Development of high-rate anaerobic ammonium-oxidizing (anammox) biofilm reactors.

Abstract

To promptly establish anaerobic ammonium oxidation (anammox) reactors, appropriate seeding sludge with high abundance and activity of anammox bacteria was selected by quantifying 16S rRNA gene copy numbers of anammox bacteria by real-time quantitative PCR (RTQ-PCR) and batch culture experiments. The selected sludge was then inoculated into up-flow fixed-bed biofilm column reactors with nonwoven fabric sheets as biomass carrier and the reactor performances were monitored over 1 year. The anammox reaction was observed within 50 days and a total nitrogen removal rate of 26.0 kg-Nm(-3)day(-1) was obtained after 247 days. To our knowledge, such a high rate has never been reported before. Hydraulic retention time (HRT) and influent NH(4)(+) to NO(2)(-) molar ratio could be important determinant factors for efficient nitrogen removal in this study. The higher nitrogen removal rate was obtained at the shorter HRT and higher influent NH(4)(+)/NO(2)(-) molar ratio. After anammox reactors were fully developed, the community structure, spatial organization and in situ activity of the anammox biofilms were analyzed by the combined use of a full-cycle of 16S rRNA approach and microelectrodes. In situ hybridization results revealed that the probe Amx820-hybridized anaerobic anammox bacteria were distributed throughout the biofilm (accounting for more than 70% of total bacteria). They were associated with Nitrosomonas-like aerobic ammonia-oxidizing bacteria (AAOB) in the surface biofilm. The anammox bacteria present in this study were distantly related to the Candidatus Brocadia anammoxidans with the sequence similarity of 95%. Microelectrode measurements showed that a high in situ anammox activity (i.e., simultaneous consumption of NH(4)(+) and NO(2)(-)) of 4.45 g-N of (NH(4)(+)+NO(2)(-))m(-2)day(-1) was detected in the upper 800 microm of the biofilm, which was consistent with the spatial distribution of anammox bacteria.

Full text

Page 1

Available at www.sciencedirect.com
journal homepage: www.elsevier.com/locate/watres
Development of high-rate anaerobic ammonium-oxidizing
(anammox) biofilm reactors
Ikuo Tsushimaa, Yuji Ogasawaraa, Tomonori Kindaichib, Hisashi Satoha, Satoshi Okabea,?
aDepartment of Urban and Environmental Engineering, Graduate School of Engineering, Hokkaido University, Sapporo 060-8628, Japan
bDepartment of Social and Environmental Engineering, Graduate School of Engineering, Hiroshima University, 1-4-1 Kagamiyama,
Higashihiroshima 739-8527, Japan
a r t i c l e i n f o
Article history:
Received 31 October 2006
Received in revised form
26 January 2007
Accepted 28 January 2007
Available online 12 March 2007
Keywords:
Anammox biofilm reactor
Biofilm structure and function
16S rRNA approach
Microelectrodes
a b s t r a c t
To promptly establish anaerobic ammonium oxidation (anammox) reactors, appropriate
seeding sludge with high abundance and activity of anammox bacteria was selected by
quantifying 16S rRNA gene copy numbers of anammox bacteria by real-time quantitative
PCR (RTQ-PCR) and batch culture experiments. The selected sludge was then inoculated
into up-flow fixed-bed biofilm column reactors with nonwoven fabric sheets as biomass
carrier and the reactor performances were monitored over 1 year. The anammox reaction
was observed within 50 days and a total nitrogen removal rate of 26.0kg-Nm?3day?1was
obtained after 247 days. To our knowledge, such a high rate has never been reported before.
Hydraulic retention time (HRT) and influent NHþ
determinant factors for efficient nitrogen removal in this study. The higher nitrogen
removal rate was obtained at the shorter HRT and higher influent NHþ
After anammox reactors were fully developed, the community structure, spatial organiza-
tion and in situ activity of the anammox biofilms were analyzed by the combined use of a
full-cycle of 16S rRNA approach and microelectrodes. In situ hybridization results revealed
that the probe Amx820-hybridized anaerobic anammox bacteria were distributed
throughout the biofilm (accounting for more than 70% of total bacteria). They were
associated with Nitrosomonas-like aerobic ammonia-oxidizing bacteria (AAOB) in the
surface biofilm. The anammox bacteria present in this study were distantly related to
the Candidatus Brocadia anammoxidans with the sequence similarity of 95%. Microelectrode
measurements showed that a high in situ anammox activity (i.e., simultaneous consump-
tion of NHþ
of the biofilm, which was consistent with the spatial distribution of anammox bacteria.
4to NO?
2molar ratio could be important
4=NO?
2molar ratio.
4and NO?
2) of 4.45g-N of (NHþ
4+NO?
2)m?2day?1was detected in the upper 800mm
& 2007 Elsevier Ltd. All rights reserved.
1.Introduction
Anaerobic ammonium oxidation (anammox) is a biological
process in which ammonium is directly converted to dinitro-
gen gas with nitrite as the electron acceptor under anoxic
conditions (Jetten et al., 1999). The anammox reaction was
first discovered in a denitrifying pilot plant reactor in Delft,
the Netherlands (Mulder et al., 1995), and today anammox
reactions have been reported from several other treatment
plants (Egli et al., 2001; Fux et al., 2002; van Dongen et al.,
2001). Anammox is carried out by chemolithoautotrophic
bacteria belonging to the order Planctomycetales. To date, two
freshwaterspecies, Candidatus
(Strous et al., 1999) and Candidatus Kuenenia stuttgartiensis
Brocadiaanammoxidans
ARTICLE IN PRESS
0043-1354/$-see front matter & 2007 Elsevier Ltd. All rights reserved.
doi:10.1016/j.watres.2007.01.050
?Corresponding author. Tel.: +810117066266; fax: +810117066266.
E-mail address: sokabe@eng.hokudai.ac.jp (S. Okabe).
WAT E R R E S E A R C H 41 (2007) 1623– 1634

