Characterization of microbial communities in the aqueous phase of a constructed model wetland treating 1,2-dichloroethene-contaminated groundwater

R E S E A R C H A R T I C L E

Characterizationofmicrobial communities in theaqueousphaseofa constructedmodelwetland treating1,2-dichloroethene-contaminatedgroundwaterGwenael Imfeld1, Cristian Estop Aragones1, Ingo Fetzer2, Eva Meszaros3, Simone Zeiger1, IvonneNijenhuis1, Marcell Nikolausz4, Sylvain Delerce1 & Hans H. Richnow1

1Department of Isotope Biogeochemistry, Helmholtz Centre for Environmental Research – UFZ, Leipzig, Germany; 2Department of Environmental

Microbiology, Helmholtz Centre for Environmental Research – UFZ, Leipzig, Germany; 3Department of Microbiology, Eotvos Lorand University of

Science, Budapest, Hungary; and 4Department of Environmental Biotechnology, Helmholtz Centre for Environmental Research – UFZ, Leipzig, Germany

Correspondence: Gwenael Imfeld,

Laboratory of Hydrology and Geochemistry

of Strasbourg (LHyGeS), University of

Strasbourg/ENGEES, CNRS 1, quai Koch BP

61039, F-67 070 Strasbourg, France. Tel.:

133 0 3 88 24 82 43; fax: 133 0 3 88 24 82

84; e-mail:

[email protected]

Received 26 May 2009; revised 16 November

2009; accepted 16 November 2009.

Final version published online 19 January 2010.

DOI:10.1111/j.1574-6941.2009.00825.x

Editor: Max Haggblom

Keywords

chlorinated ethenes; compound-specific

isotope analysis; DGGE; ordination methods;

Dehalococcoides; biodegradation.

Abstract

The dynamics and composition of microbial communities in the aqueous phase of

a model wetland supplied with cis- and trans-1,2-dichloroethenes (DCE)-con-

taminated groundwater was characterized. PCR-denaturing gradient gel electro-

phoresis analysis of water samples obtained from different parts of the wetland

revealed that changes of the bacterial community structure coincided with a

succession of the hydrochemical conditions in the wetland, from oxic towards

anoxic conditions. During this transition phase, the appearance of vinyl chloride

and ethene correlated with the presence of putative dechlorinating bacteria

(Dehalococcoides spp., Geobacter spp. and Dehalobacter spp.). Additionally, a shift

of the DCE isotopic composition indicated the progressive prevalence of reductive

dechlorination in the wetland. Although the DCE degradation processes varied

over time, biodegradation activity was maintained in the wetland system. 16S

rRNA gene libraries revealed that Proteobacteria accounted for 4 50% of 16S

rRNA genes clone libraries, whereas�17% of the sequences from the wetland were

related to sulphate reducers. Based on a multiple-method approach, this study

illustrates the linkage between microbial community dynamics and composition,

changes of hydrochemical conditions and processes of DCE degradation in a

wetland system.

Introduction

Dichloroethenes (DCE) and carcinogenic vinyl chloride

(VC) often originate from the reductive dechlorination of

tetrachloroethene (PCE) or trichloroethene (TCE) and tend

to accumulate within anoxic aquifers (McCarty & Semperini,

1994; Vogel, 1994; Maymo-Gatell et al., 1997). Over the

course of contaminant migration from ground to surface

water, the contaminant flux can be intercepted by wetlands

(Reddy & Dangelo, 1997; Stottmeister & Wiessner, 2003).

Biodegradation of DCE in wetlands is likely to be controlled

by a complex and dynamic assemblage of adjacent aerobic

and anaerobic zones that typically prevail in biogeochemical

heterogenic environments (Armenante et al., 1992; Master

et al., 2002; Meade & D’Angelo, 2005). Wetland systems

located in draining areas of groundwater may thus contri-

bute to natural attenuation of intermediate chlorinated

hydrocarbons, such as DCE and VC (McCarty & Semperini,

1994; Lorah & Olsen, 1999; Amon et al., 2007). In aerobic

zones, oxidation of DCE can occur metabolically (Bradley &

Chapelle, 2000; Coleman et al., 2002) or cometabolically

(Kim & Semprini, 2005). Alternatively, anaerobic oxidation

of DCE occurring under Mn(IV)-, Fe(III)- or SO4-reducing

conditions has been demonstrated in laboratory experiments

(Bradley & Chapelle, 1998; Hata et al., 2004). Under strictly

anaerobic conditions, DCE can additionally be subjected to

reductive dechlorination and finally transformed to ethene

(Maymo-Gatell et al., 1997). However, little is known about

microbial diversity and biogeochemical processes in wetlands

receiving water fluxes contaminated by organic chemicals.

Recently, constructed wetlands used for water remedia-

tion have been investigated with respect to their ability to

FEMS Microbiol Ecol 72 (2010) 74–88Journal compilation c� 2010 Federation of European Microbiological SocietiesPublished by Blackwell Publishing Ltd. No claim to original German government works

MIC

ROBI

OLO

GY

EC

OLO

GY

mailto:[email protected]

remove chlorinated hydrocarbons. Most investigations so

far have consisted of contaminant concentrations and

isotopic composition analyses (Kassenga et al., 2003; Keefe

et al., 2004; Amon et al., 2007; Braeckevelt et al., 2007;

Imfeld et al., 2008a). However, these analyses can be

efficiently coupled with culture-independent molecular

techniques in order to provide additional information on

the phylogenetic diversity and structure of microbial com-

munities in contaminated wetlands (Nocker et al., 2007;

Weiss & Cozzarelli, 2008). DNA fingerprinting techniques

were applied for characterization of microbial diversity and

community structure dynamics in constructed wetlands

treating domestic and dairy wastewater (Truu et al., 2005;

Ibekwe et al., 2007), as well as assessing the microbial

community composition of wetland leachates (Walsh et al.,

2002). In addition to the general overview of the microbial

diversity, specific detection of bacteria affiliated to the

genera Dehalobacter, Dehalococcoides or Geobacter degrading

chlorinated ethenes can be used as a more specific indicator

for ongoing potential dechlorinating activity (Loffler et al.,

2000; Hendrickson et al., 2002; Duhamel et al., 2004; Imfeld

et al., 2008b; Nikolausz et al., 2008). However, microbial

community diversity, structures and degradation activities

in wetland systems often vary on both spatial and temporal

scales with ambient geochemical conditions. Therefore,

statistical tools relating microbial community structures to

hydrochemical processes across ecological scales allow to

gain a more consistent insight into the contaminant

attenuation potential of heterogeneous systems (Ramette,

2007).

This study examined the dynamics of the microbial

communities in the aqueous phase of a model system

representing wetland zones located at the interface be-

tween anoxic DCE-contaminated aquifers and oxic surface

water bodies. A previous study revealed the occurrence of

biodegradation activity and major hydrochemical changes

from oxic to strongly reducing conditions in the model

wetland system (Imfeld et al., 2008a). These changes were

associated with a parallel shift in the dominant degrada-

tion mechanisms over time: the sequence was initiated by

DCE degradation under oxic conditions and progressively

changed towards reductive dechlorination under anoxic

conditions. It is hypothesized that changes of the micro-

bial diversity affected the prevailing degradation mechan-

ism in the model wetland, which in turn may influence the

system functioning over time with respect to organic

contaminant removal. Therefore, the first objective of the

present study was to characterize the wetland microbial

community during the transition phase from prevailing

aerobic to anaerobic conditions. The second objective was

to identify the potential contribution of the microbial

community to the biogeochemical processes that influ-

ence DCE transformation in the wetland system, mainly

focusing on reductive dechlorinating bacteria and their

activity.

