Ecophysiologyofabundantdenitrifying bacteria inactivated sludgeTrine Rolighed Thomsen, Yunhong Kong & Per Halkjær Nielsen

Section of Environmental Engineering, Department of Biotechnology, Chemistry and Environmental Engineering, Aalborg University, Denmark

Correspondence: Per H. Nielsen, Section of

Environmental Engineering, Department of

Biotechnology, Chemistry and Environmental

Engineering, Sohngaardsholmsvej 57,

DK-9000 Aalborg, Denmark.

Tel.: 145 96358503; fax: 145 96350558;

e-mail: phn@bio.aau.dk

Received 12 July 2006; revised 30 January

2007; accepted 30 January 2007.

First published online 28 March 2007.

DOI:10.1111/j.1574-6941.2007.00309.x

Editor: Michael Wagner

Keywords

denitrifiers; Betaproteobacteria ; activated

sludge; ecophysiology; microautoradiography.

Abstract

The abundance of potential denitrifiers in full-scale wastewater treatment plants

with biological nitrogen and phosphorus removal was investigated by FISH and

various oligonucleotide probes. The potential denitrifiers were characterized as

probe-defined populations that were able to consume radiolabelled substrate with

oxygen, nitrate and nitrite as electron acceptor as determined by microautoradio-

graphy. The most abundant potential denitrifiers were related to the genera

Aquaspirillum, Azoarcus, Thauera and Rhodocyclus, all within the Betaproteobacter-

ia. They made up 20–49% of all bacteria in most of the 17 nitrogen removal plants

investigated and were hardly present in four plants without denitrification. The

ecophysiology of Aquaspirillum, Azoarcus and Thauera-related bacteria was con-

sistent within each probe-defined group in the plants investigated. These three

groups showed distinct physiological differences, with the Aquaspirillum-related

bacteria appearing as the most specialized one, consuming only amino acids

among the substrates tested, and Thauera as the most versatile consuming some

volatile fatty acids, ethanol and amino acids. The coexistence of Aquaspirillum,

Azoarcus and Thauera-related bacteria in a range of treatment plants with

differences in wastewater, design and operation suggest that the populations

ensure a functional stability of the plants by occupying different ecological niches

related to the carbon transformation.

Introduction

Nitrogen removal from wastewater is usually performed by a

sequential nitrification-denitrification process in activated

sludge wastewater treatment plants (WWTP). Despite the

processes having been known, used and optimized for

decades, the most important nitrifying bacteria have only

recently been identified by culture-independent methods

(Juretschko et al., 1998; Daims et al., 2000), and still very

little is known about the identity of abundant denitrifiers

(Wagner & Loy, 2002; Wagner et al., 2002). The main reason

for this is that most studies of denitrifiers from activated

sludge systems have been carried out after isolation and

cultivation (e.g. Gorny et al., 1992; Scholten et al., 1999;

Gumaelius et al., 2001), and these isolates do not necessarily

represent the important denitrifiers actually present in the

WWTP (Wagner & Loy, 2002).

Some potential abundant denitrifiers have recently been

identified by using culture-independent methods in a

WWTP treating industrial wastewater (Wagner & Loy,

2002; Wagner et al., 2002) and in treatment plants treating

municipal wastewater (Thomsen et al., 2004). In the two

types of plants, the abundant denitrifiers were Azoarcus-

related bacteria and Aquaspirillum-related bacteria, respec-

tively. The abundance of these bacteria was evaluated using

FISH with specific oligonucleotide probes, and it was shown

that in some plants Aquaspirillum-related bacteria consti-

tuted 20–30% of all bacteria (Thomsen et al., 2004). Their

proposed denitrifying activity was shown by applying a

combination of microautoradiography (MAR) and FISH.

Also, bacteria belonging to the genera Thauera and Zoogloea

are assumed to be important denitrifiers in some WWTP,

as they are often present in WWTP with denitrification

(Juretschko et al., 2002; Rosello-Mora et al., 1995), and

because isolates are capable of performing full denitrifica-

tion (Harder, 1997; Foss & Harder, 1998; Scholten et al.,

1999; Mechichi et al., 2002). Their actual abundance and

importance in the denitrification process in full-scale plants

have, however, not yet been evaluated. Finally, polypho-

sphate-accumulating organisms (PAOs) have also been

suggested as potential denitrifiers in WWTP with combined

nitrogen (N) removal and enhanced biological phosphorus

removal (EBPR) (van Loosdrecht et al., 1997; Seviour et al.,

2003). Recently, it has been shown that uncultured Rhodo-

cyclus-related PAOs grown in lab-scale reactors were able to

denitrify (Zeng et al., 2003) and that in full-scale plants they

FEMS Microbiol Ecol 60 (2007) 370–382c� 2007 Federation of European Microbiological SocietiesPublished by Blackwell Publishing Ltd. All rights reserved

can also most likely perform denitrification (Kong et al.,

2004).

The physiology of these potentially important denitrifiers

is generally not very well known because most of them are

uncultured and because the isolates, which are the closest

relatives, are rather distantly related, making extrapolations

of the physiology difficult. Isolates of Azoarcus and Thauera

from activated sludge are able to grow on a large number of

organic and aromatic compounds or monoterpenes under

aerobic and denitrifying conditions (Anders et al., 1995;

Harder, 1997; Song et al., 1998, 2001; Scholten et al., 1999;

Mechichi et al., 2002). Isolates closely related to the newly

identified Aquaspirillum-related bacteria have not been

obtained so far, although they were very abundant in several

municipal treatment plants (Thomsen et al., 2004). Among

numerous substrates tested in situ by MAR, they could only

consume a mixture of amino acids and no common

substrate such as acetate, ethanol, or glucose. Thus, they

were suggested to be involved mainly in the degradation of

protein in the WWTP investigated (Thomsen et al., 2004).

The physiology of the Rhodocyclus-related PAOs is better

described than the other denitrifiers because intensive

studies in lab-scale reactors have been conducted (Seviour

et al., 2003). Largely, detailed studies of their actual ecophy-

siology in full-scale plants confirm the observations ob-

tained in lab-scale reactors (Kong et al., 2004).

The bacterial groups mentioned above all belong to the

Betaproteobacteria, which seem to include the main deni-

trifiers in activated sludge systems. Betaproteobacteria is an

abundant group in many WWTP (e.g. Juretschko et al.,

2002; Wagner et al., 2002; Klausen et al., 2004), and may

reflect that denitrification is a dominant process in these

WWTP. In this study, we have by quantitative FISH investi-

gated the abundance of potential denitrifiers targeted by

different oligonucleotide probes. Although the probes ap-

plied are not completely genus-specific, the targeted bacteria

will in this study be mentioned as Zoogloea (probe ZRA,

Rosello-Mora et al., 1995), Azoarcus (probe Azo644, Hess

et al., 1997), Thauera (probe Thau646, Lajoie et al., 2000),

Aquaspirillum (probe Aqs997, Thomsen et al., 2004) and

Rhodocyclus-related (probe PAOmix, Crocetti et al., 2000) in

several WWTP. Furthermore, we have studied various

aspects of the ecophysiology of these probe-defined groups

in order to see whether they have any physiological differ-

ences, which can be related to different ecological niches in

the WWTP investigated.

Materials and methods

Activated sludge sampling

The ecophysiology experiments were carried out with acti-

vated sludge from Aalborg East and Horsens WWTP,

Denmark. Aalborg East has carbon removal, nitrification,

denitrification, chemical and biological phosphorus re-

moval, and a mean cell residence time (sludge age) of

20–30 days. Horsens WWTP has carbon removal, nitrifica-

tion, denitrification, chemical phosphorus removal and a

sludge age of 20–25 days. The activated sludge samples were

collected from April until December 2004 from aeration

tanks and brought to the laboratory within 1–2 h.