Page 2

(Schmid et al., 2000) and three marine species, Candidatus
Scalindua sorokinii (Kuypers et al., 2003), Candidatus Scalindua
brodae (Schmid et al., 2003), and Candidatus Scalindua wagneri
(Schmid et al., 2003) have been proposed. Recently, a
mixotrophic anammox bacterium Candidatus Anammoxglo-
bus propionicus was also described (Kartal et al., 2006). Since
these bacteria have not been isolated in pure culture yet, the
current information about their physiology has been obtained
from enrichment culture studies (Egli et al., 2001; Strous et al.,
1998; Toh et al., 2002).
The anammox process is a new and promising alternative to
the conventional nitrogen removal processes. The application
of anammox to nitrogen removal would lead to a significant
reduction of costs for aeration and exogenous electron donor
as compared to the conventional nitrification–denitrification
process (van Dongen et al., 2001). However, one of the main
drawbacks common to application of the anammox process is
requirement of a long start-up period due to mainly slow
growth rates of anammox bacteria (Egli et al., 2001; van Dongen
et al., 2001) (the doubling time was reported to be approxi-
mately 11 days) (Strous et al., 1998). Additionally, since
anammox bacteria are strictly anaerobes and autotrophs, they
are very difficult to be cultured. To promptly establish
anammox reactors, appropriate seeding sludge or starter
cultures must be selected and used, and sufficient amounts
of anammox bacteria must be efficiently retained in the
reactors. However, rational procedures for start-up and opti-
mization of anammox reactors have not been developed yet.
The objectives of this study were therefore (i) to select
appropriate seeding sludge for the rapid start-up of anammox
reactors; (ii) to promptly establish the anammox reactors with
the selected sludge and optimize the reactor performance;
and (iii) to characterize the microbial community structure
and in situ activity of the anammox biofilms. To achieve these
objectives, up-flow fixed-bed biofilm column reactors with
nonwoven fabric sheets as biomass carrier were used for
cultivation of anammox bacteria, and the reactor perfor-
mance was monitored over 1 year. After anammox reactors
were fully developed, the community structure, spatial
organization and in situ activity of the anammox biofilms
were analyzed by the combined use of a full-cycle of 16S rRNA
approach and microelectrodes.
2. Materials and methods
2.1.Sludge samples
Sludge samples were collected from 11 different wastewater
treatment plants (WWTPs) to select appropriate seeding
sludge for establishment of up-flow fixed-bed biofilm reactors
(Table 1). To quantify anammox bacterial population, the 16S
rRNA gene copy numbers of anammox bacteria in the sludge
samples were determined by real-time quantitative PCR (RTQ-
PCR) with anammox specific primers as described below.
Carbon to nitrogen ratios (C/N) (g/g) of influent wastewater
ARTICLE IN PRESS
Table 1 – Summary of quantification of anammox bacterial 16S rDNA copy numbers and influent C/N ratios in sludges
taken from different wastewater treatment plants
SludgeSampling pointLocationa
Wastewater Copies/mg-dry
sludge
C/N
(g/g)
Dayb
MNRRc(kg-
Nm?3d?1)
APilot-scale, membrane
bioreactor
Full-scale, denitrifying
basin
Full-scale, nitrifying RBCe
Soseigawa WWTPDomestic2.7?107
0.9 NDd
ND
BHakodate-bay
WWTP
Teine landfill
leachate TP
Kuriyama WWTP
Hakodate-nanbu
WWTP
Kiritappu WWTP
Ebetsu WWTP
Domestic 3.1?107
5.1 2230.032
C
Landfill
leachate
Domestic
Domestic
1.7?106
2.0 1710.004
D
E
Full-scale, oxidation ditch
Full-scale, denitrifying
basin
Full-scale, oxidation ditch
Full-scale, anaerobic
digester
Full-scale, aeration tank
1.1?108
3.0?107
1.1
4.0
64
143
0.054
0.044
F
G
Domestic
Domestic
5.0?107
1.9?107
0.9
1.5
203
107
0.026
0.039
HKitasorachi night
soil TP
Esan night soil TP
Night soil2.4?107
3.0540.058
IFull-scale, denitrifying
basin
Full-scale, RBCe
Full-scale, denitrifying
basin
Night soil1.1?107
12.4 69 0.048
J
K
Chino WWTP
Higashihiroshima
WWTP
Domestic
Domestic
9.7?106
1.6?108
6.8
0.9
ND
37
ND
0.083
aAll treatment plants were located in Japan. WWTP, wastewater treatment plant; TP, treatment plant.
bThe day when complete consumptions of 30mgL?1of NH4
cMaximum nitrogen removal rates determined in anammox activity tests under anoxic conditions.
dNot detected.
eRotating biological contactor.
+-N and NO2
?-N were observed.
WAT E R R E S E A R C H 41 (2007) 1623– 1634
1624

Page 3

were also measured because organic compounds effect the
activity of anammox bacteria (van de Graaf et al., 1996).
2.2.
anammox bacteria by RTQ-PCR
Quantification of 16S rRNA gene copy number of
Direct DNA extractions were performed using the Fast DNA
spin kit for soil (BIO101, Qbiogene Inc., Carlsbad, CA)
according to the manufacturer’s protocol. RTQ-PCR assay
was performed for quantification of 16S rRNA genes of
anammox bacteria, as previously described by Tsushima et
al. (2007). Briefly, the RTQ-PCR assays were performed in
duplicate with a total volume of 25mL reaction mixture
containing 12.5mL of buffers supplied with a SYBRsgreen
PCR master mix kit (Applied Biosystems, Forster City, CA),
300nM of primers and 2.5mL of sample DNA in MicroAmp
Optical 96-well reaction plates with optical caps (Applied
Biosystems). Specific primers for anammox bacteria were
previously described by Tsushima et al. (2007). The template
DNA was amplified and monitored with an ABI Prism 7000
Sequence Detection System (Applied Biosystems). The PCR
conditions were as follows: 2min at 501C and 10min at 941C,
followed by 40 cycles of 15s at 941C and 1min at 601C. All PCR
runs included control reactions without template DNA to test
possible nonspecific amplification. Standard curves for ana-
mmox bacteria were constructed using a series of DNA
concentrations prepared from the plasmid vector carrying a
16S rRNA gene of a anammox bacterium related clone, which
was obtained from the previously constructed clone library
(Tsushima et al., 2007).
2.3.Anammox activity test
Anammox activities of the sludges taken from the WWTPs
were analyzed in standard batch cultures. Each sludge sample
was diluted to approximately 20mg-MLSSL?1
anammox nutrient medium, and 95mL of the mixed liquid
was transferred into 100-mL serum bottles. Oxygen was
removed from the mixed liquid by purging with N2 gas
(99.99%) for 30min. The serum bottles were sealed tightly
with butyl rubber caps. The anammox nutrient medium
consisted of (NH4)2SO4 (30–84mg-NL?1), NaNO2 (30–84mg-
NL?1), KHCO3(500mgL?1), KH2PO4(27mgL?1), MgSO4?2H2O
(300mgL?1), CaCl2?2H2O (180mgL?1), and 1mL of trace
element solution I and II (van de Graaf et al., 1996). pH was
adjusted to 7.5 with 1N H2SO4. Each sample was incubated at
371C in the dark and the concentrations of NHþ
NO?
medium in the serum bottles was exchanged with a fresh one
in an anaerobic chamber when the total amount of nitrogen
consumption exceeded 140mg-N per 100-mL serum bottle
(Tsushima et al., 2006).
with the
4, NO?
2and
3were periodically monitored during the incubation. The
2.4.Anammox biofilm reactors
Two up-flow fixed-bed glass biofilm column reactors (reactors
I and II) were operated in parallel (Fig. 1). The reactors had an
inner diameter of 50mm, height of 500mm and the liquid
volume of 0.8L. Nonwoven fabric sheets (12.5?2.0?0.8cm;
Japan Vilene Co., Ltd., Tokyo, Japan) were used as support
materials for biofilms. They comprised a total surface area of
500cm2and an interstitial volume of 760cm3, corresponding
to the filling ratio of 5% of total reactor volume.
Eighteen milligrams (dry weight) of the sludge with the
highest 16S rRNA gene copy numbers of anammox bacteria
among 11 sludge samples was inoculated to reactor I. Two
liters of the effluent from reactor I was collected after 3
months of operation, concentrated to 50mL (corresponding to
approximately 5mg-dry weight of solid) and inoculated into
reactor II. The temperature was maintained at 371C for both
the reactors. PharMeds
tubing (made of thermoplastic
elastomer polypropylene) was used to minimize oxygen
penetration. Both the reactors were fed with the anammox
nutrient medium as described above. The medium was
flushed with N2gas from line A for at least 1h to achieve
dissolved oxygen concentration below detection limit (oca.
0.5mgL?1) (Fig. 1). Furthermore, a N2-filled gas bag (GL
science, Tokyo, Japan) was connected with line B to avoid
oxygen getting into the medium reservoir (Fig. 1). The pH was
adjusted in the range 7.0–7.5 with 1N H2SO4. The concentra-
tions of NHþ
550mg-NL?1
and from 20 to 460mg-NL?1, respectively,
corresponding to the influent NHþ
0.75–1.25. The hydraulic retention times (HRTs) of the reactors
were reduced gradually from 8 to 1.4h in reactor I and 8 to
0.2h in reactor II, respectively. The nitrogen (NHþ
loading rates were 0.1–9.4kg-Nm?3day?1in reactor I and
0.1–58.5kg-Nm?3day?1in reactor II. Samples were obtained
from the influent and effluent lines and analyzed for the
concentrations of NHþ
carbon (DOC). The 16S rRNA gene copy numbers of anammox
bacteria in the biofilms were determined by RTQ-PCR (see
above) after the anammox reaction was clearly observed.
4and NO?
2were varied with time from 20 to
4/NO?
2
molar ratios of
4+NO?
2)-
4, NO?
2, NO?
3
and dissolved organic
2.5.Analytical procedure
The concentrations of NHþ
by using an ion-exchange chromatography (DX-100, DIONEX,
Sunnyvale, CA) with an IonPac CS3 cation column and IonPac
AS9 anion column after filtration with 0.2-mm-pore size
membranes (Advantec Co., Ltd., Tokyo, Japan). The concen-
tration of DOC was measured by a TOC-analyzer (TOC-5000A;
SHIMADZU, Kyoto, Japan) after filtration with 0.45-mm-pore
size glass fiber filters (Advantec Co., Ltd., Tokyo, Japan).
Suspended solids (SS) concentration was determined accord-
ing to Standard Methods (APHA, 1995).
4, NO?
2, and NO?
3were determined
2.6.DNA extraction and PCR amplification
DNA was extracted from biofilm samples (approximately
0.2mL) taken from the reactors with the Fast DNA spin kit
(BIO101, Qbiogene Inc., Carlsbad, CA) as described in the
manufacturer’s instructions. 16S rRNA gene fragments were
amplified from the extracted total DNA with Taq DNA
polymerase (TaKaRa Bio Inc., Otsu, Japan) using Planctomyce-
tals-specific primer set pla46f (Neef et al., 1998) and 1492r
(Weisburg et al., 1991), and bacterial primer set 11f (Kane et
al., 1993) and 1492r (Weisburg et al., 1991), respectively. The
PCR conditions targeted for bacteria were as follows: 5min
initial denaturation at 941C, 30 cycles of 1min at 941C, 1min
ARTICLE IN PRESS
WATER RESEARCH 41 (2007) 1623– 1634
1625

