.12
Ourunderstandingofmicrobialdiversityandfunctionin com-plexenvironmentshasincreasedsignificantly,inparticular theintroductionofnextgenerationsequencing(NGS).OfNGS, bothPyrosequencingandIlluminaMiseqareusedfor char-acterizesoils,13oceans,14,15mammaliangut,16aswellasthe
humanmicrobiome.17However,Illuminawouldenableusto
performindepthsequencingofhundredsofsamplesinone runatafractionofthecosts,makingitanexcellenttoolfor microbialdiversitystudies.
Thedifferentparameterscaninfluenceendophytic com-munitycompositionofplant,suchasgenotype,growthstage, physiologicalstatus,tissue,environmentalconditions,aswell asagriculturalpractices. Amongthesefactors,plant tissue andgenotypemayplaykeyrolesintheselectionofdistinct bacterialcommunitiesthatassociatewithplants.18Theaim
ofthecurrentstudywastodeterminetheendophyticbacterial communitycompositionofdifferenttissuesandcultivarsof treepeony.Tothatend,10treepeonysampleswerecollected from Luoyang National Peony Garden, in Luoyang, China,
representingrootandleaftissueoffivedifferenttreepeony cultivargroup. IlluminaMisequsing 16S rRNAgeneas the biomarkerwasconductedtoexaminethebacterialdiversityof treepeonysamples,togetabroaderoverviewofthediversity andcommunitystructureofendophyticbacteriaintreepeony tissues.Tothebestofourknowledge,thisisthefirstreportof endophyticbacterialcommunitystructureintreepeonyusing cultivation-independentmethods.
Materials
and
methods
Plantsamplingandplantsurfacesterilization
Tree peony sampleswere collected from Luoyang National PeonyGarden,inLuoyang,China.Theplantsamplesincluded fivecultivarsoriginatedfromfivedifferenttreepeony culti-vargroup(Table1).Theplanttissueswerecollectedrandomly foreachtreepeonycultivarwithsterileglovesinAugust2014. Altogether,thereare10samples:plantswerecollectedfrom5 cultivarsoftreepeony,andtherearebothrootandleavesfor eachcultivar.Foreachsample,wecollectedapproximately15 plantswithsterilegloves.Theplantsampleswererandomly dividedintothreegroups.Thesamplesineachgroup(n=5) wereplacedinoneplasticbag.All30plantsampleswere trans-portedtothelaboratoryat4◦C.Eachtreepeonytissueswere washedwithtapwatertoremoveattachedclay.Plantsurface sterilizationincludedthefollowingsteps:tissuepieceswere immersedin75%ethanolfor30sandin20%sodium hypochlo-ritefor5min.Aftereachtreatment,sampleswererinsedfive timesinsterilewater.Finally,the sampleswerethoroughly driedinalaminarflowchamber.Efficiencyofthissterilization techniquewastestedbywipingofsterilizedtissuesofeach typeacrossthesurfaceofatrypticasesoyagar(TSA)plate, whichconsistentlyyieldednobacterialcolonies.
DNAextraction,PCRandIlluminaampliconsequencing
About1.5gofthesurface-sterilizedplantsamplewasfrozen withliquidnitrogenandgroundtoafinepowderina steril-izedandprecooledmortar.DNAwasextractedusingaPlant DNAKit(OmegaBio-Tek,Norcross,USA)andtheV3-V4 hyper-variable region of the bacterial 16S rRNA gene amplified using Bac 341f (5-CCTACACGACGCTCTTCCGATCTN-3) and Univ805r(5-GACTGGAGTTCCTTGGCACCCGAGAATTCCA-3) primers.The50lPCRreactionmixturecontained100ngof DNAextract,1×Taqreactionbuffer,20pmolofeachprimer, 200MeachdNTPand1.5UofTaqDNApolymerase(Sangong Biotech,China).DNAaliquotswerePCR-amplifiedwith5min denaturationat95◦C,30cyclesof1minat94◦C,50sat55◦C, and72◦Cfor1min,afterwhichafinalelongationstepat72◦C for5minwasperformed.Allsampleswereamplified. Repli-catePCRproductsofthesamesamplewereassembledwithin aPCRtube.Thentheywerevisualizedonagarosegels(2%in TBEbuffer)containingethidiumbromide,andpurifiedwitha DNAgelextractionkit(SangongBiotech,China).After purifica-tion,theconcentrationofthePCRproductswasmeasuredon Qubit®2.0FluorometerusingthePicoGreen®dsDNA quantita-tionassay(Invitrogen,Carlsbad,CA).DNAconcentrationwas
Table1–Treepeonysamplecodesandtheirorigin.
Samples Cultivars Cultivargroup Tissue Latitude Longitude Location
rPr1 ‘Xuelian’(P.rockii) ChineseNorthwest Root
34◦43N 112◦24E
Luoyang National Peony Garden rPs2 ‘Luoyanghong’(P.suffruticosa) ChineseCentralPlains Root
rPs3 ‘Kaoh’(P.suffruticosa) Japan Root
rPl4 ‘HighNoon’(P.suffruticosa×P.delavayivar.lutea) America Root
rPsl5 ‘Kinkaku’(P.suffruticosa×P.lutea) French Root
lPr6 ‘Xuelian’(P.rockii) ChineseNorthwest Leaf
lPs7 ‘Luoyanghong’(P.suffruticosa) ChineseCentralPlains Leaf
lPs8 ‘Kaoh’(P.suffruticosa) Japan Leaf
lPl9 ‘HighNoon’(P.suffruticosa×P.delavayivar.lutea) America Leaf
lPsl10 ‘Kinkaku’(P.suffruticosa×P.lutea) French Leaf
adjustedto1ng/ml.Theampliconlibrarieswerepreparedby pooling10ngofeachPCR.