FISH and probe specificity

The FISH procedure was performed on fixed sludge as de-

scribed by Amann (1995), and the following oligonucleotide

probes were used: EUB338, EUB338-II and EUB338-III, called

EUBmix (all Bacteria; Amann et al., 1990; Daims et al., 1999),

BET42a (Betaproteobacteria; Manz et al., 1992), BONE23a and

competitor (Betaone subgroup; Amann et al., 1996), Azo644

(targeting most Azoarcus; Hess et al., 1997), ZRA (Zoogloea

ramigera; Rosello-Mora et al., 1995), PAOmix consisting of

probe PAO462, PAO651 and PAO846 (Candidatus ‘Accumuli-

bacter phosphatis’, a polyphosphate-accumulating bacterium,

also called Rhodocyclus-related PAO; Crocetti et al., 2000),

OTU4-645 (targeting Alcaligenes latus and some activated

sludge clones; Juretschko et al., 2002) and OTU6-178 (targeting

Brachymonas denitrificans and some activated sludge clones;

Juretschko et al., 2002). Details about the probes can be found

in probeBase (Loy et al., 2003). In addition Thau646 (Thauera

spp.; Lajoie et al., 2000) and Aqs997, including competitors

(targeting Aquaspirillum-like sequences; Thomsen et al., 2004),

were used. Probe Aqs997 has a perfect match with the Aqua-

spirillum-like sequences found by Thomsen et al. (2004);

Aquaspirillum delicatum ATCC14667 (recently reclassified as

Curvibacter delicatus comb. nov.; Ding & Yokota, 2004),

Pseudomonas lanceolata AB021390 (recently reclassified as

Curvibacter lanceolatus comb. nov.; Ding & Yokota, 2004) and

other bacteria in the Curvibacter genus. Also, a few other

sequences of Comamonas sp. and clones related within the

family Comamonadaceae are targeted (some Rhodoferax, Acid-

ovorax and Comamonas). Probe Thau644 targets different

Thauera but also sequences in other genera within the Rhodo-

cyclales order and sequences within the Burkholderiales, Hydro-

genophilales and Nitrosomonadales orders. Many of these are

sequences of uncultured bacteria. Probe Azo644 targets most

bacteria in the Azoarcus genus but also a few uncultured

bacteria in other genera within the family Rhodocyclaceae (some

Rhodocyclus and Dechloromonas). PAOmix targets Candidatus

‘Accumulibacter phosphatis’ and other primarily uncultured

bacteria within the Rhodocyclales order. The cells were simulta-

neously hybridized with EUBmix, a probe targeting the Beta-

proteobacteria, as well as a more specific probe. In Fig. 1 the

specificity of probe ZRA, Azo644, Thau646, Aqs997 and

PAOmix is illustrated, showing representative isolates from the

various genera targeted by the probes. Azoarcus sp., AF011344 is

FEMS Microbiol Ecol 60 (2007) 370–382 c� 2007 Federation of European Microbiological SocietiesPublished by Blackwell Publishing Ltd. All rights reserved

371Ecophysiology of denitrifying bacteria in activated sludge

targeted by both probe Thau646 and Azo644, which is illu-

strated with a � and an imprecise border between the two

probes in Fig. 1.

Oligonucleotides were 50-labelled with 5(6)-carboxyfluores-

cein-N-hydroxy-succinimide ester (FLUOS) or with sulphoin-

docyanine dyes (Cy5 and Cy3) (Thermo Hybaid, Germany).

For the study of the ecophysiology, the identity was always

linked to the function. FISH was combined either by MAR,

enzyme-labelled fluorescence (ELF), microspheres’ adhesion to

cells (MAC) or staining for storage products (PHA and

polyphosphate). This was conducted by dual staining, where

FISH was performed either before (PHA) or after (MAR, ELF,

MAC and polyphosphate) the different methods mentioned.

An automatic stage controller in the LSM 510 software allowed

saved positions on a cover glass to be relocated.

Autoradiography

MAR experiments were performed according to the proce-

dure described by Nielsen & Nielsen (2002a) with 2 mM

final concentration of each substrate, 10 mCi [3H]-labelled

substrate (propionate, formate, and pyruvate were 14C-

labelled) and 2 mL of diluted activated sludge (1 g SS L�1).

Under aerobic conditions, uptake of the following substrates

was investigated: formate, acetate, propionate, lactate, pyr-

uvate, oleic acid, ethanol, glucose, galactose, mannose,

glycine, leucine, aspartate, glutamate and thymidine (Amer-

sham, NEN/Perkin-Elmer, American radiolabelled chemi-

cals). Furthermore, an [3H]-labelled amino acid mixture

(ICN Biomedicals) was investigated with each of the 16

amino acids (alanine, arginine, aspartic acid, glutamic acid,

glycine, histidine, isoleucine, leucine, lysine, phenylalanine,

proline, serine, threonine, tyrosine and valine) in a concen-

tration of 0.1 mM each, and the specific activity varied from

6.8 to 28.8mCi mL�1. The uptake of acetate incubated with

unlabelled amino acid mixture and thymidine with unla-

belled acetate was also tested under aerobic conditions. The

potential use of electron acceptors other than oxygen was

investigated by studying the uptake of selected substrates

with nitrate (2 mM) or nitrite (0.5 mM) as electron acceptor

Aqs997

Azo644

ZRAZoogloea sp., DQ413151

Zoogloea ramigera, D14257

Azoarcus evansii, X77679

Azoarcus tolulyticus, AF229857

Azoarcus anaerobius, Y14701

Azoarcus sp., AF011344*

Thauera sp., AF110005

Thauera aromatica, AF170281

Thauera aromatica, AF229881

Thauera terpenica, AJ005817

Azonexus sp., AB166882

Dechloromonas sp., AF170356

Candidatus ”Accumulibacter phosphatis”, AY962316

uncultured Betaproteobacterium, AY062125

Leptothrix sp., AF385528Aquabacterium parvum, AF035052

Aquabacterium sp., AF089857

Comamonas sp., AF188304

Acidovorax sp., AY258065

Comamonas sp., AF501881

Aquaspi clone C, AY322153

Aquaspi clone B, AY322152

Aquaspi clone A, AY322151

Pseudomonas lanceolata (reclassified as Curvibacter lanceolata), AB021390

Aquaspirillum delicatum (reclassified as Curvibacter delicatum), AF078756

0.10

Thau646

PAOmix

Fig. 1. A phylogenetic tree illustrating the specificity of the applied gene probes Azo644, Thau646, ZRA, Aqs997 and PAOmix. Pure culture

representatives of the different genera targeted by the probes are shown. The tree was calculated using default settings for the maximum likelihood

algorithm in ARB. The scale bar corresponds to 10% estimated sequence divergence. �Azoarcus sp. (AF011344) is targeted by both probe Azo644 and

Thau646.

FEMS Microbiol Ecol 60 (2007) 370–382c� 2007 Federation of European Microbiological SocietiesPublished by Blackwell Publishing Ltd. All rights reserved

372 T.R. Thomsen et al.

in the absence of oxygen or under anaerobic conditions

without nitrate/nitrite present. In experiments with nitrate

or nitrite as electron acceptor and in anaerobic experiments,

a pre-incubation step of 2 h with unlabelled organic sub-

strate (and nitrate/nitrite when tested as electron acceptor)

was included in order to ensure that substrate uptake could

be ascribed to direct growth or very large storage capabil-

ities. In some experiments the simultaneous uptake of two

substrates was investigated as described by Kong et al.

(2004). As a control for chemography, sludge was pasteur-

ized at 70 1C for 10 min just before the incubation. After 3 h

incubation, the activated sludge was fixed, and the FISH

procedure was performed before applying the photographic

emulsion (Lee et al., 1999).

