See

discussions, stats, and author profiles for this publication at: https://www.researchgate.net/publication/273154942

Nitrogen transforming community in a

horizontal subsurface-flow constructed
wetland

Article in Water Research · February 2015

DOI: 10.1016/j.watres.2015.02.018 · Source: PubMed

CITATIONS READS

21 362

7 authors, including:

Oksana Coban Marion Martienssen

NASA Brandenburg University of Technology Cottb…
4 PUBLICATIONS 30 CITATIONS 52 PUBLICATIONS 351 CITATIONS

SEE PROFILE SEE PROFILE

Mike S M Jetten
Radboud University
797 PUBLICATIONS 29,907 CITATIONS

SEE PROFILE

Some of the authors of this publication are also working on these related projects:

Development and testing of a novel photocatalytic system for efficient cogeneration of clean water
and hydrogen for ecosafe agriculture View project

NITROLIMIT - Stickstofflimitation in Binnengewässern; www.nitrolimit.de View project

All content following this page was uploaded by Oksana Coban on 19 March 2015.

The user has requested enhancement of the downloaded file. All in-text references underlined in blue are added to the original document
and are linked to publications on ResearchGate, letting you access and read them immediately.
 w a t e r r e s e a r c h 7 4 ( 2 0 1 5 ) 2 0 3 e2 1 2

Available online at www.sciencedirect.com

ScienceDirect

journal homepage: www.elsevier.com/locate/watres

Nitrogen transforming community in a horizontal

subsurface-flow constructed wetland

Oksana Coban a,*, Peter Kuschk b, Uwe Kappelmeyer b, Oliver Spott c,

Marion Martienssen d, Mike S.M. Jetten e, Kay Knoeller a
a
Department of Catchment Hydrology, UFZ e Helmholtz Centre for Environmental Research, Theodor-Lieser-Str. 4,
06120 Halle/Saale, Germany
b
Department of Environmental Biotechnology, UFZ e Helmholtz Centre for Environmental Research, Permoserstr.
15, 04318 Leipzig, Germany
c
Department of Soil Physics, UFZ e Helmholtz Centre for Environmental Research, Theodor-Lieser-Str. 4, 06120
Halle/Saale, Germany
d
Department of Biotechnology for Water Treatment, BTU-Cottbus-Senftenberg, Cottbus, Germany
e
Department of Microbiology, Institute for Water and Wetland Research, Radboud University of Nijmegen,
Nijmegen, The Netherlands

article info abstract

Article history: Constructed wetlands are important ecosystems with respect to nitrogen cycling. Here we
Received 21 December 2014 studied the activity and abundance of nitrogen transforming bacteria as well as the spatial
Received in revised form distribution of nitrification, anaerobic ammonium oxidation (anammox), and denitrifica-
9 February 2015 tion processes in a horizontal subsurface-flow constructed wetland. The functional genes
Accepted 10 February 2015 of the nitrogen cycle were evenly distributed in a linear way along the flow path with
Available online 19 February 2015 prevalence at the superficial points. The same trend was observed for the nitrification and
denitrification turnover rates using isotope labeling techniques. It was also shown that
Keywords: only short-term incubations should be used to measure denitrification turnover rates.
Nitrification Significant nitrate consumption under aerobic conditions diminishes nitrification rates and
Anammox should therefore be taken into account when estimating nitrification turnover rates. This
Aerobic denitrification nitrate consumption was due to aerobic denitrification, the rate of which was comparable
Abundance to that for anaerobic denitrification. Consequently, denitrification should not be considered
Activity as an exclusively anaerobic process. Phylogenetic analysis of hydrazine synthase (hzsA)
Constructed wetland gene clones indicated the presence of Brocadia and Kuenenia anammox species in the
constructed wetland. Although anammox bacteria were detected by molecular methods,
anammox activity could not be measured and hence this process appears to be of low
importance in nitrogen transformations in these freshwater ecosystems.
© 2015 Elsevier Ltd. All rights reserved.

