ARTICLE

Diversity of endophytic and rhizoplane bacterialcommunities associated with exotic Spartina alterniflora andnative mangrove using Illumina amplicon sequencingYouwei Hong, Dan Liao, Anyi Hu, Han Wang, Jinsheng Chen, Sardar Khan, Jianqiang Su,and Hu Li

Abstract: Root-associated microbial communities are very important for biogeochemical cycles in wetland eco-systems and help to elaborate the mechanisms of plant invasions. In the estuary of Jiulong River (China), Spartinaalterniflora has widely invaded Kandelia obovata-dominated habitats, offering an opportunity to study the influenceof root-associated bacteria. The community structures of endophytic and rhizosphere bacteria associated withselected plant species were investigated using the barcoded Illumina paired-end sequencing technique. Thediversity indices of bacteria associated with the roots of S. alterniflora were higher than those of the transitionstands and K. obovata monoculture. Using principal coordinate analysis with UniFrac metrics, the comparison of�-diversity showed that all samples could be significantly clustered into 3 major groups, according to the bacteriacommunities of origin. Four phyla, namely Proteobacteria, Bacteroidetes, Chloroflexi, and Firmicutes, were enriched inthe rhizoplane of both salt marsh plants, while they shared higher abundances of Cyanobacteria and Proteobacteriaamong endophytic bacteria. Members of the phyla Spirochaetes and Chloroflexi were found among the endophyticbacteria of S. alterniflora and K. obovata, respectively. One of the interesting findings was that endophytes weremore sensitive in response to plant invasion than were rhizosphere bacteria. With linear discriminate analysis, wefound some predominant rhizoplane and endophytic bacteria, including Methylococcales, Pseudoalteromonadacea,Clostridium, Vibrio, and Desulfovibrio, which have the potential to affect the carbon, nitrogen, and sulfur cycles. Thus,the results provide clues to the isolation of functional bacteria and the effects of root-associated microbialgroups on S. alterniflora invasions.

Key words: Spartina alterniflora, mangrove, endophytic bacteria, diversity, Illumina amplicon sequencing.

Résumé : Les communautés microbiennes associées aux racines revêtent une importance capitale dans les cyclesbiogéochimiques d’écosystèmes de milieux humides et permettent de préciser les mécanismes d’invasions deplantes. Dans l’estuaire du fleuve Jiulong (Chine), Spartina alterniflora a envahi une grande partie des habitatsdominés par Kandelia obovata, ce qui donne le loisir d’étudier l’influence des bactéries associées aux racines. Aumoyen de la technique de séquençage d’extrémités appariées a code-barres d’Illumina, on a examiné les structurescommunautaires des bactéries endophytes et rhizosphériques associées avec des espèces de plantes particulières.Les indices de diversité des bactéries associées aux racines de S. alterniflora étaient supérieurs a ceux de lots detransition et de monoculture de K. obovata. Une comparaison de �-diversité par analyse en coordonnées principalesavec paramètres UniFrac a révélé que tous les échantillons pouvaient être clairement regroupés en 3 groupesprincipaux, selon les communautés bactériennes d’origine. Quatre phylums, soit Proteobacteria, Bacteroidetes,Chloroflexi et Firmicutes, étaient enrichis dans les rhizoplans des plantes de marais salants, tandis qu’on retrouvaitdavantage de Cyanobacteria et de Proteobacteria chez les bactéries endophytes. Les phylums des Spirochaetes et desChloroflexi se sont retrouvés chez les bactéries endophytes de S. alterniflora et K. obovata, respectivement. Une

Received 7 February 2015. Revision received 7 July 2015. Accepted 7 July 2015.

Y. Hong. Key Laboratory of Urban Environment and Health, Institute of Urban Environment, Chinese Academy of Sciences, 1799Jimei Road, Xiamen 361021, People’s Republic of China; Department of Environmental Sciences, University of California, Riverside,CA 92521, USA.D. Liao. Xiamen Huaxia College, Xiamen 361024, People’s Republic of China.A. Hu, J. Chen, and J. Su. Key Laboratory of Urban Environment and Health, Institute of Urban Environment, Chinese Academy ofSciences, 1799 Jimei Road, Xiamen 361021, People’s Republic of China.H. Wang. College of Ecology and Resources Engineering, Wuyi University, Wuyishan City 354300, People’s Republic of China.S. Khan. Department of Environmental Sciences, University of Peshawar, 25120, Pakistan.H. Li. Key Laboratory of Urban Environment and Health, Institute of Urban Environment, Chinese Academy of Sciences, 1799 JimeiRoad, Xiamen 361021, People’s Republic of China; University of Chinese Academy of Sciences, Beijing 100049, People’s Republic ofChina.Corresponding author: Youwei Hong (e-mail: ywhong@iue.ac.cn).

Pagination not final (cite DOI) / Pagination provisoire (citer le DOI)

1

Can. J. Microbiol. 61: 1–11 (2015) dx.doi.org/10.1139/cjm-2015-0079 Published at www.nrcresearchpress.com/cjm on 7 July 2015.

Can

. J. M

icro

biol

. Dow

nloa

ded

from

ww

w.n

rcre

sear

chpr

ess.

com

by

Inst

itute

of

Urb

an E

nvir

onm

ent C

hine

se A

cade

my

of S

cien

ces

on 0

7/29

/15

For

pers

onal

use

onl

y.

trouvaille intéressante fut que les endophytes étaient plus sensibles en réponse a l’invasion de plantes, par rapportaux bactéries de la rhizosphère. Une analyse discriminante linéaire a permis de faire ressortir des bactériesprédominantes du rhizoplan et de la microflore endophyte, dont les Methylococcales, les Pseudoalteromonadacea, etles genres Clostridium, Vibrio et Desulfovibrio, qui auraient potentiellement un impact sur les cycles du carbone, del’azote et du soufre. Ainsi, les résultats fournissent des indices relatifs a l’isolation de bactéries fonctionnelles eta l’influence de groupes de microbes associés aux racines sur les invasions de S. alterniflora. [Traduit par laRédaction]

Mots-clés : Spartina alterniflora, mangrove, bactéries endophytes, diversité, séquençage d’amplicons sur plateformeIllumina.

IntroductionCoastal wetland research has broadly focused on pres-

ervation and restoration aspects, owing to a range ofecosystem services such as carbon (C) sequestration andclimate regulation (Mcleod et al. 2011; Hensel and Silliman2013). In subtropical and tropical coastlines, mangrovesare considered C-rich ecosystems and are beneficial tobiodiversity, hydrology, and global biogeochemistry (Alongi2014). However, the functions of mangrove ecosystemsare increasingly threatened by human activities, includ-ing eutrophication, organic and inorganic pollution, andthe introduction of invasive species.