Processingofsequencingdata
AmpliconsequencingwasperformedontheIlluminaMiSeq platforms at Sangon Biotech (Shanghai) Co., Ltd. Raw sequence data derived from the sequencing process was transferred into FASTA files for each sample, along with sequencingqualityfiles.Fileswereaccessedusingthe bioin-formaticssoftware the Quantitative InsightsInto Microbial Ecology (QIIME) where they were processed and analyzed followinggeneralprocedures recommended byCaporaso.19
Briefly, sequences were denoised, and trimmed to remove barcodes and primers. Sequences representing chloroplast ormitochondrial DNAwere eliminatedfromfurther analy-sis.Theobtainedsequences wereassignedintooperational taxonomicunits (OTUs) using 97% identityclustering, and the mostabundantsequence from each OUT wasselected astherepresentativesequenceforthatOUT.Sequencedata havebeendepositedintheNationalCenterforBiotechnology Information(NCBI)SequenceReadArchive(SRA)underthe accessionnumberSRP090142.
Bacterialdiversity,richnessandtaxonomicdistributionof taxa
BasedontheresultsofOTU,bacterialtaxonomywasassigned usingtheRibosomalDatabaseProjectClassifieratconfidence level of80%.20 OTUs were definedat the specieslevel (3%
sequencedivergence)usingtheCompleteLinkageClustering toolofRDP.21Alphadiversityistheanalysisofspeciesdiversity
inasinglesample,includingchao1,Shannonandthe Simp-sonindices.Rarefactioncurvesweregeneratedbasedonthese threemetricsbyQIIME.Venndiagramswereemployedto char-acterizethesharedbacterialcommunitiesamongallsamples. For ˇ diversity analysis, dissimilarity of bacterial commu-nitieswas determined using principal componentanalysis (PCA)onweightedUniFracdistancesamongallsamples.The PICRUStsoftwarepackagewasusedtopredictfunctionaland metabolicprofiles ofthebacterialcommunitybasedonthe 16SrRNAgenesequencesfromeachdataset.22Thetaxonomic
profileswerenormalizedbythe16SrRNAgenecopynumber andthe functionalreference profileswere computedbased on400bpreads.PICRUStmetagenomepredictionswere cal-culated.
Statisticalanalysis
Meanandstandarddeviationforeachsetofdatawere cal-culated.ANOVAwasusedtotesttheeffectof‘cultivars’and ‘organ’ontherelativeabundanceofthemembersofthecore community.One-wayanalysisofvariance(ANOVA)was per-formedbySPSS(version19.0).
Results
Analysisofsequencingdata
After processing, 149,842high quality sequences remained withan averagelengthof450basesacrossall 10 samples. Thenumberofretainedsequences persamplevariedfrom 6405to25,166.Atotalof21,463OTUswere generatedafter clustering at a 97% similarity level. The number of OTUs ofper samplewas rangingfrom 1152to3452.The rarefac-tioncurvestendedtoapproachthesaturationplateauinall the10samples(Fig.1).Rarefactioncurvesdemonstratedthat OTU abundancewere diversebetweendifferent treepeony cultivars. Interesting,cultivars‘Xuelian’and ‘Luoyanghong’ originatedinChinarevealedthehighernumberofOTUs(6602 and5303),whilecultivars‘Kaoh’,‘HighNoon’and‘Kinkaku’
Observed_species: SampleID 2000 1P19 1Pr6 1Ps7 1Ps8 1Ps110 rP14 rPr1 rPs2 rPs3 rPs15 1500 1000 500 0 0 1000 2000 3000
Sequences per sample
Raref action measure: obser v e d_species 4000 5000 6000 7000
Fig.1–Rarefactioncurvesdepictingtheeffectof3% dissimilarityonthenumberofOTUsidentified.Notethe comparativelyhighspeciesrichnessofthetreepeony samples.
Table2–NumberofOUTsandAlphadiversityofendophyticbacteriaintreepeony.
Samples Numberofsequences ObservedOTUs Alphadiversity
Chao1 Shannon Simpson
rPr1 23,571 3452 3641.34 8.79 0.98 rPs2 10,993 1917 2549.91 7.74 0.97 rPs3 10,755 1721 2462.44 7.99 0.97 rPl4 9008 2118 2961.33 8.65 0.98 rPsl5 9902 1152 1373.30 6.85 0.95 lPr6 22,517 3150 3316.65 8.12 0.96 lPs7 25,166 3386 3378.13 8.40 0.96 lPs8 11,284 1540 1722.25 9.04 0.99 lPl9 20,241 1608 1620.18 7.50 0.97 lPsl10 6405 1419 2215.33 7.77 0.96 2658 IPr6 1162 114 27 126 19 12 32 18 18 279 84 145 502 1395 IPs7 189 138 116 27 121 114 78 145 38 51 525 33 439 IPs8 IPI9 39 22 125 273 IPsl10 rPs2 472 320230 364 1523 rPr1 8 20 16 9 9 87 276 74 50 37 54 173 19 150 24 4 19 46 341 115 54 rPsI5 rPI4 579 60 63 120 182 rPs3 Leaves Roots 3931 1567
Fig.2– VenndiagramsshowingthenumberofOTUssharedanduniqueamongdifferentsamples.
introduced from Japan, American and French showed the lowerrichness,with3261,3726and2571OTUs,respectively
(Table2).TheOTUrichnessassessedfromrootandleaf
tis-suesoffivecultivarsshowed slightlydifference.Exceptthe cultivar‘Kinkaku’and‘Luoyanghong’,theOTUabundanceof roottissueswashigherthanleaftissuesacrossother three cultivars(Table2).AccordingtotheresultsofVennanalysis, consistentoverlappatternsofOTUclustersamongdifferent sampleswereobtained(Fig.2).Forendophyticbacteria,root samplesandleavessamplesharbored1567and2658unique OTUs,respectively,whilesharing3931OTUs.Moreover,inroot samples,rPr1,rPs2,rPs3,rPl4 andrPsl5harbored1523,472, 182,579and54uniqueOTUs,respectively,whiletheyshared 276OTUs.lPr6,lPs7,lPs8,lPl9andlPsl10harbored1162,1395, 439,525and273uniqueOTUs,respectively,whiletheyshared 279OTUs.Theresultshowedthatthenumberofsharedand
unique OTUs was differentin treepeonysamples and the distributionofOTUswasinfluencedbyplanttissuesand cul-tivars.