Storage compounds

The lipidic storage granules of the different bacteria investi-

gated were studied by Nile Blue staining of polyhydroxyalk-

anoates (PHA) (Ostle & Holt, 1982). Polyphosphate

granules were stained with methylene blue (Neisser stain)

as described in Eikelboom & van Buijsen (1983). Fixed

sludge was spread out on gelatine-coated cover slides,

hybridized with specific oligonucleotide probes, and images

were taken. Subsequently, the slides were washed for 10 min

with 70% ethanol, and the Nile Blue or Neisser stain was

performed. Both stains were performed on sludge directly

from the WWTP. In addition, Nile Blue stain was performed

on sludge incubated for 3 h with acetate or amino acid

mixture under aerobic conditions with nitrate or nitrite as

electron acceptor, and under anaerobic conditions. Forma-

tion of intracellular PHA granules by probe Azo644,

Thau646 and Aqs997 probe-defined cells was then investi-

gated. Furthermore, selected substrates (alanine, arginine,

aspartic acid, glutamic acid, histidine, isoleucine, leucine,

lysine, phenylalanine, proline, serine, threonine, tyrosine

and valine) were incubated (2 mM) with sludge for 3 h

followed by dual staining with the Aqs997 oligonucleotide

probe and Nile Blue.

ELF and MAC

The presence of exo-enzyme activity was determined using

ELF (ELF-97s, Molecular Probes; www.probes.invitrogen.

com), where, after enzymatic cleavage, substrates form a

fluorescent precipitate on the surfaces of bacteria. The same

procedure as described by Nielsen et al. (2002) and Krage-

lund et al. (2005) was used. The following enzymes were

evaluated: ELFs 97 esterase (ELFs 97 acetate), ELFs 97

lipase (ELFs 97 palmitate), ELFs 97 b-D-galactosidase

(ELFs 97 b-D-galactopyranoside), ELFs 97 b-D-glucuroni-

dase (ELFs 97 b-D-glucuronide), ELFs 97 chitinase/N-

acetylglucosaminidase (ELFs 97 N-acetylglucosaminide;

ELFs 97 NAG) and an ELFs 97 Endogenous Phosphatase

detection kit.

The hydrophobicity of the bacteria was investigated by

MAC with a modified version of Zita & Hermansson, (1997)

and Nielsen et al. (2001) as described by Kragelund et al.

(2005).

Microscopy

A model LSM 510-Meta confocal laser scanning microscope

(Carl Zeiss, Oberkochen, Germany) was used for the detec-

tion of oligonucleotide probe positive cells combined with

various in situ methods. For quantitative FISH, 32 images

were taken of each of the five probe-defined populations in

all sludge samples investigated and the images were subse-

quently analysed using IMAGE ANALYSIS software (ImageJ

1.33 s, Rasband W, National Institutes of Health, USA,

http://rsb.info.nih.gov/ij/) with a designed macro. Numbers

are given as average � standard deviation. For ecophysio-

logical studies, a minimum of 20 microcolonies of Aqua-

spirillum, Azoarcus or Thauera-related bacteria were

investigated visually to estimate the percentage of cells

exhibiting a certain physiological property. Most experi-

ments concerning ecophysiology were performed with

sludge from both WWTP investigated (see Table 2).

Results

Abundance and morphology revealed by FISHprobing

Bacteria related to Aquaspirillum, Azoarcus, Thauera and

Rhodocyclus-related PAOs were present in most WWTP

investigated (Table 1). In the 16 plants treating mainly

domestic wastewater with biological nitrogen (N) and/or

phosphorus (P) removal, the predominant group was the

Aquaspirillum-related bacteria, which constituted 11–29%

of all bacteria detected with the EUBmix. Azoarcus and

Thauera were also abundant, representing 3–16% of the

biovolume. Rhodocyclus-related PAOs were present in EBPR

plants only, where they constituted 3–9%.

In treatment plants without denitrification, only very

small numbers of Thauera, Azoarcus, Aquaspirillum-related

bacteria and Rhodocyclus-related PAOs were observed. In

contrast, these groups of bacteria constituted the vast

majority of the Betaproteobacteria present in most munici-

pal denitrifying WWTP, where they constituted up to 49%

of the total biovolume (Table 1). Other potential denitri-

fiers, Zoogloea-related bacteria, were observed but only in

small numbers (Table 1).

The Aquaspirillum-like bacteria showed quite distinct

morphology, being relatively large, coccoid cells with a

diameter of 1–1.5 mm. The size of the microcolonies varied,

with the majority being 10–15 mm in diameter, but colonies

FEMS Microbiol Ecol 60 (2007) 370–382 c� 2007 Federation of European Microbiological SocietiesPublished by Blackwell Publishing Ltd. All rights reserved

373Ecophysiology of denitrifying bacteria in activated sludge

of up to 40 mm in diameter were observed. In addition, the

specific probe also targeted some filamentous bacteria in

some WWTP, which have been investigated in detail in

another study (Thomsen et al., 2006). Thauera-related

bacteria made characteristic zoogloeal clusters of various

sizes (5–40mm). The cell size was typically 1.0–2.5 mm, and

the cell morphology was rod-like. Zoogloea-related bacteria

made characteristic clusters similar to the Thauera. Most

cells were rod-like (1–1.5 mm� 1 mm). Different Azoarcus

morphotypes were observed in most WWTP, usually as

rather large (1.0mm� 2.0 mm) or small (0.5mm� 1.0 mm)

rods growing either as single cells or in relatively small

microcolonies (5–20mm). Rhodocyclus-related bacteria were

coccobacilli with a diameter of 0.4–1.2 mm. Only very few

cells hybridizing with probe OUT4-645 (Alcaligenes latus)

and probe OTU6-178 (Brachymonas denitrificans) were

found.

Ecophysiology

Two WWTP, Aalborg East and Horsens, were studied in

detail to reveal possible physiological differences and simila-

rities among bacteria from the three gene probe-defined

groups investigated (Aquaspirillum, Azoarcus, Thauera). The

studies included substrate uptake profiles under different

electron acceptor conditions, presence of storage products

and exoenzymes, and presence of specific surface properties.

Rhodocyclus-related PAOs have previously been studied

(Kong et al., 2004) and Zoogloea–related organisms were

not abundant enough in the WWTP to be investigated.

The substrate uptake pattern, as determined by MAR, was

almost identical for each probe-defined population in the

two WWTP investigated. The pattern was, however, differ-

ent for the three populations investigated (Tables 2 and 3).

The Aquaspirillum-related bacteria were tested for uptake of

16 different substrates under aerobic, denitrifying and

anaerobic conditions (Thomsen et al., 2004; Tables 2 and 3).