The invasion of exotic plant species may alter ecosys-tem functions through a variety of mechanisms, such asreducing plant and animal biodiversity, altering wetlandhydrology, and changing C or nitrogen (N) cycling (Williamsand Grosholz 2008). Spartina alterniflora invasion wasshown to affect the community structures of methano-gens and sulfate-reducing bacteria in Phragmites australis-vegetated sediments (Zeleke et al. 2013). On the otherhand, plants may benefit from their association with di-verse root-associated microbial communities that re-spond rapidly to environmental changes (Lau and Lennon2012; Bai et al. 2013). The microbiota colonizing the rhi-zoplane and endophytic environment contribute to plantgrowth, productivity, C sequestration, and phytoreme-diation (Badri et al. 2009; Weyens et al. 2009; Bulgarelliet al. 2012). Biological interactions, including resourceavailability and enemy release, have been considered asthe 2 common hypotheses of plant invasion mechanisms(Blumenthal et al. 2009).

The perennial salt marsh grass S. alterniflora, whichoriginated from the northeastern United States, was in-troduced to China in 1979. Due to its extensive expan-sion, S. alterniflora is widely distributed in 9 coastalprovinces in East and South China, and its displacementof the native species has caused a number of ecologicalimpacts (Wan et al. 2009; Zhang et al. 2012). Some studiesinvestigated the effects of S. alterniflora invasion on man-grove ecosystems, including biomass of mangrove seed-lings, macroinvertebrate communities, microeukaryoticdiversity, and the composition of ammonia oxidizers(Zhang et al. 2012, 2013; Yu et al. 2014; Zhao et al. 2014).Rhizospheric bacterial diversity of S. alterniflora was alsoinvestigated by means of denaturing gradient gel electro-

phoresis (DGGE) of PCR-amplified nifH gene or 16S rDNAfragments (Lovell et al. 2000; Nie et al. 2010; Thomas et al.2014). However, little attention has been paid to the bac-terial diversity and structure of endophytic bacteria forboth exotic and native plants (Debbab et al. 2013). Under-standing community structures of root-associated bacte-ria from exotic plants would benefit the study of theirecological functions and invasion mechanisms.

For analysis of microbial community structure, thereare several culture-independent methods used in paststudies, such as 16S rRNA gene clone libraries, terminalrestriction fragment length polymorphism (TRFLP), andDGGE (Theron and Cloete 2000; Zhang and Xu 2008; FrancoDias et al. 2011). However, these molecular approachescould not reveal the details of the highly diverse microbialcommunities (Zhang and Xu 2008; Vanwonterghem et al.2014). In recent years, pyrosequencing and Illumina tech-niques have been widely used to get a more detailedpicture of the microbial communities in various environ-ment matrices (Degnan and Ochman 2012; Hu et al.2014). Therefore, this study investigated the endophyticand rhizoplane bacterial community structure and diver-sity from 3 sampling stands covered by monocultures ofS. alterniflora, Kandelia obovata, and both plants (transitionstands). These results would ultimately help in betterunderstanding the potential impacts of S. alterniflora in-vasion on the ecological function of mangrove.

Materials and methods

Study sites and samplingSpartina alterniflora (n = 6), Kandelia obovata (n = 3), and

sediment (n = 9) samples were collected from JiulongRiver Estuary Mangrove Nature Reserve (24°24=N, 117°55=E),China, in July 2012. In the sampled habitat, the 3 man-grove plant species Kandelia candel, Aegiceras corniculatum,and Avicennia marina were present. In 2007, S. alterniflorainvaded the mudflat zone and the K. candel habitat andthen formed a monoculture and a transition zone in thelow tidal wetland. The details of the different types ofsamples are given in Table 1. For rhizosphere sedi-ment (RS), roots of S. alterniflora and K. obovata were dugup and shaken gently and then rinsed with saline waterto remove the root-adhering sediments. RS samples werestored into clean brown glass bottles. The top 5 cm layerof the mudflat sediment (MS) was carefully placed into abucket using a stainless steel spoon. All types of samples

Pagination not final (cite DOI) / Pagination provisoire (citer le DOI)

2 Can. J. Microbiol. Vol. 61, 2015

Published by NRC Research Press

Can

. J. M

icro

biol

. Dow

nloa

ded

from

ww

w.n

rcre

sear

chpr

ess.

com

by

Inst

itute

of

Urb

an E

nvir

onm

ent C

hine

se A

cade

my

of S

cien

ces

on 0

7/29

/15

For

pers

onal

use

onl

y.

were taken in triplicates, stored at 4 °C in a car refriger-ator (CF-110DC, WAECO, Germany), and then transportedto the laboratory at the Institute of Urban Environment,Chinese Academy of Sciences, for further analyses.Physiochemical properties, such as pH, total C, total N,and total sulfur (S), of the sediment have already beenreported in a previous study (Zhang et al. 2011). SedimentpH ranged from 6.12 to 6.99. Total C and total N concen-trations in the mudflat were 12.4 and 1.33 g·kg−1, respec-tively, while in S. alterniflora zones they were 12.4 and1.49 g·kg−1, respectively. These values were obviouslylower than those (18.2 and 1.81 g·kg−1) observed in nativeK. obovata zones.

Roots of S. alterniflora and K. obovata were thoroughlywashed with deionized water, and rhizoplane bacteriawere isolated by vigorously shaking root segments (2 g)in 200 mL of PBS buffer (140 mmol·L–1 NaCl, 2.5 mmol·L–1

KCl, 10 mmol·L–1 Na2HPO4, 1.5 mmol·L–1 KH2PO4 (pH 7.4))for 1 h (Mendes et al. 2007). Microbial cells were collectedby centrifugation. Samples of endophytic bacteria inroots were collected using the method adopted in ourprevious studies (Hong et al. 2015). Briefly, washed rootsegments (5 g) were surface sterilized by sequential wash-ing in 70% ethanol for 1 min, 2% (v/v) sodium hypochloritefor 3 min, and 70% ethanol for 30 s, and were rinsed5 times with ample sterilized distilled water. By coatingonto tryptic soy agar plates, the final rinse was used toverify surface sterilization. Plates were incubated at28 °C, and no colonies were found after 10 days.

DNA extraction, PCR, and Illumina amplicon sequencingSurface-sterilized root segments were ground in a

mortar with liquid nitrogen. Total DNA from sedimentand strains was extracted and purified with a FastDNASpin kit for soil (MP Biomedicals, California, USA) and aDNA purification kit (Tiangen, China), according to themanufacturers’ recommendations. The V3 region ofbacterial 16S rRNA gene was amplified using primers338F (5=-ACTCCTACGGGAGGCAGCAG-3=) and 533R (5=-TTACCGCGGCTGCTGGCAC-3=) with identified barcodes(Huse et al. 2008). PCR amplifications were performedusing a thermocycler (Eppendorf, Hamburg, Germany)in 50 �L reaction volumes containing 25 �L of DreamTaqGreen PCR Master Mix (2×) (Thermo Scientific Co., USA),

0.5 �L of 1% bovine serum albumin (BSA), 0.2 �mol·L–1

(each) primer, 40–50 ng of template DNA, and 20.5 �L ofsterile water. PCR was carried out with the followingtemperature profiles: step one heated to 94 °C (3 min),30 s of denaturation at 94 °C, 30 s at the primers anneal-ing temperature (55 °C), 30 s of elongation at 72 °C. PCRproducts were purified using a Universal DNA purifica-tion kit (Tiangen, China) following the instructions ofthe manufacturer. The PCR products from all sampleswere combined and stood in 0.1 volume of 3 mol·L–1 so-dium acetate and 3 volumes of 100% ethanol overnight.The pellet of combined DNA was collected by centrifuga-tion, washed with 75% ethanol, eluted in 50 �L of water,and submitted to BGI (Beijing Genomics Institute, Shen-zhen, China) for sequencing (pair-end) on Illumina HiSeq2000 platform. Duplicates in endophytic bacteria ofK. obovata (EBM) were only included in the Fig. 3 data setdue to sample loss.