* Corresponding author. Tel.: þ49 341 235 1696; fax: þ49 341 235 45 1696.
E-mail address: oksana.voloshchenko@ufz.de (O. Coban).
http://dx.doi.org/10.1016/j.watres.2015.02.018
0043-1354/© 2015 Elsevier Ltd. All rights reserved.
204 w a t e r r e s e a r c h 7 4 ( 2 0 1 5 ) 2 0 3 e2 1 2
1. Introduction
Nitrogen (N) compounds, such as ammonium (NHþ 4 ) and ni-
trate (NO 3 ), are some of the most widespread contaminants in
ground- and wastewater (Harrington and McInnes, 2009;
Singleton et al., 2007). For decades it was believed that the
only N transformations responsible for returning fixed nitro-
gen to the atmosphere were nitrification, the aerobic stepwise
oxidation of NHþ  
4 via nitrite (NO2 ) to NO3 , and denitrification,
anaerobic NO 3 or NO 
2 reduction with nitrous oxide (N2O) or
dinitrogen gas (N2) as a final product. In the meantime, how-
ever, the process of anaerobic ammonium oxidation (anam-
mox) was discovered, extending the N cycle by showing that
NHþ 4 can also be oxidized under anaerobic conditions using
NO 2 as the electron acceptor to produce N2 (Mulder et al.,
1995).
Constructed wetlands (CWs) are man-made systems
designed to include specific modified characteristics of
wetland ecosystems for improved treatment capacity (Kadlec
and Wallace, 2008). CWs can provide an effective contaminant
degradation zone due to enhancement of microbial activity in
the plant's rhizosphere (Stottmeister et al., 2003). CWs can be
used to treat a wide range of contaminants including N-con-
taining compounds. To date, only few studies have focused on
the microorganisms performing the nitrification and denitri-
fication processes in CWs, despite their crucial role in N-
removal (Chon et al., 2011; Correa-Galeote et al., 2013; Song
et al., 2012). While the overall N turnover in ecosystems can be
investigated by a wide range of methods, measurements of
microbial activity and its correlation with the copy numbers of
specific functional genes will add to our understanding of the
systems (Ruiz-Rueda et al., 2009; Song et al., 2012; Wang et al.,
2013). Furthermore, there are only a few reported measure-
ments on nitrification and denitrification turnover rates in
CWs using isotope labeling techniques (Erler et al., 2008; Scott
et al., 2008; Stepanauskas et al., 1996).
Denitrification is commonly considered to be an anaerobic
process and even small amounts of oxygen are reported to
inhibit the activity of the denitrifying enzymes and suppress
their synthesis in a stepwise manner (Bryan, 1981). However, a
number of laboratory studies with batch cultures revealed
that denitrification can also take place in the presence of ox-
ygen, a process referred to as aerobic denitrification
(Robertson et al., 1995; Robertson and Kuenen, 1984). In CWs,
nitrification occurs in the plant's rhizosphere, as the plant's
roots deliver oxygen, and/or in the top layer of the water body.
This would also be the zone where aerobic denitrification
could potentially occur. However, studies in natural and near-
natural environments such as CWs to verify and quantify
rates of aerobic denitrification are lacking (Gao et al., 2010).
Anammox accounts for over 50% of N loss in marine eco-
systems (Kuypers et al., 2003; Thamdrup and Dalsgaard, 2002).
However, to date, the role of anammox in freshwater eco-
systems has not yet been explored in depth (Humbert et al.,
2010; Jasper et al., 2014; Zhu et al., 2010). So far, anammox
research has focused mainly on quantifying turnover rates
(Burgin et al., 2011; Erler et al., 2008) and/or detecting anam-
mox bacteria using molecular biological tools (qPCR, FISH) in
various ecosystems in order to reveal if and where this
process occurs naturally (Burgin et al., 2011). While anammox
bacteria have been detected in a range of aquatic ecosystems
(Jetten et al., 2009), almost no investigations were done in
terrestrial ecosystems, despite molecular evidence of the
presence of anammox bacteria in diverse soils (Hu et al., 2011;
Humbert et al., 2010). Due to the mosaic of aerobic and
anaerobic zones as well as the usually low dissolved oxygen
concentration, CWs are assumed to offer favorable conditions
for anammox (Zhu et al., 2010). Although anammox organ-
isms have been detected in natural wetland soils (Humbert
et al., 2012), information about their activity and contribu-
tion to N turnover, particularly in CWs, is rather limited. Some
research has been done on enrichment and enhancement of
anammox processes in CWs (Paredes et al., 2007; Tao et al.,
2011; Wang and Li, 2011; Zhu et al., 2011). Anammox's
contribution to total N2 production has also been investigated,
but only in the sediments of full-scale surface flow CWs
treating domestic sewage (Erler et al., 2008). Such systems are
broadly representative of CWs together with horizontal
subsurface-flow (HSSF) CWs, and the HSSF CWs are charac-
terized by lower dissolved oxygen concentrations (Vymazal
and Kro € pfelova
 , 2008), so this could potentially increase
anammox contributions. Thus the role of anammox in N
removal in systems such as HSSF CWs as well as its rela-
tionship with other N transformations needs to be investi-
gated (Zhu et al., 2010).
The objectives of the present study were: i) to explore the
spatial distribution of both activity and bacterial abundance of
nitrification and denitrification, ii) to quantify rates of aerobic
denitrification, and iii) to investigate the role of anammox in N
removal in HSSF CW.
2. Materials and methods
2.1. Constructed wetland design
The CW in the study site was built as a part of CoTra
(Compartment Transfer) project in 2007 at Leuna near Leipzig,
Germany. Leuna has served as a major chemical
manufacturing site since the beginning of the 20th century,
and accidental spills, improper handling, and damages due to
heavy bombing during the World War II have left the under-
lying groundwater highly contaminated with benzene,
methyl-tert-butyl ether (MTBE), and NHþ 4 (Martienssen et al.,
2006).
The CW consisted of a basin (length 5 m  width
1.1 m  depth 0.6 m), was operated as horizontal subsurface
flow (HSSF), and planted with common reed (Phragmites aus-
tralis). The system was filled with gravel (grain size 2e3.2 mm)
up to a height of 0.5 m and the water level was set to 0.4 m,
resulting in a vadose zone of 0.1 m. The CW was operated in
continuous flow regime with flow rate of 7 L h1 and the
theoretical hydraulic retention time (assuming no water loss)
was 6.88 days. Inflow water was pumped in from adjacent
contaminated groundwater and had the following chemical
composition: NHþ 23.4 ± 5.0 mg L1, benzene
4 eN
4.1 ± 4.0 mg L , MTBE 0.4 ± 0.4 mg L1, TOC 17.5 ± 7.0 mg L1,
1
COD 45.0 ± 22.0 mg L1, BOD5 21.0 ± 14.0 mg L1, and pH
 w a t e r r e s e a r c h 7 4 ( 2 0 1 5 ) 2 0 3 e2 1 2 205

7.2 ± 0.4. Nitrate was absent in inflow and only a trace amount
(0.2 ± 0.2 mg L1) of NO2 was detected.
Physico-chemical parameters and N removal efficiencies
were described by Coban et al. (2014). Briefly, the average
seasonal air temperature was 18.0  C in summer, 12.7  C in
autumn, and 16.3  C in spring. The water loss calculated from
inflow and outflow streams varied between 14% in spring and
99% in summer. Redox was always negative and fluctuated
between 71 at the top sampling points (0.2 m) and
119 mV at the deepest sampling points (0.4 m). The CW had
the following average NHþ þ
4 eN removal: 0.63 g NH4 eN m
2 1
d facility (Macrogen Inc., Amsterdam, Netherlands). The hzsA
þ 2 1
in summer, 0.37 g NH4 eN m d in autumn, and 0.25 g
NHþ 4 eN m
2 1
d in spring.
2.2. DNA extraction and quantitative polymerase chain
reaction assay
Samples of gravel and roots in the HSSF CW were taken from 9
sampling points according to distance from the flow path (1 m,
2.5 m, and 4 m), and according to depth (0.2 m, 0.3 m, and
0.4 m). DNA was isolated from 0.82 ± 0.06 g of mixed sample of
gravel and roots using the Fast DNA® SPIN Kit for Soil and the
FastPrep® Instrument (MP Biomedicals, Santa Ana, CA). The
copy numbers of nitrogen transformation genes were deter-
mined by the quantitative polymerase chain reaction (qPCR)
method. qPCR was performed using the StepOnePlus Real-
Time PCR System (Applied Biosystems) and published
primer sets for hzsA (Harhangi et al., 2012), nirS, nirK, amoA and
16S rRNA (for sequences, coding genes and annealing tem-
perature, see Table 1). A dilution series of cloned amoA, nirS,
nirK and hzsA gene fragments (pGEM-T easy; Promega, Madi-
son, WI, USA) were used to prepare DNA standards with
known quantities of target DNA (insert source see Table 1).
Reactions were carried out using KAPA™ SYBR® FAST qPCR
MasterMix. PCR efficiency for the different assays ranged be-
tween 84 and 98%. Diluting 1:10 and 1:100 of DNA extracts was
performed in order to avoid any inhibition of PCR by inhibitors
contained in the gravel material.
2.3. Cloning, sequencing, and phylogenetic analysis of
anammox bacteria
For the cloning, three locations inside the CW were chosen
based on the highest hzsA gene abundances: sampling point
(SP) 1 is located at a distance of 1 m from the inflow at a depth
of 0.2 m, SP2 is 1 m from the inflow at a depth of 0.3 m, and SP3
is 4 m from the inflow at a depth of 0.2 m. The qPCR products
targeting the hzsA gene were purified using the MinElute PCR
Purification Kit (Qiagen, Chatsworth, CA), ligated, and cloned
using the pGEM-T Easy Vector System according the manu-
factures protocol (Promega, Madison, WI, USA). Sixty-eight
colonies were picked up, amplified with M13 primers (M13-f
GTTTTCCCAGTCACGAC, M13-r CAGGAAACAGCTATGAC)
and sequencing was performed at the Macrogen sequencing
gene sequences obtained in this study are available in NCBI
under the accession numbers KP943049eKP943112. Sequences
were aligned using BioEdit Sequence Alignment Editor pro-
gram (version BioEdit v7.2.5; North Carolina State University,
Raleigh, NC). Phylogenetic analysis was carried out using the
Mega 6.0 (Tamura et al., 2013).
2.4. Measuring anammox and denitrification rates with
the 15N isotopic tracing method
To estimate potential anammox and denitrification turnover
rates we used a sampling scheme similar to that used for the
molecular biology investigations. This time samples were
taken from 3 points: at a distance of 1 m from the inflow at a
depth of 0.2 m, 2.5 m from the inflow at a depth of 0.3 m, and
4 m from the inflow at a depth of 0.4 m. The presence, activity,
and potential of anammox and denitrifying bacteria were
measured as described by Risgaard-Petersen et al. (2004). Ho-
mogenized samples of gravel and roots of known weight and
density (16.28 ± 1.27 g, 1.98 ± 0.14 g cm3) were transferred to
22 mL glass vials together with inflow water, purged with N2,
and sealed. The experiment was performed in three repli-
cates. The slurries were then pre-incubated for 24 h to remove
all ambient O2 and NO x in the incubation media. Subse-
quently, 100 ml of N2-purged stock solution of each isotopic
mixture, i.e. (1) 15NHþ 15
4 ( N at.%: 98), (2)
15
NO 15
3 ( N at.%: 98),
15 þ 15 
and (3) NH4 ( N at.%: 98) þ NO3 was added resulting in the
final concentration of about 100 mM N. Incubations in the dark
and at 20  C were stopped at intervals 0 h, 2 h, 4 h, and 6 h by
adding 200 ml of a 7 M ZnCl2 solution. The experiment was
repeated using a set NHþ 4 1 mM þ
15
NO
3 200 mM and longer
incubation times of 12 h, 24 h, and 48 h. Incubations with
15
NHþ 4 were used as control to ensure that anoxic conditions