Page 4

at 501C, and 110s at 721C. Final extension was carried out for
4min at 721C. The PCR conditions targeted for anammox
bacteria were as follows: 5min initial denaturation at 941C, 25
cycles of 1min at 941C, 1min at 501C, and 70s at 721C. Final
extension was carried out for 4min at 721C. The PCR products
were electrophoresed on a 1% (wt/vol) agarose gel.
2.7.
phylogenetic analysis
Cloning and sequencing of 16S rRNA gene and
PCR products were ligated into a pCR-XL-TOPOsvector and
transformed into ONE SHOT Escherichia coli cells according to
the manufacturer’s instructions (TOPO XL PCR cloning;
Invitrogen, Carlsbad, CA). Nucleotide sequencing was per-
formed with an automatic sequencer (ABI Prism 3100 Avant
Genetic Analyzer; Applied Biosystems). All sequences were
checked for chimeric artifacts by the CHECK_CHIMERA
program in the Ribosomal Database Project (Maidak et al.,
1997). Almost full-length sequences (ca. 1500bp) were com-
pared with similar sequences of the reference organisms by
BLAST search (Altschul et al., 1990). Sequences with more
than 97% sequence similarity were grouped into the same
operational taxonomic unit (OTU) by using Similarity Matrix
program from the Ribosomal Database Project (Maidak et al.,
1997). The sequences of each representing OTU were aligned
with the CLUSTAL W package (Thompson et al., 1994) and
used for phylogenetic analysis. A phylogenetic tree was
constructed by the neighbor-joining method (Saito and Nei,
1987). Bootstrap resampling analysis for 100 replicates was
performed to estimate the confidence of tree topologies.
2.8. Microelectrode measurements
The concentration profiles of O2, NHþ
were measured using microelectrodes as described by Okabe
et al. (1999). Clark-type microelectrodes for O2 with a tip
diameter of approximately 15mm and 90% response time of
shorter than 1s were prepared and calibrated as described by
Revsbech (1989). The LIX-type microelectrodes for NHþ
NO?
ing to the protocol described by de Beer et al. (1997) and Satoh
et al. (2003).
4, NO?
2, NO?
3, and pH
4, NO?
2,
3, and pH were constructed, calibrated, and used accord-
ARTICLE IN PRESS
500 mm
50 mm
Flow
Effluent
Gas bag
Nutrient
medium
Nonwoven
fabric carrier
Line ALine B
P
Sampling
Fig. 1 – Schematic drawing of an up-flow fixed-bed column reactor. The nutrient medium was purged with N2gas from line A
for at least 1h to achieve the concentration of dissolved oxygen below 0.5mgL?1. A N2-filled gas bag was connected with line
B to prevent O2penetration during the operation.
WAT E R R E S E A R C H 41 (2007) 1623– 1634
1626

Page 5

The microelectrode measurements in anammox biofilms
were performed in a glass column reactor. The reactor
volume was 64cm3(length 13.5cm, inner diameter 2.5cm).
A sampling port (1.0cm height, inner diameter 1.0cm) was
installed in the reactor (at ca. 6cm from the inlet side), which
was filled with 5% agarose gel. The microelectrodes were
directly inserted into the biofilm through this agarose gel-
filled samplingport.Nonwoven
2.0?0.8cm) with fully developed anammox biofilms were
taken from reactor I on day 392 and installed in the glass
column reactor and pre-incubated for a week. The same
anammox nutrient medium containing 100mgL?1of NHþ
and NO?
was 0.8h and the nitrogen-loading rate was maintained at ca.
6.0kg-Nm?3day?1during the measurements. The tempera-
ture was maintained at 371C.
Based on the measured NHþ
profiles, the total NHþ
2
(J (mg-Ncm?2h?1)) were calculated using Fick’s first law of
diffusion. Net volumetric NHþ
(R (mg-Ncm?3h?1)) in the biofilms were also calculated using
Fick’s second law of diffusion as described previously by
Lorenzen et al. (1998). The molecular diffusion coefficients
used for the calculations were 1.47?10?5cm2s?1for NHþ
1.30?10?5cm2s?1for NO?
water at 371C.
fabricsheets(12.5?
4-N
2-N was fed into the reactor. The HRT of the reactor
4, NO?
and NO?
2and NO?
3concentration
conversion rates
4, NO?
3
4, NO?
2and NO?
3conversion rates
4,
2, and 1.30?10?5cm2s?1for NO?
3in
2.9.Fixation and cryosectioning of biofilm samples
Biofilm samples obtained from the reactor were fixed in 4%
paraformaldehyde solution for 24h at 41C, washed three
times with phosphate-buffered saline (PBS) (10mM sodium
phosphate buffer, 130mM sodium chloride; pH 7.2), and
embedded in Tissue-Tek OCT compound (Sakura Finetek,
Torrance, CA) overnight to infiltrate the OCT compound into
the biofilm, as described previously (Okabe et al., 1999). After
rapid freezing at ?211C, 30-mm-thick vertical thin sections
were prepared with a cryostat (Reichert–Jung Cryocut 1800,
Leica, Bensheim, Germany).
2.10.
hybridization (FISH)
Oligonucleotide probes and fluorescence in situ
The 16S rRNA-targeted oligonucleotide probes used in this
study were EUB338 (Amann et al., 1990), EUB338-II (Daims et
al., 1999), EUB338-III (Daims et al., 1999) for mostly eubacteria,
Amx820 (Schmid et al., 2000) for anammox bacteria, Nso190
(Mobarry et al., 1996) and Nse1472 (Juretschko et al., 1998) for
aerobic ammonia-oxidizing bacteria (AAOB). To detect all
bacteria, the probes were used in the equimolar mixture
together with probes EUB338, EUB338II, and EUB338III. The
probes were labeled with fluorescein isothiocyanate (FITC) or
tetramethylrhodamine 5-isothiocyanate (TRITC) at the 50end.
In situ hybridization was performed according to the proce-
dure described by Amann (1995) and Okabe et al. (1999). A
model LSM510 confocal laser-scanning microscope (CLSM,
Carl Zeiss, Oberkochen, Germany), equipped with an Ar ion
laser (488nm) and HeNe laser (543nm), was used. The average
surface area fraction of probe-hybridized cells was deter-
mined from at least 20 randomly chosen LSM projection
images of each cross section of the biofilm samples using
image analysis software provided by Zeiss (Okabe et al., 1999).
2.11.Nucleotide sequence accession numbers
The GenBank/EMBL/DDBJ accession numbers for the 16S
rRNA gene sequences of the three OTU sequences used for
the phylogenetic tree analysis are AB269933–AB269935.
3. Results
3.1. Screening of seeding sludge
To select an appropriate seeding sludge for rapid start-up of
anammox reactors, the 16S rRNA gene copy numbers of
anammox bacteria in the sludge samples obtained from 11
different treatment plants were quantified by RTQ-PCR assay
(Table 1). The 16S rRNA gene copy numbers of anammox
bacteria and their correlation with the average C/N ratios of
the influent wastewater are summarized in Table 1. The 16S
rRNA gene copy numbers were in the range of 106–108copies
per mg of dry sludge. The highest copy number (1.6?108
copies per mg of dry sludge) was detected in sludge K
obtained from a denitrifying basin in a domestic WWTP
located in Higashihiroshima, Japan. The influent C/N ratio of
sludge K was the lowest (0.9).
Anammox activity tests revealed that sludge K completely
consumed 30mg-NL?1of NHþ
the initial 37 days, which was the fastest among 11 sludge
samples analyzed (Table 1). The maximum nitrogen removal
rate (MNRR) reached 0.083kg-Nm?3day?1after day 100,
which was also the highest among 11 sludge samples. In
the activity tests using sludge C that had the lowest copy
number (1.7?106copies per mg of dry sludge), no anammox
reaction was detected during more than 100 days of incuba-
tion (Table 1). In addition, in the case of sludge J (9.6?106
copies per mg of dry sludge), no anammox reaction was
detected at all. Thus, we decided to use sludge K as an
inoculum for following the anammox reactor experiments.
4and NO?
2, respectively, within
3.2. Performance of anammox reactors
Reactor I, which was inoculated with sludge K, was operated
for 392 days (Fig. 2A). On day 40, the simultaneous removal of
NHþ
the anammox reaction. The total nitrogen removal rate
gradually increased with increasing the nitrogen-loading rate
after 55 days. After around 60 days, the color of biomass
changed from light brown to the characteristic red color. The
16S rRNA gene copy number of anammox bacteria attached
on the nonwoven fabric sheet was 8.371.2?1012copies cm?3
(n ¼ 4) on day 392. The maximum total nitrogen removal rate
of 6.2kg-Nm?3day?1was obtained on day 392 when the
nitrogen-loading rate was 9.4kg-Nm?3day?1(HRT, 1.4h).
In reactor II, which was inoculated with the effluent
biomass of reactor I after 3-month operation, the simulta-
neous removal of NHþ
(Fig. 2B). This is because a small amount of biomass (5mg dry
weight of solid) was inoculated into reactor II. Anammox
4and NO?
2was clearly observed, indicating occurrence of
4and NO?
2was observed after 50 days
ARTICLE IN PRESS
WATER RESEARCH 41 (2007) 1623– 1634
1627