Materials and methods

System design and sampling procedure

The model wetland was a rectangular chamber

(201� 60� 5 cm), filled with quartz sand and planted with

common rush (Juncus effusus, L.) (Fig. 1). The wetland was

continuously supplied with groundwater from a stainless-

steel tank. The contaminated groundwater was collected

bimonthly from the Bitterfeld/Wolfen site and kept under

anoxic conditions (N2 atmosphere). Regular geochemical

analysis of the hydrogeochemistry revealed a stable hydro-

chemistry of the supplied groundwater over time, with cis-

and trans-1,2-DCE as the dominant contaminants (see

Supporting Information, Appendix S1). Additional quanti-

ties of DCE were spiked into the tanks during groundwater

sampling in order to increase DCE average concentrations to

6.5 and 1.5 mg L�1 for cis- and trans-DCE, respectively. The

system was operated in a flow-through mode with a rate of

1.8 L day�1 contaminated groundwater, corresponding to a

retention time of approximately 15 days (for more details

regarding the system design, refer to Imfeld et al., 2008a).

The system was equipped with a cooling system maintaining

the tank and the wetland temperature constantly at

11� 2 1C.

The biogeochemical development of the wetland was

followed for a period of 430 days, as described in Imfeld

et al. (2008a). However, pore water sampling for the micro-

bial investigations presented in this study started when

indications of the establishment of anaerobic conditions in

the wetland were observed. Samples were retrieved at days

199, 227, 255, 283 and 311. Water samples were collected

from the tank, the inflow chamber, the pond, and at four

vertical profiles across the sand compartment to conduct

hydrogeochemical and microbial analyses in parallel. Verti-

cal profiles in the sand compartment were taken at 6, 49, 94

and 139 cm from the inflow via the three sampling ports at

20, 32 and 44 cm depths (Fig. 1).

For the microbial analyses, pore water samples were

retrieved from the wetland at days 199, 227, 255, 283 and

311 from the wetland using sterile syringes. The samples

collected at each sampling day were then pooled in two

integrative samples. Each of these integrative pore water

samples consisted of six 20-mL subsamples, which were

separately pooled in sterile vials. The first series of six pore

water subsamples was retrieved from the two vertical

profiles at the inflow side of the sand compartment (6 and

49 cm from the inflow). The second series of six subsamples

was retrieved from the two vertical profiles at the pond side

of the sand compartment (94 and 139 cm). This integrative

FEMS Microbiol Ecol 72 (2010) 74–88 Journal compilation c� 2010 Federation of European Microbiological SocietiesPublished by Blackwell Publishing Ltd. No claim to original German government works

75Microbial communities in a wetland treating dichloroethenes

sampling procedure accounted for the dominant spatial

development of the hydrogeochemical gradients within the

model wetland that occurred along the flow path across the

wetland rather than over depth (see Supporting Informa-

tion, Fig. S1). In parallel, 0.75-L water samples were col-

lected in cleaned and sterile Schott bottles from the tank, the

inflow and the outflow pond for microbial investigations.

The sampled material was immediately cooled to 4 1C to

slow down further transformation processes, and samples

for microbial analyses were filtered within o 3 h after

sampling.

Hydrogeochemical analysis and compound-specific carbon isotope analysis

Quantification of cis-, trans-1,2-dichloroethene, VC and

ethene was performed with a gas chromatograph equipped

with a flame ionization detector (Varian Chrompack CP-

3800, Middelburg, the Netherlands), with detection limits of

50, 30, 5 and 5 mg L�1, respectively, according to Nijenhuis

et al. (2007). Geochemical parameters (pH, redox potential,

Cl�, NH41, PO43�, Fe21, total Fe, O2, SO4

2� and total

sulphide) were determined according to DIN and DIN EN

ISO standards, following the laboratory procedures de-

scribed in Imfeld et al. (2008a). Stable carbon isotope

compositions of DCE were measured using a GC-combus-

tion-isotope ratio MS system (GC-C-IRMS) (see Nijenhuis

et al., 2007; Imfeld et al., 2008b). The carbon isotope ratio

for an individual compound is reported in d-notation (%)

relative to the Vienna Pee Dee Belemnite standard (V-PDB,

IAEA-Vienna) (Coplen et al., 2006). The analytical error

is � 0.5 d unit and incorporates both the accuracy and the

reproducibility on at least three replicate measurements of

the sample.

Molecular analysis

DNA extraction

In order to avoid perturbation of the system, only water

samples have been retrieved during the course of the

experiment. Water samples were filtered through a sterile

0.2 mm membrane (MoBio Water DNA kit, Carlsbad, CA).

The membranes were stored at � 20 1C until extraction.

DNA was extracted by disrupting microorganisms with a

bead beater (Fast Prep System, Qbiogene, Irvine, CA),

applying a FastDNA spin kit for DNA extraction (BIO101,

La Jolla, CA) and elution in 50 mL nuclease-free water.

16S rRNA gene-targeted PCR

PCR was used to amplify part of the 16S rRNA genes from

Bacteria and Archaea. The PCR mix per reaction contained

1� PCR buffer (with 1.5 mM MgCl2) (Qiagen, Hilden,

(a)

50

600

530

480

505

Inflow Outflow

Pond

(b)

590450 450490

1390940

1980

60

(mm)

Fig. 1. Scheme of the model-constructed

wetland (a) and location of the pore water

sampling devices (b). The model horizontal

subsurface flow wetland was filled with

quartz sand (average depth = 54 cm;

kfaverage = 2.27� 0.14�10�4 m s�1;

grain size = 0.40–0.63 mm) and planted

with common rush (Juncus effusus, L.). A

50-cm-long water pond remained in direct

contact with the atmosphere. Groundwater was

continuously pumped from the 50 L tank at a

constant flow rate (0.4 mL min�1). Supplied

groundwater (50 L tank) was maintained under

anaerobic conditions at a constant N2 pressure

(0.5 mbar). (b) Pore water sampling devices (�)

were located at 6, 49, 94 and 139 cm from the

inflow. At these distances, the sampling ports

were mounted at 20, 32 and 44 cm depths from

the surface.


76 G. Imfeld et al.

Germany), 0.2 mM (each) dNTP (Qiagen), 0.5 mM (each)

forward and reverse primer (Invitrogen), 1.5 U of HotStar-

Taq DNA polymerase (Qiagen), 1 : 10 v : v of DNA template

and molecular-grade water (Promega, Madison, WI). Eu-

bacterial primers 27f (Lane, 1991) and 1378r (Heuer et al.,

1997) were used to amplify almost the complete 16S rRNA

gene using the following amplification program: 95 1C

(15 min), followed by 30 cycles of 95 1C (30 s), 51 1C (30 s)

and 72 1C (50 s), completed with an additional 30 min at

72 1C. The second-round PCR for denaturing gradient gel

electrophoresis (DGGE) analysis used universal primers

GC968f (Nubel et al., 1996) and 1378r. The conditions for

PCR amplifications were as follows: 95 1C (15 min), fol-

lowed by 30 cycles of 95 1C (30 s), 60 1C (1 min), 60–55 1C

(0.5 1C min�1), 72 1C (1 min) and a final extension for

30 min at 72 1C. In parallel, to test the presence of the genera

Dehalobacter (Schlotelburg et al., 2002) and Geobacter

(Duhamel & Edwards, 2006) in the model system, a second

round of PCR with specific primers was performed using the

PCR products from the universal 16S rRNA gene amplifica-

tion as a template (Imfeld et al., 2008b). A Dehalococcoides-

specific amplification protocol according to Imfeld et al.

(2008b) was used for detecting Dehalococcoides-affiliated

bacteria, using specific primers Fp DHC 1 and Rp DHC

1377 for the first round of PCR, and the second PCR was

carried out with a nested Dehalococcoides-specific primer set

(Fp DHC 774 and Rp DHC 1212) (Hendrickson et al.,

2002). The specificity of the Dehalobacter- and Dehalococ-

coides-specific primer sets was previously tested in a pre-

vious study using DNA samples from the same groundwater

that was used for this experiment (Nijenhuis et al., 2007).

PCR for amplification of the 16S rRNA genes of Archaea

used Archaea-specific primers ARC21f and ARC958r ac-

cording to Bano et al. (2004). The second-round PCR for

DGGE used primers GC-ARC344f and UNIV517r under the

conditions described previously (Bano et al., 2004).