Table 1 e Primers and standard information of performed qPCRs.

Gene Primer Sequence (50 -30 ) Annealing Reference Standard clone
temperature ( C)
16S rRNA Nad F TCCTACGGGAGGCAGCAGT 57 Nadkarni et al. (2002) Pseudomonas putida
Nad R GGACTACCAGGGTATCT
AATCCTGTT
Hydrazine synthase, hzsA hzsA 1597F WTYGGKTATCARTATGTAG 55 Harhangi et al. (2012) Kuenenia stuttgartiensis
hzsA 1857R AAABGGYGAATCATARTGGC
ɑ subunit of ammonium amoA 1F GGGGTTTCTACTGGTGGT 57 Rotthauwe et al. (1997) Nitrosomonas europaea
monooxigenase, amoA amoA 2R CCCCTCKGSAAAGCCTTCTTC
Dissimilatory nitrite nirS cd3AF GTSAACGTSAAGGARACSGG 57 Throback et al. (2004) Pseudomonas stutzeri
reductase, nirS nirS R3cd GASTTCGGRTGSGTCTTGA
Dissimilatory nitrite nirK 1F GGMATGGTKCCSTGGCA 45 Braker et al. (1998) Blastobacter denitrificans
reductase, nirK nirK 5R GCCTCGATCAGRTTRTGGTT
206 w a t e r r e s e a r c h 7 4 ( 2 0 1 5 ) 2 0 3 e2 1 2
were created, incubations with 15NO 3 were used for denitri-
fication turnover quantification, and incubations with
15
NHþ  þ
4 þ NO3 as well as NH4 þ
15
NO3 were used to determine
the denitrification and anammox turnover rates.
The analyses of 15N2 abundance in the samples were con-
ducted using a GCMS-QP2010Plus (Shimadzu). The system is
equipped with a ShinCarbon ST column (100/120 mesh,
1.33 m  1 mm ID, Restek) connected downstream via an
automatic 6-port-valve (Valco Instruments Co. Inc.) to a Mol-
sieve 5A column (10 m  0.53 mm ID, 0.50 mm film, Agilent) and
helium as the carrier gas. Gas samples were applied via a
sample loop attached to a second automatic 6-port-valve
(Valco Instruments Co. Inc.), which is assembled between the
injector outlet and ShinCarbon ST inlet. The concentration of
N2 was calculated based on a calibration function gained from
standard gas calibration (N2 2530 ppm, loop size: 0.01, 0.5, and
2 mL, n ¼ 5 for each loop size). For N2 analysis a gas aliquot of
500 mL was withdrawn from the headspace of a sample vial by
gas-tight syringe and flushed through a 100 mL sample loop
(attached to the 6-port-valve). Afterwards, the sample loop
was immediately switched into the GCMS carrier gas stream
and mass 28N2, 29N2, and 30N2 were determined. The system
set-up for N2 analysis was as follows: 40  C oven temperature,
200  C I/F and ion source temperature, total flow 14 mL min1,
column flow 8 mL min1.
The rates of anammox and denitrification were calculated
from the mole fraction of 29N2 and 30N2 in the sampled vial
atmosphere of samples amended with 15NHþ 4 þ
14
NO 3 and
15  of variance (ANOVA F-test). The statistical analyses were
NO3 respectively using the equations of Spott and Stange
(2011). Given that the HSSF CW had maximum N removal in
summer, the incubation temperature for the calculation of
potential turnover rates under simulated in situ conditions in
this system was chosen based on summer average air tem-
perature (þ20  C) (Coban et al., 2014).
2.5. Potential ammonia oxidation rates
To determine the potential turnover rates for ammonia
oxidation, we used the pool enrichment/dilution method ac-
cording to Wessel and Tietema (1992). The sampling points
were chosen as for the denitrification and anammox in-
cubations. Homogenized samples of gravel and roots from
HSSF CW of known weight and density (178.55 ± 16.78 g,
1.98 ± 0.14 g cm3) were placed in 500 mL Schott bottles with
0.1 L of inflow water. The experiment was performed in three
replicates. Samples were closed with air access and incubated
at þ20  C in the dark with shaking. Subsequently, two sets of
incubations were set up. To the first set 5 mL of 10% 15NHþ 4 was
added to the final concentration of 177.38 mM NHþ 4 ; to the
second 1 mL of 10% 15NO 3 was added resulting in the final
concentration of 7.08 mM NO 3.
Between 6 and 8 mL of sample was taken with a syringe
immediately after substrate addition, and then again after 3,
12, and 24 h. Samples were filtered (filter 0.2 mm) and stored in
þ4  C until analysis. In all incubations 15N abundance and
concentration of NO-3 were measured. For measurements, the
SPINMAS technique was used which is a direct coupling of an
SPIN (sample preparation unit for inorganic nitrogen) to a
common quadropole Mass Spectrometer (GAM 400, InProcess
Instruments GmbH, Bremen, Germany) (Stange et al., 2007).
2.6. Calculation method and data analysis
In the ammonia oxidation rates experiment, the net nitrifi-
cation rates in 15NHþ4 amendments were determined using the
isotope mixing equation (e.g. Spott et al. (2006)) and the 15NO
3
amendments were used to determine the gross nitrification
rate and the NO 3 consumption rates by the pool dilution
approach using Equations (1) and (2) from Wessel and Tietema
(1992):
lnðft kÞ
f0 k W0  Wt
p¼ ln Wt
 (1)
W0
t
2 3
lnðft kÞ
6 7 W W
6 f k 7 0 t
c ¼ 61 þ ln0 Wt 7  (2)
4 5 t
W 0
where p e gross production rate of the enriched pool,
c e gross consumption rate of the enriched pool,
f e 15N abundance of the enriched pool,
k e naturally present 15N abundance,
W e amount of 14N plus 15N in the enriched pool,
and t e time.
Differences between different sample treatments over
distance and depth were determined using one-way analysis
performed by IBM SPSS Statistics 21 software, and the differ-
ences were regarded as significant at p < 0.05. Error bars
represent ±1 standard deviation (SD).
3. Results and discussion
3.1. Ammonia oxidizing bacterial abundance and
activity
Abundance of ammonia oxidizing genes was quantified by
qPCR targeting a subunit of ammonium monooxygenase
(amoA), which is the key enzyme in the aerobic ammonia
oxidation process. The genomic material of ammonia
oxidizing bacteria was detected at all points in the CW with
the highest abundance at the depth of 0.2 m, 2.2  106 copy
number g1 sample, and the lowest at 0.4 m depth, 6.8  104
copy number g1 sample (F(2,33) ¼ 12.968, p ¼ 0.000) (Fig. 1).
The abundance of the amoA gene varied with the distance
from the inflow along the flow path at each depth (p < 0.05).
Also, the net nitrification rates in 15NHþ
4 amendments were
determined using the isotope mixing equation and the 15NO 3
amendments were used to determine the gross nitrification
rate and the NO 3 consumption rates using Equations (1) and
(2) respectively. In the latter case, NHþ 4 was not additionally
added or primarily present in incubations and therefore it
derived from the mineralization of the organic material. Thus,
in these incubations the nitrification rate is equal to miner-
alization. The results are presented in Table 2. Like the dis-
tribution of the ammonia oxidizing genes, the gross
nitrification rates depended on the location of the sample, i.e.
 w a t e r r e s e a r c h 7 4 ( 2 0 1 5 ) 2 0 3 e2 1 2 207