Data processing and statistical analysisThe raw reads were assembled following the barcoded

Illumina PE sequencing (BIPES) pipeline to reduce se-quence and PCR errors (Zhou et al. 2011) and were filteredby BGI in a prebioinformatics analysis. All the sequenceswere analyzed using Quantitative Insights Into Micro-bial Ecology (version 1.60) (Caporaso et al. 2010). Asmentioned in previous studies (Xie et al. 2015), briefly,operational taxonomic units (OTUs) were picked at 97%sequence similarity, and their representative sequenceswere chosen for alignment and taxonomic assignmentwith RDP classifier. Chimeric sequences, mitochondrial,chloroplast, and singleton OTUs were removed. We usedphylogenetic diversity index, Shannon diversity index,and an abundance-based coverage estimators Chao1 asmeasures of �-diversity according to our previous study(Hu et al. 2014). Rarefaction, �-diversity, and �-diversityanalyses were conducted according to the OTU table withminimal sequencing number (24 270) of the samples. For�-diversity analysis, dissimilarity of bacterial communi-ties was determined using principal coordinate analysison unweighted and weighted UniFrac distances amongall samples. Venn diagrams were employed to character-ize the shared bacterial communities among samplegroups and to generate core microbiome in different

Table 1. Details of different samples used in this study.

TypeSampledesignation Description

Spartina alterniflora RBSA Rhizoplane bacteria in root of S. alterniflora monocultureEBSA Endophytic bacteria in root of S. alterniflora monocultureRBSAM Rhizoplane bacteria in root of S. alterniflora from the transition standsEBSAM Endophytic bacteria in root of S. alterniflora from the transition stands

Kandelia obovata RBM Rhizoplane bacteria in root of K. obovata monocultureEBM Endophytic bacteria in root of K. obovata monoculture

Sediment RSSA Rhizospheric sediment in the S. alterniflora monocultureRSSAM Rhizospheric sediment in the S. alterniflora from the transition standsMS Mudflat sediment

Pagination not final (cite DOI) / Pagination provisoire (citer le DOI)

Hong et al. 3

Published by NRC Research Press

Can

. J. M

icro

biol

. Dow

nloa

ded

from

ww

w.n

rcre

sear

chpr

ess.

com

by

Inst

itute

of

Urb

an E

nvir

onm

ent C

hine

se A

cade

my

of S

cien

ces

on 0

7/29

/15

For

pers

onal

use

onl

y.

samples. Linear discriminate analysis (LDA) effect sizewas employed to identify indicator taxa (from genus tophylum level) associated with different sampling groups(Segata et al. 2011; Hu et al. 2014). All the sequences wereuploaded to the NCBI SRA database under the accessionNo. SRX525654.

Mean and standard deviation for each set of data werecalculated. One-way analysis of variance (ANOVA) andleast significant difference were performed by SPSS (ver-sion 19.0) for Windows (IBM Co., USA), and the diversityof endophytic, rhizoplane, and rhizosphere bacteria fromdifferent wetland plants was compared.

Results and discussion

Diversity of endophytic and rhizoplane bacterialcommunities

A total of 1 362 696 high-quality reads were obtainedfrom 27 samples, and clustered into 44 094 OTUs at equalsequencing depth. RS samples presented the highestdiversity of OTUs followed by MS, rhizoplane, and endo-phytic bacteria. (Table 2). �-Diversity using phylogeny-based metrics showed a significant difference betweensediment and root-associated samples (ANOVA test,P < 0.01). The root itself is generally considered as a morestable niche, while RS represents more complex habitatsproviding microorganisms with a large variety of C sources,including amino acids, organic acids, and carbohydrates(Berg and Smalla 2009). In this study, diversity indices(Chao1 and Shannon) indicated higher diversity of rhizo-plane and endophytic bacteria in the S. alterniflora mon-oculture compared with the K. obovata stands (ANOVAtest, P < 0.05) (Table 2 and Fig. 1). After S. alterniflorainvasion, the diversity indices of its root-associated bac-terial community decreased by approximately 20% ow-ing to the interactions between exotic and native plantsto a certain extent. Blumenthal et al. (2009) reported thatplants classified as competitors hosted more than 4 timesas many viruses and fungi as did stress tolerators. It alsosuggested that enemy release primarily contributes toinvasion by fast-growing species adapted to resource-rich environments. In addition, the numbers of OTUs

from the 2 rhizosphere sediments (RSSA and RSSAM)were higher than that from MS, indicating the “rhizo-sphere effect”. Root exudates released from the exoticplants might cause changes in sediment parameters,such as pH, C, N, and S, which significantly shape rhizo-spheric bacterial community structure (Haichar et al.2008; Berg and Smalla 2009; Zhang et al. 2011).

Overlapping of OTUsAccording to the results of Venn analysis, consistent

overlap patterns of OTU clusters among different sam-ples were obtained (Fig. 2). For endophytic bacteria, EBSAand EBSAM harbored 4795 and 2905 unique OTUs, re-spectively, while sharing 250 and 612 OTUs with EBM.Moreover, EBM shared more OTUs with EBSAM thanthose with EBSA. The results indicated the potential im-pact of K. obovata on the colonization of endophytic bac-teria in roots of S. alterniflora. A similar variation patternwas observed for rhizoplane bacteria (Fig. 2). These find-ings could be corroborated by the decreasing diversityindices of root-associated bacteria of S. alterniflora afterits invasions. In comparison with MS, RSSA and RSSAMhad a higher number of OTUs overlap, which suggestedthe impact of plantation on the microbial communitystructure in the sediments. Sediment and rhizoplanebacterial communities showed a higher number of OTUsoverlap, as compared with those existing between endo-phytic and rhizoplane samples. Meanwhile, endophyticbacterial communities shared 4434 OTUs with rhizoplane

Table 2. Estimated operational taxonomic unit (OTU) rich-ness and diversity indices of different types of samples col-lected from coastal wetland.

Type Sample*Phylogeneticdiversity Chao1 OTUs

Spartina alterniflora RBSA 333 14 168 5577EBSA 279 9 858 4213RBSAM 285 11 901 4524EBSAM 232 7 980 3220

Kandelia obovata RBM 241 9 837 3865EBM 162 5 791 2142

Sediment RSSA 453 20 753 8382RSSAM 448 20 280 8371MS 388 15 337 6837

Note: Data are the means (n = 3).*See Table 1 for description of samples.