Only the amino acid mixture was taken up by some, but not

all, microcolonies (20–30%). Substrate was taken up with

oxygen, nitrate or nitrite as electron acceptor, but not under

anaerobic conditions without the addition of external

electron acceptors, thus confirming their denitrifying po-

tential. Several amino acids were also tested as single

substrates, but no uptake was observed by MAR. All

substrates were taken up by other bacteria in the flocs

investigated, hence serving as positive controls. The uptake

of amino acids was also confirmed by PHA staining (Nile

blue) after incubation with the amino acid mixture for 3 h

(Table 4). Large PHA inclusions were observed in many of

the Aquaspirillum-related bacteria. These cells clearly con-

tained more PHA than cells in fresh sludge taken directly

from the WWTP, where Aquaspirillum-related bacteria only

Table 1. Abundance of bacteria related to Aquaspirillum, Azoarcus, Thauera, Zoogloea and Rhodocyclus-related PAOs in different wastewater

treatment plants determined by quantitative FISH

Plant Type wastewater Process design Aqs997 (%) Azo644 (%) Thau646 (%) ZRA (%) PAOmix (%)

Aalborg East Municipal C, N, DN, EBPR, CP 28.7� 4.9 4.5�2.3 7.1� 4.3 1.5� 2.9 6.7�1.8

Aalborg West Municipal C, N, DN, EBPR, CP 18.9� 8.5 5.5�4.9 4.3� 2.5 2.4� 4.8 5.5�1.9

Naestved Municipal C, N, DN, EBPR, CP 24� 8.3 3.9�3.3 4.5� 2.2 1.8� 2 8.5�0.7

Bjerringbro Municipal C, N, DN, EBPR, CP 11� 5.5 2.1�2.9 2.9� 2.1 1� 1.7 5.9�1.5

Hjoerring Municipal C, N, DN, EBPR, CP 21.9� 3.5 6.4�5 5.9� 4.6 0 3.0�0.7

Usseroed Municipal C, N, DN, EBPR, CP 12.4� 5 2.6�3.5 4� 2.3 1.1� 2.4 2.6�0.7

Vedbaek Municipal C, N, DN, (EBPR), CP 15.7� 7.3 3.8�2.1 4.9� 2.2 o 1 o 1

Egaa Municipal C, N, DN, EBPR 23.6� 11.9 3.5�2.1 5.8� 2.1 o 1 4.6�0.6

Horsens Municipal C, N, DN, CP 22.3� 8.4 5�1 11.4� 5.5 o 1 o 1

Aabybro Municipal C, N, DN, CP 14.4� 5.7 2.5�1.5 2.1� 1.2 o 1 0

Viborg Municipal C, N, DN, CP 14.7� 6.5 1.9�1.6 2.2� 1.3 o 1 0

Svendborg Municipal C, N, DN, CP 20.8� 7.4 4.6�4.2 3.9� 3.2 o 1 o 1

Hjallerup Municipal C, N, DN, CP 13.7� 8.4 o 1 3.5� 2.8 0 0

Fjerritslev Municipal C, N, DN, CP 18.7� 9.4 1.6�1.5 1.1� 1.7 0 0

Vadum Municipal C, N, DN 13.2� 5.3 2.3�1.7 3.4� 1.5 o 1 0

Skovlund Industrial/municipal C, N, DN, EBPR, CP 11.7� 5.8 3�2.9 3� 2.5 o 1 6.6�1.6

DynCNR1 Industrial C, N, DN o 1 o 1 o 1 o 1 0

DynBio23 Industrial C, N, CP o 1 o 1 0 0 0

DynTNO12 Industrial C, N, CP o 1 o 1 o 1 2.9� 4.9 0

DynTNO32 Industrial C, N, CP o 1 o 1 o 1 3.7� 2.3 0

DynCNR20 Industrial C o 1 0 0 0 0

The probes applied were Aqs997, Azo644, Thau646, ZRA and PAOmix, respectively. The abundance was compared to bacteria hybridizing with the

EUBmix.

C, carbon removal; N, nitrification; DN, denitrification; EBPR, enhanced biological phosphorus removal; CP, chemical phosphorus removal.

FEMS Microbiol Ecol 60 (2007) 370–382c� 2007 Federation of European Microbiological SocietiesPublished by Blackwell Publishing Ltd. All rights reserved

374 T.R. Thomsen et al.

occasionally contained a few PHA granules. Subsequently,

the potential of utilizing one or more of the specific amino

acids present in the mixture was tested by incubating

activated sludge for 3 h with single amino acids followed by

dual staining with oligonucleotide probes and Nile Blue.

Only incubations with phenylalanine resulted in large

intracellular inclusions of PHA in many Aquaspirillum-

related bacteria, strongly indicating that this compound

was a suitable sole substrate. Aquaspirillum-related bacteria

in sludge from Aalborg East did not contain polyphosphate,

but a few did in Horsens WWTP. Surface-associated exoen-

zymatic activity was only exhibited by a few Aquaspirillum-

related bacteria as detected by the applied ELF-substrates.

A few galactosidase-positive cells were found in sludge from

Aalborg East WWTP, while none were found in the two

plants with esterase, lipase, chitinase, glucoronidase or

phosphatase activity (Fig. 2c). When the surface properties

were investigated using MAC, both hydrophobic and hydro-

philic structures were found in the flocs as indicated by

many and few attached fluorescent hydrophobic micro-

spheres, respectively. No hydrophobic microspheres were

adsorbed on the Aquaspirillum-related bacteria in any of

the two sludges, indicating a rather hydrophilic surface

(Fig. 2d).

The probe-defined Azoarcus bacteria were only able to

take up a limited number of substrates with oxygen as

electron acceptor (Table 2). Acetate, ethanol and the amino

acid mixture were taken up by 20–60% of the bacteria, while

fewer took up pyruvate. No uptake of formate, propionate

or lactate was observed. Leucine could only be taken up

when acetate was added as a cosubstrate. The same sub-

strates were generally taken up with nitrate or nitrite as

electron acceptor, although the active fraction of probe-

defined bacteria was lower (Table 3), indicating that some of

them may not be able to perform full denitrification. No

Azoarcus bacteria were able to take up substrate under

anaerobic conditions, all were Neisser-negative (no poly-

phosphate granules), and they did not contain PHA gran-

ules in samples taken directly from the treatment plants.

Only a few Azoarcus cells containing PHA granules were

Table 2. Uptake of organic substrates by Aquaspirillum, Azoarcus, and Thauera-related bacteria under aerobic conditions in Aalborg East WWTP and

Horsens WWTP as investigated by MAR

Aquaspirillum (%)� Azoarcus (%) Thauera (%)

Aalborg East Horsens Aalborg East Horsens Aalborg East Horsens

Sugars

Glucose – – – – 10–20 5–10

Galactose – – – – – –

Mannose – – – – – –

Fatty acids

Formate – – – – ND –

Acetate – – 20–30 30–40 50 50

Propionate – – – – 30–40 60–70

Lactate – ND – ND – ND

Pyruvate – – 5–10 10–20 10–20 50

Oleic acid – – – – 10 25

Alcohol

Ethanol – – 20–30 20–30 10–20 30–40

Amino acids

Amino acid mixture 20–30 20–30 50–60 50–60 15–25 20–30

Glycine – – – – – –

Leucine ND – ND – ND –

Aspartate – ND – ND – ND

Glutamate – ND – ND – ND

Nucleoside

Thymidine – – – – – –

Mixed incubations

Thymidine and unlabelled acetate ND ND – ND – ND

Acetate and unlabelled amino acid mixture – ND ND ND ND ND

Leucine and unlabelled acetate ND ND 50–70 ND 10–20 ND

�Some data are already published in Thomsen et al. (2004) and Thomsen et al. (2006).

A minimum of 20 microcolonies was investigated visually to estimate the percentage of gene-probe defined microcolonies clearly MAR-positive. The

number 20–30% indicates e.g. that 20–30% of the Aqs997 probe-defined population was able to take up amino acids under the mentioned

incubation: ‘–’, o 5% takes up substrate; ND, not determined.

FEMS Microbiol Ecol 60 (2007) 370–382 c� 2007 Federation of European Microbiological SocietiesPublished by Blackwell Publishing Ltd. All rights reserved

375Ecophysiology of denitrifying bacteria in activated sludge

observed after 3-hour incubations with amino acid mixture

(Table 4). Esterase activity was observed for some Azoarcus

bacteria in sludge from Aalborg East and Horsens, whereas

no other exoenzymes were detected. All Azoarcus investi-

gated had hydrophilic surfaces.

Probe-defined Thauera bacteria were versatile in their

uptake of organic substrates under aerobic conditions as

they were able to use glucose, acetate, propionate, pyruvate,

oleic acid, ethanol and the amino acid mixture (Table 2, Fig.