The Alpha diversity estimation demonstrated that the
diversity ofendophytic bacteria is abundantin treepeony
(Table2).HigherShannonandChao1indicesindicatedthat
thebacterialcommunitydiversityoftreepeonytissues.The resultssuggestedthattheselibrariesdetectedalarge major-ityoftheendophyticbacterialdiversityinthesamplesused inourstudy.
Taxonomiccompositionanalysis
Allsequenceswereclassifiedfromphylumtogenusaccording totheprogramQIIMEusingthedefaultsetting.Thesequences wereclassifiedinto32differentphyla,75classes,156orders,
WS3 Verrucomicrobia Unassigned;Other TM7 Tenericutes Proteobacteria Planctomycetes OD1 Nitrospirae Gemmatimonadetes Fusobacteria Firmicutes Euryarchaeota Cyanobacteria Chloroflexi Chlorobi Chlamydiae Bacteroidetes Armatimonadetes Actinobacteria Acidobacteria 1.0 0.8 0.6 0.4 0.2 0.0 Relativ e ab undance Roots Leaves
rPr1 rPs2 rPs3 rPI4 rPsI5 IPr6 IPs7 IPs8 IPI9 IPsI10
Fig.3–Relativeabundanceofmostdominantbacterialphylaassociatedwithtreepeonyrootsandleaves(taxarepresented withaveragerelativesequenceabundancesoccurredat>0.01%).
292families,and542genera.Theoverallbacterialcomposition ofthedifferentsampleswassimilar,whilethedistributionof eachphylumvariedinallsamples.Proteobacteria,Firmicutes, Bacteroidetes,AcidobacteriaandActinobacteriawerethefivemost dominantphyla, accountingfor 86.00%ofthe reads, while theremaining14%involvedtwenty-sevenverylow-abundant phyla (Fig. 3). Proteobacteria were the most abundant divi-sion,comprisingapproximately42.40–64.70%readsacrossall samples,whereasthemembersfromFirmicutes,Bacteroidetes, AcidobacteriaandActinobacteriamadeup17.60%,6.50%,5.30% and 4.60% of the whole libraries respectively. In the root endophyticcommunities,sequencesassignedtoProteobacteria
(54.20%),Acidobacteria(5.60%)andActinobacteria(5.30%)were moreabundantcomparedtoleafsamples.Intheleaf endo-phyticcommunities,sequencesassignedtoFirmicutes(14.30%)
and Bacteroidetes(5.90%) were moreabundantcompared to
rootsamples(Fig.3).TheaveragereadsofNA(Norank)group accountsfor0.70%,butfluctuatesindifferentsamples.
Wenotice that predominance at the phylum and class levelisdrivenbythe high abundanceofseveralOTUs. For example, Gammaproteobacteria were dominant in the root andleafcommunityduetothelargenumberofsequences ofAeromonadales,EnterobacterialesandPseudomonadales. Sim-ilarly, Firmicutes were mostly represented by the OTUs,
ClostridialesandBacillales.ThephylumBacteroidetesismainly representedbysequencesbelongingto2OTUs:Bacteroidales
and Saprospirales. The Actinobacteria were mostly
repre-sentedbytheOTUsActinomycetales.Therefore,thedominant orderswereAeromonadales,Enterobacteriales,Pseudomonadales, Clostridiales,Bacteroidales,Bacillales,Lactobacillales,Rhizobiales, Sphingomonadalesand Burkholderialesinall the samples.For eachdominantorder,itsdistributionamongthetensamples
waseithervariedorconsistent.Foreachsample,percentages oftheseorderswerehighlydiversified(Fig.4).
Coregeneradistribution
Amongthe542definedOTUsacrossall thesamplesatthe genuslevel,acorerootandleafendophyticmicrobiomewas observed.Ourdatarevealsthatall67(12.36%)detected gen-era were sharedbyall thesamples, othergenerawere not detected in all samples. For example, Streptomyces can be detectedinroot,whichwasnotdetectedinleaf,while Methy-lobacteriumcanbedetectedinleaf,whichwasnotdetectedin root.Burkholderiaonlywasdetectedincultivars‘Kaoh’rootand leaf(SupplementaryTableS1).Thetop20coregenera(with minimumrelativeabundance0.2%)contributedthemostto communityabundanceamongalltensampleswerelistedin
Table3.ItwasworthnotingSuccinivibriowasthemost
pre-dominantgenusintheleafandendophyticcommunities,with relativeabundancerangingfrom6.30%to20.80%.
Coreendophyticmicrobiomefortissueofhostplantswas alsoexplored.Therelativeabundanceofcoregenerastrongly differedbetweentherootandleaftissueoftreepeony.The relative abundanceof Acinetobacter, unknown genusfrom
Clostridiales, other genus from Ruminococcaceae, unknown
genus from Lachnospiraceae, Lachnospira and Prevotella were significantly differences in roots compared with leaves (p=0.001;p=0.017;p=0.001;p=0.005;p=0.041;p=0.011).We alsoanalyzedthecoreendophyticmicrobiomeforeach sam-pleatthegenuslevel.Theresultsshowedthatthedistribution ofeachdominantgenusamongthetensampleswaseither variedorconsistent, andpercentagesofthesegenerawere highlydiversifiedforeachsample.Succinivibrio(11.00%)was
1.0 Xanthomonadales TM7-1 Syntrophobacterales Sva0725 Sphingomonadales Solirubrobacterales Solibacterales SJA-28 SC-I-84 Saprospirales Rhodospirillales Rhodobacterales Rhizobiales RF39 RB41 Pseudomonadales Pirellulates Pedosphaerales Other OPB56 Nitrospirales Myxococcales MND1 Lactobacillales iii1-15 Gemmatimonadetes Gemmatales Gemm-1 Gaiellales Flavobacteriales Fimbriimonadales Erysipelotrichales Entrerobacteriales Cytophagales CW040 Clostridiales Chthoniobacterales Caulobacterales Campylobacterales Burkholderiales Betaproteobacteria Bacteroidales Bacillales Aeromonadales Actinomycetales Acidobacteriales Acidobacteria-5 Acidimicrobiales 0.8 0.6 0.4 0.2 0.0 rPr1 rPs2 rPs3 Roots Relativ e ab undance Leaves
rPI4 rPsI5 IPr6 IPs7 IPs8 IPI9 IPsI10
Fig.4–Relativeabundanceofmostdominantbacterialordersassociatedwithtreepeonyrootsandleaves(taxarepresented withaveragerelativesequenceabundancesoccurredat>0.2%).