Page 6

bacteria were enriched at only the inlet side of nonwoven
fabric sheets when HRT was 8h, while anammox bacteria
gradually proliferated throughout the reactor after HRT was
reduced to 1h (increasing the nitrogen-loading rate). As a
result, the maximum total nitrogen removal rate of 26.0kg-
Nm?3day?1was attained after 247 days when the nitrogen-
loading rate was 58.5kg-Nm?3day?1(HRT, 0.24h). Thereafter,
the nitrogen removal rate decreased to 5.8kg-Nm?3day?1
because HRTwas increased to 1h. The nitrogen removal rate
gradually increased to 14.0kg-Nm?3day?1again by increas-
ing the influent NHþ
at 1h). However, the nitrogen removal rate no longer
increased (actually decreased to 9.8kg-Nm?3day?1) when
the influent nitrogen (NHþ
creased up to 1000mg-NL?1(the nitrogen-loading rate was
24.0kg-Nm?3day?1).
Fig. 3 showed the nitrogen removal rates in reactor II
attained at various influent nitrogen concentrations during
days 259–362 (HRT was fixed at 1h). The nitrogen removal
ratesteadilyincreasedwith
nitrogen concentration up to 900mg-NL?1, above which it
decreased. This indicated that there was a limitation to
enhance the nitrogen removal rate by only increasing the
influent nitrogen concentration at a fixed HRT. Fig. 4 showed
the nitrogen removal rates and produced DOC concentrations
in reactor II at different HRTs with fixed influent nitrogen
(NHþ
(M/M)). As expected, the nitrogen removal rate increased as
HRT was reduced (Fig. 4). Reduction of HRT resulted in
4and NO?
2concentrations (HRTwas fixed
4+NO?
2) concentration was in-
increasingtheinfluent
4+NO?
2) concentration (900mg-NL?1; NHþ
4/NO?
2¼ ca. 1.2
decrease in the DOC concentrations in the reactor effluent
due to a dilution effect. Fig. 5 showed the nitrogen
removal rates at different influent NHþ
a fixed HRT (8h) and a fixed influent nitrogen (NHþ
4/NO?
2molar ratios at
4+NO?
2)
ARTICLE IN PRESS
400
0
2
4
6
8
10
Loading, removal rates
(kg-N m-3 day-1)
100
Nitrogen removal ratio (%)
60
80
40
20
0
10
0
HRT (hours)
6
8
4
2
050 150200 250300 350100
Time (day)
0
Nitrogen removal ratio (%)
100
10
0
HRT (hours)
6
8
4
2
60
80
40
20
Loading, removal rates
(kg-N m-3 day-1)
0
20
30
40
50
60
10
050150200250300 350100
Time (day)
400
Fig. 2 – Rates of influent nitrogen (NHþ
reactor II (B). The filled triangles represents the nitrogen (NHþ
represents the nitrogen removal rate; and the gray area represents the nitrogen removal efficiency in the reactors. Solid line
represents HRT in the reactor. The effluent of reactor I after a 3-month operation was inoculated into reactor II.
4, NO?
2, and NO?
3)-loading and effluent nitrogen removal in the column reactor I (A) and
4, NO?
2, and NO?
3)-loading rate into the reactor; the empty circles
Influent nitrogen concentration (mg-N L-1)
400500600 700800900 1000
4
8
0
Loading, Nitrogen removal rate
(kg-N m-3 day-1)
12
16
20
24
28
0
50
100
150
200
250
Residual nitrite concentration
(mg-N L-1)
Fig. 3 – The nitrogen removal rates at different influent
nitrogen (NHþ
between day 259 and 362 when HRTwas fixed at 1h. Error
bars indicate the standard deviations (n ¼ ¼ 3–6). The whole
bars represent total nitrogen-loading rates and white parts
represent nitrogen removal rates. Line plots (K) represent
residual nitrite concentrations. Nitrogen removal
efficiencies were 57% (400mg-NL?1), 61% (500mg-NL?1),
63% (600mg-NL?1), 64% (700mg-NL?1), 62% (800mg-NL?1),
58% (900mg-NL?1), and 42% (1000mg-NL?1), respectively.
4-N+NO?
2-N) concentrations in reactor II
WAT E R R E S E A R C H 41 (2007) 1623– 1634
1628