2a and b). Leucine could only be taken up when acetate was

added as a cosubstrate. However, for several substrates, only

a fraction of the bacteria was active, ranging from 5–10% for

glucose to 60–70% for propionate. Nitrate or nitrite could

be used as electron acceptor for uptake of most substrates,

except glucose and oleic acid. For propionate, an identical

fraction of Thauera was active with oxygen and nitrate as

electron acceptor, while only less than half of them were

active with nitrite as electron acceptor, indicating that some

of them were unable to perform full denitrification. Under

anaerobic conditions, pyruvate was consumed by a few of

the Thauera in Horsens WWTP. Thauera had only a small

amount of polyphosphate granules in a few cells, but

contained some PHA granules in samples taken directly

from the WWTP. However, a much higher PHA storage

potential was observed when the sludge was incubated with

amino acid mixture or acetate for 3 h (Table 4). Surface-

associated esterase and glucoronidase activity was observed

for only very few Thauera in sludge from Aalborg East.

All microcolonies belonging to Thauera had hydrophilic

surfaces.

Other bacteria than the probe-defined Aquaspirillum-,

Azoarcus- and Thauera-related bacteria were observed

Table 4. Formation of intracellular PHA granules in Aquaspirillum, Azoarcus and Thauera after 3 h incubation with amino acid mixture or acetate in

sludge from Aalborg East WWTP

Aquaspirillum (%) Azoarcus (%) Thauera (%)

Oxygen Nitrate Nitrite Anaerobic Oxygen Nitrate Nitrite Anaerobic Oxygen Nitrate Nitrite Anaerobic

Amino acid mixture 40–50 40–50 40–50 – 10 10 10 – 20 20 20 –

Acetate – – ND – – – ND – 40–50 40–50 ND –

The percent PHA-positive gene-probe defined cells are shown.

ND, not determined.

Table 3. Uptake of selected organic substrates by Aquaspirillum, Azoarcus and Thauera under denitrifying and anaerobic conditions in Aalborg East

WWTP and Horsens WWTP

Aquaspirillum (%) Azoarcus (%) Thauera (%)

Nitrate Nitrite Anaerobic Nitrate Nitrite Anaerobic Nitrate Nitrite Anaerobic

Glucose

Aalborg East – ND – ND ND ND – ND –

Horsens ND ND ND ND ND ND – – –

Acetate

Aalborg East – ND – 25–35 ND – 40–50 ND –

Horsens – – – 10–20 10–20 – 10–20 10–20 –

Propionate

Aalborg East ND ND ND ND ND ND 20–30 ND –

Horsens – ND – – ND – 50–60 20–30 –

Pyruvate

Aalborg East – ND – 5–15 ND – 15–25 ND –

Horsens – – – 10–20 5–15 – 15–25 15–25 5–10

Oleic acid

Aalborg East ND ND ND ND ND ND – ND –

Horsens – – – – – – – – –

Ethanol

Aalborg East ND ND ND – – – 5–10 5–10 –

Horsens ND ND ND – – – 5–10 5–10 –

Amino acid mix

Aalb East 20–30 20–30 – 25–35 25–35 – 15–25 15–25 –

Horsens 15–25 15–25 – 25–35 25–35 – 15–25 15–25 –

Only substrates which were taken up under aerobic conditions were analysed further: ‘–’, o 5% takes up substrate; ND, not determined.

FEMS Microbiol Ecol 60 (2007) 370–382c� 2007 Federation of European Microbiological SocietiesPublished by Blackwell Publishing Ltd. All rights reserved

376 T.R. Thomsen et al.

taking up substrates under denitrifying conditions in the

two sludges investigated, indicating the presence of other

denitrifying bacteria. Some of these were Rhodocyclus-re-

lated PAOs.

Discussion

Bacteria related to the Aquaspirillum, Azoarcus, Thauera and

Rhodocyclus, all affiliating in the Betaproteobacteria, were

shown to be abundant potential denitrifiers in the municipal

WWTP with N- and P-removal investigated. These results,

together with the results of Wagner & Loy (2002), Wagner

et al. (2002), Thomsen et al. (2004) and Thomsen et al.

(2006) strongly suggest that abundant denitrifiers in many

types of WWTP with N-removal belong to these three

groups and, in plants with EBPR, also to the genus Rhodo-

cyclus. The genus Zoogloea has been reported to be present in

large numbers in very high-loaded conventional systems

(Rosello-Mora et al., 1995; Juretschko et al., 2002), but was

not found in large numbers in any of the plants investigated

here. There may still be several uncharacterized denitrifiers

present. Nielsen & Nielsen (2002b) found by MAR that

Fig. 2. MAR-FISH micrographs of an Aalborg East WWTP sludge sample incubated with ethanol under denitrifying conditions with nitrite as electron

acceptor and dual hybridized with FLUOS-labelled EUBmix and Cy3-labelled Thau646. Thauera-related cells appear yellow, other bacteria appear green

(a). The bright-field MAR image in (b) is showing the Thauera-related cells taking up ethanol (illustrated with circle). (c) illustrates glucuronidase activity

in an Aalborg East WWTP sludge sample hybridized with Cy3-labelled Aqs997. Aquaspirillum-related cells appear red, and glucoronidase activity by ELF

appears green. Arrows indicate typical Aquaspirillum-related microcolonies. In (d) a micrograph of a sludge sample from Horsens WWTP is illustrating

hydrophobic surfaces by MAC-FISH. Hydrophobic surfaces are covered with a layer of green fluorescent microspheres, and Aquaspirillum-related cells

are hybridized with Cy3-labelled Aqs997 and are red. The scale bars correspond to 10 mm.

FEMS Microbiol Ecol 60 (2007) 370–382 c� 2007 Federation of European Microbiological SocietiesPublished by Blackwell Publishing Ltd. All rights reserved

377Ecophysiology of denitrifying bacteria in activated sludge

about 70% of the biomass in Aalborg East WWTP is likely to

denitrify, which is in accordance with observations in this

study that many, but not all, nitrate and nitrite-respiring

organisms have been identified.

The denitrifying capacity of the probe-defined popula-

tions was shown by MAR-FISH in the study. It is difficult to

link the denitrifying activity to specific species in environ-

mental samples (Philippot & Hallin, 2005), so their poten-

tial denitrifying activities were here defined as being able to

take up organic substrates in the presence of nitrate as well

as nitrite as sole electron acceptor. Furthermore, they were

unable to consume any substrates under strictly anaerobic

conditions without nitrite or nitrate present. The fraction of

bacteria able to use nitrite was identical for Aquaspirillum-

related bacteria and Azoarcus to the fraction using nitrate as

electron acceptor, suggesting a full denitrification. This was

also the case for Thauera except with propionate as electron

acceptor, where only half of them were able to use nitrite

(Table 3). This may indicate a partial denitrification (to

nitrite) due to the presence of several subpopulations as

discussed below.

Ecophysiology of potential denitrifiers

Some interesting physiological differences between the three

groups were observed. The results showed, in agreement

with previous studies (Thomsen et al. 2004, 2006), that

Aquaspirillum-related bacteria apparently are very specia-

lized, only consuming very few substances. Surprisingly,

none of the substrates assumed to be commonly present in

activated sludge systems, such as acetate, propionate and

lactate (Nielsen et al., 1992; Lie & Welander, 1997; Henze

et al., 2002), could be consumed. Bacteria related to

Azoarcus were slightly more versatile and able to consume

acetate and a few other substrates, but not propionate or

glucose. Bacteria related to Thauera were the most versatile

ones and able to consume most substrates tested. It is,

however, remarkable that none of the groups studied were

able to assimilate glycine, leucine or formate, which are

likely to be present in many types of wastewater as hydrolysis

or fermentation products from protein and other macro-

molecules. However, leucine could be taken up when acetate

was added as a co-substrate, as has also been observed by the

Rhodocyclus-related PAOs (Kong et al., 2004). It is unclear

whether the uptake of leucine under these conditions is

mainly as an N or C-source. The overall results suggest that

acetate and some amino acids potentially are key com-

pounds for the removal of nitrate in many WWTP.