Fig. 2 e Comparison of measured 29N2 data with modelled

29
N2 data when assuming an increasing contribution of
Fig. 1 e The spatial variations of ammonia oxidizing either only denitrification N2 or only anammox N2 to N2
bacterial abundance targeting the amoA gene and the atmosphere with natural 15N abundance in the HSSF CW
nitrification rates calculated using the mixing equation incubations with NHþ 4 þ
15
NO¡
3 (98 at%).

(net) and using Equation (1) (gross) in samples of gravel and

roots taken from the HSSF CW at three distances along the
flow path and at three depths for each distance.
high NO3 consumption can also take place and this should be
accounted for when calculating the nitrification rates.
Therefore, the approach of Wessel and Tietema (1992) should
depth and distance along the flow path (F(2,23) ¼ 5.428, be used to measure the nitrification turnover rate.
p ¼ 0.012) (Fig. 1). Furthermore, the highest nitrification rates
and ammonia oxidizing bacterial abundance were detected at 3.2. Potential anammox activity
the 0.2 and 0.3 m depths, which can be explained by the
enhanced microbial activity in zones with the highest root To determine potential activity for anammox and denitrifi-
density, i.e. the top sampling points. This could imply that the cation, the isotope labeling experiments were performed with
plant's rhizosphere plays an important role in oxygen delivery three sets of incubations (15NHþ 4,
15
NO3 , and
15
NHþ 
4 þ NO3 ) and
þ 15 
as no oxygen is permitted into the system via the surface in a subsequent set NH4 þ NO3 . The isotope incubations
the HSSF CW (Vymazal and Kro € pfelova
 , 2008). amended with only 15NHþ 4 had no accumulation of
29
N2 or
30 
The gross nitrification rates were significantly higher than N2, indicating that all ambient O2 and NOx had been
the net nitrification rates (F(1,49) ¼ 15.599, p ¼ 0.000). This can consumed during the 24 h preincubations. When both 15NHþ 4
be explained by the fact that the applied isotope mixing and NO-3 were added, no accumulation of 29N2 or 30N2 could be
equation does not consider a possible NO 3 consumption. In observed (i.e. no 15N enrichment in sampled N2 could be
contrast, the nitrification estimation method from Wessel and detected) in any incubation, which implies that 15NHþ 4 was not
Tietema (1992) produces the result as ‘gross nitrification’ and utilized for N2 production within the incubation period, and
thus accounts for a possible NO 3 consumption. At depths of hence hinted at an absence of an active anammox process
0.2 m and 0.3 m, where high NO 3 consumption were detected, (Fig. 2). In fact, when using the equations of Spott and Stange
the nitrification rates calculated on the basis of those two (2011) to calculate the fraction of anammox N2 (i.e. hybrid N2)
methods were quite different. Likewise, at 0.4 m depth, where in treatments with NHþ 4 and
15
NO 3 , the resulting contributions
NO 3 consumption was not significant, there was also no dif- were 0.08% to the total N2 mixture, which is just below the
ference between net nitrification and gross nitrification rates. detection limit of 0.1%. Moreover, the observed accumulation
However, at this depth the nitrification rates were also much of 29N2 within sampled N2 was extremely weak (0.78 %e0.84
lower. It can be concluded that under a high nitrification rate %), which also indicated the absence of an active fraction of

Table 2 e Calculated turnover rates for N-turnover processes at different points in HSSF CW.
Distance Depth Net nitrification Gross nitrification Nitrate consumption Anaerobic Anaerobic
from [m] (mixing equation) nmol NO 3 h
1 1
g nmol NO 3 h
1 1
g denitrification denitrification
inflow [m] nmol NO 3 h
1 1
g sample sample (short-term) nmol (long-term) nmol
sample N2 h1 g1 sample N2 h1 g1 sample
1 0.2 1.6 ± 0.9 7.1 ± 0.3 109.3 ± 89.0 3.8 ± 2.7 1.1 ± 0.7
2.5 0.3 2.1 ± 0.2 6.8 ± 3.7 164.2 ± 201.0 3.4 ± 0.7 0.7 ± 0.4
4 0.4 1.3 ± 0.1 1.3 ± 0.3 7.3 ± 12.2 0.6 ± 0.2 0.7 ± 0.3
208 w a t e r r e s e a r c h 7 4 ( 2 0 1 5 ) 2 0 3 e2 1 2

anammox bacteria in the HSSF CW (Fig. 2). The possible in-

hibition of the anammox bacteria could be the use of NO 3 as
substrate instead of NO 2 . However, given that anammox
bacteria perform dissimilatory NO -
3 reduction to NO2 (Kartal
 