Page 7

concentrations (800mg-NL?1). The total nitrogen removal
rate increased from 1.2 to 1.5kg-Nm?3day?1with increasing
the influent NHþ
(po0.05).
4/NO?
2molar ratio from 0.75 to 1.25 (M/M)
3.3.Phylogenetic analysis
Two 16S rRNA gene clone libraries (an anammox bacterial
clone library and a bacterial clone library) were constructed
from the biofilms taken from reactor I on day 392 using the
Planctomycetals specific primer set (Pla46f and 1492r) and
bacterial specific primer set (11f and 1492r), respectively. No
chimeric sequence was observed in both clone libraries.
Sixteen clones were randomly selected from the anammox
bacterial clone library and sequenced. All clones were
grouped into one OTU (OTU1 in Fig. 6) on the basis of more
than 97% sequence similarity. The clones grouped into OTU1
were related to the Candidatus Brocadia anammoxidans be-
longing to the order Planctomycetales with 95% sequence
similarity (Fig. 6). Moreover, 41 clones were randomly selected
from the bacterial clone library. Of these 41 clones, 24 clones
were related to anammox bacteria and grouped into two
OTUs (OTU2 and OTU3 in Fig. 6). The clones grouped into
OTU2 were also related to the Candidatus Brocadia anammox-
idans with 95% sequence similarity (Fig. 6). The clone
sequences of OTU1 and OTU2 were closely related to each
other (more than 99.6% similarity). OTU3 was also closely
related to OTU1 and OTU2 with 98.9% and 97.4% sequence
similarity, respectively. We also obtained clone sequences
closely related to OTU1, OTU2, and OTU3 (more than 98.5%
similarity) from the original sludge K and reactor II on day
247. This indicates that the enriched anammox bacteria in
reactor I was originated from sludge K that was inoculated.
The remaining 17 clones were affiliated with uncultured
clones belonging to the Betaproteobacteria such as Thauera sp.
and Acidovorax sp. with 90–99% sequence similarity (data not
shown). These bacteria were not observed from other
anammox-enriched cultures (Fujii et al., 2002; Strous et al.,
2006).
3.4.In situ anammox activity
Fig. 7A shows the concentration profiles of O2, NHþ
NO?
mined under as realistic conditions (i.e., water flow, water
chemistry, temperature and so on). The concentrations of
NHþ
(at ca. 6cm from the inlet port) were 25.6, 18.8, and 16.8mg-
NL?1, respectively. The average nitrogen removal rate of the
reactor was 3.0kg-Nm?3day?1during the measurements. O2
concentration was under the detection limit throughout the
biofilm at this measuring point. NHþ
profiles indicated that both NO?
simultaneously decreased in the biofilm, indicating occur-
rence of the anammox reaction. The NHþ
however, gradually increased below a depth of 800mm due to
probably anaerobic mineralization of organic compounds.
The NO?
3
concentration steadily increased in the upper
800mm. pH increased from 7.3 to 7.8 in the upper 700mm of
the biofilm. The similar concentration profiles were repeat-
edly obtained when we conducted several measurements
under the same condition.
The spatial distributions of net volumetric NHþ
NO?
NHþ
4, NO?
2,
3, and pH in the anammox biofilm, which were deter-
4, NO?
2, and NO?
3in the bulk water at the measuring point
4and NO?
2and NHþ
2concentration
4concentrations
4concentration,
4, NO?
2and
3consumption rates showed occurrence of simultaneous
4and NO?
2consumption and NO?
3production in the upper
ARTICLE IN PRESS
0
0.5
1.0
1.5
2.0
2.5
Produced DOC in the reactors
(mg-C L-1)
8
20
24
16
12
8
4
0
Loading, Nitrogen removal rate
(kg-N m-3 day-1)
14
HRT (hour)
Fig. 4 – The nitrogen removal rates and produced DOC
concentrations in reactor II at different HRTs (1, 4, and 8h).
The whole bars represent loading rates and the white parts
represent the nitrogen removal rates. Line plots (K)
represent the produced DOC concentrations in the reactor.
Error bars indicate the standard deviations (n ¼ ¼ 10). The
produced DOC concentration was obtained by subtracting
the effluent DOC concentration from the influent DOC
concentration. The nitrogen removal efficiencies and the
residual nitrite concentrations were 118mg-NL?1and 57%
at HRT 1.0h, 53mg-NL?1and 69% at HRT 4.0h, 129mg-
NL?1, and 60% at HRT 8.0h, respectively.
0.75-0.850.95-1.05 1.15-1.25
Influent NH4+/NO2- ratio
Eff.Inf.Eff.Inf.Eff.Inf.
800
600
400
200
0
Nitrogen compounds concentrations
(mg-N L-1)
NH4+
NO2-
NO3-
Fig. 5 – The nitrogen concentrations in influent and effluent
at different influent NHþ
HRTwas fixed at 8h and the influent total nitrogen
concentration was fixed at 800mg-NL?1(black; ammonium,
gray; nitrite, white; nitrate concentrations, respectively).
Error bars indicate the standard deviations (n ¼ ¼ 8). The
nitrogen removal rates and nitrogen removal efficiencies
were 1.2kg-Nm?3day?1and 50% (NHþ
1.3kg-Nm?3day?1and 55% (0.95–1.05), 1.5kg-Nm?3day?1
and 63% (1.15–1.25), respectively.
4/NO?
2molar ratios in reactor II when
4/NO?
2; 0.75–0.85),
WATER RESEARCH 41 (2007) 1623– 1634
1629

Page 8

800mm (data not shown). Based on the microprofiles, the total
consumption rates of NHþ
be 2.04, 2.41, and ?0.67g-Nm?2day?1(1:1.2:?0.33) in the
anammox zone.
4, NO?
2, and NO?
3were calculated to
3.5. Spatial distribution of anammox bacteria
The
biofilm obtained from reactor I on day 392 was examined
by FISH using EUB338 probe mixture (EUB338mix) and
Amx820 probe specific to anammox bacteria. A vertical
cross-section image ofthe
geneous structure consisting of bacterial cells and void
spaces (Fig. 8A). As shown in Fig. 8B, almost all bacteria
detected with EUB338mix (EUB338+EUB338II+EUB338III) probe
were simultaneously hybridized with Amx820 probe and were
distributed throughout the biofilm. The Amx820 probe-
hybridized cells were mainly present in the form of spherical
dense microcolonies (Fig. 8C). The cells hybridized with
Amx820 accounted for 89%, 74%, and 72% of the total bacteria
detected with EUB338mix probe at depths of 0–1, 1–3, and
3–4mm from the biofilm surface, respectively. The bacteria
hybridized with EUB338mix probe but not hybridized with
Amx820 probe (shown in green in Fig. 8B) coexisted with
anammox bacteria mainly in the surface layer and the
biomass–liquid interface. To identify these bacteria, the
probes specific for AAOB were applied. As a result, AAOB
hybridized with NSO190 was detected mainly in the surface
biofilm. In addition, the NSO190-hybridized AAOB were also
hybridized with Nse1472, indicating that they were most
likely Nitrosomonas eutropha, N. europaea or N. halophila
(shown in red in Fig. 8C). The presence of the unidentified
eubacteria hybridized with only EUB338mix probe (except
AAOB) became evident in the deeper part of the biofilm
(shown in green in Fig. 8D).
spatial distribution ofanammox bacteriainthe
biofilm revealedahetero-
ARTICLE IN PRESS
Uncultured anoxic sludge bacterium KU1 [AB054006]
Candidatus Brocadia anammoxidans [AF375994]
Anaerobic ammonium-oxidizing planctomycete [AJ131819]
Uncultured Planctomycetales bacterium [AB176696]
Planctomycete KSU-1 [AB057453]
Anaerobic ammonium-oxidizing planctomycete [AJ250882]
Candidatus Kuenenia stuttgartiensis [AF375995]
Candidatus Scalindua wagneri [AY254882]
Candidatus Scalindua brodae [AY254883]
Candidatus Scalindua sorokinii [AY257181]
Clone HU-BU2 OTU2 (21/41) [AB269933]
Clone HU-PU4 OTU1 (16/16) [AB269935]
Clone HU-BU14 OTU3 (3/41) [AB269934]
Candidatus Brocadia fulgida [DQ459989]
Candidatus Jettenia asiatica [DQ301513]
Candidatus Anammoxoglobus propionicus [DQ317601]
0.05
Fig. 6 – Phylogenetic tree of anammox bacteria showing the positions of the clones obtained from the biofilm in reactor I after
392-day operation. The tree was generated by using 1429bp of the 16S rRNA and neighbor-joining method. The scale bar
represents 5% sequence divergence. The filled and empty circles at the nodes represent bootstrap values higher than 95% and
80%, respectively (100 times resampling analysis). The Aquifex aeolicus sequence served as the outgroup for rooting the tree.
Numbers in parentheses indicate the frequency of appearance of the identical clones in the total clones analyzed with
specific primer set.
Fig. 7 – Steady-state concentration profiles of O2, NHþ
NO?
is at a depth of 0mm.
4, NO?
2,
3and pH in the anammox biofilm. Surface of the biofilm
WAT E R R E S E A R C H 41 (2007) 1623– 1634
1630