Fig. 1. Shannon and equitability index of bacterialcommunities of sediment, rhizoplane, and endophyteassociated with Spartina alterniflora and Kandelia obovata.Error bars denote the standard deviation. Lowercaseletters above bars indicate significant difference atP < 0.05 for different samples, ANOVA with LSD test.RB, rhizoplane bacteria; EB, endophytic bacteria;SA, S. alterniflora monoculture; SAM, S. alterniflora from thetransition stands; RBM, rhizoplane bacteria in root ofK. obovata monoculture; EBM, endophytic bacteria in rootof K. obovata monoculture; RS, rhizospheric sediment;MS, mudflat sediment.

Pagination not final (cite DOI) / Pagination provisoire (citer le DOI)

4 Can. J. Microbiol. Vol. 61, 2015

Published by NRC Research Press

Can

. J. M

icro

biol

. Dow

nloa

ded

from

ww

w.n

rcre

sear

chpr

ess.

com

by

Inst

itute

of

Urb

an E

nvir

onm

ent C

hine

se A

cade

my

of S

cien

ces

on 0

7/29

/15

For

pers

onal

use

onl

y.

and sediment samples, suggesting the colonization ofendophytes in root partly originated from rhizosphericsediment environment.

Structure of bacterial communitiesOrdination plots based on principal coordinate analy-

ses were constructed using unweighted and weightedUniFrac distances to identify the pattern of communitystructure of endophyte, rhizoplane, and sediment bacte-ria (Fig. 3). RS and MS samples grouped tightly, and en-dophytic bacteria could be separated into 2 lineages: oneincluded K. obovata samples, while the other groupconsisted of monocultures of S. alterniflora and transi-tion stands. Rhizoplane bacteria from monoculturesof S. alterniflora, K. obovata, and transition stands werealso grouped. Rhizoplane samples of S. alterniflora werecloser to that of K. obovata, indicating the major influenceof sediment properties. However, there were obvious dif-ference in endophytic bacteria between S. alterniflora andK. obovata. This may be related to the difference betweensalt marsh plants. The result also showed the differenceof endophyte and rhizoplane bacteria in the monocul-ture and transition stands of S. alterniflora, due to theinfluence of K. obovata (Fig. 3). Overall, the clustered pat-terns of all samples were consistent with the OTUs and�-diversity analysis results.

Taxonomic characteristicsAll representative sequences clustered into different

groups (phyla or classes) according to the taxonomic clas-sification of the EzTaxon-e database (Fig. 4). In this study,12 known phyla with a predominance of 5 major phyla(Proteobacteria, Cyanobacteria, Bacteroidetes, Firmicutes, andChloroflexi) were observed. It is obvious that Proteobacteriawas the predominant phylum and contributed to 41%–88% of the total tags. These results suggested high bacte-rialdiversityacrosstherhizosphere,rhizoplane,endophyteofS. alterniflora and K. obovata. The number of phyla ob-served here was much greater than those reported inprevious studies using the 16S rRNA clone library (Li et al.2014).

In S. alterniflora monoculture, the major phyla ofrhizoplane bacteria were Proteobacteria (78.0% ± 4.6%),Bacteroidetes (5.6% ± 1.0%), Chloroflexi (3.3% ± 0.4%), andFirmicutes (1.6% ± 0.5%), while Proteobacteria (45.0% ± 1.9%),Cyanobacteria (23.7% ± 3.2%), Bacteroidetes (11.4% ± 1.8%),Firmicutes (4.3% ± 1.5%), and Spirochaetes (4.1% ± 0.6%) werepredominant in the endophytic bacteria of roots (Fig. 4).In contrast, dominant phyla in the rhizoplane bacterialcommunity of K. obovata included Proteobacteria (88.0% ±7.7%), Bacteroidetes (2.1% ± 0.5%), Chloroflexi (2.0% ± 0.3%),and Firmicutes (1.1% ± 0.2%), while most of the endophyticbacteria belong to Cyanobacteria (46.9% ± 1.2%), Proteobacteria

Fig. 2. Venn diagrams showing the number of operational taxonomic units shared among different groups of samples.RB, rhizoplane bacteria; EB, endophytic bacteria; SA, Spartina alterniflora monoculture; SAM, S. alterniflora from the transitionstands; RBM, rhizoplane bacteria in root of Kandelia obovata monoculture; EBM, endophytic bacteria in root of K. obovatamonoculture; RS, rhizospheric sediment; MS, mudflat sediment.

Pagination not final (cite DOI) / Pagination provisoire (citer le DOI)

Hong et al. 5

Published by NRC Research Press

Can

. J. M

icro

biol

. Dow

nloa

ded

from

ww

w.n

rcre

sear

chpr

ess.

com

by

Inst

itute

of

Urb

an E

nvir

onm

ent C

hine

se A

cade

my

of S

cien

ces

on 0

7/29

/15

For

pers

onal

use

onl

y.

Fig. 3. Principal coordinate (PC) analysis of bacterial community structures using the unweighted (a) and weighted (b) UniFrac distances. RB, rhizoplane bacteria;EB, endophytic bacteria; SA, Spartina alterniflora monoculture; SAM: S. alterniflora from the transition stands; RBM, rhizoplane bacteria in root of Kandelia obovatamonoculture; EBM, endophytic bacteria in root of K. obovata monoculture; RS, rhizospheric sediment; MS, mudflat sediment.

Pagination

notfinal(cite

DO

I)/

Pagination

provisoire(citer

leD

OI)

6C

an.J.M

icrobiol.Vol.61,2015

Publish

edby

NR

CR

esearchPress

Can

. J. M

icro

biol

. Dow

nloa

ded

from

ww

w.n

rcre

sear

chpr

ess.

com

by

Inst

itute

of

Urb

an E

nvir

onm

ent C

hine

se A

cade

my

of S

cien

ces

on 0

7/29

/15

For

pers

onal

use

onl

y.

(41.7% ± 6.3%), Firmicutes (6.7% ± 4.5%), and Chloroflexi(1.1% ± 0.8%). The results indicated that the Cyanobacteriaphylum was largely enriched in endophytic bacterial com-munities of both plants, while Spirochaetes only existed inS. alterniflora. The invasion and growth of S. alterniflorawithin the K. obovata forest may have increased the per-centage of Cyanobacteria and decreased the abundance ofBacteroidetes in rhizoplane and endophytic bacteria. Forthe abundance of Cyanobacteria, a significant difference(P < 0.001) was observed between EBSA and EBSAM, sug-gesting the influence of S. alterniflora invasion on the abun-dance of endophytic bacterial communities. This ispossibly attributed to the impact of K. obovata stands,including root exudates, rhizosphere bacteria, and theproperties of sediments. Previous studies suggested thecomposition and temporal variation of soil microbialcommunities were associated with habitat characteris-tics and vegetation types, such as the influence of rootexudates (Lau and Lennon 2012; Jiang et al. 2013). Mean-while, soil type and host genotype also determined thecomposition of root-inhabiting bacterial communities,especially for soil-derived root endophytes (Badri et al.2009; Reinhold-Hurek and Hurek 2011; Bulgarelli et al.2012). In this study, compared with MS, the abundancesof Proteobacteria and Bacteroidetes in RS were enhancedwhile numbers of Chloroflexi were inhibited. However, RSshowed no difference of bacterial communities betweenS. alterniflora monoculture and its ecotone with K. obovata,mainly including Proteobacteria (55.7% ± 4.0%), Chloroflexi(10.3% ± 1.4%), Bacteroidetes (4.7% ± 0.3%), and Firmicutes

(2.0% ± 0.3%). These results suggested a more sensitiveresponse of endophytes to environment change com-pared with rhizosphere bacteria (Bulgarelli et al. 2012).These limited samples reflect the variation of typical bac-terial species in roots of the 2 plants, but we recommendthat more samples be collected to identify the interac-tions of root-associated bacteria between different plants.