Rhodocyclus-related PAOs were present in the EBPR

plants, constituting 3–9% of the total biovolume. Some of

these PAOs are also able to denitrify (Kong et al., 2004) and

may constitute a significant part of the denitrifiers. Their

substrate uptake profile with oxygen, nitrate or nitrite as

electron acceptor in Aalborg East WWTP and two other

plants is slightly more diverse than that of Azoarcus, but less

than that of Thauera; there is uptake of acetate, propionate

and pyruvate, but not glycine, leucine, ethanol or oleic acid

(Kong et al., 2004). In contrast to the other denitrifiers they

are also able to take up substrate under anaerobic condi-

tions, store it as PHA and subsequently take up orthopho-

sphate under denitrifying and aerobic conditions for storage

of polyphosphate.

In common for all probe-defined populations was the

observation that a fraction of the bacteria appeared inactive

and did not take up labelled substrate. Some FISH-positive

cells may have been viable but not active, or, as the three

probes applied in this study are not very specific and

hybridize with several species within the genera described

(see ‘Material and methods’), variations in physiology may

be due to an undescribed microdiversity covered by the

probes. For Azoarcus and Thauera, it seems most plausible

that several subpopulations with different substrate uptake

profiles were present, as different morphologies existed.

However, a clone library study of Aalborg East sludge

showed a rather low diversity of bacteria targeted by probe

Thau644, indicating the probe seems not to target many

unspecific bacteria in this specific sludge (data not shown).

For the microcolony-forming Aquaspirillum-related bacter-

ia, which all had a very characteristic morphology, 30–50%

were metabolically active (in MAR and PHA experiments),

indicating that a certain fraction was inactive or that

morphologically identical subpopulations were present.

Alternatively, we have not yet found the primary substrates,

especially for the Aquaspirillum-related bacteria, but possi-

bly also for subpopulations of the two other groups. Some of

these might uptake substrates not tested in this study, e.g.

aromatic compounds and monoterpenes, which are usually

not present in high concentration in influent to most

WWTP (Henze et al., 2002). However, we also observed that

a certain fraction of probe-defined filamentous Alphapro-

teobacteria had a very low activity in a WWTP (Nielsen

et al., 2003) and that 10–50% of denitrifying Rhodocyclus-

related PAOs were inactive in different EBPR plants (Kong

et al., 2004). It was suggested that the operation of the

treatment plants caused inactivity for a part of the bacteria

due to poor growth conditions, but whether this explana-

tion is also likely for the denitrifiers investigated here is

unknown.

Intracellular storage of PHA is a well-known feature of

several activated sludge bacteria (Seviour & Blackall, 1999),

but PHA was hardly present in any of the probe-defined

bacteria studied here in fresh sludge. Rhodocyclus-related

PAOs usually have it when sampled from the anaerobic

tanks (Kong et al., 2004). However, Aquaspirillum and

Thauera had a high PHA storage potential, as shown when

they were incubated with suitable substrates, so under the

FEMS Microbiol Ecol 60 (2007) 370–382c� 2007 Federation of European Microbiological SocietiesPublished by Blackwell Publishing Ltd. All rights reserved

378 T.R. Thomsen et al.

dynamic conditions that occur in most full-scale N-removal

plants this metabolic feature can be a competitive advantage.

Bacteria from all three groups besides the PAOs only showed

storage capacity when incubated under aerobic and deni-

trifying conditions and not under anaerobic conditions,

which correlates well with the MAR results. Furthermore,

the three groups had no or very few Neisser-positive

granules, indicating an absence of large amounts of poly-

phosphate, thus confirming that they were not PAOs.

The presence of specific surface-associated exoenzymes

on different probe-defined bacteria has been observed in

activated sludge flocs (Kloeke & Geesey, 1999; Nielsen et al.,

2002). Such enzymes were only detected to a very small

extent in the three groups investigated. This indicates that all

three groups were mainly involved in the consumption of

soluble, easily consumable compounds and not in the

hydrolysis of macromolecules.

The microcolonies of Aquaspirillum, Azoarcus and

Thauera all had hydrophilic surfaces. Other bacteria in the

floc had hydrophobic surfaces, revealing some heterogeneity

in accordance with earlier studies. Cell surface properties are

important to the floc strength and stability (Zita & Her-

mansson, 1997; Olofsson et al., 1998; Nielsen et al., 2004),

and they can be important to the accessibility of substrates

to the cells (Nielsen et al., 2002). There was no detectable

difference between the surface properties of the three groups

investigated. It is known that Thauera can form gels with a

high content of water (Lajoie et al., 2000), but it was not

further investigated in this study.

Comparison with physiological data from purecultures

A number of isolates of Azoarcus and Thauera are available.

Most of them are not isolated from activated sludge, so their

identity, and thereby their physiology, may be different from

those in the sludges investigated. Azoarcus isolates can grow

on many aromatic compounds, some amino acids such as

alanine, on acetate, lactate, pyruvate, methanol and ethanol

(Anders et al., 1995; Song et al., 2001; Mechichi et al., 2002).

This generally corresponds well to the findings in this study,

although lactate could not be used. Furthermore, several

isolates of Azoarcus can denitrify with nitrate and nitrite,

and they can accumulate and grow on PHA (Mechichi

et al., 2002). There are no reports of the presence of

polyphosphate.

Recently, Ginige et al. (2005) showed that Thauera and

other members of the families Comamonadaceae and Rho-

docyclaceae are acetate-utilizing denitrifiers in activated

sludge by combining stable isotope probing and a full-cycle

rRNA analysis, and it corresponds well with the data

obtained in this study. Isolates of Thauera use nitrate, nitrite

or oxygen as electron acceptor, and many utilize aromatic

compounds, monoterpenes, amino acids, organic substrates

such as sugars, acetate, lactate, pyruvate and ethanol (Foss &

Harder, 1998; Song et al., 1998, 2001; Mechichi et al., 2002).

The relatively high versatility is in accordance with the data

obtained in this study. Interestingly, however, lactate was not

taken up in situ, although isolates can grow on this substrate.

This indicates the presence of several uncultured subpopu-

lations of this genus, or it illustrates differences between

bacteria growing in pure culture and in situ. Some isolates of

Thauera accumulate PHA (Anders et al., 1995; Scholten

et al., 1999), which was also observed in this study. T.

linealoolentis and T. terpenica are able to use sugar com-

pounds under aerobic conditions, and in accordance with

data obtained in our study these isolates are not able to grow

on sugars under denitrifying conditions (Foss & Harder,

1998). Such a restriction of the range of substrate to be

consumed with nitrate or nitrite as electron acceptor instead

of oxygen was also recently observed in situ by some

filamentous Alphaproteobacteria in activated sludge (Krage-

lund et al., 2005).

Recently 199 denitrifying bacteria were isolated from

activated sludge, where the majority belonged to the Beta-

proteobacteria (Heylen et al., 2006). Some Thauera, one

Zoogloea and one Aquaspirillum-related bacterium were

isolated. The isolated Aquaspirillum bacterium was distantly

related to the Aquaspirillum-related sequences used for the

design of probe Aqs997 (Thomsen et al., 2004), and the

probe does not target the isolate. No other pure culture of

the Aquaspirillum-related bacteria exists, so this study pre-

sents the first more detailed study of the physiology of these

organisms. The closest isolated relatives are Aquaspirillum

delicatum and Pseudomonas lanceolata, which grow on a

variety of substrates under aerobic conditions, but are

unable to perform full denitrification (the strains can only

make a partial reduction of nitrate to nitrite; Leifson, 1962;

Krieg, 1984; Pot et al., 1991). This physiology is very

different from that found in this in situ study of the

Aquaspirillum-related bacteria, where only uptake of pheny-

lalanine and perhaps a few other amino acids were observed

and they most likely carried out full denitrification.