Page 9

4. Discussion
4.1.Anammox capacity of the reactors
In this study, we have successfully developed anammox
reactors promptly and achieved a total nitrogen removal rate
of 26.0kg-Nm?3day?1within 250 days. To our knowledge,
such a high volumetric nitrogen removal rate has never been
reported before. To date, Sliekers et al. (2003) have reported a
nitrogen conversion rate of 8.7kg-Nm?3day?1in a gas-lift
reactor, which was one third of our rate. Such a high
anammox rate was probably attributed to the high density
of anammox bacteria (ca. 16gVSSL?1, more than 70% of total
bacteria was anammox bacteria) retained in the reactors by
using nonwoven fabric sheets as biofilm carrier as well as
applying the high total nitrogen loading rate (especially
nitrite), compared to other reactors (van de Graaf et al.,
1996; Sliekers et al., 2003). The high nitrogen-loading rate
prevents substrate transport limitation in the biofilms
throughout the reactor (Nicolella et al., 2000). The specific
nitrogen removal rate in this study was calculated to be at
1.6kg-Nkg-VSS?1day?1, which is also higher than the rates
reported in the literature (Third et al., 2005). This high activity
should be correlated to the growth rate. In our previous study,
we have determined the doubling time of the anammox
bacteria enriched from the same sludge K to be 3.6–5.4 days
(Tsushima et al., 2007), which were shorter than the
ARTICLE IN PRESS
Fig. 8 – Phase contrast image (A) and CLSM images (B and D) of vertical sections (30-lm thick) of the anammox biofilm. (B)
FISH with TRITC-labeled EUB338 probe mixture (green) and FITC-labeled Amx820 probe (red) (specific to anammox bacteria;
Candidatus Brocadia anammoxidans and Candidatus Kuenenia stuttgartiensis). Yellow signals result from binding both probes
to one cell. (C) FISH with TRITC-labeled Amx820 probe (green) and FITC-labeled Nse1472 probe (red) (specific to Nitrosomonas
eutropha, N. europaea and N. halophila). (D) FISH with TRITC-labeled EUB338 probe mixture (green) and FITC-labeled Amx820
probe. The biofilm surfaces are at the top of all images.
WATER RESEARCH 41 (2007) 1623– 1634
1631

Page 10

previously reported value (ca. 11 days) (Strous et al., 1999). In
addition, the appropriate seeding sludge was selected and
used after quantifying 16S rRNA gene copy number of
anammox bacteria followed by anammox activity test. HRT
and influent NHþ
obtaining efficient nitrogen removal in this study. The
nitrogen removal rate was decreased when the influent
nitrogen concentration was increased to 1000mgL?1(Fig. 3).
This is probably due to the toxicity of the residual nitrite
(224710mg-NL?1), which was above the inhibitory nitrite
concentration levels (70–180mg-NL?1) suggested previously
by Strous et al. (1999), or possibly an accumulation of
unknown by-products derived from the anammox reaction.
Shortening HRT led to the higher nitrogen-loading rates,
which could prevent a possible nitrite limitation. An alter-
native explanation could be that unknown by-products
derived from anammox reaction, which possibly cause self-
inhibition of anammox bacterial activity, were washed out at
shorter HRTs (Fig. 4). Other studies also reported that the
performance of anammox reactors was enhanced by reducing
HRT stepwise (Sliekers et al., 2003). In fact, although Rouse
et al. (2003) operated an anammox reactor with recycling the
effluent, the nitrogen removal rate could hardly increase even
though sufficient nitrogen (NHþ
maintained. Furthermore, the higher anammox rate could be
achieved by the higher influent NHþ
This is probably because the high influent NHþ
the lower nitrite-loading rate, which led to less nitrite
inhibition as demonstrated by the lower nitrite concentration
in the effluent (Fig. 5). In addition, AAOB present in the
surface biofilm could convert a part of NHþ
anammox bacteria convert the remaining NHþ
NO?
further oxidation of NO?
4/NO?
2molar ratio could play a vital role in
4and NO?
2)-loading rate was
4/NO?
2molar ratio (Fig. 5).
4/NO2ratio gave
4to NO?
4and produced
2, and then
2to N2. Maintaining NHþ
4concentration high also prevents
2to NO?
3.
4.2.In situ anammox activity
The microelectrode measurements clearly demonstrated that
anammox reaction (simultaneous consumption of NHþ
NO?
results of 16S rRNA gene cloning and FISH analyses. There
was, however, on partial nitrification (visible peak of NO?
within the surface biofilm although AAOB were present (Fig. 8C).
This is because the rate of aerobic ammonia oxidation was
lower than the rate of anammox reaction due to the
limitation of O2flux. Similarly, no visible peak of NO?
detected in anammox bacterial granules in oxygen-limited
sequencing batch reactors (SBRs) (Nielsen et al., 2005). The
stoichiometric ratio of total NHþ
sumption and NO?
which was similar to the previously reported value for the
anammox reaction (1:1.3170.06:0.2270.02) (van de Graaf et
al., 1996). The slightly lower NO?
attributed to partial nitrification (i.e., conversion of NHþ
NO?
The NO?
2
consumption rate (2.41g-Nm?2day?1) in the
biofilm in this study was higher than the rates (0.2–0.5g-
Nm?2day?1) of anammox bacterial granules in the oxygen-
limited SBRs as determined by microelectrodes (Nielsen et al.,
2005). The NHþ
4
consumption rate of the biofilm (2.04g-
4and
2) occurred in the biofilm, which was consistent with the
2)
2was
4consumption, NO?
2con-
3production in the biofilm was 1:1.2:0.33,
2 consumption could be
4to
2) in the biofilm.
Nm?2day?1) was in the same order of magnitude as aerobic
NHþ
4
oxidation rats of nitrifying biofilms in the previous
studies, in which 2.4g-N of (NO?
(Schramm et al., 1996) and 0.4g-N of NHþ
et al., 1999) were detected, respectively. These high rates were
probably attributed to the inherent high specific conversion
rate of anammox bacteria (Jetten et al., 1999) and the higher
density of anammox bacteria in the reactors. In addition, the
electron acceptor for anammox reaction (i.e., NO?
be limited in the biofilm due to much higher solubility of NO?
than that of O2for AAOB, which could be advantageous for
treating wastewater containing high NHþ
FISH analysis clearly showed that AAOB and anammox
bacteria coexisted mainly in the surface biofilm. AAOB
probably consumed a trace amount of oxygen (o0.5mgL?1)
and oxidized NHþ
suitable microenvironments. While anammox bacteria con-
verted the toxic NO?
distribution of AAOB and anammox bacteria was also found
in microbial granules in the oxygen-limited SBRsfc (Sliekers
et al., 2002; Nielsen et al., 2005). The spatial distribution and
types of AAOB could be governed by O2concentration in the
bulk liquid. FISH and phylogenetic analyses also revealed the
presence of other bacteria that were not affiliated with
anammox bacteria and AAOB in the deeper part of the
biofilm. Since no organic compounds except EDTA was
supplied to the reactor, these bacteria (probably heterotrophs
based on the 16S rRNA gene sequence analysis) could utilize
the organic compounds derived from dead cells or produced
by anammox bacteria as the electron donor for denitrifica-
tion. This was speculated from the decrease in NO?
tration in the deeper part of the biofilm (Fig. 7). Otherwise,
anammox bacterial activity might be self-inhibited by accu-
mulation of organic compounds. Further study is necessary to
study the interaction between anammox bacteria and other
heterotrophic bacteria in the biofilm. We are presently
investigating if the coexisting heterotrophic bacteria can
utilize the metabolites of anammox bacteria by using micro-
autoradiography-combined with FISH (MAR-FISH) technique
(Kindaichi et al., 2004; Okabe et al., 2004).
2
plus NO?
4m?2day?1(Okabe
3)m?2day?1
2) could not
2
4concentrations.
4to NO?
2, providing anammox bacteria with
2and remaining NHþ
4to N2gas. A similar
3concen-
4.3.Application of anammox process
Our findings will provide fundamental understanding on how
we could promptly develop and efficiently operate high-rate
anammox biofilm reactors. Application of anammox process
as the major nitrogen removal process in wastewater treat-
ment systems is advantageous in terms of reduced O2(i.e.,
aeration) and organic carbon demands. In future, energy-
saving anammox reactors with small footprint will attain a
sustainable nitrogen removal process.
Furthermore, the results of the present study suggested
that the up-flow fixed-bed biofilm column reactor used in this
study is suited for the combined aerobic ammonia oxidation
(partial oxidation to NO?
single biofilm reactor because of its high biomass retention
capacity. Since the solubility of oxygen in water is relatively
low, the penetration of oxygen in the biofilm will also be
limited. Thus it is likely that when introducing oxygen partial
nitrification will occur in the outer layers of the biofilm and
2) and anammox process within a
ARTICLE IN PRESS
WAT E R R E S E A R C H 41 (2007) 1623– 1634
1632