At the class level, Alphaproteobacteria, Deltaproteobacteria,and Gammaproteobacteria were the major groups forall samples (Fig. 4). In the S. alterniflora monoculture,Gammaproteobacteria (52.2% ± 4.6%) was significantlymore prominent in rhizoplane bacteria, followed byDeltaproteobacteria (15.9% ± 0.2%) and Alphaproteobacteria(6.9% ± 1.3%). In contrast, there was a different patternof endophytic bacteria, showing the distribution ofDeltaproteobacteria (16.4% ± 1.9%), Gammaproteobacteria (13.5% ±1.0%), and Alphaproteobacteria (12.1% ± 1.3%). In the transi-tion stands, Alphaproteobacteria, Gammaproteobacteria, andDeltaproteobacteria in endophytic bacteria of S. alterniflora ac-counted for 23.0% ± 5.4%, 8.9% ± 0.6%, and 8.2% ± 1.4%, respec-tively. In addition, Deltaproteobacteria and Gammaproteobacteriawere greatly enriched in sediment samples with planta-tions, although there was no obvious difference forAlphaproteobacteria between RS and MS.

Indicator bacterial taxa and their ecological functionalimplication

In addition to �- and �-diversity, LDA effect size wasused to demonstrate potentially discriminating taxaamong 3 major habitats, including S. alterniflora and

Fig. 4. Bacterial community composition in sediment, rhizoplane, and endophyte of Spartina alterniflora and Kandelia obovata.Minor phyla that accounting for <0.5% of total sequences are summarized in the group “Other and unclassified bacteria”.Error bars denote the standard deviation. RB, rhizoplane bacteria; EB, endophytic bacteria; SA, S. alterniflora monoculture;SAM, S. alterniflora from the transition stands; RBM, rhizoplane bacteria in root of K. obovata monoculture; EBM, endophyticbacteria in root of K. obovata monoculture; RS, rhizospheric sediment; MS, mudflat sediment.

Pagination not final (cite DOI) / Pagination provisoire (citer le DOI)

Hong et al. 7

Published by NRC Research Press

Can

. J. M

icro

biol

. Dow

nloa

ded

from

ww

w.n

rcre

sear

chpr

ess.

com

by

Inst

itute

of

Urb

an E

nvir

onm

ent C

hine

se A

cade

my

of S

cien

ces

on 0

7/29

/15

For

pers

onal

use

onl

y.

K. obovata stands (Fig. 5). There were only 3 replicates forrhizoplane bacteria; therefore, we combined the LDAscore of exotic and native plants to determine their dif-ference with other bacterial groups. The results showedthat there were 33 taxa (4 phyla, 9 classes, 12 orders,11 families, and 7 genera) distinguishing the 2 groups onthe basis of LDA scores >3.0. Phylum Proteobacteria and sev-eral Gammaproteobacteria-associated taxa were significantlypredominant in root-associated bacteria of S. alternifloraand K. obovata (RB group). For the endophytic bacterialcommunity in the roots of S. alterniflora, Bacteroideteswere enriched from different classified levels, such asBacteroidia, Flavobacteria, and Sphingobacteriia. In addition,EBSA contained a greater abundance of Alphaproteobacteria(Proteobacteria), Clostridia (Firmicutes), Spirochaetes (Spirochaetes),and several taxa of Deltaproteobacteria than those in othergroups. In comparison, the phylum Firmicutes was signif-icantly more prominent in the endophytic bacterial com-munity in the roots of K. obovata, including Clostridiaceae,Pseudoalteromonadaceae, Oceanospirillaceae, and Planococcaceae.

However, Proteobacteria was enriched in the rhizoplanebacterial community in root of both plants.

Ecological functional implications of root-associatedbacteria of S. alterniflora and K. obovata were discussed inthis study. In the past, some studies reported that inva-sion of S. alterniflora would alter the community struc-ture of related functional microorganisms, and affect theC, N, and S cycles in the salt marshes ecosystem (Liaoet al. 2007; Thomas et al. 2014). In this study, the commu-nity structure of rhizoplane and endophyte were inves-tigated using samples from different stands covered byS. alterniflora and K. obovata (Fig. 5), so as to further under-stand their impacts on natural ecology of salt marsh.

Nitrogen-fixing cyanobacteria are able to form sym-biotic associations with various plants, such as wheatand potato (Gantar et al. 1995; Ringelberg et al. 2012).Fuernkranz et al. (2008) also reported that leaf-associateddiazotrophic bacterial communities primarily belong toCyanobacteria, which may provide significant N input intothis rainforest ecosystem. In this study, Cyanobacteria pres-

Fig. 5. Indicator bacterial groups associated with different groups of samples identifying using linear discriminateanalysis (LDA) effect size algorithm (LDA > 3). EBM, endophytic bacteria in root of Kandelia obovata; EBS, endophytic bacteria inroot of Spartina alterniflora; RB, rhizoplane bacteria of both plants.

Pagination not final (cite DOI) / Pagination provisoire (citer le DOI)

8 Can. J. Microbiol. Vol. 61, 2015

Published by NRC Research Press

Can

. J. M

icro

biol

. Dow

nloa

ded

from

ww

w.n

rcre

sear

chpr

ess.

com

by

Inst

itute

of

Urb

an E

nvir

onm

ent C

hine

se A

cade

my

of S

cien

ces

on 0

7/29

/15

For

pers

onal

use

onl

y.

ence in root endophytes is valuable for the 2 plants toincorporate N into plant biomass. These results indi-cated that endophytic bacteria from these salt marshplants may provide an advantage as N-fixing agents. Wealso found the genus Clostridium, which is considered aubiquitous endophytic bacterium in gramineous plantsand has exhibited N2-fixing capability in association withnondiazotrophic endophytes (Miyamoto et al. 2004).Nitrogen-fixing bacteria have been isolated from man-grove rhizosphere and identified as members of the gen-era Azospirillum, Azotobacter, Rhizobium, Klebsiella, Vibrio,Phyllobacterium, Arthrobacter, Corynebacterium, and Oceanomonas(Holguin et al. 1992; Flores-Mireles et al. 2007). Thus, theyhave a significant impact on the N cycles in the coastalwetland ecosystem. In Fig. 5, some rhizosphere bacte-ria of salt marsh plants belonged to the families ofAlteromonadacea, Pseudoalteromonadacea, Vibrionacea, andMethylophilacea, which play key roles in the decomposi-tion processes of dissolved organic matter. Pseudomon-ads are often found in contaminated aquifers because oftheir ability to use a large number of substances as Csources (Moore et al. 2006).