et al., 2007), not only NO2 but also NO3 should be utilizable
in combination with NHþ 4 for anammox. Instead of anammox,
however, in all 15NO 3 treatments high denitrifying activity
was detected.
Analysis of operating parameters in our HSSF CW and
sufficient operating time of six years to establish the microbial
community indicated the presence of all suitable conditions
for the anammox process (Zhu et al., 2010). It was shown by
Musat et al. (2010) that anammox process can be coupled to
nitrate-dependent hydrocarbon attenuation, thus the pres-
ence of benzene and MTBE should not have a negative impact
on the anammox rates. However, given the high organic car-
bon of this inflow of contaminated groundwater and the car-
bon from the rhizodeposition products, NO 2 produced by
ammonia oxidizing bacteria may be predominantly used for
denitrification instead of anammox. As it was shown by
Reiche et al. (2010), low concentrations of contaminants and a
low C/N ratio of below 2 stimulates anammox, and in our
study a C/N ratio of about 2 was relatively high, which would
create conditions that favor growth of denitrifying bacteria.
Therefore, the anammox process might play a role in CWs,
albeit primarily in systems where they will not be out-
competed by denitrifying bacteria, i.e. under low C/N ratio (He
et al., 2012).
One advantage of treating contaminated waters with
anammox in CWs would be the potential mitigation of N2O
emissions because, unlike denitrification, anammox does not
produce N2O as a byproduct (Van de Graaf et al., 1997). So far,
only two studies in CWs detected active anammox process
without stimulation and/or enhancement (Erler et al., 2008;
Jasper et al., 2014). However, in the study of Erler et al. (2008)
it was shown that even when anammox contributed up to
30% to N removal, high N2O emissions were detected. Alter-
natively, CWs were considered to be a minor source of N2O
emissions and it was suggested that if all global domestic
wastewater were treated by wetlands the share of the trace
gas emission budget would still be less than 1% (Teiter and
Mander, 2005). This means that absence of anammox pro-
cess in CWs for waste- and groundwater treatment should not
have a negative impact on global warming.
3.3. Anammox bacterial abundance
The abundance of anammox genes in the biofilms attached to
the gravel and roots from the HSSF CW was quantified with
qPCR using hzsA primers, the most diagnostic phylogenetic
marker for anammox bacteria (Harhangi et al., 2012). The re-
sults are shown in Fig. 3. The genomic material of anammox
bacteria was detected at all points in the CW and the resulting
copy number was similar to the copy number of amoA gene,
between 3.7  104 and 1.4  106 copy number g1 sample.
Likewise the ammonia oxidizing gene copy number, hzsA gene
copies showed that bacteria were most abundant at 0.2 m and
least abundant at 0.4 m depth (F(2,24) ¼ 7.157, p ¼ 0.004). The
abundance of the hzsA gene varied with the distance from the
inflow along the flow path at each depth (p < 0.05).
Fig. 3 e The spatial variations of anammox bacterial
abundance targeting diagnostic hzsA gene in the samples
of gravel and roots from the HSSF CW at three distances
along the flow path and at three depths for each distance.
The hzsA primers have a preference over traditional 16S
rRNA anammox primers as anammox primers may target not
only the anammox group but other Planctomycetes as well
(Humbert et al., 2010). Using traditional 16S rRNA anammox
primers for the same samples gave us two orders of magni-
tude of higher copy numbers of anammox bacteria (data not
shown), and these results are similar to those of Bale et al.
(2014). However, even when using a highly specific target for
anammox, the abundance of the bacteria does not correlate
with its activity (Wang et al., 2012). While Bale et al. (2014)
detected anammox activity already at 4.2  105e1.4  106
copies of the hzsA gene g1 in a marine environment, the ac-
tivity of anammox bacteria in our samples could not be
measured, even though the anammox bacterial abundance
was similar.
3.4. Biodiversity of anammox bacteria
The presence of anammox bacteria and diversity of the
anammox community were studied by analysis of the 68 hzsA
gene sequences obtained from the gravel and root samples
from the HSSF CW. Phylogenetic analysis of the hzsA genes
(Fig. 4) showed that most of the sequences in both SP1 and SP3
were related to “Ca. Brocadia fulgida”. One hzsA clone sequence
from the SP1 and three hzsA clone sequences from the SP3
were closely related to the sequence of “Ca. Brocadia anam-
moxidans”, while two other SP1 clones with two SP2 clones
clustered together with “Ca. Kuenenia stuttgartiensis”. While
Brocadia could be found at locations SP1 and SP3 (i.e. 1 and 4 m
distance to inflow), Kuenenia was detected at SP1 and SP2 (i.e.
1 m distance to inflow). This reflects the presence of the
various microniches in CWs suitable for different anammox
species (Zhu et al., 2010). Other studies also showed high
biodiversity of anammox in terrestrial ecosystems (Hu et al.,
2011; Humbert et al., 2010; Zhu et al., 2013).
3.5. Denitrifying bacterial abundance and activity
Abundance of denitrifying genes was quantified with qPCR
targeting nirS and nirK, which encode for key enzymes in
 w a t e r r e s e a r c h 7 4 ( 2 0 1 5 ) 2 0 3 e2 1 2
Fig. 4 e Phylogenetic analysis of hzsA gene sequences from
samples of gravel and roots from HSSF CW (68 in total). The
evolutionary history was inferred using the neighbor-
joining method. The optimal tree with the sum of branch
length ¼ 1.055 is shown. Bootstrap values of ≥50 (500
replicates) are shown at the branches. The tree is drawn to
scale, with branch lengths in the same units as those of the
evolutionary distances used to infer the phylogenetic tree.
The evolutionary distances were computed using the
JukeseCantor method and are in the units of the number of
base substitutions per site. The bar represents 5%
sequence divergence. All positions containing gaps and
missing data were eliminated. There were a total of 258
positions in final dataset.
denitrification that are responsible for NO 2 reduction. The
results showed high copy numbers of denitrifying genes, be-
tween 1.4  108 and 7.7  108 g1 sample (Fig. 5). Furthermore,
denitrifying bacteria were most abundant at the depth of
0.2 m and least abundant at the depth of 0.4 m
(F(2,32) ¼ 12.278, p ¼ 0.000). The nirS and nirK genes did not vary
with distance from the inflow along the flow path at each
depth (p > 0.05).
In the next step, the potential denitrification rate was
calculated with the isotope labeling technique using the
approach of Spott and Stange (2011). The results are presented
in Table 2. Interestingly, there was a significant difference
between the rates calculated from the short-term and long-
term incubations (F(1,50) ¼ 17.157, p ¼ 0.000) (Fig. 5). With
short-term incubation, the maximum denitrification rates
were detected at the 0.2 m depth and the minimum at the
0.4 m depth (F(2,22) ¼ 6.714, p ¼ 0.005), showing a similar trend
to the distribution of the denitrifying genes. However, in the
long-term experimental set-up there was no difference be-
tween depths (F(2,24) ¼ 1.181, p ¼ 0.324) and the rates were
lower. This might be attributable to a decrease of turnover
rates over time in a batch experiment with a decrease of
substrate concentration, according to MichaeliseMenten ki-
netics. This would also smooth out the difference between
different sampling points. Therefore, only short-term in-
cubations (up to 6 h) should be used to measure denitrification
rates in future experiments.
209
Given that the highest denitrification rates were observed
at 0.2 m and 0.3 m depths, i.e. in zones with higher root den-
sity, we can assume that root exudates play the main role in
organic carbon delivery for the denitrification process.
Another possible conclusion is that denitrification is not
inhibited by the presence of oxygen. This will be discussed
further.
3.6. Aerobic denitrification potential
While measuring nitrification rates in 15NO 3 amendments,
significant NO 3 consumption under aerobic conditions was
observed (Table 2). There are several possible routes through
which these NO 
3 losses can occur. Dissimilatory NO3 reduc-
tion (DNRA) is a microbial process that transforms NO3 to NHþ