Table3–Therelativeabundanceoftop20coregeneraineachsample.
Phylum Genus Rootrelativeabundance(%) Leavesrelativeabundance(%) ResultofANOVAtest rPr1 rPs2 rPs3 rPl4 rPsl5 lPr6 lPs7 lPs8 lPl9 lPsl10 Organ Cultivar F p F p Proteobacteria Succinivibrio 11.00 10.50 16.70 11.40 13.50 18.60 20.80 6.30 15.9 20.0 1.655 0.234 0.246 0.905 Serratia 3.50 2.70 7.10 2.8 7.00 8.70 1.90 2.30 8.90 2.50 0.016 0.902 0.892 0.503 Acinetobacter 4.10 3.10 3.40 3.30 4.90 5.50 6.30 5.20 5.70 6.40 26.034 0.001 0.161 0.953 Enterobacteriaceae;g 1.40 3.40 4.20 4.30 15.90 2.30 1.00 1.80 1.30 0.80 2.902 0.127 2.028 0.166 Pseudomonadaceae;g 1.00 1.70 1.00 1.30 1.70 1.60 1.90 1.10 1.10 1.60 0.293 0.603 0.750 0.580 Enterobacteriaceae;Other 0.10 12.80 0.20 1.70 0.70 0.50 0.40 1.10 0.40 0.40 1.079 0.329 3.529 0.048 Comamonadaceae;g 2.70 4.10 2.90 1.50 0.40 0.40 0.30 2.10 0.30 0.40 5.014 0.055 1.987 0.173 Kaistobacter 1.60 1.00 1.00 1.10 0.70 1.10 1.10 1.40 0.40 0.70 0.377 0.556 1.039 0.434 Syntrophobacteraceae;g 1.30 1.40 0.80 1.10 0.70 0.90 1.10 1.20 0.40 0.80 0.853 0.383 0.692 0.614 Pseudomonas 1.40 1.50 1.10 8.30 9.90 0.20 0.30 0.50 0.10 0.10 4.776 0.060 2.046 0.163 Sphingomonas 0.20 0.20 0.50 0.50 0.10 0.20 0.20 1.00 3.00 0.80 2.009 0.194 3.635 0.045 Firmicutes Clostridiales;f ;g 1.80 1.70 2.50 2.00 3.30 3.10 3.30 4.10 3.50 2.80 9.016 0.017 0.224 0.919 Ruminococcaceae;g 1.90 1.80 1.50 1.90 0.90 2.80 3.40 2.70 2.40 2.70 22.857 0.001 0.231 0.915 Bacillaceae;g 3.30 1.90 0.20 0.10 1.00 0.60 0.90 0.70 1.30 6.30 0.282 0.610 2.687 0.093 Lachnospiraceae;g 0.80 0.70 1.00 0.90 0.40 1.30 1.20 1.40 1.00 1.30 15.158 0.005 0.227 0.917 Lachnospira 0.50 0.30 0.80 0.40 0.80 0.80 0.90 0.80 0.70 1.00 5.939 0.041 0.572 0.689 Bacteroidetes S24-7;g 0.80 0.90 1.20 0.90 0.80 1.70 1.70 1.30 1.40 1.70 31.508 0.001 0.025 0.999 Prevotella 0.70 0.60 1.20 0.80 1.10 1.40 1.40 1.30 1.10 1.30 10.756 0.011 0.193 0.937 Chitinophagaceae;g 0.80 0.40 1.70 1.10 0.70 0.60 0.70 1.40 0.70 0.30 0.493 0.503 3.120 0.066 Actinobacteria Micrococcaceae;g 3.50 0.60 0.40 1.30 0.10 0.10 0.20 0.40 0.10 0.20 2.538 0.150 2.426 0.117 ‘g ’ represents genusnot grouped intoany known generawithin these families/groups; ‘Other’ represents othergenus withinthese families/groups.
0.10 0.05 0.00 -0.05 -0.10 0.10 0.05 0.00 -0.05 -0.10 PC1(30.5%) PC2 (27 .1 % ) rPr1 rPs2 rPs3 rPl4 rPsl5 lPr6 lPs7 lPs8 lPl9 lPsl10
Fig.5–Principalcomponentanalysisillustratesdifferences betweenbacterialcommunitiesintreepeonyrootsand leavesat5cultivars.Twofirstcomponents(PC1andPC2) wereplottedandrepresented57.6%ofwholeinertia.
the mostabundant genus followed byAcinetobacter (4.10%) andSerratia(3.5%)in‘Xuelian’roots.In‘Luoyanghong’roots,
Enterobacteriaceae (12.80%), Succinivibrio (10.50%) and while in the culitivar ‘Kinkaku’ roots, Enterobacteriaceae (15.90%),
Succinivibrio(13.50%),Pseudomonas(9.90%)andSerratia(7.00%) werethedominantgenusorfamilies.Therelativeabundance ofthe same genus was significant difference between the samples.Forexample,Pseudomonaswasmuchmoreenriched inthe‘Kinkaku’and‘Highnoon’roots,andtherelative abun-dancereachedto9.90%and8.30%ofthereads,respectively, whiletheaverageabundanceintheothereightsampleswas ≤1.5%.TherelativeabundanceofSphingomonaswas3.00%in ‘Highnoon’leaf,whiletheabundanceoftheothersamples variedfrom0.10%to1.00%.Therelativeabundanceof Enter-obacteriaceaeandSphingomonasweresignificantlydifferences indifferenttreepeonycultivars(p=0.048;p=0.045).