It is well documented that Deltaproteobacteria is a majorgroup of S-reducing bacteria in anaerobic environments(Castro et al. 2000; Zeleke et al. 2013), which may explainhigher S concentrations in S. alterniflora-invaded zones thanin native plant and unvegetated zones (Zhou et al. 2009).In this study, after the invasion of S. alterniflora withinK. obovata forest, the percentage of Gammaproteobacteria inrhizoplane bacteria increased, while the abundance ofAlphaproteobacteria in endophytic bacteria decreased. ForK. obovata, Gammaproteobacteria (75.0% ± 7.7%) was observedas the more significant and prominent rhizoplane bac-teria, followed by Deltaproteobacteria (7.4% ± 3.9%) andAlphaproteobacteria (3.0% ± 0.4%) (Fig. 4). This was con-sistent with the fact that Gammaproteobacteria andDeltaproteobacteria are the 2 major classes of the phylumProteobacteria in the rhizosphere of mangrove (Jiang et al.2013). These abundant classes were clearly habitat-specificowing to the different nutrition patterns. As shown inFig. 5, some endophytic bacteria in roots of S. alterniflorawere members of the order Desulfovibrionales. The genusDesulfovibrio, belonging to sulfate-reducing bacteria, isknown to oxidize acetate and other organic compounds(Muyzer and Stams 2008). These species are considered asnumerically important members on macrophyte rootsurfaces, such as Phragmites australis (Vladar et al. 2008).Desulfovibrio spp. also has a strong affinity for sulfate un-der sulfate-limited conditions (Laanbroek et al. 1984;Basso et al. 2005). Thus, the root-associated bacteria ofS. alterniflora might make a great contribution to S accu-mulation in S. alterniflora-invaded stands.

On the basis of analysis of microbial community, root-associated bacteria in S. alterniflora roots have the poten-tial to affect nutrient metabolism in wetland ecosystems,especially with regard to the N and S cycles, and the

removal of some organic matter. Variations in commu-nity structures of root-associated bacteria also providesome clues for S. alterniflora invasions. However, theculture-independent method cannot provide direct infor-mation on the function of the individual communitymembers; therefore, further work is necessary to im-prove our understanding of the mechanisms throughthe isolation of endophytic bacteria in salt marsh plants.

AcknowledgementsThis work was supported by the National Natural Science

Foundation of China (No. 41006064), the Fujian Youth Sci-ence Foundation (No. 2012J05081), the Natural ScienceFoundation of Ningbo City (No. 2014A610100), and the StateScholarships Foundation (No. 201304910080). We alsothank Shen Yu from Institute of Urban Environment, Chi-nese Academy of Sciences, for his valuable comments, andYongshan Chen for collection of samples.

ReferencesAlongi, D.M. 2014. Carbon cycling and storage in mangrove for-

ests. Annu. Rev. Mar. Sci. 6: 195–219. doi:10.1146/annurev-marine-010213-135020. PMID:24405426.

Badri, D.V., Weir, T.L., van der Lelie, D., and Vivanco, J.M. 2009.Rhizosphere chemical dialogues: plant–microbe interac-tions. Curr. Opin. Biotechnol. 20(6): 642–650. doi:10.1016/j.copbio.2009.09.014. PMID:19875278.

Bai, S.J., Li, J.W., He, Z.L., Van Nostrand, J.D., Tian, Y., Lin, G.H.,et al. 2013. GeoChip-based analysis of the functional genediversity and metabolic potential of soil microbial commu-nities of mangroves. Appl. Microbiol. Biotechnol. 97(15):7035–7048. doi:10.1007/s00253-012-4496-z. PMID:23135227.

Basso, O., Caumette, P., and Magot, M. 2005. Desulfovibrio putealissp. nov., a novel sulfate-reducing bacterium isolated from adeep subsurface aquifer. Int. J. Syst. Evol. Microbiol. 55: 101–104. doi:10.1099/ijs.0.63303-0. PMID:15653861.

Berg, G., and Smalla, K. 2009. Plant species and soil type coop-eratively shape the structure and function of microbial com-munities in the rhizosphere. FEMS Microbiol. Ecol. 68(1):1–13. doi:10.1111/j.1574-6941.2009.00654.x. PMID:19243436.

Blumenthal, D., Mitchell, C.E., Pysek, P., and Jarosík, V. 2009.Synergy between pathogen release and resource availabilityin plant invasion. Proc. Natl. Acad. Sci. U.S.A. 106(19): 7899–904. doi:10.1073/pnas.0812607106. PMID:19416888.

Bulgarelli, D., Rott, M., Schlaeppi, K., van Themaat, E.V.L.,Ahmadinejad, N., Assenza, F., et al. 2012. Revealing structureand assembly cues for Arabidopsis root-inhabiting bacterialmicrobiota. Nature, 488(7409): 91–95. doi:10.1038/nature11336.PMID:22859207.

Caporaso, J.G., Kuczynski, J., Stombaugh, J., Bittinger, K.,Bushman, F.D., Costello, E.K., et al. 2010. QIIME allows anal-ysis of high-throughput community sequencing data. Nat.Methods, 7(5): 335–336. doi:10.1038/nmeth.f.303. PMID:20383131.

Castro, H.F., Williams, N.H., and Ogram, A. 2000. Phylogeny ofsulfate-reducing bacteria. FEMS Microbiol. Ecol. 31(1): 1–9.doi:10.1016/S0168-6496(99)00071-9. PMID:10620713.

Debbab, A., Aly, A.H., and Proksch, P. 2013. Mangrove derivedfungal endophytes — a chemical and biological perception.Fungal Divers. 61(1): 1–27. doi:10.1007/s13225-013-0243-8.

Degnan, P.H., and Ochman, H. 2012. Illumina-based analysis ofmicrobial community diversity. ISME J. 6(1): 183–194. doi:10.1038/ismej.2011.74. PMID:21677692.

Flores-Mireles, A.L., Winans, S.C., and Holguin, G. 2007. Molec-

Pagination not final (cite DOI) / Pagination provisoire (citer le DOI)

Hong et al. 9

Published by NRC Research Press

Can

. J. M

icro

biol

. Dow

nloa

ded

from

ww

w.n

rcre

sear

chpr

ess.

com

by

Inst

itute

of

Urb

an E

nvir

onm

ent C

hine

se A

cade

my

of S

cien

ces

on 0

7/29

/15

For

pers

onal

use

onl

y.

ular characterization of diazotrophic and denitrifying bac-teria associated with mangrove roots. Appl. Environ.Microbiol. 73(22): 7308–7321. doi:10.1128/AEM.01892-06.PMID:17827324.

Franco Dias, A.C., Dini-Andreote, F., Taketani, R.G., Tsai, S.M.,Azevedo, J.L., de Melo, I.S., and Andreote, F.D. 2011. Archaealcommunities in the sediments of three contrasting man-groves. J. Soils Sediments, 11(8): 1466–1476. doi:10.1007/s11368-011-0423-7.