Cloning sequencing

Two samples were selected for setting up 16S rRNA gene

libraries. The first sample consisted of groundwater supplied

to the wetland between days 200 and 315. The second

sample consisted of pore water retrieved at day 227 from

the sand compartment at the vertical profiles located at 6

and 49 cm from the inflow. The latter sample was selected in

order to assess the bacterial composition associated with the

developing front of iron sulphide mineral precipitates in the

wetland (Fig. S2). PCR products obtained with primers 27f

and 1378r were ligated into a pGEM-TEasyTM plasmid

(Promega), and were transformed into competent Escher-

ichia coli JM109 cells. Following plasmid extraction and

amplification using vector-specific M13f and M13r primers

(Stratagen, La Jolla, CA), the PCR products of clones were

separately digested with 1 U of the tetrameric site restriction

endonuclease Hin6I and BsuRI for amplified rDNA restric-

tion analysis (Fermentas, St. Leon-Rot, Germany). Clones

with the same pattern with both enzymes were grouped into

an operating taxonomic unit (OTU) (Massol-Deya et al.,

1995). Before sequencing, the 16S rRNA genes of each OTU

were purified using an ABI PRISM BigDye Terminator Cycle

Sequencing Kit V. 3.0 (Applied Biosystems, Foster City, CA)

and sequenced using an ABI PRISM 3100 DNA analyzer

(Applied Biosystems). Putative chimeric sequences (about

10% of the sequences) were discarded from the dataset after

detection analysis performed using the Chimera Detection

Program of the RDP-II (Cole et al., 2009). Examination of

phylogenetic relationships and taxonomic assignments at a

confidence level of 50% was performed using the naıve

Bayesian rRNA Classifier and the Sequence Match tools of

the RDP-II (release 10, http://rdp.cme.msu.edu/), respec-

tively (Wang et al., 2007; Cole et al., 2009). Bacterial

nucleotide sequence data are available in the European

Molecular Biology Laboratory database under accession

numbers FM205929–FM206115.

Molecular fingerprinting analysis ofmicrobial communities

Amplicons obtained from the amplification of bacterial or

archaeal 16S rRNA genes were separated using the DCode

Universal Mutation Detection System (BioRad, Munich,

Germany) DGGE system (for protocol details see Imfeld

et al., 2008b). The DGGE fingerprints were normalized

according to the reference patterns and compared with the

GELCOMPARII software (Applied Maths, Kortrijk, Belgium).

DGGE banding patterns were then converted into a binary

presence–absence matrix subsequently used for further

statistical analysis. Distinct bands of 16S rRNA gene-DGGE

gels for Archaea were excised from the gels and placed in

30-mL sterile nanopure water overnight to elute DNA. Direct

sequencing of the bands was subsequently performed as

described in Cloning sequencing. Archaeal sequence data of

DGGE bands are available under accession numbers

FM211566–FM211590.

Data analysis

Statistical data analyses were performed using the R software

ver. 2.7.2 (R Development Core Team, 2008) with additional

package ‘VEGAN’ (Oksanen et al., 2008) for multivariate

analyses. Principal component analysis (PCA) was applied

to assess the spatio-temporal dynamics of the hydrochemi-

cal variables (pH, Eh, Cl�, NH41, PO43�, Fe21, O2, SO4

2�,

total sulphide, cis- and trans-DCE concentrations and iso-

topic composition). The PCA was based on the Euclidian

correlation matrix of data obtained at days 227, 255, 283 and



http://rdp.cme.msu.edu/




311. Correspondingly, a nonmetric multidimensional scal-

ing (nMDS) was used to evaluate the changes in bacterial

community structures from day 227 to 311. Before the

analyses, the binary numerical matrix of DGGE bands was

converted into a Euclidean dissimilarity matrix. This con-

version was chosen as the most appropriate for the data

according to the highest rank-order dissimilarity test with

gradient separation after a comparison of indices (Oksanen

et al., 2008). The relationship between the community

profiles and the hydrogeochemical variables was interpreted

by fitting the environmental vectors a posteriori onto the

nMDS. Hydrochemical vectors were fitted onto the com-

munity matrix by maximizing their correlations with the

ordination configuration. In the final plot, the vectors, each

representing a hydrochemical variable, point towards the

direction that corresponds to the largest change of the

variable value. Hence, hydrochemical variables significantly

associated with the observed changes in the bacterial com-

munity structures can be revealed. The significance of the

fitted vectors was assessed by conducting a Monte-Carlo

permutation test with 1000 permutation steps. Only hydro-

chemical variables with a significance level of Po 0.05 were

considered significant.

Results

Hydrochemistry

Detailed hydrogeochemical data are provided separately for

each sampling date in Tables S1–S8. The main geochemical

changes in the wetland occurring from days 199 to 430 are

characterized by a progressive decrease of dissolved oxygen

concentrations (0–3.63 mg L�1) accompanied by an increase

of dissolved ferrous iron (from 0.01 to 4.18 mg L�1) and

sulphide (from o 5 to 1573 mM) concentrations (Fig. 2).

The concentration values of VC and ethene increased in the

wetland between days 225 and 430 from o 5 to 518� 13

and o 5 to 102� 5 mg L�1, respectively. This indicates the

occurrence of ongoing reductive dechlorination in the wet-

land. The concentrations of VC and ethene increased and

coincided with significant and increasing isotopic enrich-

ment of both trans- and cis-DCE (Dd13C4 0.5%) between

the inflow and the end of the sand compartment ongoing

from day 225 (Fig. 2c and d). This suggests that DCE

oxidation, associated with a lower isotope fractionation,

was progressively replaced over time by reductive dechlori-

nation, associated with larger isotope fractionation (Imfeld

et al., 2008a). Thus, the progressive development of anoxic

conditions in the wetland reflected a shift of the prevailing

DCE degradation process.

PCA of hydrogeochemical variables (pH, Eh, Cl�, NH41,

PO43�, Fe21, O2, SO4

2�, total sulphide, cis- and trans-DCE

concentrations and isotopic composition) underscored the

major hydrogeochemical trends in the wetland from days

227 to 311 (Fig. 3). This period corresponds to the period of

microbial community structure characterization. Hydroche-

mical profiles of the tank and the inflow clustered together

over time (Fig. 3a). This emphasizes that the quality of the

supplied water did not significantly change over the investi-

gation period. In contrast, hydrochemical profiles from the

0

2

4

6

8

10

12

0

500

1000

1500

2000

0

5

10

15

20

25

30

35

199 227 255 283 311 340 396 430 199 227 255 283 311 340 396 430

Δδ13

C (

‰)

Δδ13

C (

‰)

Fe(

II) (

mg

L–1)

Sul

phid

e (μ

M)

Time (day) Time (day)

(a)

(b)

(c)

(d)

Inflow 6 and 49 cm 94 and 139 cm Outflow

0

2

4

6

8

10

199 227 255 283 311 340 396 430199 227 255 283 311 340 396 430

Fig. 2. Iron(II) (a) and sulphide (b) concentrations

and the mean carbon isotopic signature of

trans- (c) and cis- (d) 1,2-dichloroethenes at the

inflow, the sand compartment (vertical profiles at

6, 49 and 94, 139 cm from the inflow) and the

outflow in the model wetland between days 199

and 430. Error bars for iron(II) and sulphide mean

concentration values indicate the SDs of the six

concentration values retrieved at the sampling

ports from the sand compartment (at 6, 49 and

94, 139 cm from the inflow, respectively). The

dashed lines indicate the time period of microbial

community structure analysis.


78 G. Imfeld et al.

tank and the inflow could be distinguished from the profiles

of the sand compartment, and differed from those of the

effluent collecting pond (Fig. 3a). The larger the distance

from the inflow at which a sample was collected, the higher

the variations of the hydrochemical profiles from one

sampling date to another. This indicates that prominent

hydrochemical changes occurred on the flow path across the

wetland. On the variables plot (Fig. 3b), scores of PC1

correlated positively to sulphide, phosphorus, cis- and

trans-DCE carbon isotope composition shifts (Dd13C), and

ferrous iron, as well as with time. Hence, these variables

largely changed over the investigation period in the wetland.