Organandcultivarstypedifferentiatecommunities
Core genera distribution results indicated the presence of diverseendophyticbacteriainleafandrootoftreepeony,and therelativeabundanceofthesamecoregenuswasdifference indifferentcultivarsoftreepeony.Tofurtherelucidatethe microbialpopulationcorrelateswiththetissuesandcultivars oftreepeony,Principalcomponentanalysis(PCA)wasusedto illustratetheextentofvariationofendophyticbacteria pop-ulationinthedifferentsamples.Dataarepresentedasa2D plottobetterillustratetherelationship.PCA(Fig.5)identified twoprincipalcomponentfactors(PCF)inrelationto percent-ageabundanceofgroups,explaining30.5%and27.1%ofthe variation,respectively.ThePCAanalysisrevealedthattissue wasastronginterpretivefactorforthevariationincommunity compositionofendophytesoftreepeony.Rootsampleshada significantlyhigherPC1value(exceptforrPsl5),andleaf sam-pleshadahigherPC2value(exceptforlPs8).Thefirstprincipal coordinateseparatedthesamplesbasedonthecultivars,the samplerPs2,rPs3andrPl4wererelativelysimilar,whichthe samplelPr6,lPs7andlPsl10wererelativelysimilar.Thesample rPsl5andlPs8clearlyseparatedfromtheothersamplesalong
PC1,whichshowedanobviousdifferenceaboutbacteria com-munitycompositionamongcultivaroftreepeony.Itindicated thatthecultivarwasalsoafactortoshapethecommunity composition.
Predictivemetagenomeanalysis
PICRUSt approach was applied to infer the metagenomic contentofthesamples,andtoevaluatethefunctional poten-tialofthe bacterialcommunity’smetagenomefrom its 16S profile.Basedon thepredicted metagenomes, 47 oflevel3 KEGGOrthology(KO)groupswerefoundandthegene fami-liesbelongingtoaminoacidmetabolism,energymetabolism, Two-ComponentSystemandbacterialmotilityproteinwere identifiedasthemajorgenefamilies(Fig.6).Similarlytothe phylogenetic beta-diversity,thetissuesoftreepeony repre-sentedastrongstructuringfactorforconstructedfunctional. Incontrast,byPICRUStinference,thefunctionalabundanceof ‘aminoacidmetabolism’,‘gluconeogenesis’,‘photosynthesis’ were significantly greater in leaves tissues than roots tis-suesoftreepeony,whiletheabundanceof‘Two-Component System’,‘Bacterialmotilityproteins’,‘flagellarassembly’and ‘secondarymetabolitebiosynthesis’weregreaterinroots tis-suesoftreepeony.
Discussion
Thisstudywasundertakentoexploretheendophytic bacte-riadiversityinrootand leafoftreepeonyinLuoyangcity. UsingIlluminasequencingoftheV3-V4hypervariableregion ofbacterial16SrRNAgenesandmetagenomiclibraryanalysis, weobservedconspicuousdifferencesinendophyticbacterial communitystructureofdifferenttissuesandcultivarsoftree peony.Toourknowledge,thisisthefirstimplementationof PCR-based Illumina Miseq technologyfor investigating the diversityofendophyticbacteriaintreepeonytissues.
Thecapacity oftheIlluminaplatform togenerate enor-mousdatasetsisundoubtedly anadvantage.23,24 Themost
heavily sequenced members of the bacterial communities associatedwithtreepeonyrootsandleaveswererelatedto describedspecies,83%ofoursequencescouldbeassignedat thegenuslevel.Ourabilitytoassignsequencesatthegenus levelishigherthanresultsobtainedinapyrosequencingstudy ofArabidopsisthalianarhizosphereandphyllosphere(60%).25
Therarefactioncurvealsotendedtoplateau,suggestingthe librarywaslargeenoughtoreflecttheendophyticbacterial diversityoftreepeony.
Proteobacteria, Firmicutes, Bacteroidetes, Acidobacteria andActinobacteriawerethefivemostabundantphyla associ-atedwithtreepeony,allphylathataretypicaloftherootand leafendophyticbacteria,26–28 suggestingsubstantial overlap
inthe key communitymembers across hostspecies. How-ever,therearemanybacterialgroupscommononotherhosts thatwedidnotobserveontreepeony.Forexample,theroots of Spartinaalterniflora and Kandeliaobovata were heavily populated byCyanobacteria,29 thisphylum wasnotexisted
in all samples. Gammaproteobacteria was the most abun-dantclassofProteobacteriaintheleafendophyticcommunity, which wasconsistentwithother studies.30,31 Moreover,we
95% confidence intervals Leaf Root p-v a lue (corrected) 8.01e-3 0.025 0.040 0.021 0.013 3.95e-3 7.82e-3 9.53e-3 0.032 0.012 7.07e-3 0.049 8.78e-3 1-12e-3 0.029 2.28e-3 0.031 3.53e-3 0.036 1.97e-3 3.27e-3 4.50e-3 0.012 1.38e-3 0.012 0.015 0.025 3.29e-3 0.021 5.21e-4 0.026 0.034 5.10e-4 2.90e-3 6.63e-3 0.039 0.015 2.30e-3 3.95e-3 0.038 0.044 0.023 2.31e-3 0.022 0.033 0.022 8.99e-3 Ribosome Two-component system Bacterial motility proteins DNA repair and recombination proteins Pyrimidine metabolism DNA replication proteins Chromosome Aminoacyl-tRNA biosynthesis Sporulation Amino acid related enzymes Purine metabolism Flagellar assembly Homologous recombination Nitrogen metabolism Beta-Alanine metabolism Methane metabolism Geraniol degradation Lysine biosynthesis Cell cycle - Caulobacter DNA replication Thiamine metabolism Protein export Mismatch repair Cytoskeleton proteins Glycolysis / Gluconeogenesis Valine, leucine and isoleucine biosynthesis Starch and sucrose metabolism RNA degradation Inorganic ion transport and metabolism One carbon pool by folate Translation factors Alanine, aspartate and glutamate metabolism Photosynthesis Pantothenate and CoA biosynthesis Photosynthesis proteins Biosynthesis of unsaturated fatty acids Ascorbate and aldarate metabolism C5-Branched dibasic acid metabolism RNA polymerase Metabolism of cofactors and vitamins Carbon fixation in photosynthetic organisms Biosynthesis of siderophore group nonribosomal pep... Alpha-Linolenic acid metabolism Tropane, piperidine and pyridine alkaloid biosynth... Ether lipid metabolism Isoquinoline alkaloid biosynthesis Indole alkaloid biosynthesis
Mean proportion (%) Difference in mean proportions (%)
0.0 2.5 –0.3 –0.2 –0.1 0.0 0.1 0.2 0.3
didnot observe many sequences for Erwinia, Pantoeaand Xanthomonas, which were abundant in the four kinds of commercial saladleafvegetables. Onthe other hand, Suc-cinivibrio wasthe mostabundantgenus inthe tree peony rootsand leaves,which was notpresent in other kinds of plantinvestigatedintheliterature.Succinivibrioisan anaer-obic,Gram-negative,curved,spiralbacilluswithapolar,were thoughttooccurasorganismsintherumensofherbivorous animals.32 Moreresearch isrequired toelucidatewhy this
genusisabundantintreepeonytissues.