Implications for full-scale treatment plant

It was shown that the abundant denitrifiers in a wide range

of WWPT with biological N and P-removal belonged to

relatively few genera; Aquaspirillum, Azoarcus, Thauera and

Rhodocyclus. Also, relatively few genera are dominant for

other groups of functional importance in full-scale activated

sludge systems, e.g. for the ammonium oxidizers (Purkhold

et al. 2000) and the PAO (Kong et al., 2005). It is assumed

that a large diversity ensures a better functional stability and

robustness against changes, leading to a more stable treat-

ment plant. The species richness of nitrifiers, for example,

FEMS Microbiol Ecol 60 (2007) 370–382 c� 2007 Federation of European Microbiological SocietiesPublished by Blackwell Publishing Ltd. All rights reserved

379Ecophysiology of denitrifying bacteria in activated sludge

varies in different WWTP, where, in some cases, it is

dominated by one single species, in others by at least four

different species (Purkhold et al. 2000). Factors such as

differences in substrate affinity and growth rates may con-

trol their presence. Among PAOs typically three (or more, so

far unknown) groups are usually present and differences in

substrate utilization capabilities seems to determine their

presence (Kong et al., 2005). It seems also to be the case for

the groups of denitrifiers investigated in this study. The

physiology was rather consistent within each probe-defined

group, and each group showed distinct differences, with the

Aquaspirillum–related bacteria as the most specialized and

Thauera as the most versatile (generalist). These differences

indicate that they occupy different nutritional niches related

to the carbon flow in the WWTP although differences in

storage capabilities (PHA and polyphosphate) and maybe

also in other kinetic parameters were present. Usually, the

heterogeneity in space and/or time can explain the degree of

diversity in an ecosystem (Kassen & Rainey, 2004). The

heterogeneity in space (within the floc) is probably limited,

but the dynamic nature of substrate supply, aerobic and

anaerobic phases, and a high degree of physiological differ-

ences supported the presence of four to five coexisting

denitrifying groups with an unknown microdiversity in

most WWTP investigated. Most likely several other species

of denitrifiers were present in low numbers in the WWTP

investigated indicating that the total diversity was large, but

it seems that the diversity of the functional abundant species

was much smaller.

Besides being important to the denitrification process, the

dominating species are, simply due to their high abundance

(up to 49% of all biovolume), also important to floc

formation, floc properties and thus solid–liquid separation.

Thauera and Zoogloea, for example, are known to produce

high amounts of water-containing extracellular polymeric

substances (EPS) that can cause problems with sludge

compaction and dewatering (Easson et al., 1987; Lajoie

et al., 2000). The Aquaspirillum–related bacteria, on the

other hand, seem not to be related to poor floc properties

(Thomsen & Nielsen, unpublished results). Therefore, better

knowledge about the ecophysiology of the different species

might help to control excessive growth of Thauera and

Zoogloea to ensure presence of denitrifiers with good floc

properties.

Acknowledgements

The Danish Technical Research Council supported this

study under the framework program ‘Activity and Diversity

in Complex Microbial Systems’. We thank M. Stevenson,

M. Fredsgaard and S. Bielidt for their valuable technical

assistance. S. Fuereder from Vienna University is greatly

acknowledged for investigating targets of gene probes used

in this study on a clone library on Aalborg East sludge.

References

Amann RI (1995) In situ identification of microorganisms by

whole cell hybridization with rRNA- targeted nucleic acid

probes. Molecular Microbial Ecological Manual (Akkermans

ADL, van Elsas JD & de Bruijn FJ, eds), pp. 1–15. Kluwer

Academic Publications, London, UK.

Amann RI, Binder BJ, Olson RJ, Chisholm SW, Devereux R &

Stahl D (1990) Combination of 16S rRNA-targeted

oligonucleotide probes with flow cytometry for analyzing

mixed microbial populations. Appl Environ Microbiol 56:

1919–1925.

Amann R, Snaidr J, Wagner M, Ludwig W & Schleifer K-H (1996)

In situ visualization of high genetic diversity in a natural

microbial community. J Bacteriol 178: 3496–3599.

Anders H, Kaetzke A, Kampfer P, Ludwig W & Fuchs G (1995)

Taxonomic position of aromatic-degrading denitrifying

pseudomonad strains K 172 and KB 740 and their description

as new members of the genera Thauera, as Thauera aromatica

sp. nov., and Azoarcus, as Azoarcus evansii sp. nov.,

respectively, members of the beta subclass of the

Proteobacteria. Int J Syst Bacteriol 45: 327–333.

Crocetti GR, Hugenholtz P, Bond PL, Schuler A, Keller J, Jenkins

D & Blackall LL (2000) Identification of polyphospate-

accumulating organisms and design of 16S rRNA-directed

probes for their detection and quantification. Appl Environ

Microbiol 66: 1175–1182.

Daims H, Bruhl 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.

Daims H, Nielsen PH, Nielsen JL, Juretschko S & Wagner M

(2000) Novel Nitrosospira-like bacteria as dominant nitrite-

oxidizers in biofilms from wastewater treatment plants:

diversity and in situ physiology. Wat Sci Technol 41: 85–90.

Ding L & Yokota A (2004) Proposals of Curvibacter gracilis gen.

nov., sp. nov. and Herbaspirillum putei sp. nov. for bacterial

strains isolated from well water and reclassification of

[Pseudomonas] huttiensis, [Pseudomonas] lanceolata,

[Aquaspirillum] delicatum and [Aquaspirillum] autotrophicum

as Herbaspirillum huttiense comb. nov., Curvibacter lanceolatus

comb. nov., Curvibacter delicatus comb. nov. and

Herbaspirillum autotrophicum comb. nov. Int J Syst Evol

Microbiol 54: 2223–2230.

Easson DD, Peoples OP & Sinskey AJ (1987). Biopolymer

engineering: genetic control of exopolysaccharide biosynthesis.

Industrial Polysaccharides: Genetic Engineering, Structure/

Property Relations and Applications: 57–64.

Eikelboom DH & van Buijsen HJJ (1983) Microscopic Sludge

Investigation Manual, 2nd edn. TNO Research Institute for

Environmntal Hygiejne, Delft.

FEMS Microbiol Ecol 60 (2007) 370–382c� 2007 Federation of European Microbiological SocietiesPublished by Blackwell Publishing Ltd. All rights reserved

380 T.R. Thomsen et al.

Foss S & Harder J (1998) Thauera linaloolentis sp. nov. and

Thauera terpenica sp. nov., isolated on oxygen-containing

monoterpenes (linealool, menthol, and eucalyptol) and

nitrate. Syst Appl Microbiol 21: 365–373.

Ginige MP, Keller J & Blackall LL (2005) Investigation of an

acetate-fed denitrifying microbial community by stable

isotope probing, full-cycle rRNA analysis, and fluorescent in

situ hybridization-microautoradiography. Appl Environ

Microbiol 71: 8683–8691.

Gorny N, Wahl G, Brune A & Schink B (1992) A strictly anaerobic

nitrate-reducing bacterium growing with resorcinol and other

aromatic compounds. Arch Microbiol 158: 48–53.