Page 11

anammox will occur in the inner layers. It is, therefore,
relatively easy to maintain high anammox activity in such
biofilm reactors. This combined process is the so-called
completely autotrophic nitrogen removal over nitrite (CA-
NON) process (Third et al., 2001; Sliekers et al., 2002) and has
been suggested to be a promising nitrogen removal process
for wastewater characterized by a low content of organic
materials (Schmidt et al., 2003). The microbiology, feasibility,
and optimization of the single-reactor CANON process have
been investigated in SBRs (Nielsen et al., 2005; Third et al.,
2001, 2005; Sliekers et al., 2002) and gas-lift reactors (Sliekers
et al., 2003). Similarly, oxygen limited autotrophic nitrifica-
tion/denitrification (OLAND) processes have been reported in
moving-bed reactors (Szatkowska et al., 2006; Rosenwinkel
and Cornelius, 2005) and rotating disk contactors (Pynaert et
al., 2004). However, these processes have never been applied
to fixed-bed biofilm column reactors. Further study on the
feasibility and optimization of the CANON process in the
fixed-bed biofilm reactors is definitely needed.
5. Conclusions
In conclusion, we could successfully develop high-rate
anammox biofilm reactors using the up-flow fixed-bed
biofilm column reactor by inoculating seeding sludge with
high abundance of anammox bacteria for short periods,
operate steadily and achieve a high nitrogen removal rate of
26.0kg-Nm?3day?1within 250 days. Community structure,
spatial distribution and in situ metabolic activity of anammox
bacteria in the biofilm were also analyzed. The results will
provide profound insights into optimal design and operation
of anammox biofilm reactors.
Acknowledgements
This research was partly supported by Grant-in Aid (K1740)
for Research and Technology Development on Waste Manage-
ment from the Ministry of the Environment of Japan. Ikuo
Tsushima was financially supported by the 21st Century
Center Of Excellence (COE) program ‘‘Sustainable Metabolic
System of Water and Waste for Area-Based Society’’ Hokkaido
University, Sapporo Japan.
R E F E R E N C E S
Altschul, S.F., Gish, W., Miller, W., Myers, E.W., Lipman, D.J., 1990.
Basic local alignment search tool. J. Mol. Biol. 215, 403–410.
APHA, 1995. Standard Methods for the Examination of Water and
Wastewater, 19th ed. American Public Health Association.
Amann, R.I., 1995. In situ identification of micro-organisms by
whole cell hybridization with rRNA-targeted nucleic acid
probes. In: Akkerman, A.D.L., van Elsas, J.D., de Bruijn, F.J.
(Eds.), Molecular Microbial Ecology Manual. Kluwer Academic
Publishers, Dordrecht, The Netherlands, pp. 1–15.
Amann, R.I., Binder, B.J., Olson, R.J., Chisholm, S.W., Devereux, R.,
Stahl, D.A., 1990. Combination of 16S rRNA-targeted oligonu-
cleotide probes with flow cytometry for analyzing mixed
microbial populations. Appl. Environ. Microbiol. 56, 1919–1925.
Daims, H., Bru ¨hl, A., Amann, R., Schleifer, K.-H., Wagner, M., 1999.
The domain-specific probe EUB338 is insufficient for the
detection of all bacteria: development and evaluation of a
more comprehensive probe set. Syst. Appl. Microbiol. 22,
434–444.
de Beer, D., Schramm, A., Santegoeds, C.M., Ku ¨hl, M., 1997. A
nitrite microsensor for profiling environmental biofilms. Appl.
Environ. Microbiol. 63, 973–977.
Egli, K., Fanger, U., Alvarez, P.J.J., Siegrist, H., van der Meer, J.R.,
Zehnder, A.J.B., 2001. Enrichment and characterization of an
anammox bacterium from a rotating biological contactor
treating ammonium-rich leachate. Arch. Microbiol. 175,
198–207.
Fujii, T., Sugino, H., Rouse, J.D., Furukawa, K., 2002. Characteriza-
tion of the microbial community in an anaerobic ammonium-
oxidizing biofilm cultured on a nonwoven biomass carrier. J.
Biosci. Bioeng. 94, 412–418.
Fux, C., Boehler, M., Huber, P., Brunner, I., Siegrist, H., 2002.
Biological treatment of ammonium-rich wastewater by partial
nitritation and subsequent anaerobic ammonium oxidation
(anammox) in a pilot plant. J. Biotechnol. 99, 295–306.
Jetten, M.S.M., Strous, M., van de Pas-Schoonen, K.T., Schalk, J.,
van Dongen, U.G.J.M., van de Graaf, A.A., Logemann, S.,
Mkuyzer, G., van Loosdrecht, M.C.M., Kuenen, J.G., 1999. The
anaerobic oxidation of ammonium. FEMS Microbiol. Rev. 22
(5), 421–437.
Juretschko, S., Timmermann, G., Schmid, M., Schleifer, K.H.,
Pommerening-Ro ¨ser, A., Koops, H.P., Wagner, M., 1998. Com-
bined molecular and conventional analyses of nitrifying
bacterium diversity in activated sludge: nitrosococcus mobilis
and Nitrospira-like bacteria as dominant populations. Appl.
Environ. Microbiol. 64, 3042–3051.
Kane, M.D., Poulsen, L.K., Stahl, D.A., 1993. Monitoring the
enrichment and isolation of sulfate-reducing bacteria by
using oligonucleotide hybridization probes designed from
environmentally derived 16S ribosomal RNA sequences. Appl.
Environ. Microbiol. 59, 682–686.
Kartal, B., Rattray, J., van Niftrik, L.A., van de Vossenberg, J.,
Schmid, M.C., Webb, R.I., Schouten, S., Fuerst, J.A., Damste, J.S.,
Jetten, M.S.M., Strous, M., 2006. Candidatus ‘‘Anammoxoglo-
bus propionicus’’ a new propionate oxidizing species of
anaerobic ammonium oxidizing bacteria. Syst. Appl. Micro-
biol. 30, 39–49.
Kindaichi, T., Ito, T., Okabe, S., 2004. Eco-physiological interaction
between nitrifying bacteria and heterotrophic bacteria in
autotrophic nitrifying biofilms as determined by MAR-FISH.
Appl. Environ. Microbiol. 70, 1641–1650.
Kuypers, M.M.M., Sliekers, A.O., Lavik, G., Schmid, M., Jørgensen,
B.B., Kuenen, J.G., Sinninghe Damste ´, J.S., Strous, M., Jetten,
M.S.M., 2003. Anaerobic ammonium oxidation by anammox
bacteria in the Black Sea. Nature 422, 608–611.
Lorenzen, J., Larsen, L.H., Kjar, T., Revsbech, N.P., 1998. Biosensor
detection of the microscale distribution of nitrate, nitrate
assimilation, nitrification, and denitrification in a diatom-
inhabited freshwater sediment. Appl. Environ. Microbiol. 64,
3264–3269.
Maidak, B.L., Olsen, G.L., Larsen, N., Overbeek, R., McCaughey,
M.J., Woese, C.R., 1997. The RDP (ribosomal database project).
Nucleic Acids Res. 25, 109–110.
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, 177–183.
Mobarry, B.K., Wagner, M., Urbain, V., Rittmann, B.E., Stahl, D.A.,
1996. Phylogenetic probes for analyzing abundance and spatial
organization of nitrifying bacteria. Appl. Environ. Microbiol.
62, 2156–2162.
Neef, A., Amann, R., Schlesner, H., Schleifer, K.-H., 1998.
Monitoring a widespread bacteria group: in situ detection of
ARTICLE IN PRESS
WATER RESEARCH 41 (2007) 1623– 1634
1633