Comparisonofthecommunitiesassociatedwiththeleaves androotsrevealsbothubiquitousandorganspecificgroups.
Serratiawasmostabundantgenusinbothrootandleaf asso-ciatedcommunities,whichhasbeenisolatedfromavariety ofplantsamplesandiswidelyknownfortheirplantgrowth promoting(PGP)andbiocontrolqualitiesaswellasfortheir endophyticlifestyle.33Acinetobacter,acommonsoilbacterium,
couldcolonizeplantsbytheirrootsandmultiplyextensively in the pathogen-infected tissues.34 Enterobacteriaceae,
Pseu-domonas,Comamonadaceaesequencesare commoninallten
sampletypes,buttheyaremostabundantintheroot endo-phyticcommunity.Ithasbeenpreviouslyobservedthat, in manycases,PseudomonasandmembersofEnterobacteriaceae
are abundant in both the soil environment and the plant interior.35–37 Thissuggestssomeofthetaxafoundinrootof
treepeonymaycomefromsoil.Apreviousstudyreportedthat
Pseudomonaswasabundantonrhizosphereoftreepeonybased onculture-dependentmethod,suggestingsoilmicroorganism orrhizosphericspeciesare themainsourcesoftree peony endophytes.38Otherwise,leavesandrootsoftreepeonyare
colonizedbysomeofthesamegenera,albeitindifferent pro-portions.Thissuggeststhatsomeendophyticbacteriaspecies canmigratefromrootstoleaves.
Thecorebacterialcommunitymaybeimportantin main-tainingessentialfunctions. When comparingthe predicted functionsofthecorecommunitytotheentirecommunity,we foundthatthefunctionsarehighlycorrelated.Acinetobacter, EnterobacteriaceaeandPseudomonasweremostabundantinthe rootendophyticcommunity.Asthemainsourcesoftreepeony endophytes,PICRUStanalysisalsoshowed‘Two-Component System’,‘Bacterialmotilityproteins’and‘flagellarassembly’ weremorerelativeabundanceinroottissues.Bacterial Two-component signal transduction systems are widespread in prokaryotesand playextensiverolesinadaptationto envi-ronmentalchanges,whichisveryimportantforendophytic plantcolonization.39Cellmotilitymightbeimportantinplant
invasionandinthespreadingofendophyticmicroorganisms throughoutplantorgansandtissues.40Otherwise,theresults
showed bacterial auxin phytohormone indole-3-acetic acid (IAA)synthesisandsiderophoreproductionplaysakeyrolein plantgrowthpromotingtraitsanditisnecessaryforefficient rhizospherecolonization.41Inourexperiment,theproduction
ofindole-3-aceticacid,isoquinolinealkaloid,andsiderophore productionalso showed the beneficial interaction between endophytesandtreepeony.Thissuggeststhatthecore com-munityisresponsibleforfunctionsthatareimportantforthe endophytessymbiosistopersist.
Plant-associated habitats are dynamic environments in whichmanyfactorsaffectthespeciescompositionsof micro-bialcommunities.Avarietyofreportsindicatethatstructure
ofendophyticcommunityisinfluencedbyabioticandbiotic factorssuchasenvironmentconditionsandhostgenotype. The host plant genotype is one of the major influencing factors.For instance,Adams andKloepperinvestigatedthe impact of cottonplant genotype on the inherent bacterial population oftheir seeds,seedling stems and roottissues. Theyfoundthatthedistinctgenetic,morphologicaland phys-iologicalcharacteristics ofindividual cottoncultivarsledto differencesintheendophyticbacterialcommunitystructure amongthecultivars.42Thediversityofbacterialcommunities
associated with rice roots across 10 cultivarswas investi-gated.Theresultsindicatedthatdifferentricecultivarsselect specificfractionsofthebacterialcommunitiesinsometimes highlydifferentandatothertimessimilarways.Thatis,rice genotype determined,toalargeextent,thecompositionof thedifferentbacterialcommunitiesacrosscultivars.43Inour
study,wepresentasurveyof5treepeonycultivars,inwhich endophyticbacterialcommunitiesassociatedwithrootsand leaves were assessed. Our results indicated that different treepeonycultivarshadsomedifferencesinthedensitiesof endophyticmicrobialcommunities.Forexample,roots sam-plesofChinacultivarshadthehighernumberofOTUsthan introducedcultivars.Thus,itispossiblethatChinacultivars haveabetteradaptabilitytothelocalenvironmentconditions andhaveformmoremutualisticassociationsbetweenplant andbacteria.Furthermore,wefoundnoticeabledifferencesin endophyticbacterialcommunitiesacross5treepeony culti-vars.Forexample,Pseudomonaswasmuchmoreenrichedin the‘Kinkaku’and‘Highnoon’rootthanothersamples.These results suggest that ‘Kinkaku’ and ‘High noon’ selectively recruitedandpromotedthegrowthofcertainPseudomonasin its rootinterior.Thebasisforthe differencesbetween cul-tivarsisnotclear.Itisalsolikelythatcultivar-specificroot exudates may be shaping the endophytecommunity. Root exudates wouldalter themicrobial community inintimate contactwithplantroots.Itisworthmentioningthattheroot barkoftreepeonycanproducesomechemicalmaterialsuch as paeonoland paeoniflorin.44 Therootexudates chemical
compositionandproportioncaninfluencethecompositionof endophyticcommunityacross5treepeonycultivars. Other-wise,5treepeonycultivarswhichhaddifferentgeneticand morphologicalphysiological characteristic,whichmay thus affectbacterialspectra.Basedonourresults,wehypothesize thatrootexudatesandgeneticcharacteristicsmayresultin distinctchangesinendophyticbacterialpopulationsoftree peony.