Until day 255, the concentrations of ferrous iron re-

mained generally one order of magnitude lower than those

of total iron (Fig. 2 and Table S1, S2). This indicates the

prevalence of the ferric form in the model wetland. Mobili-

zation of ferrous iron was revealed by concentration values

increasing with distance from the inflow between days 283

and 311 (Fig. 2a). Mobilization of ferrous iron also corre-

sponded to larger phosphate concentration values from day

255 (generally 4 0.1 mg L�1). Indeed, under reducing con-

ditions, the progressive reduction of Fe(III) may result in the

release of phosphate ions into pore water derived from

phosphorus bound to Fe(III)-hydroxides, as observed pre-

viously (Gosselink & Turner, 1978; Buffle, 1988; Kalbitz

et al., 2000). PCA also underlined that a stepwise increase

of sulphide concentrations throughout the investigation

period was inversely correlated to a decline with sulphate

concentrations over the flow path (Fig. 3b). This consis-

tently indicated that sulphate reduction occurred over time.

From day 311, large concentrations of ferrous iron and

sulphide were simultaneously detected. From day 199,

apparent precipitation of both species in the form of iron

sulphide mineral started, and a visible front of black mineral

precipitates, very likely iron sulphide, was formed within the

sand compartment (Fig. S2). A progression of the front of

iron sulphide minerals occurred horizontally across the

model wetland, which indicated the progressive prevalence

of an anoxic milieu.

Analysis of the bacterial community structures

DGGE of PCR-amplified partial 16S rRNA genes was

applied to assess changes in bacterial community structures

during the described transition of hydrochemical condi-

tions. DGGE patterns from wetland samples collected be-

tween days 227 and 311 provided complex and

heterogeneous microbial fingerprints (Fig. S3). To analyse

changes in bacterial community structures, a nonmetric

multidimentional scaling of 16S rRNA gene-DGGE patterns

was carried out (Fig. 4). In the nMDS plot, the distance

between the points reflects the degree of similarity of the

DGGE profiles. Hence, samples displaying a similar com-

munity structure are found close to each other in the plot.

The 16S rRNA gene-DGGE fingerprints of the groundwater

samples retrieved from the tank for all time points in the

(a) (b)

–4 –2 0 2 4

–4

–2

0

2

4

PC1 (33.4%)

PC

1 (2

6.7%

)

–0.4 –0.2 0 0.2 0.4–0.4

–0.2

0

0.2

0.4

Fig. 3. PCA ordination plot of (a) hydrogeochemical characteristics of water samples and of (b) hydrogeochemical variables collected in the model

wetland between days 227 and 311. (a) Values on the axes indicate % of the total variation explanation by the corresponding axis (PC 1, principal

component axis 1; PC 2, principal component axis 2). The first and second principal components accounted for 60.1% of the variance in the data set.

Objects are labelled according to the section of the wetland they were collected from (m, tank; ’, inflow; black and white B, sand compartment at

6–49 and 94–139 cm from the inflow, respectively; �, pond) and numbered according to the sampling date (days 227, 255, 283 and 311). (b)

Description vectors correspond to: cis-DCE, 1,2-cis-dichloroethene; trans-DCE, trans-1,2-dichloroethene; d13Ctrans-DCE, trans-DCE isotopic composition;

PO43�, P-phosphate; Fe21, ferrous iron; d13Ccis-DCE; Cl�, chloride; O2, oxygen; SO4

2�, sulphate; Eh, redox potential; and NH41, ammonium.



observed period were found to be close together. This

emphasizes the similarity of community structures in the

supplied groundwater throughout the investigation period.

However, the analysis revealed that the bacterial community

structure changed over time at the inflow, the sand com-

partment and the pond. The sand compartment and pond

samples from day 227 to 255 were separated from samples

from days 283 to 311 (Fig. 4). This suggests that relatively

large changes in the bacterial community structure occurred

during the first investigation period (days 227–253),

whereas later only smaller changes were observed.

The relationship between the bacterial community dy-

namics and the observed hydrochemical development in the

wetland was inferred statistically. The resultant characteris-

tic variables such as time, phosphate and sulphide vectors

positively correlated with changes in the microbial commu-

nity structures, whereas the sulphate vector correlated

negatively (Po 0.05) (Fig. 4). Thus, the observed changes

in the microbial community structures over time statistically

correlated with sulphate reduction in the wetland. The

initial changes in the bacterial community structures corre-

sponded to the mobilization of both ferrous iron and

sulphide within the system, and thus reflected major varia-

tions of the prevailing redox conditions.

Detection of potential reductive dechlorinatingbacteria

An indication of the presence of putative reductive dechlori-

nating bacteria in the wetland was provided by taxon-

specific assays. The presence of Dehalobacter, Dehalococcoides

and Geobacter DNA at the inflow, the sand compartment

and the pond of the model wetland was tested from day 199

to 430 (Table 1). Geobacter spp. DNA was detected in all

samples retrieved from the model system throughout the

investigation period. The genus Geobacter sp. contains

metabolically versatile iron-reducers, including members

capable of dechlorination of chlorinated ethenes during

iron reduction (Sung et al., 2006). Dehalobacter spp. was

observed from day 311 and coincided with the presence of

Dehalococcoides spp. DNA. DNA of all the targeted genera

could be detected across the system at day 430, suggesting

the existence of a consortium of reductive dechlorinating

bacteria. The presence of Dehalococcoides-like bacteria was

detected from day 253 in the sand compartment; this

presence correlated with both the detection of VC and

ethene as well as a larger shift in the carbon isotope

composition of both cis- and trans-DCE (Table 1; Fig. 2c

and d).

Sequence analysis of the clone libraries

Two 16S rRNA gene clone libraries were constructed to gain

an insight into the composition of bacterial communities of

the supplied groundwater and the model wetland at day 227

(Table 2). Details on the phylogenetic affiliation and se-

quence similarities of retrieved 16S rRNA gene clones are

provided in Fig. 5. A total of 138 and 165 clones were

obtained and restriction pattern analysis further refined

these sequences, resulting in, respectively, 89 and 97 OTUs

for the groundwater and wetland samples. Because the

ΣS2–

–15 –10 –5 0 5 10 15

–10

–5

0

5

10

–15

NMDS 1

NM

DS

2TimeTime

ΣS2–

SO42–SO42–

PO34–PO34–

311

255

283

311

283

255

227

311

283

255

227

255

283

311

255

311

283

:6-49 cm :Pond:Tank

Sand compartment

:Inflow :94-139 cm

Fig. 4. nMDS plot (two dimensional) of the 16S rDNA gene DGGE

patterns of the bacterial community from water samples, showing the

community changes (stress value: 18.5%). Objects are labelled according

to location (m, tank; ’, inflow; black and white B, sand compartment

at 6–49 and 94–139 cm from the inflow, respectively; �, pond) and

numbered according to the sampling date (day 227, 255, 283 and 311).

Vector arrows of maximum correlation with the corresponding DGGE

patterns are superimposed depicting the direction, magnitude and

correlation of four most important characterizing components (SO42�,

PO34�, time and SS2�). The significance of fitted vectors is assessed using

permutation of all assessed variables a posteriori by permutation of

variables at Po 0.05.

Table 1. Detection of Dehalococcoides, Dehalobacter and Geobacter

DNA in the model wetland using taxon-specific assays

Taxon-specific assay

Time (days)

225 253 280 309 430

Geobacter spp. a, b, c, d a, b, c, d a, b, c, d a, b, c, d a, b, c, d

Dehalobacter spp. – – – a, b a, b, c, d

Dehalococcoides spp. – a, b a a, b, c a, b, c, d

� , absence of the targeted genera. Letters correspond to the part of the

wetland at which samples were retrieved: a, inflow; b, sand compart-

ment at 6 and 49 cm from the inflow; c, sand compartment at 94 and

139 cm from the inflow; d, pond.