Page 12

planctomycetes with 16S rRNA-targeted probes. Microbiology
144, 3257–3266.
Nicolella, C., van Loosdrecht, M.C.M., Heijnen, S.J., 2000. Particle-
based biofilm reactor technology. Trends Biotechnol. 18,
312–320.
Nielsen, M., Bollmann, A., Sliekers, O., Jetten, M.S.M., Schmid, M.,
Strous, M., Schmidt, I., Larsen, L.H., Nielsen, L.P., Revsbech,
N.P., 2005. Kinetics, diffusional limitation and microscale
distribution of chemistry and organisms in a CANON reactor.
FEMS Microbiol. Ecol. 51, 247–256.
Okabe, S., Satoh, H., Watanabe, Y., 1999. In situ analysis of
nitrifying biofilms as determined by in situ hybridization and
the use of microelectrodes. Appl. Environ. Microbiol. 65,
3182–3191.
Okabe, S., Kindaichi, T., Ito, T., 2004. Fate of 14C-labeled microbial
products derived from nitrifying bacteria in autotrophic
nitrifying biofilms. Appl. Environ. Microbiol. 71, 3987–3994.
Pynaert, K., Smets, B.F., Beheydt, D., Verstraete, W., 2004. Start-up
of autotrophic nitrogen removal reactors via sequential
biocatalyst addition. Environ. Sci. Technol. 38, 1228–1235.
Revsbech, N.P., 1989. An oxygen microelectrode with a guard
cathode. Limnol. Oceanogr. 34, 474–478.
Rosenwinkel, K.-H., Cornelius, A., 2005. Deammonification in the
moving-bed process for the treatment of wastewater with
high ammonia content. Chem. Eng. Technol. 28, 49–52.
Rouse, J.D., Yoshida, N., Hatanaka, H., Imajo, U., Furukawa, K.,
2003. Continuous treatment studies of anaerobic oxidation of
ammonium using a nonwoven biomass carrier. Jpn. J. Water
Treat. Biol. 39, 33–41.
Saito, N., Nei, M., 1987. The neighbor-joining method: a new
method for constructing phylogenetic trees. Mol. Biol. Evol. 4,
406–425.
Satoh, H., Okabe, S., Yamaguchi, Y., Watanabe, Y., 2003. Evaluation
of the impact of bioaugmentation and biostimulation by in situ
hybridization and microelectrode. Water Res. 37, 2206–2216.
Schmidt, I., Sliekers, O., Schmid, M., Bock, E., Fuerst, J., Kuenen,
J.G., Jetten, M.S.M., Strous, M., 2003. New concepts of microbial
treatment processes for the nitrogen removal in wastewater.
FEMS Microbiol. Rev. 27, 481–492.
Schmid, M., Twachtmann, U., Klein, M., Strous, M., Juretschko, S.,
Jetten, M., Metzger, J., Schleifer, K.-H., Wagner, M., 2000.
Molecular evidence for genus level diversity of bacteria
capable of catalyzing anaerobic ammonia oxidation. Syst.
Appl. Microbiol. 23, 93–106.
Schmid, M., Walsh, K., Webb, R., Rijpstra, W.I., van de Pas-
Schoonen, K., Verbruggen, M.J., Hill, T., Moffett, B., Fuerst, J.,
Schouten, S., Damste, J.S., Harris, J., Shaw, P., Jetten, M., Strous,
M., 2003. Candidatus ‘‘Scalindua brodae’’ sp. nov., Candidatus
‘‘Scalindua wagneri’’ sp. nov., two new species of anaerobic
ammonium oxidizing bacteria. Syst. Appl. Microbiol. 26,
529–538.
Schramm, A., Larsen, L.H., Revsbech, N.P., Amann, R.I., Schleifer,
K.-H., 1996. Structure and function of a nitrifying biofilm as
determined by in situ hybridization and the use of microelec-
trodes. Appl. Environ. Microbiol. 62, 4641–4647.
Sliekers, A.O., Derwort, N., Campos Gomez, J.L., Strous, M.,
Kuenen, J.G., Jetten, M.S.M., 2002. Completely autotrophic
nitrogen removal over nitrite in one single reactor. Water Res.
36, 2475–2482.
Sliekers, A.O., Third, K.A., Abma, W., Kuenen, J.G., Jetten, M.S.M.,
2003. CANON and anammox in a gas-lift reactor. FEMS
Microbiol. Lett. 218, 339–344.
Strous, M., Heijnen, J.J., Kuenen, J.G., Jetten, M.S.M., 1998. The
sequencing batch reactor as a powerful tool to study very
slowly growing micro-organisms. Appl. Microbiol. Biotechnol.
50, 589–596.
Strous, M., Kuenen, J.G., Jetten, M.S.M., 1999. Key physiology of
anaerobic ammonium oxidation. Appl. Environ. Microbiol. 65,
3248–3250.
Strous, M., Pelletier, E., Mangenot, S., Rattei, T., Lehner, A., Taylor,
M.W., Horn, M., Daims, H., Bartol-Mavel, D., Wincker, P., Barbe,
V., Fonknechten, N., Vallenet, D., Segurens, B., Schenowitz-
Truong, C., Medigue, C., Collingro, A., Snel, B., Dutilh, B.E., Op
den Camp, H.J., van der Drift, C., Cirpus, I., van de Pas-
Schoonen, K.T., Harhangi, H.R., van Niftrik, L., Schmid, M.,
Keltjens, J., van de Vossenberg, J., Kartal, B., Meier, H.,
Frishman, D., Huynen, M.A., Mewes, H.W., Weissenbach, J.,
Jetten, M.S.M., Wagner, M., Le Paslier, D., 2006. Deciphering the
evolution and metabolism of an anammox bacterium from a
community genome. Nature 440, 790–794.
Szatkowska, B., Cema, G., Plaza, E., Trela, J., Hultman, B., 2006.
One-stage system with partial nitritation and anammox
processes in moving-bed biofilm reactor. In: Proceedings of the
International Conference on Biofilm Systems VI—Amsterdam,
pp. 49–58.
Third, K.A., Sliekers, A.O., Kuenen, J.G., Jetten, M.S.M., 2001. The
CANON system (completely autotrophic nitrogen-removal
over nitrite) under ammonium limitation: Interaction and
competition between three groups of bacteria. System. Appl.
Microbiol. 24, 588–596.
Third, K.A., Paxman, J., Schmid, M., Strous, M., Jetten, M.S.M.,
Cord-Ruwisch, R., 2005. Treatment of nitrogen-rich waste-
water using partial nitrification and anammox in the CANON
process. Water Sci. Technol. 52, 47–54.
Thompson, J.D., Higgins, D.G., Gibson, T.J., 1994. CLUSTAL W:
improving the sensitivity of progressive multiple sequence
alignment through sequence weighting, position-specific gap
penalties and weight matrix choice. Nucleic Acids Res. 22,
4673–4680.
Tsushima, I., Kindaichi, T., Okabe, S., 2007. Quantification of
anaerobic ammonium-oxidizing bacteria in enrichment cul-
tures by real-time PCR. Water Res. 41, 785–794.
Toh, S.K., Webb, R.I., Ashbolt, N.J., 2002. Enrichment of auto-
trophic anaerobic ammonium-oxidizing consortia from var-
ious wastewaters. Microb. Ecol. 43, 154–167.
van Dongen, U., Jetten, M.S.M., van Loosdrecht, M.C.M., 2001. The
SHARON-anammox process for treatment of ammonium rich
wastewater. Water Sci. Technol. 44 (1), 153–160.
van de Graaf, A.A., de Bruijn, P., Robertson, L.A., Jetten, M.S.M.,
Kuenen, J.G., 1996. Autotrophic growth of anaerobic ammo-
nium-oxidizing micro-organisms in a fluidized bed reactor.
Microbiology 142, 2187–2196.
Weisburg, W.G., Barns, S.M., Pelletier, D.A., Lane, D.J., 1991. 16S
ribosomal DNA amplification for phylogenetic study. J. Bac-
teriol. 173, 697–703.
ARTICLE IN PRESS
WAT E R R E S E A R C H 41 (2007) 1623– 1634
1634