Conclusions
Insummary,thediversityofendophyticbacteriawashighin tree peony.Thecommunityofendophyticbacteria showed obvious changes in different tissues and cultivars. The host genotypes determine the endophytic bacterial com-munity. Further studies are needed to understand the mechanismsoftheinteractions betweenrootexudatesand endophytic community. Otherwise, the endophytic bacte-rial functional diversity will be further studied in tree peony to lay the foundation for application of tree peony endophytes.
Conflicts
of
interest
Theauthors declarethattheyhave noconflictsofinterest concerningthisarticle.
Acknowledgments
This work was supported by the National Natural Science FoundationofChina(GrantNo.31500008),andtheProjectof ScienceandTechnologyDepartment(GrantNo.152102210328) andEducationDepartment(GrantNo.14A180024)ofHenan Province.Theauthorsgreatlyappreciatethesupportprovided byNationalPeonyGardenofLuoyangcityinChina.
Appendix
A.
Supplementary
data
Supplementarydataassociatedwiththisarticlecanbefound, intheonlineversion,atdoi:10.1016/j.bjm.2017.02.009.
r
e
f
e
r
e
n
c
e
s
1. SchulzB,BoyleC,eds.WhatareEndophytes:MicrobialRoot
Endophytes.Springer-Verlag:Berlin;2006.
2. LodewyckC,VangronsveldJ,PorteousF,MooreERB,Taghavi
S,MezgeayM.Endophyticbacteriaandtheirpotential
application.CritRevPlantSci.2002;21:583–606.
3. StrobelG,DaisyB,CastilloU,HarperJ.Naturalproductsfrom
endophyticmicro-organisms.JNatProd.2004;67:257–268.
4. RyanRP,GermaineK,FranksA,RyanDJ,DowlingDN.
Bacterialendophytes:recentdevelopmentsand
applications.FEMSMicrobiolLett.2008;278:1–9.
5. StaniekA,WoerdenbagHJ,KayserO.Endophytes:exploiting
biodiversityfortheimprovementofnaturalproduct-based
drugdiscovery.JPlantInteract.2008;3:75–93.
6. CompantS,DuffyB,NowakJ,ClementC,BarkaEA.Useof
plantgrowth-promotingbacteriaforbiocontrolofplant
diseases:principles,mechanismsofaction,andfuture
prospects.ApplEnvironMicrobiol.2005;71:4951–4959.
7. QinS,ZhangYJ,YuanB,etal.IsolationofACC
deaminase-producinghabitat-adaptedsymbioticbacteria
associatedwithhalophyteLimoniumsinense(Girard)Kuntze
andevaluatingtheirplantgrowth-promotingactivityunder
saltstress.PlantSoil.2014;374:752–766.
8. GolinskaP,WypijM,AgarkarG,RathodD,DahmH,RaiM.
Endophyticactinobacteriaofmedicinalplants:diversityand
bioactivity.AntonieVanLeeuwenhoek.2015;108:267–289.
9. El-DeebB,FayezK,GherbawyYA.Isolationand
characterizationofendophyticbacteriafromPlectranthus
tenuiflorusmedicinalplantinSaudiArabiadesertandtheir
antimicrobialactivities.JPlantInteract.2012;8:56–64.
10.ChengFY.Advancesinthebreedingoftreepeoniesanda
cultivarsystemforthecultivargroup.IntJPlantBreeding.
2007;1:89–104.
11.ChungHS,KangM,ChoC,etal.Inhibitionofnitricoxideand
tumornecrosisfactor-alphabymoutancortexinactivated
mouseperitonealmacrophages.BiolPharmBull.
2007;30:912–916.
12.BulgariD,CasatiP,QuaglinoF,BiancoPA.Endophytic
bacterialcommunityofgrapevineleavesinfluencedby
samplingdateandphytoplasmainfectionprocess.BMC
Microbiol.2014;14:1–11.
13.FiererN,JacksonRB.Thediversityandbiogeographyofsoil
bacterialcommunities.ProcNatlAcadSciUSA.
2006;103:626–631.
14.HuberJA,WelchDBM,MorrisonHG,etal.Microbial
populationstructuresinthedeepmarinebiosphere.Science.
2007;318:97–100.
15.CaporasoJG,LauberCL,WaltersWA,etal.Globalpatternsof
16SrRNAdiversityatadepthofmillionsofsequencesper
sample.ProcNatlAcadSciUSA.2011;108:4516–4522.
16.LeyRE,HamadyM,LozuponeC,etal.Evolutionofmammals
andtheirgutmicrobes.Science.2008;320:1647–1651.
17.CostelloEK,LauberCL,HamadyM,FiererN,GordonJI,
KnightR.Bacterialcommunityvariationinhumanbody
habitatsacrossspaceandtime.Science.2009;326:
1694–1697.
18.AndreoteFD,RochaUND,AraújoWL,AzevedoJL,van
OverbeekLS.Effectofbacterialinoculation,plantgenotype
anddevelopmentalstageonroot-associatedandendophytic
bacterialcommunitiesinpotato(Solanumtuberosum).Antonie
VanLeeuwenhoek.2010;97:389–399.