80 G. Imfeld et al.

rarefaction curves did not indicate a clear tendency towards

saturation, the discovery of additional sequences is expected

by increasing the number of investigated clones (Fig. S4). It

is therefore very likely that the current sampling strategy did

not allow detection of minor populations of the bacterial

communities, including those potentially involved in con-

taminant degradation. Consequently, accurate quantitative

estimates of the bacterial groups present in these samples

could not be provided in this study, and relative clone

frequencies mainly apply to dominant populations of the

communities. Clone sequences from the groundwater and

the wetland could be affiliated with, respectively, eight and

11 classes of the domain Bacteria (Table 2). Overall, riboso-

mal sequences affiliated to the Proteobacteria largely domi-

nated, and accounted for about half of both clone libraries.

About 25% of the clones could not be assigned to any

bacterial phylum at the defined confidence level of 50%.

Sequences affiliated to the Beta- and Deltaproteobacteria

were retrieved in higher numbers than the other groups in

both libraries. The majority of sequences within the b-

subgroup in the groundwater were similar to sequences

affiliated to the neutrophilic and chemolithotrophic iron-

oxidizing Gallionella ferruginea, in agreement with the

hydrogeochemistry of the supplied groundwater. Geobacter

sp. accounted for 7% of the wetland clone library. Sequences

affiliated to Geobacter sp. had a close relationship with

sequences retrieved from Fe(III)-reducing enrichment cul-

tures obtained from contaminated sediments (Holmes et al.,

2004; Scala et al., 2006; Kittelmann & Friedrich, 2008).

Sequences affiliated to Verrumicrobia, Chloroflexi and Spiro-

chaetes were exclusively detected in the constructed wetland.

Within the phylum Chloroflexi, one clone sequence was

affiliated to the Levilinea genus. The presence of non-

Dehalococcoides Chloroflexi populations also raises the even-

tuality of their involvement in reductive dechlorination

(Watts et al., 2005). Recently, the existence of putative

reductive dechlorinating phylotypes within Chloroflexi that

includes uncultured microorganisms was revealed (Fager-

vold et al., 2005). While Dehalococcoides is so far the only

genus containing only members dependent on reductive

dehalogenation as an energy-gaining process, the involve-

ment of Chloroflexi members in the degradation process

cannot be excluded. In contrast, no Dehaloccocoides-

affiliated sequences could be retrieved, which underscores

the higher sensitivities of the taxon-specific PCR assay

(Hendrickson et al., 2002) over the cloning-sequencing

procedure. Lack of Dehaloccoides-like sequences in the clone

library, despite the obvious reductive dehalogenation activ-

ity in the wetland, could be due to the relatively high

degradation capacity of Dehalococcoides compared with the

slow-growing nature of these organisms (with a doubling

time of approximately 1 day) (He et al., 2005; Tang et al.,

2009). An additional explanation is that the Dehalococcoides

strains described so far have only one copy of the ribosomal

genes, while the predominant Proteobacteria have many

more (up to 13, 4.17 in average), leading to lower sensitivity

for their detection (Lee et al., 2009). Noticeably, sequences

related to the TM7 candidate division accounted for about

9% of the groundwater clone library, but could not be

retrieved from the wetland. Some members of this division

were suggested to be associated with TCE and cis-DCE

cometabolic oxidation (Lowe et al., 2002; Connon et al.,

2005).

To gain insights into Archaea diversity, the community

composition of the Archaea was assessed at day 255, when

methane production in the wetland was observed (Imfeld

et al., 2008a). This assessment also aimed at identifying

methanogen candidates in the wetland. High archaeal

diversity (15–19 bands) with variation between the segments

of the wetland was observed, suggesting changes in the micro-

bial community composition over the flow path (Fig. S5).

Bands with a high intensity within the DGGE pattern were

excised from the gel to identify predominant Archaea. The

detailed outcome of sequence analysis of selected bands is

Table 2. Relative clone frequencies in major phylogenetic groups of the

clone libraries from the original groundwater (supplied to the wetland

between days 200 and 310) and the wetland pore water (sampling day 227)

Group

Relative frequency (%)

Groundwater Wetland

Proteobacteria 57 34

Alphaproteobacteria 2 7

Magnetospirillum spp. 0 2

Betaproteobacteria 35 16

Gallionella spp. 17 1

Deltaproteobacteria 10 7

Geobacter spp. 0 7

Desulfovibrio spp. 1 15

Epsilonproteobacteria 2 2

Gammaproteobacteria 7 2

Dokdonella 0 1

Thiothrix spp. 1 1

Unclassified Proteobacteria 1 0

Firmicute

‘Clostridia’ Clostridiales 2 4

TM7 Division 9 0

Bacteroidetes 2 4

Lentisphaerae 1 5

OD1 division 3 2

Actinobacteria 1 2

Verrucomicrobia 0 2

Cloroflexi 1 2

Acidobacteria 1 1

Spirochaetes 0 1

Unclassified Bacteria 24 18

Number of clones 138 165

Number of OTUs 89 97



(a) (b)

Candidatedivision OD1

Fig. 5. Neighbour-joining tree of 16S rRNA gene sequences depicting the relationships among community members of the (a) groundwater and (b)

constructed wetland as revealed by comparative analysis of 16S rRNA gene sequences and those stored in the ARB database and GenBank. 16S rRNA

gene bacterial sequences determined in this study are highlighted in grey and were deposited in GenBank under accession numbers

FM205929–FM206115. Bootstrap values of 4 50% are indicated at the node of the branch. Scale bar represents a 10% estimated change.


82 G. Imfeld et al.

provided in Fig. 6. Comparative analyses of sequences

retrieved from GenBank revealed that all 11 examined

sequences proved to belong to Euryarchaeota, including

eight sequence types affiliated to hydrogenotrophic metha-

nogenic genera (Methanomicrobiales, Methanococcales and

Methanopyrales). Two sequences were related to the Archaeo-

globi class and one sequence was identified to be a member

of the Halobacteria class.

Discussion

In this study, we assessed the biogeochemical development

of a wetland treating DCE-contaminated groundwater over

time, with a particular emphasis on the microbial character-

ization. Microbial and hydrochemical investigations were

combined to characterize in parallel the wetland microbial

community and ambient hydrochemistry in the wetland.

Microbial analyses performed in this study were exclusively

based on liquid-phase samples, allowing a temporal and

spatial assessment of the microbial community with mini-

mal disturbance of the system during sampling. Owing to

the dimensions of the wetland systems, parallel sediment

collections would have resulted in a significant disturbance

of ongoing biogeochemical changes over space and time.

However, this liquid sampling design very likely results in an

incomplete view of the wetland microbiology as the liquid

phase is generally less heterogeneously distributed than the

solid material, and specific populations or activities may be

asymmetrically distributed between the pore water and the

sediment phases (Lehman & O’Connell, 2002; Lehman,

2007). Nevertheless, changes of the hydrochemistry in the

wetland could be identified and corresponded to a temporal

shift in the prevailing pathway of DCE degradation, from

degradation reactions under oxic conditions, to reductive

dechlorination under anoxic conditions. These changes

reflected variations of both the bacterial community struc-

tures and the composition of the guild of putative dechlori-

nating bacteria.

Microbial respiration coupled to degradation of organic

carbon and hydrogen consumption by various groups of

reducing bacteria is assumed to be a dominant degradation

mechanism in various anaerobic soils and sediments (Hol-

mer & Storkholm, 2001). The high availability of electron

donors due to decaying plant material was evidenced by

high DOC concentrations in the wetland (Imfeld et al.,

2008a), which sustained the coexistence of iron and sulphate

reduction, in agreement with previous observations in

freshwater wetlands (Alewell et al., 2004). In the wetland

Fig. 6. Neighbour-joining tree of 16S rRNA gene sequences depicting the phylogenic relationship of 16S rRNA gene archaeal sequences from excised

DGGE bands to other related organisms. 16S rRNA gene archaeal sequences determined in this study are highlighted in grey and are designated

according to their origin in the wetland (tank, inflow, sand compartment or outflow) (GenBank accession numbers in parentheses). Bootstrap values of

4 50% are indicated at the node of the branch. Scale bar represents 5% estimated change.



system, the spatial and temporal development of zones of

iron sulphide precipitation was likely sustained by (1) the

maintenance of reducing conditions created by continuous

supply of anoxic groundwater, (2) the constant supply of

nonreduced precursors [Fe(III) and sulphate] in the sup-

plied groundwater and (3) the high availability of electron

donors from decaying plant organic matter within the

wetland. These spatial and temporal changes in the wetland

hydrochemical properties likely shaped the ecological habi-

tats within the porous sand compartment. For instance, the

accumulation of decaying plant organic material and the

visible deposits of iron sulphide mineral in the wetland

likely resulted in the reduction of hydraulic conductivity and

void space of the porous medium, oxygen supply and

increased dispersion, as described previously (Tanner &

Sukias, 1995; Garcia et al., 2004). Changes in both the

hydrochemistry and the hydrodynamic properties of the

sand compartment can in turn affect both the structures and

the diversity of the microbial community by changing fluxes

and transport patterns of organic substrates and nutrients.