4
via formation of NO 2 in anaerobic or low O2 environments.
DNRA was shown to be relevant under reduced conditions
(Eh ¼ 200 mV) (Buresh and Patrick, 1981). However, today
there is growing evidence that DNRA can also take place in the
presence of O2, especially under high C:N ratios of about 10
and low NO 3 concentrations (Fazzolari et al., 1998), and it is
stimulated by presence of macrophytes (Nijburg and
Laanbroek, 1997). Still, the rates of DNRA in CWs related to
the total NO 3 consumption reported in the literature are in the
2e9 % range (Fazzolari et al., 1998) and therefore could not be
of high importance in our incubations considering the sig-
nificant NO 3 consumption. Immobilization (microbial uptake)
could be another NO þ
3 consumption route. However, NH4 is a

preferred N source for microorganisms and NO3 uptake by
microorganisms is inhibited when both NHþ 
4 and NO3 are
present (Recous et al., 1990). Taking into account the occur-
rence of nitrification in these incubations, NHþ 4 was present
and therefore, NO 3 uptake should be negligible. Denitrifica-
tion was assumed to be an exclusively anaerobic process
which occur only with absence of oxygen (Bryan, 1981).
However, in the laboratory studies evidence was found for
denitrification to occur under aerobic conditions as well
(Martienssen and Schops, 1999; Robertson et al., 1995) and to
Fig. 5 e The spatial variations of denitrification bacterial
abundance targeting the nirS and nirK gene and the
denitrification rates for the short-term and long-term
experimental set-ups in the samples of gravel and roots
from the HSSF CW at three distances along the flow path
and at three depths for each distance.
210 w a t e r r e s e a r c h 7 4 ( 2 0 1 5 ) 2 0 3 e2 1 2
be persistent at high O2 levels (Lloyd et al., 1987). Aerobic
denitrification or co-respiration implies the simultaneous use
of both O2 and NO 3 as oxidizing agents and can be performed
by various genera of microorganisms (Robertson and Kuenen,
1984). Gao et al. (2010) concluded that O2 dynamics did not
strongly affect N consumption by denitrification in the pres-
ence of abundant NO
x , but rather denitrification coexisted
with O2 respiration. Therefore, the NO
3 consumption rates
observed under aerobic conditions in our incubations could be
attributed to aerobic denitrification.
Studies on aerobic denitrification rates in natural envi-
ronments are largely missing, except for one study in marine
sediments (Gao et al., 2010). Ours is the first report of aerobic
denitrification rates in freshwater ecosystems. At each depth
examined in the incubations, the potential denitrification rate
under aerobic conditions was much higher than that
measured under anaerobic conditions. However, the rates of
aerobic denitrification might be overestimated given that: 1)
the starting NO 3 concentrations for the aerobic incubations
were an order of magnitude higher than for the anaerobic
ones; 2) the aerobic denitrification (NO 3 consumption) is
expressed in nmol NO 1
g1 sample and the anaerobic
3 h
denitrification in nmol N2 h1 g1 sample; and 3) 2 mol of NO 3
transform to 1 mol of N2. The previous findings in the sea
sediments reported quite similar rates of denitrification under
aerobic and anaerobic conditions (Gao et al., 2010). The high-
est denitrification rate was not observed in the deepest depth
interval with the lowest redox potential, but rather in the
surface 0.2 m depth. It is well related to the nitrification rates
and implies simultaneous occurrence of nitrification and
denitrification, as has been demonstrated by the stable
isotope approach in Coban et al. (2014). Our CW was fed with
NHþ 4 rich contaminated groundwater and therefore it would
be consistent to assume that denitrifying bacteria should be
located in close proximity to nitrifying bacteria in order to
obtain NO 3 from them. Biofilm formation is advantageous for
this as it creates an oxiceanoxic interface and therefore pro-
vides process stratification (Fukada et al., 2004).
4. Conclusions
In this study, the contribution and diversity of various nitro-
gen cycle processes in a horizontal subsurface-flow con-
structed wetland were determined with molecular markers
and stable isotope tests. The functional genes of the nitrogen
cycle were abundant along the flow path with prevalence at
the superficial points. The same trend was observed for the
nitrification and denitrification turnover rates. Furthermore,
significant nitrate consumption under aerobic conditions was
detected and this should be taken into account when esti-
mating nitrification rates. This nitrate consumption was
apparently due to aerobic denitrification. Therefore, in con-
structed wetlands denitrification should not be considered as
an exclusively anaerobic process. Although anammox bacte-
ria were detected by molecular-biological techniques, anam-
mox activity was absent and therefore this process appeared
to be of low importance in nitrogen transformations in the
studied ecosystem of constructed wetland due to presence of
oxygen and high concentrations of organic carbon.
Acknowledgments
This research was completed within the framework of the
Marie Curie Initial Training Network ADVOCATE e Advancing
sustainable in situ remediation for contaminated land and
groundwater, funded by the European Commission, Marie
Curie Actions Project No. 265063, SAFIRA project and the
Helmholtz Interdisciplinary Graduate School for Environ-
mental Research (HIGRADE). MSMJ was supported by ERC AG
232937 and 339880. We are also grateful to Ines Ma€ usezahl of
the molecular biology laboratory for her help with the mo-
lecular biological analysis of the samples. Thank also goes to
Anne Carney for her help with English proofreading of this
manuscript.
references
Bale, N.J., Villanueva, L., Fan, H.X., Stal, L.J., Hopmans, E.C.,
Schouten, S., Damste, J.S.S., 2014. Occurrence and activity of
anammox bacteria in surface sediments of the southern North
Sea. FEMS Microbiol. Ecol. 89 (1), 99e110.
Braker, G., Fesefeldt, A., Witzel, K.P., 1998. Development of PCR