19.CaporasoJG,KuczynskiJ,StombaughJ,etal.QIIMEallows
analysisofhigh-throughputcommunitysequencingdata.
NatMethods.2010;7:335–336.
20.ColeJR,WangQ,CardenasE,etal.TheRibosomalDatabase
Project:improvedalignmentsandnewtoolsforrRNA
analysis.NucleicAcidsRes.2009;37:141–145.
21.SchlossPD,HandelsmanJ.IntroducingDOTUR,acomputer
programfordefiningoperationaltaxonomicunitsand
estimatingspeciesrichness.ApplEnvironMicrobiol.
2005;71:1501–1506.
22.LangilleMGI,ZaneveldJ,CaporasoJG,etal.Predictive
functionalprofilingofmicrobialcommunitiesusing
16SrRNAmarkergenesequences.NatBiotechnol.2013;31:
814–821.
23.LazarevicV,WhitesonK,HuseS,etal.Metagenomicstudyof
theoralmicrobiotabyIlluminahigh-throughputsequencing.
JMicrobiolMethods.2009;79:266–271.
24.GloorGB,HummelenR,MacklaimJM,etal.Microbiome
profilingbyIlluminasequencingofcombinatorial
sequence-taggedPCRproducts.PLoSONE.2010;5:e15406.
25.BodenhausenN,HortonMW,BergelsonJ.Bacterial
communitiesassociatedwiththeleavesandtherootsof
Arabidopsisthaliana.PLoSOne.2013;8:e56329.
26.ManterDK,DelgadoJA,HolmDG,StongRA.Pyrosequencing
revealsahighlydiverseandcultivarspecificbacterial
endophytecommunityinpotatoroots.MicroEcolog.
2010;60:157–166.
27.BulgarelliD,RottM,SchlaeppiK,etal.Revealingstructure
andassemblycuesforArabidopsisrootinhabitingbacterial
microbiota.Nature.2012;488:91–95.
28.JacksonCR,RandolphKC,OsbornSL,TylerHL.Culture
dependentandindependentanalysisofbacterial
communitiesassociatedwithcommercialsaladleaf
vegetables.BMCMicrobiol.2013;13:1–12.
29.HongYW,LiaoD,HuAY,etal.Diversityofendophyticand
rhizoplanebacterialcommunitiesassociatedwithexotic
SpartinaalternifloraandnativemangroveusingIllumina
ampliconsequencing.CanJMicrobiol.2015;61:723–733.
30.GottelNR,CastroHF,KerleyM,etal.Populusdeltoidsroots
harbordistinctmicrobialcommunitieswithinthe
endosphereandrhizosphereacrosscontrastingsoiltypes.
ApplEnvironMicrobiol.2011;77:5934–5944.
31.RomeroFM,MaríaM,PieckenstainFL.Thecommunitiesof
tomato(SolanumlycopersicumL.)leafendophyticbacteria,
analyzedby16S-ribosomalRNAgenepyrosequencing.FEMS
MicrobiolLett.2014;351:187–194.
32.O’HerrinSM,KenealyWR.Glucoseandcarbondioxide
metabolismbySuccinivibriodextrinosolvens.ApplEnviron
33.PetersenLM,TisaLS.Friendorfoe?Areviewofthe
mechanismsthatdriveSerratiatowardsdiverselifestyles.
CanJMicrobiol.2013;59:627–640.
34.KayE,BertollaF,VogelTM,SimonetP.Opportunistic
colonizationRalstoniasolanacearuminfectedplantsby
Acinetobacterspanditsnaturalcompetencedevelopment.
MicrobEcol.2002;43:291–297.
35.SpiersAJ,BucklingA,RaineythenPB.Thecausesof
Pseudomonasdiversity.Microbiology.2000;146:2345–2350.
36.HallmannJ,Quadt-HallmannA,MahaffeeWF,KloepperJW.
Bacterialendophytesinagriculturalcrops.CanJMicrobiol.
1997;43:895–914.
37.SicilianoSD,TheoretCM,deFreitasJR,HuclPJ,GermidaJJ.
Differencesinthemicrobialcommunitiesassociatedwith
therootsofdifferentcultivarsofcanolaandwheat.CanJ
Microbiol.1998;44:844–851.
38.HanJ,SongY,LiuZ,HuY.Culturablebacterialcommunity
analysisintherootdomainsoftwovarietiesoftreepeony
(Paeoniaostii).FEMSMicrobiolLett.2011;322:15–24.
39.SheibanitezerjiR,RatteiT,SessitschA,TrognitzF,MitterB.
TranscriptomeprofilingoftheendophyteBurkholderia
phytofirmansPsJNindicatessensingoftheplantenvironment
anddroughtstress.Mbio.2015;6:e00621–e715.
40.CompantS,ClémentC,SessitschA.Plantgrowth-promoting
bacteriaintherhizo-andendosphereofplants:theirrole,
colonization,mechanismsinvolvedandprospectsfor
utilization.SoilBiolBiochem.2010;42:669–678.
41.Zú ˜nigaA,PoupinMJ,DonosoR,etal.Quorumsensingand
indole-3-aceticaciddegradationplayaroleincolonization
andplantgrowthpromotionofArabidopsisthalianaby
BurkholderiaphytofirmansPsJN.MolPlantMicrobeInteract.
2013;26:546–553.
42.AdamsPD,KloepperJW.Effectofhostgenotypeon
indigenousbacterialendophytesofcotton(Gossypium
hirsutumL.).PlantSoil.2002;241:181–189.
43.HardoimPR,AndreoteFD,Reinhold-HurekB,SessitschA,
vanOverbeekLS,vanElsasJD.Riceroot-associatedbacteria:
insightsintocommunitystructuresacross10cultivars.FEMS
MicrobiolEcol.2011;77:154–164.
44.YanD,SaitoK,OhmiY,FujieN,OhtsukaK.Paeoniflorin,a
novelheatshockprotein-inducingcompound.CellStress