Fuernkranz, M., Wanek, W., Richter, A., Abell, G., Rasche, F.,and Sessitsch, A. 2008. Nitrogen fixation by phyllosphere bac-teria associated with higher plants and their colonizing epi-phytes of a tropical lowland rainforest of Costa Rica. ISME J.2(5): 561–570. doi:10.1038/ismej.2008.14. PMID:18273066.

Gantar, M., Rowell, P., Kerby, N.W., and Sutherland, I.W. 1995.Role of extracellular polysaccharide in the colonization ofwheat (Triticum vulgare L.) roots by N2-fixing cyanobacteria.Biol. Fertil. Soils, 19(1): 41–48. doi:10.1007/Bf00336345.

Haichar, F.E., Marol, C., Berge, O., Rangel-Castro, J.I., Prosser, J.I.,Balesdent, J., et al. 2008. Plant host habitat and root exudatesshape soil bacterial community structure. ISME J. 2(12): 1221–1230. doi:10.1038/ismej.2008.80. PMID:18754043.

Hensel, M.J.S., and Silliman, B.R. 2013. Consumer diversityacross kingdoms supports multiple functions in a coastalecosystem. Proc. Natl. Acad. Sci. U.S.A. 110(51): 20621–20626.doi:10.1073/pnas.1312317110. PMID:24297926.

Holguin, G., Guzman, M.A., and Bashan, Y. 1992. Two newnitrogen-fixing bacteria from the rhizosphere of mangrovetrees: their isolation, identification and in vitro interactionwith rhizosphere Staphylococcus sp. FEMS Microbiol. Ecol.101(3): 207–216. doi:10.1111/j.1574-6968.1992.tb05777.x.

Hong, Y., Liao, D., Chen, J., Khan, S., Su, J., and Li, H. 2015. Acomprehensive study of the impact of polycyclic aromatichydrocarbons (PAHs) contamination on salt marsh plantsSpartina alterniflora: implication for plant–microbe interac-tions in phytoremediation. Environ. Sci. Pollut. Res. 22(9):7071–7081. doi:10.1007/s11356-014-3912-6. PMID:25501539.

Hu, A.Y., Yang, X.Y., Chen, N.W., Hou, L.Y., Ma, Y., and Yu, C.P.2014. Response of bacterial communities to environmentalchanges in a mesoscale subtropical watershed, SoutheastChina. Sci. Total Environ. 472: 746–756. doi:10.1016/j.scitotenv.2013.11.097. PMID:24333997.

Huse, S.M., Dethlefsen, L., Huber, J.A., Welch, D.M., Relman, D.A.,and Sogin, M.L. 2008. Exploring microbial diversity andtaxonomy using SSU rRNA hypervariable tag sequencing.PLoS Genet. 4(11): e1000255. doi:10.1371/journal.pgen.1000255.PMID:19023400.

Jiang, X.-T., Peng, X., Deng, G.-H., Sheng, H.-F., Wang, Y.,Zhou, H.-W., and Tam, N.F.-Y. 2013. Illumina sequencing of16S rRNA tag revealed spatial variations of bacterial commu-nities in a mangrove wetland. Microb. Ecol. 66(1): 96–104.doi:10.1007/s00248-013-0238-8. PMID:23649297.

Laanbroek, H.J., Geerligs, H.J., Sijtsma, L., and Veldkamp, H.1984. Competition for sulfate and ethanol among Desulfobacter,Desulfobulbus, and Desulfovibrio species isolated from intertidalsediments. Appl. Environ. Microbiol. 47(2): 329–334. PMID:16346474.

Lau, J.A., and Lennon, J.T. 2012. Rapid responses of soil micro-organisms improve plant fitness in novel environments.Proc. Natl. Acad. Sci. U.S.A. 109(35): 14058–14062. doi:10.1073/pnas.1202319109. PMID:22891306.

Li, H., Liao, D., Su, J.Q., Huang, F.Y., and Hong, Y.W. 2014. Diver-sity and function of endophytic bacteria in roots of exoticplant Spartina alterniflora. Chin. J. Appl. Environ. Biol. 20(5):856–862. doi:10.3724/SP.J.1145.2014.03027. [In Chinese.]

Liao, C.Z., Luo, Y.Q., Jiang, L.F., Zhou, X.H., Wu, X.W., Fang, C.M.,et al. 2007. Invasion of Spartina alterniflora enhanced ecosys-tem carbon and nitrogen stocks in the Yangtze Estuary,

China. Ecosystems, 10(8): 1351–1361. doi:10.1007/s10021-007-9103-2.

Lovell, C.R., Piceno, Y.M., Quattro, J.M., and Bagwell, C.E. 2000.Molecular analysis of diazotroph diversity in the rhizosphereof the smooth cordgrass, Spartina alterniflora. Appl. Environ.Microbiol. 66(9): 3814–3822. doi:10.1128/AEM.66.9.3814-3822.2000. PMID:10966395.

Mcleod, E., Chmura, G.L., Bouillon, S., Salm, R., Bjork, M.,Duarte, C.M., et al. 2011. A blueprint for blue carbon: towardan improved understanding of the role of vegetated coastalhabitats in sequestering CO2. Front. Ecol. Environ. 9(10): 552–560. doi:10.1890/110004.

Mendes, R., Pizzirani-Kleiner, A.A., Araujo, W.L., andRaaijmakers, J.M. 2007. Diversity of cultivated endophyticbacteria from sugarcane: Genetic and biochemical character-ization of Burkholderia cepacia complex isolates. Appl. Envi-ron. Microbiol. 73(22): 7259–7267. doi:10.1128/AEM.01222-07.PMID:17905875.

Miyamoto, T., Kawahara, M., and Minamisawa, K. 2004. Novelendophytic nitrogen-fixing clostridia from the grass Miscanthussinensis as revealed by terminal restriction fragment lengthpolymorphism analysis. Appl. Environ. Microbiol. 70(11):6580–6586. doi:10.1128/AEM.70.11.6580-6586.2004. PMID:15528521.

Moore, F.P., Barac, T., Borrernans, B., Oeyen, L., Vangronsveld, J.,van der Lelie, D., et al. 2006. Endophytic bacterial diversity inpoplar trees growing on a BTEX-contaminated site: the char-acterisation of isolates with potential to enhance phytoreme-diation. Syst. Appl. Microbiol. 29(7): 539–556. doi:10.1016/j.syapm.2005.11.012. PMID:16919907.

Muyzer, G., and Stams, A.J.M. 2008. The ecology and biotechnol-ogy of sulphate-reducing bacteria. Nat. Rev. Microbiol. 6(6):441–454. doi:10.1038/nrmicro1892. PMID:18461075.

Nie, M., Gao, L.X., Yan, J.H., Fu, X.H., Xiao, M., Yang, J., and Li, B.2010. Population variation of invasive Spartina alterniflora candifferentiate bacterial diversity in its rhizosphere. Plant Ecol.209(2): 219–226. doi:10.1007/s11258-009-9687-z.