Although substantial variations in the bacterial community

structures were observed throughout the investigation per-

iod, the contaminant biodegradation capacity was main-

tained. Nevertheless, changes in the microbial community

structures coincided with a shift in the DCE carbon isotope

composition, which indicated changes in the DCE degrada-

tion pathway.

Although changes in the microbial community structures

coincided with a shift in the DCE degradation pathway as

suggested by compound-specific isotope analysis, degrading

bacterial populations were also expected to change. Detec-

tion of dechlorinating bacteria provided an additional line

of evidence that reductive dechlorination represented a key

mechanism for contaminant removal in the wetland. As

suggested by the isotopic composition analysis, reductive

dechlorinating activity likely has become more important

over time, until it completely overcame oxidative degrada-

tion. Decreasing availability of electron acceptors for aerobic

(i.e. O2) or anaerobic oxidation (i.e. Fe31) may have

progressively limited oxidative degradation of DCE or VC

during the course of the investigation period (Bradley &

Chapelle, 1998). Although no quantitative approach was

used in the study, Geobacter sp. likely dominated at the

beginning of the transition phases, when iron reduction

occurred. However, the involvement of this genus in reduc-

tive dechlorination in the wetland cannot be proven with the

presented data. The stepwise detection of members of the

reductive dechlorinating guild over time also reflected

hydrogeochemical changes in the wetland. Under anoxic

conditions, favourable conditions for microbial reductive

dechlorination of DCE were progressively achieved within

the wetland. Hence, the detection of ethene along with the

simultaneous presence of putative reductive dechlorinating

bacteria under strongly reducing conditions emphasized

that reductive dechlorination of DCE prevails at the end of

the investigation period. Although cultivated members of

the genera Geobacter and Dehalobacter are capable of partial

dechlorination of PCE and TCE to cis-DCE, Dehalococcoides

is the only group known so far to be capable of complete

dechlorination of PCE/TCE to ethene (Maymo-Gatell et al.,

1997; Seshadri et al., 2005). In the case of a hypothetical

change of the prevailing groundwater contaminants, the

simultaneous presence of these reductive dechlorinating

bacteria also suggests that the wetland system could sustain

reductive dechlorination of PCE or TCE up to ethene.

The 16S rRNA gene clone library indicated that the model

wetland sustained diverse microbial populations capable of

various terminal electron-accepting processes. A large pro-

portion of recovered microorganisms were putatively driv-

ing iron and sulphate reduction, in agreement with the

hydrochemical observations. Interactions among the var-

ious populations of the wetland system can result in changes

of the hydrochemical conditions and may affect members of

the reductive dechlorinating guild. This in turn can influ-

ence the wetland DCE-degrading capacity. For instance,

sulphate-reducing and reductive dechlorinating micro-

organisms are able to thrive at similar H2 levels (Heimann

et al., 2005). Sulphate-reducing bacteria were relevant in the

wetland when the transition from an oxic to a prevailing

anoxic regime started, and possibly compete for an electron

donor or its fermentation products, thus determining the

rate and extent of DCE dechlorination (Aulenta et al., 2002;

Heimann et al., 2007). However, reductive dechlorinating

activity was observed concurrently with sulphate reduction,

and even increased progressively throughout the investiga-

tion period. This suggests that dechlorinators could com-

pete for electron usage, in agreement with previous

observations (Aulenta et al., 2007). Although considered as

ubiquitous in anaerobic systems, cometabolic dechlorinat-

ing processes (presumably fortuitous) are generally incap-

able of mediating complete reduction to ethene (McCarty &

Semperini, 1994). Nevertheless, if the present reductive

dechlorinating bacteria do not display enzymatic capabil-

ities to use iron or sulphate, competitive exclusion may

influence organisms involved in the cometabolic dechlor-

ination of cis-DCE. These include methanogens, present in

the wetland, and acetogens, both of which are highly

influenced by H2 consumption, as well as Archaea. Although

no Archaea have been identified in dechlorination processes,

their direct or indirect role in dechorination is not well

known. Examples of the syntrophic association of dechlor-

inating bacteria within a methanogenic microbial consor-

tium have been reported. Species also capable of

cometabolic dechlorination include Methanosarcina thermo-

phila, Methanosarcina mazei, which can dechlorinate PCE to

TCE, and Methanobacterium thermoautotrophicum and


84 G. Imfeld et al.

Methanothrix soehngenii, both capable of reductively

dechlorinating cis-DCE to chloroethane (Fantroussi et al.,

1998; Holliger et al., 2003).

Though the DCE degradation mechanism varied over

time, the biodegrading function of the model wetland was

maintained. Although coupling reduction and oxidation

processes in wetland systems to reach an efficient transfor-

mation of some chlorinated hydrocarbons and their trans-

formation products may be of interest, possible changes in

the degradation mechanism over the lifespan of a wetland

system need to be carefully considered. Further studies

integrating hydrochemical and microbial approaches are

required to understand how biodegradation mechanisms

and processes influence the transformation of chlorinated

hydrocarbons over the lifespan of wetland systems.

Acknowledgements

G.I. and E.M. were supported by a European Union Marie

Curie Early Stage Training Fellowship (AXIOM, contract no.

MEST-CT-2004-8332). C.E.A. was supported by a Leonardo

Da Vinci exchange grant. We thank the Analytical Chemistry

Department of the UFZ for measurements of the anorganic

parameters and anonymous reviewers for critical review and

helpful comments. This work was supported by the Helm-

holtz Centre for Environmental Research – UFZ in the scope

of the SAFIRA II Research Programme (Revitalization of

Contaminated Land and Groundwater at Megasites, project

‘Compartment Transfer’).

References

Alewell C, Lischeid G, Hell U & Manderscheid B (2004) High

temporal resolution of ion fluxes in semi-natural ecosystems –

gain of information or waste of resources? Biogeochemistry 69:

19–35.

Amon JP, Agrawal A, Shelley ML et al. (2007) Development of a

wetland constructed for the treatment of groundwater

contaminated by chlorinated ethenes. Ecol Eng 30: 51–66.

Armenante PM, Kafkewitz D, Lewandowski G & Kung CM

(1992) Integrated anaerobic–aerobic process for the

biodegradation of chlorinated aromatic-compounds. Environ

Prog 11: 113–122.

Aulenta F, Majone M, Verbo P & Tandoi V (2002) Complete

dechlorination of tetrachloroethene to ethene in presence of

methanogenesis and acetogenesis by an anaerobic sediment

microcosm. Biodegradation 13: 411–424.

Aulenta F, Pera A, Rossetti S, Papini MP & Majone M (2007)

Relevance of side reactions in anaerobic reductive

dechlorination microcosms amended with different electron

donors. Water Res 41: 27–38.

Bano N, Ruffin S, Ransom B & Hollibaugh JT (2004)

Phylogenetic composition of Arctic Ocean archaeal

assemblages and comparison with antarctic assemblages.

Appl Environ Microb 70: 781–789.

Bradley PM & Chapelle FH (1998) Microbial mineralization of

VC and DCE under different terminal electron accepting

conditions. Anaerobe 4: 81–87.

Bradley PM & Chapelle FH (2000) Aerobic microbial

mineralization of dichloroethene as sole carbon substrate.