primer systems for amplification of nitrite reductase genes
(nirK and nirS) to detect denitrifying bacteria in environmental
samples. Appl. Environ. Microbiol. 64 (10), 3769e3775.
Bryan, B.A., 1981. Physiology and Biochemistry of Denitrification.
Buresh, R.J., Patrick, W.H., 1981. Nitrate reduction to ammonium
and organic nitrogen in an estuarine sediment. Soil Biol.
Biochem. 13 (4), 279e283.
Burgin, A.J., Yang, W.H., Hamilton, S.K., Silver, W.L., 2011. Beyond
carbon and nitrogen: how the microbial energy economy
couples elemental cycles in diverse ecosystems. Front. Ecol.
Environ. 9 (1), 44e52.
Chon, K., Chang, J.S., Lee, E., Lee, J., Ryu, J., Cho, J., 2011.
Abundance of denitrifying genes coding for nitrate (narG),
nitrite (nirS), and nitrous oxide (nosZ) reductases in estuarine
versus wastewater effluent-fed constructed wetlands. Ecol.
Eng. 37 (1), 64e69.
Coban, O., Kuschk, P., Wells, N., Strauch, G., Knoeller, K., 2014.
Microbial nitrogen transformation in constructed wetlands
treating contaminated groundwater. Environ. Sci. Pollut. Res.
1e11.
Correa-Galeote, D., Marco, D.E., Tortosa, G., Bru, D.,
Philippot, L., Bedmar, E.J., 2013. Spatial distribution of N-
cycling microbial communities showed complex patterns
in constructed wetland sediments. FEMS Microbiol. Ecol. 83
(2), 340e351.
Erler, D.V., Eyre, B.D., Davison, L., 2008. The contribution of
anammox and denitrification to sediment N2 production in a
surface flow constructed wetland. Environ. Sci. Technol. 42
(24), 9144e9150.
Fazzolari, E., Nicolardot, B., Germon, J.C., 1998. Simultaneous
effects of increasing levels of glucose and oxygen partial
pressures on denitrification and dissimilatory nitrate
reduction to ammonium in repacked soil cores. Eur. J. Soil Biol.
34 (1), 47e52.
Fukada, T., Hiscock, K.M., Dennis, P.F., 2004. A dual-isotope
approach to the nitrogen hydrochemistry of an urban aquifer.
Appl. Geochem. 19 (5), 709e719.
Gao, H., Schreiber, F., Collins, G., Jensen, M.M., Kostka, J.E.,
Lavik, G., de Beer, D., Zhou, H.Y., Kuypers, M.M.M., 2010.
Aerobic denitrification in permeable Wadden Sea sediments.
ISME J. 4 (3), 417e426.
 w a t e r r e s e a r c h 7 4 ( 2 0 1 5 ) 2 0 3 e2 1 2
Harhangi, H.R., Le Roy, M., van Alen, T., Hu, B.L., Groen, J.,
Kartal, B., Tringe, S.G., Quan, Z.X., Jetten, M.S.M., Op den
Camp, H.J.M., 2012. Hydrazine synthase, a unique
phylomarker with which to study the presence and
biodiversity of anammox bacteria. Appl. Environ. Microbiol. 78
(3), 752e758.
Harrington, R., McInnes, R., 2009. Integrated Constructed
Wetlands (ICW) for livestock wastewater management.
Bioresour. Technol. 100 (22), 5498e5505.
He, Y.L., Tao, W.D., Wang, Z.Y., Shayya, W., 2012. Effects of pH
and seasonal temperature variation on simultaneous partial
nitrification and anammox in free-water surface wetlands. J.
Environ. Manag. 110, 103e109.
Hu, B.L., Rush, D., van der Biezen, E., Zheng, P., van Mullekom, M.,
Schouten, S., Damste, J.S.S., Smolders, A.J.P., Jetten, M.S.M.,
Kartal, B., 2011. New anaerobic, ammonium-oxidizing
community enriched from peat soil. Appl. Environ. Microbiol.
77 (3), 966e971.
Humbert, S., Tarnawski, S., Fromin, N., Mallet, M.P., Aragno, M.,
Zopfi, J., 2010. Molecular detection of anammox bacteria in
terrestrial ecosystems: distribution and diversity. ISME J. 4 (3),
450e454.
Humbert, S., Zopfi, J., Tarnawski, S.E., 2012. Abundance of
anammox bacteria in different wetland soils. Environ.
Microbiol. Rep. 4 (5), 484e490.
Jasper, J.T., Jones, Z.L., Sharp, J.O., Sedlak, D.L., 2014. Nitrate
removal in shallow, open-water treatment wetlands. Environ.
Sci. Technol. 48 (19), 11512e11520.
Jetten, M.S.M., van Niftrik, L., Strous, M., Kartal, B., Keltjens, J.T.,
Op den Camp, H.J.M., 2009. Biochemistry and molecular
biology of anammox bacteria. Crit. Rev. Biochem. Mol. Biol. 44
(2e3), 65e84.
Kadlec, R.H., Wallace, S.D., 2008. Treatment Wetlands, second ed.
CRC Press.
Kartal, B., Kuypers, M.M.M., Lavik, G., Schalk, J., den Camp, H.,
Jetten, M.S.M., Strous, M., 2007. Anammox bacteria disguised
as denitrifiers: nitrate reduction to dinitrogen gas via nitrite
and ammonium. Environ. Microbiol. 9 (3), 635e642.
Kuypers, M.M.M., Sliekers, A.O., Lavik, G., Schmid, M.,
Jorgensen, B.B., Kuenen, J.G., Damste, J.S.S., Strous, M.,
Jetten, M.S.M., 2003. Anaerobic ammonium oxidation by
anammox bacteria in the Black Sea. Nature 422 (6932),
608e611.
Lloyd, D., Boddy, L., Davies, K.J., 1987. Persistence of bacterial
denitrification capacity under aerobic conditions: the rule
rather than the exception. FEMS Microbiol. Lett. 45 (3),
185e190.
Martienssen, M., Fabritius, H., Kukla, S., Balcke, G.U.,
Hasselwander, E., Schirmer, M., 2006. Determination of
naturally occurring MTBE biodegradation by analysing
metabolites and biodegradation by-products. J. Contam.
Hydrol. 87 (1e2), 37e53.
Martienssen, M., Schops, R., 1999. Population dynamics of
denitrifying bacteria in a model biocommunity. Water Res. 33
(3), 639e646.
Mulder, A., van de Graaf, A.A., Robertson, L.A., Kuenen, J.G.,
1995. Anaerobic ammonium oxidation discovered in a
denitrifying fluidized bed reactor. FEMS Microbiol. Ecol. 16
(3), 177e184.
Musat, F., Wilkes, H., Behrends, A., Woebken, D., Widdel, F., 2010.
Microbial nitrate-dependent cyclohexane degradation coupled
with anaerobic ammonium oxidation. ISME J. 4 (10),
1290e1301.
Nadkarni, M.A., Martin, F.E., Jacques, N.A., Hunter, N., 2002.
Determination of bacterial load by real-time PCR using a
broad-range (universal) probe and primers set. Microbiol.-SGM
148, 257e266.
211
Nijburg, J.W., Laanbroek, H.J., 1997. The fate of N-15-nitrate in
healthy and declining Phragmites australis stands. Microb. Ecol.
34 (3), 254e262.
Paredes, D., Kuschk, P., Stange, F., Muller, R.A., Koser, H., 2007.
Model experiments on improving nitrogen removal in
laboratory scale subsurface constructed wetlands by
enhancing the anaerobic ammonia oxidation. Water Sci.
Technol. 56 (3), 145e150.
Recous, S., Mary, B., Faurie, G., 1990. Microbial immobilization of
ammonium and nitrate in cultivated soils. Soil Biol. Biochem.