Reinhold-Hurek, B., and Hurek, T. 2011. Living inside plants:bacterial endophytes. Curr. Opin. Plant Biol. 14(4): 435–443.doi:10.1016/j.pbi.2011.04.004. PMID:21536480.

Ringelberg, D., Foley, K., and Reynolds, C.M. 2012. Bacterial en-dophyte communities of two wheatgrass varieties followingpropagation in different growing media. Can. J. Microbiol.58(1): 67–80. doi:10.1139/w11-122. PMID:22220581.

Segata, N., Izard, J., Waldron, L., Gevers, D., Miropolsky, L.,Garrett, W.S., and Huttenhower, C. 2011. Metagenomic bio-marker discovery and explanation. Genome Biol. 12(6): R60.doi:10.1186/gb-2011-12-6-r60. PMID:21702898.

Theron, J., and Cloete, T.E. 2000. Molecular techniques for de-termining microbial diversity and community structure innatural environments. Crit. Rev. Microbiol. 26(1): 37–57. doi:10.1080/10408410091154174. PMID:10782339.

Thomas, F., Giblin, A.E., Cardon, Z.G., and Sievert, S.M. 2014.Rhizosphere heterogeneity shapes abundance and activity ofsulfur-oxidizing bacteria in vegetated salt marsh sediments.Front. Microbiol. 5: 309. doi:10.3389/fmicb.2014.00309. PMID:25009538.

Vanwonterghem, I., Jensen, P.D., Ho, D.P., Batstone, D.J., andTyson, G.W. 2014. Linking microbial community structure,interactions and function in anaerobic digesters usingnew molecular techniques. Curr. Opin. Biotechnol. 27: 55–64. doi:10.1016/j.copbio.2013.11.004. PMID:24863897.

Vladar, P., Rusznyak, A., Marialigeti, K., and Borsodi, A.K. 2008.Diversity of sulfate-reducing bacteria inhabiting the rhizo-sphere of Phragmites australis in Lake Velencei (Hungary) re-vealed by a combined cultivation-based and molecularapproach. Microb. Ecol. 56(1): 64–75. doi:10.1007/s00248-007-9324-0. PMID:18066486.

Wan, S.W., Qin, P., Liu, J.N., and Zhou, H.X. 2009. The positive

Pagination not final (cite DOI) / Pagination provisoire (citer le DOI)

10 Can. J. Microbiol. Vol. 61, 2015

Published by NRC Research Press

Can

. J. M

icro

biol

. Dow

nloa

ded

from

ww

w.n

rcre

sear

chpr

ess.

com

by

Inst

itute

of

Urb

an E

nvir

onm

ent C

hine

se A

cade

my

of S

cien

ces

on 0

7/29

/15

For

pers

onal

use

onl

y.

and negative effects of exotic Spartina alterniflora in China.Ecol. Eng. 35(4): 444–452. doi:10.1016/j.ecoleng.2008.05.020.

Weyens, N., van der Lelie, D., Taghavi, S., and Vangronsveld, J.2009. Phytoremediation: plant–endophyte partnerships takethe challenge. Curr. Opin. Biotechnol. 20(2): 248–254. doi:10.1016/j.copbio.2009.02.012. PMID:19327979.

Williams, S.L., and Grosholz, E.D. 2008. The invasive specieschallenge in estuarine and coastal environments: marryingmanagement and science. Estuaries Coasts, 31(1): 3–20. doi:10.1007/s12237-007-9031-6.

Xie, W.Y., Su, J.Q., and Zhu, Y.G. 2015. Phyllosphere bacterialcommunity of floating macrophytes in paddy soil environ-ments as revealed by Illumina high-throughput sequencing.Appl. Environ. Microbiol. 81(2): 522–532. doi:10.1128/AEM.03191-14. PMID:25362067.

Yu, Z., Yang, J., Yu, X., Liu, L., and Tian, Y. 2014. Abovegroundvegetation influences belowground microeukaryotic com-munity in a mangrove nature reserve. Wetlands, 34(2): 393–401. doi:10.1007/s13157-013-0506-x.

Zeleke, J., Sheng, Q., Wang, J.-G., Huang, M.-Y., Xia, F., Wu, J.-H.,and Quan, Z.-X. 2013. Effects of Spartina alterniflora invasionon the communities of methanogens and sulfate-reducingbacteria in estuarine marsh sediments. Front. Microbiol. 4:243. doi:10.3389/fmicb.2013.00243. PMID:23986751.

Zhang, L., and Xu, Z. 2008. Assessing bacterial diversity in soil.J. Soils Sediments, 8(6): 379–388. doi:10.1007/s11368-008-0043-z.

Zhang, Q.-F., Peng, J.-J., Chen, Q., Li, X.-F., Xu, C.-Y., Yin, H.-B.,and Yu, S. 2011. Impacts of Spartina alterniflora invasion onabundance and composition of ammonia oxidizers in estua-rinesediment. J. SoilsSediments,11(6): 1020–1031.doi:10.1007/s11368-011-0369-9.

Zhang, Q., Peng, J., Chen, Q., Yang, X., Hong, Y., and Su, J. 2013.Abundance and composition of denitrifiers in response toSpartina alterniflora invasion in estuarine sediment. Can. J.Microbiol. 59(12): 825–836. doi:10.1139/cjm-2013-0516. PMID:24313455.

Zhang, Y., Huang, G., Wang, W., Chen, L., and Lin, G. 2012.Interactions between mangroves and exotic Spartina in an an-thropogenically disturbed estuary in southern China. Ecology,93(3): 588–597. doi:10.1890/11-1302.1. PMID:22624213.

Zhao, C., Liu, X., Bai, J., Lü, F., and Li, J. 2014. Impact of Spartinaalterniflora on benthic macro-invertebrates communities onmangrove wetland in Xicungang Estuary, Guangxi. Biodiv-ers. Sci. 22(5): 630–639. doi:10.3724/SP.J.1003.2014.00109.

Zhou, C.F., An, S.Q., Deng, Z.F., Yin, D.Q., Zhi, Y.B., Sun, Z.Y.,et al. 2009. Sulfur storage changed by exotic Spartina alterniflorain coastal saltmarshes of China. Ecol. Eng. 35(4): 536–543. doi:10.1016/j.ecoleng.2008.01.004.

Zhou, H.W., Li, D.F., Tam, N.F.Y., Jiang, X.T., Zhang, H., Sheng, H.F.,et al. 2011. BIPES, a cost-effective high-throughput method forassessing microbial diversity. ISME J. 5(4): 741–749. doi:10.1038/ismej.2010.160. PMID:20962877.

Pagination not final (cite DOI) / Pagination provisoire (citer le DOI)

Hong et al. 11

Published by NRC Research Press

Can

. J. M

icro

biol

. Dow

nloa

ded

from

ww

w.n

rcre

sear

chpr

ess.

com

by

Inst

itute

of

Urb

an E

nvir

onm

ent C

hine

se A

cade

my

of S

cien

ces

on 0

7/29

/15

For

pers

onal

use

onl

y.