Environ Sci Technol 34: 221–223.

Braeckevelt M, Rokadia H, Imfeld G et al. (2007) Assessment of

in situ biodegradation of monochlorobenzene in

contaminated groundwater treated in a constructed wetland.

Environmental Pollut 148: 428–437.

Buffle J (1988) Complexation properties of homologous

comlexants and choice of measuring methods. Complexation

Reactions in Aquatic Systems: An Analytical Approach

(Chalmers RA & Masson M, eds), pp. 304–383. Ellis Horwood

Series in Analytical Chemistry, Chichester, UK.

Cole JR, Wang Q, Cardenas E et al. (2009) The ribosomal

database project: improved alignments and new tools for

rRNA analysis. Nucleic Acids Res 37: D141–D145 DOI:

10.1093/nar/gkn879.

Coleman NV, Mattes TE, Gossett JM & Spain JC (2002)

Biodegradation of cis-dichloroethene as the sole carbon source

by a beta-proteobacterium. Appl Environ Microb 68: 2726–2730.

Connon SA, Tovanabootr A, Dolan M, Vergin K, Giovannoni SJ

& Semprini L (2005) Bacterial community composition

determined by culture-independent and -dependent methods

during propane-stimulated bioremediation in

trichloroethene-contaminated groundwater. Environ Microbiol

7: 165–178.

Coplen TB, Brand WA, Gehre M, Groning M, Meijer HAJ, Toman

B & Verkouteren RM (2006) New guidelines for delta C-13

measurements. Anal Chem 78: 2439–2441.

Duhamel M & Edwards EA (2006) Microbial composition of

chlorinated ethene-degrading cultures dominated by

Dehalococcoides. FEMS Microbiol Ecol 58: 538–549.

Duhamel M, Mo K & Edwards EA (2004) Characterization of a

highly enriched Dehalococcoides-containing culture that grows

on vinyl chloride and trichloroethene. Appl Environ Microb 70:

5538–5545.

Fagervold SK, Watts JEM, May HD & Sowers KR (2005)

Sequential reductive dechlorination of meta-chlorinated

polychlorinated biphenyl congeners in sediment microcosms

by two different Chloroflexi phylotypes. Appl Environ Microb

71: 8085–8090.

Fantroussi SE, Naveau H & Agathos SN (1998) Anaerobic

dechlorinating bacteria. Biotechnol Progr 14: 167–188.

Garcia J, Chiva J, Aguirre P, Alvarez E, Sierra JP & Mujeriego R

(2004) Hydraulic behaviour of horizontal subsurface flow

constructed wetlands with different aspect ratio and granular

medium size. Ecol Eng 23: 177–187.

Gosselink JG & Turner RE (1978) The role of hydrology in

freshwater wetland ecosystems. Freshwater Wetlands, Ecological

Processes and Management Potential (Good RE, Whigham DF

& Simpson RL, eds), pp. 63–78. Academic Press, New York.



DOI: 10.1093/nar/gkn879

DOI: 10.1093/nar/gkn879

Hata J, Miyata N, Kim ES, Takamizawa K & Iwahori K (2004)

Anaerobic degradation of cis-1,2-dichloroethylene and vinyl

chloride by Clostridium sp strain DC1 isolated from landfill

leachate sediment. J Biosci Bioeng 97: 196–201.

He J, Sung Y, Krajmalnik-Brown R, Ritalahti KM & Loffler FE

(2005) Isolation and characterization of Dehalococcoides sp.

strain FL2, a trichloroethene (TCE)- and 1,2-dichloroethene-

respiring anaerobe. Environ Microbiol 7: 1442–1450.

Heimann AC, Friis AK & Jakobsen R (2005) Effects of sulfate on

anaerobic chloroethene degradation by an enriched culture

under transient and steady-state hydrogen supply. Water Res

39: 3579–3586.

Heimann AC, Friis AK, Scheutz C & Jakobsen R (2007) Dynamics

of reductive TCE dechlorination in two distinct H-2 supply

scenarios and at various temperatures. Biodegradation 18:

167–179.

Hendrickson E, Payne J, Young R et al. (2002) Molecular analysis

of Dehalococcoides 16S ribosomal DNA from chloroethene-

contaminated sites throughout North America and Europe.


Heuer H, Krsek M, Baker P, Smalla K & Wellington EMH (1997)

Analysis of actinomycete communities by specific

amplification of genes encoding 16S rRNA and gel-

electrophoretic separation in denaturing gradients. Appl

Environ Microb 63: 3233–3241.

Holliger C, Regeard C & Dietert G (2003) Dehalogenation by

anaerobic bacteria. Dehalogenation: Microbial Processes and

Environmental Applications (Haggblom MM & Bossert ID,

eds), pp. 115–157. Kluwer Academic Publishers, Boston, MA.

Holmer M & Storkholm P (2001) Sulphate reduction and

sulphur cycling in lake sediments: a review. Freshwater Biol 46:

431–451.

Holmes DE, Nevin KP & Lovley DR (2004) In situ expression of

nifD in Geobacteraceae in subsurface sediments. Appl Environ

Microb 70: 7251–7259.

Ibekwe AM, Lyon SR, Leddy M & Jacobson-Meyers M (2007)

Impact of plant density and microbial composition on water

quality from a free water surface constructed wetland. J Appl

Microbiol 102: 921–936.

Imfeld G, Estop C, Zeiger S et al. (2008a) Tracking in situ

biodegradation of 1,2-dichloroethenes in a model wetland.

Environ Sci Technol 42: 7924–7930.

Imfeld G, Nijenhuis I, Nikolausz M et al. (2008b) Assessment of

in situ degradation of chlorinated ethenes and bacterial

community structure in a complex contaminated groundwater

system. Water Res 42: 871–882.

Kalbitz K, Solinger S, Park JH, Michalzik B & Matzner E (2000)

Controls on the dynamics of dissolved organic matter in soils:

a review. Soil Sci 165: 277–304.

Kassenga GR, Pardue JH, Blair S & Ferraro T (2003) Treatment of

chlorinated volatile organic compounds in upflow wetland

mesocosms. Ecol Eng 19: 305–323.

Keefe SH, Barber LB, Runkel RL & Ryan JN (2004) Fate of volatile

organic compounds in constructed wastewater treatment

wetlands. Environ Sci Technol 38: 2209–2216.

Kim Y & Semprini L (2005) Cometabolic transformation of cis-

1,2-dichloroethyleneand cis-1,2-dichloroethylene epoxide by a

butane-grown mixed culture. Water Sci Technol 52: 125–131.

Kittelmann S & Friedrich MW (2008) Identification of novel

perchloroethene-respiring microorganisms in anoxic river

sediment by RNA-based stable isotope probing. Environ

Microbiol 10: 31–46.

Lane DJ (1991) 16S/23S rRNA sequencing. Nucleic Acid

Techniques in Bacterial Systematics (Stackebrandt E &

Goodfellow M, eds), pp. 115–148. Wiley, New York.

Lee ZMP, Bussema C & Schmidt TM (2009) rrnDB: documenting

the number of rRNA and tRNA genes in bacteria and archaea.

Nucleic Acids Res 37: D489–D493.

Lehman RM (2007) Understanding of aquifer microbiology is

tightly linked to sampling approaches. Geomicrobiol J 24:

331–341.

Lehman RM & O’Connell SP (2002) Comparison of extracellular

enzyme activities and community composition of attached

and free-living bacteria in porous medium columns. Appl

Environ Microb 68: 1569–1575.

Loffler F, Sun Q, Li J & Tiedje J (2000) 16S rRNA gene-based

detection of tetrachloroethene-dechlorinating Desulfuromonas

and Dehalococcoides species. Appl Environ Microb 66:

1369–1374.

Lorah MM & Olsen LD (1999) Degradation of 1,1,2,2-

tetrachloroethane in a freshwater tidal wetland: field and

laboratory evidence. Environ Sci Technol 33: 227–234.

Lowe M, Madsen EL, Schindler K, Smith C, Emrich S, Robb F &

Halden RU (2002) Geochemistry and microbial diversity of a

trichloroethene-contaminated Superfund site undergoing

intrinsic in situ reductive dechlorination. FEMS Microbiol Ecol

40: 123–134.