22 (7), 913e922.
Reiche, N., Lorenz, W., Borsdorf, H., 2010. Development and
application of dynamic air chambers for measurement of
volatilization fluxes of benzene and MTBE from constructed
wetlands planted with common reed. Chemosphere 79 (2),
162e168.
Risgaard-Petersen, N., Meyer, R.L., Schmid, M., Jetten, M.S.M.,
Enrich-Prast, A., Rysgaard, S., Revsbech, N.P., 2004. Anaerobic
ammonium oxidation in an estuarine sediment. Aquat.
Microb. Ecol. 36 (3), 293e304.
Robertson, L.A., Dalsgaard, T., Revsbech, N.P., Kuenen, J.G., 1995.
Confirmation of aerobic denitrication in batch cultures, using
gas-chromatography and 15N mass-spectrometry. FEMS
Microbiol. Ecol. 18 (2), 113e119.
Robertson, L.A., Kuenen, J.G., 1984. Aerobic denitrification e a
controversy revived. Arch. Microbiol. 139 (4), 351e354.
Rotthauwe, J.H., Witzel, K.P., Liesack, W., 1997. The ammonia
monooxygenase structural gene amoA as a functional marker:
molecular fine-scale analysis of natural ammonia-oxidizing
populations. Appl. Environ. Microbiol. 63 (12), 4704e4712.
Ruiz-Rueda, O., Hallin, S., Baneras, L., 2009. Structure and
function of denitrifying and nitrifying bacterial communities
in relation to the plant species in a constructed wetland. FEMS
Microbiol. Ecol. 67 (2), 308e319.
Scott, J.T., McCarthy, M.J., Gardner, W.S., Doyle, R.D., 2008.
Denitrification, dissimilatory nitrate reduction to ammonium,
and nitrogen fixation along a nitrate concentration gradient in
a created freshwater wetland. Biogeochemistry 87 (1), 99e111.
Singleton, M.J., Esser, B.K., Moran, J.E., Hudson, G.B.,
McNab, W.W., Harter, T., 2007. Saturated zone denitrification:
potential for natural attenuation of nitrate contamination in
shallow groundwater under dairy operations. Environ. Sci.
Technol. 41 (3), 759e765.
Song, K.Y., Kang, H., Zhang, L., Mitsch, W.J., 2012. Seasonal and
spatial variations of denitrification and denitrifying bacterial
community structure in created riverine wetlands. Ecol. Eng.
38 (1), 130e134.
Spott, O., Russow, R., Apelt, B., Stange, C.F., 2006. A 15N-aided
artificial atmosphere gas flow technique for online
determination of soil N2 release using the zeolite Ko € strolith
SX6®. Rapid Commun. Mass Spectrom. 20 (22), 3267e3274.
Spott, O., Stange, C.F., 2011. Formation of hybrid N2O in a
suspended soil due to co-denitrification of NH2OH. J. Plant
Nutr. Soil Sci. 174 (4), 554e567.
Stange, C.F., Spott, O., Apelt, B., Russow, R.W.B., 2007. Automated
and rapid online determination of 15N abundance and
concentration of ammonium, nitrite, or nitrate in aqueous
samples by the SPINMAS technique. Isot. Environ. Health
Stud. 43 (3), 227e236.
Stepanauskas, R., Davidsson, E.T., Leonardson, L., 1996. Nitrogen
transformations in wetland soil cores measured by N-15
isotope pairing and dilution at four infiltration rates. Appl.
Environ. Microbiol. 62 (7), 2345e2351.
Stottmeister, U., Wiessner, A., Kuschk, P., Kappelmeyer, U.,
Kastner, M., Bederski, O., Muller, R.A., Moormann, H., 2003.
Effects of plants and microorganisms in constructed wetlands
for wastewater treatment. Biotechnol. Adv. 22 (1e2), 93e117.
212 w a t e r r e s e a r c h 7 4 ( 2 0 1 5 ) 2 0 3 e2 1 2
Tamura, K., Stecher, G., Peterson, D., Filipski, A., Kumar, S., 2013.
MEGA6: molecular evolutionary genetics analysis version 6.0.
Mol. Biol. Evol. 30 (12), 2725e2729.
Tao, W.D., Wen, J.F., Huchzermeier, M., 2011. Batch operation of
biofilter e free-water surface wetland series for enhancing
nitritation and anammox. Water Environ. Res. 83 (6), 541e548.
Teiter, S., Mander, U., 2005. Emission of N2O, N2, CH4, and CO2
from constructed wetlands for wastewater treatment and
from riparian buffer zones. Ecol. Eng. 25 (5), 528e541.
Thamdrup, B., Dalsgaard, T., 2002. Production of N2 through
anaerobic ammonium oxidation coupled to nitrate reduction
in marine sediments. Appl. Environ. Microbiol. 68 (3),
1312e1318.
Throback, I.N., Enwall, K., Jarvis, A., Hallin, S., 2004. Reassessing
PCR primers targeting nirS, nirK and nosZ genes for community
surveys of denitrifying bacteria with DGGE. FEMS Microbiol.
Ecol. 49 (3), 401e417.
Van de Graaf, A.A., de Bruijn, P., Robertson, L.A., Jetten, M.S.M.,
Kuenen, J.G., 1997. Metabolic pathway of anaerobic
ammonium oxidation on the basis of 15N studies in a fluidized
bed reactor. Microbiol.-UK 143, 2415e2421.
Vymazal, J., Kro€ pfelova
 , L., 2008. In: Vymazal, J. (Ed.), Wastewater
Treatment, Plant Dynamics and Management in Constructed
and Natural Wetlands. Springer, Netherlands, pp. 311e317.
Wang, C.X., Zhu, G.B., Wang, W.D., Yin, C.Q., 2013. Preliminary
study on the distribution of ammonia oxidizers and their
contribution to potential ammonia oxidation in the plant-bed/
View publication stats
ditch system of a constructed wetland. J. Soils Sediments 13
(9), 1626e1635.
Wang, L., Li, T., 2011. Anaerobic ammonium oxidation in
constructed wetlands with bio-contact oxidation as
pretreatment. Ecol. Eng. 37 (8), 1225e1230.
Wang, S.Y., Zhu, G.B., Peng, Y.Z., Jetten, M.S.M., Yin, C.Q., 2012.
Anammox bacterial abundance, activity, and contribution in
riparian sediments of the Pearl river estuary. Environ. Sci.
Technol. 46 (16), 8834e8842.
Wessel, W.W., Tietema, A., 1992. Calculating gross N
transformation rates of 15N pool dilution experiments with
acid forest litter: analytical and numerical approaches. Soil
Biol. Biochem. 24 (10), 931e942.
Zhu, G., Wang, S., Wang, W., Wang, Y., Zhou, L., Jiang, B., Op den
Camp, H.J.M., Risgaard-Petersen, N., Schwark, L., Peng, Y.,
Hefting, M.M., Jetten, M.S.M., Yin, C., 2013. Hotspots of
anaerobic ammonium oxidation at land-freshwater
interfaces. Nat. Geosci. 6 (2), 103e107.
Zhu, G.B., Jetten, M.S.M., Kuschk, P., Ettwig, K.F., Yin, C.Q., 2010.
Potential roles of anaerobic ammonium and methane
oxidation in the nitrogen cycle of wetland ecosystems. Appl.
Microbiol. Biotechnol. 86 (4), 1043e1055.
Zhu, G.B., Wang, S.Y., Feng, X.J., Fan, G.N., Jetten, M.S.M.,
Yin, C.Q., 2011. Anammox bacterial abundance, biodiversity
and activity in a constructed wetland. Environ. Sci. Technol.
45 (23), 9951e9958.