Massol-Deya AA, Odelson DA, Hickey RF & Tiedje JM (1995)

Bacterial community fingerprinting of amplified 16S and

16–23S ribosomal DNA gene sequences and restriction

endonuclease analysis (ARDRA). Molecular Microbial Ecology

Manual, Vol. 3.3.2 (Akkermans ADL, van Elsas JD & de Ruijn

FJ, eds), pp. 1–8. Kluwer Academic Publishers, Dordrecht, The

Netherlands.

Master ER, Lai VWM, Kuipers B, Cullen WR & Mohn WW

(2002) Sequential anaerobic–aerobic treatment of soil

contaminated with weathered aroclor 1260. Environ Sci

Technol 36: 100–103.

Maymo-Gatell X, Chien YT, Gossett JM & Zinder SH (1997)

Isolation of a bacterium that reductively dechlorinates

tetrachloroethene to ethane. Science 276: 1568–1571.

McCarty P & Semperini L (1994) Groundwater treatment for

chlorinated solvents. Handbook of bioremediation (Norris RD,

Hinchee RE, Brown R, McCarty PL, Semprini L, Wilson DH,

Kampbell M, Reinhard EG, Bouwer R, Borden C, Vogel TM,

Thomas JM, Ward CH, eds), pp. 87–116. Lewis Publisher,

Boca Raton, FL.

Meade T & D’Angelo EM (2005) [C-14]Pentachlorophenol

mineralization in the rice rhizosphere with established

oxidized and reduced soil layers. Chemosphere 61: 48–55.


86 G. Imfeld et al.

Nijenhuis I, Nikolausz M, Koth A et al. (2007) Assessment of the

natural attenuation of chlorinated ethenes in an anaerobic

contaminated aquifer in the Bitterfeld/Wolfen area using stable

isotope techniques, microcosm studies and molecular

biomarkers. Chemosphere 67: 300–311.

Nikolausz M, Chatzinotas A, Palatinszky M, Imfeld G, Martinez P

& Kastner M (2008) Single-nucleotide primer extension assay

for detection and sequence typing of ‘Dehalococcoides’ spp.


Nocker A, Burr M & Camper AK (2007) Genotypic microbial

community profiling: a critical technical review. Microb Ecol

54: 276–289.

Nubel U, Engelen B, Felske A et al. (1996) Sequence

heterogeneities of genes encoding 16S rRNAs in Paenibacillus

polymyxa detected by temperature gradient gel electrophoresis.

J Bacteriol 178: 5636–5643.

Oksanen J, Kindt R, Legendre P, O’Hara RB, Simpson GL,

Solymos P, Stevens MHH & Wagner H (2008) Vegan:

community ecology package version 1.15-1. Available at http://

cran.r-project.org/

Ramette A (2007) Multivariate analyses in microbial ecology.

FEMS Microbiol Ecol 62: 142–160.

R Development Core Team (2008) R: A Language and

Environment for Statistical Computing. R Foundation for

Statistical Computing, Vienna, Austria. http://www.R-project.

org.

Reddy KR & Dangelo EM (1997) Biogeochemical indicators to

evaluate pollutant removal efficiency in constructed wetlands.

Water Sci Technol 35: 1–10.

Scala DJ, Hacherl EL, Cowan R, Young LY & Kosson DS (2006)

Characterization of Fe(III)-reducing enrichment cultures and

isolation of Fe(III)-reducing bacteria from the Savannah River

site, South Carolina. Res Microbiol 157: 772–783.

Schlotelburg C, von Wintzingerode C, Hauck R, von

Wintzingerode F, Hegemann W & Gobel U (2002) Microbial

structure of an anaerobic bioreactor population that

continuously dechlorinates 1,2-dichloropropane. FEMS

Microbiol Ecol 39: 229–237.

Seshadri R, Adrian L, Fouts D et al. (2005) Genome sequence of

the PCE-dechlorinating bacterium Dehalococcoides

ethenogenes. Science 307: 105–108.

Stottmeister U & Wiessner A (2003) Effects of plants and

microorganisms in constructed wetlands for wastewater

treatment. Biotechnol Adv 22: 93–117.

Sung Y, Fletcher KF, Ritalaliti KM et al. (2006) Geobacter lovleyi sp

nov strain SZ, a novel metal-reducing and tetrachloroethene-

dechlorinating bacterium. Appl Environ Microb 72: 2775–2782.

Tang YJJ, Yi S, Zhuang WQ, Zinder SH, Keasling JD & Alvarez-

Cohen L (2009) Investigation of carbon metabolism in

‘Dehalococcoides ethenogenes’ strain 195 by use of isotopomer

and transcriptomic analyses. J Bacteriol 191: 5224–5231.

Tanner CC & Sukias JP (1995) Accumulation of organic solids in

gravel-bed constructed wetlands. Water Sci Technol 32:

229–239.

Truu J, Nurk K, Juhanson J & Mander U (2005) Variation of

microbiological parameters within planted soil filter for

domestic wastewater treatment. J Environ Sci Heal A 40:

1191–1200.

Vogel TM (1994) Natural bioremediation of chlorinated solvents.

Handbook of Bioremediation (Norris RD, Hinchee RE, Brown

R, et al., eds), pp. 201–225. Lewis Publisher, Boca Raton, FL.

Walsh KA, Hill TCJ, Moffett BF, Harris JA, Shaw PJ & Wallace JS

(2002) Molecular characterisation of bacteria in a wetland

used to remove ammoniacal-N from landfill leachate. Waste

Manage Res 20: 529–535.

Wang Q, Garrity GM, Tiedje JM & Cole JR (2007) Naive Bayesian

classifier for rapid assignment of rRNA sequences into the new

bacterial taxonomy. Appl Environ Microb 73: 5261–5267.

Watts JEM, Fagervold SK, May HD & Sowers KR (2005) A PCR-

based specific assay reveals a population of bacteria within the

Chloroflexi associated with the reductive dehalogenation of

polychlorinated biphenyls. Microbiol-SGM 151: 2039–2046.

Weiss JV & Cozzarelli IM (2008) Biodegradation in contaminated

aquifers: incorporating microbial/molecular methods. Ground

Water 46: 305–322.

Supporting Information

Additional Supporting Information may be found in the

online version of this article:

Appendix S1. Additional information and results.

Fig. S1. Prevailing spatial hydrogeochemical gradients de-

velopment in the sand compartment of the model wetland

at days 227, 283 and 311 of the investigation.

Fig. S2. Iron sulphide minerals precipitation in the model

subsurface horizontal-flow model wetland treating cis- and

trans-DCE-contaminated groundwater at day 227.

Fig. S3. Example of the DGGE patterns of Bacteria 16S rRNA

gene PCR products amplified from water-DNA extracts

obtained across the model wetland at the tank, the inflow,

the sand compartment (Sample AB: pooled samples retrieved

at 6 and 49 cm from the inflow; and CD: pooled samples

retrieved at 94 and 139 cm from the inflow) and the pond.

Fig. S4. Rarefaction curve of bacterial 16S rRNA gene clones

recovered from groundwater (m) and from the wetland (�).

Fig. S5. DGGE patterns of Archaea 16S rRNA gene PCR

products amplified from water-DNA extracts obtained

across the model wetland at the tank, the inflow, the sand

compartment (pooled samples retrieved at 6 and 49 cm

from the inflow) and the pond at day 255.

Table S1. Results of the geochemical and contaminant

analyses in the sand compartment of the model constructed

wetland at day 199 of the microbial investigation period.






http://cran.r-project.org/

http://cran.r-project.org/

http://www.R-project.org

http://www.R-project.org



















Please note: Wiley-Blackwell is not responsible for the

content or functionality of any supporting materials sup-

plied by the authors. Any queries (other than missing

material) should be directed to the corresponding author

for the article.


88 G. Imfeld et al.

Documents

Characterization of microbial communities in the aqueous phase of a constructed model wetland treating 1,2-dichloroethene-contaminated groundwater