Gumaelius L, Magnusson G, Pettersson B & Dalhammer G (2001)

Comamonas denitrificans sp. nov., an efficient denitrifying

bacterium isolated from activated sludge. Int J Syst Evol

Microbiol 51: 999–1006.

Harder J (1997) Anaerobic degradation of cyclohexane-1,2-diol

by a new Azoarcus species. Archives Microbiol 168: 199–204.

Henze M, Harremoes P, Jansen JlC & Arvin E (2002) Wastewater

Treatment. Biological and Chemical Processes, 3rd ed. Springer,

Berlin, Germany.

Hess A, Zarda B, Hahn D, Haner A, Stax D, Hohener P & Zeyer J

(1997) In situ analysis of denitrifying toluene- and m-xylen-

degrading bacteria in a diesel fuel-contaminated laboratory

aquifer column. Appl Environ Microbiol 63: 2136–2141.

Heylen K, Vanparys B, Wittebolle L, Verstraete W, Boon N & De

Vos P (2006) Cultivation of denitrifying bacteria: optimization

of isolation conditions and diversity study. Appl Environ

Microbiol 72: 2637–2643.

Juretschko S, Timmermann G, Schmid M, Schleifer K-H,

Pommerening-Roser A, Koops H-P & Wagner M (1998)

Combined molecular and conventional analysis of nitrifying

bacterium diversity in activated sludge: Nitrosococcus mobilis

and Nitrosospira-like bacteria as dominant populations. Appl

Environ Microbiol 64: 3042–3051.

Juretschko S, Loy A, Lehner A & Wagner M (2002) The microbial

community composition of a nitrifying-dentrifying activated

sludge from an industrial sewage treatment plant analyzed by

the full-cycle rRNA approach. Syst Appl Microbiol 25: 84–99.

Kassen R & Rainey PB (2004) The ecology and genetics of

microbial diversity. Annu Rev Microbiol 58: 207–231.

Klausen MM, Thomsen TR, Nielsen JL, Mikkelsen LH & Nielsen

PH (2004) Variations in microcolony strength of probe-

defined bacteria in activated sludge flocs. FEMS Microbiol Ecol

50: 123–132.

Kloeke FO & Geesey GG (1999) Localization and identification of

populations of phosphatase-active bacterial cells associated

with activated sludge flocs. Microbial Ecol 38: 201–214.

Kong Y, Nielsen JL & Nielsen PH (2004) Microautoradiographic

study of Rhodocyclus-related poly-P accumulating bacteria in

full-scale EBPR plants. Appl Environ Microbiol 70: 5383–5390.

Kong Y, Nielsen JL & Nielsen PH (2005) Identity and

ecophysiology of uncultured actinobacterial polyphosphate-

accumulating organisms in full-scale enhanced biological

phosphorus removal plants. Appl Environ Microbiol 71:

4076–4085.

Kragelund C, Nielsen JL, Thomsen TR & Nielsen PH (2005) In

situ physiology of the filamentous Alphaproteobacterium

Meganema perideroedes in activated sludge. FEMS Microbiol

Ecol 54: 111–122.

Krieg NR (1984) Aerobic/microaerophilic, motile, helical/vibroid

gram-negative bacteria. Bergey’s Manual of Systematic

Bacteriology, Vol. 1 (Krieg NR & Holt JG, eds), pp. 104–110.

Williams & Wilkins, Baltimore, MD.

Lajoie CA, Layton AC, Gregory IR, Sayle GS, Taylor DE & Meyers

AJ (2000) Zoogleal clusters and sludge dewatering potential in

an industrial activated-sludge wastewater treatment plant. Wat

Environ Res 72: 56–64.

Lee N, Nielsen PH, Andreasen KH, Juretschko S, Nielsen JL,

Schleifer KH & Wagner M (1999) Combination of fluorescent

in situ hybridization & microautoradiography – a new tool for

structure function analyses in microbial ecology. Appl Environ

Microbiol 65: 1289–1297.

Leifson E (1962) The bacterial flora of distilled and stored water.

III. New species of the genera Corynebacteria, Flavobacterium,

Spirillum and Pseudomonas. Int Bull Bacteriol Nomencl Taxon

12: 161–170.

Lie E & Welander T (1997) A method for determination of the

readily fermentable organic fraction in municipal wastewater.

Wat Res 31: 1269–1274.

Loy A, Horn M & Wagner M (2003) probeBase – an online

resource for rRNA-targeted oligonucleotide probes. Nucleic

Acids Res 31: 514–516.

Manz W, Amann R, Ludwig W, Wagner M & Schleifer K-H

(1992) Phylogenetic oligodeoxynucleotide probes for the

major subclasses of proteobacteria: problems and solutions.

Syst Appl Microbiol 15: 593–600.

Mechichi T, Stackebrandt E, Gad’on N & Fuchs G (2002)

Phylogenetic and metabolic diversity of bacteria degrading

aromatic compounds under denitrifying conditions, and

description of Thauera phenylacetica sp. nov., Thauera

aminoaromatica sp. nov., and Azoarcus buckelii sp. nov. Arch of

Microbiol 178: 26–35.

Nielsen JL & Nielsen PH (2002a) Enumeration of acetate-

consuming bacteria by microautoradiography under oxygen-

and nitrate respiring conditions in activated sludge. Wat Res

36: 421–428.

Nielsen JL & Nielsen PH (2002b) Quantification of functional

groups in activated sludge by microautoradiography. Wat Sci

Technol 46: 389–395.

Nielsen PH, Raunkjær K, Norsker NH, Jensen NA & Hvitved-

Jacobsen T (1992) Transportation of wastewater in sewer

systems – a review. Wat Sci Technol 25: 17–31.

Nielsen JL, Mikkelsen LH & Nielsen PH (2001) In situ detection

of cell surface hydrophobicity of probe-defined bacteria in

activated sludge. Water Sci Technol 43: 97–103.

Nielsen PH, Roslev P, Dueholm TE & Nielsen JL (2002)

Microthrix parvicella, a specialized lipid consumer in

FEMS Microbiol Ecol 60 (2007) 370–382 c� 2007 Federation of European Microbiological SocietiesPublished by Blackwell Publishing Ltd. All rights reserved

381Ecophysiology of denitrifying bacteria in activated sludge

anaerobic-aerobic activated sludge plants. Water Sci Technol

46: 73–80.

Nielsen JL, Christensen D, Kloppenborg M & Nielsen PH (2003)

Quantification of cell-specific substrate uptake by probe-

defined bacteria under in situ conditions by

microautoradiography and fluorescence in situ hybridization.

Environ Microbiol 5: 202–211.

Nielsen PH, Thomsen TR & Nielsen JL (2004) Bacterial

composition of activated sludge–importance for floc and

sludge properties. Water Sci Technol 49: 51–58.

Olofsson AC, Zita A & Hermansson M (1998) Floc stability and

adhesion of green-fluorescent-protein-marked bacteria to

flocs in activated sludge. Microbiol 144: 519–528.

Ostle AG & Holt JG (1982) Nile blue A as a fluorescent stain for

poly-b-hydroxybutyrate. Appl Environ Microbiol 44: 238–241.

Philippot L & Hallin S (2005) Finding the missing link between

diversity and activity using denitrifying bacteria as a model

functional community. Curr Opinion Microbiol 8: 1–6.

Pot B, Gillis M & de Ley J (1991) The genus Aquaspirillum. The

Prokaryotes: A Handbook of the Biology of Bacteria:

Ecophysiology, Isolation, Identification, Application (Balows A,

Truper HG, Dworkin M, Harder W & Schleifer KH, eds), pp

2569–2582. Springer-Verlag, New York.

Purkhold U, Pommerening-Roser A, Juretschko S, Koops HP &

Wagner M (2000) Phylogeny of all recognized species of

ammonia-oxidizers based on comparative 16S rRNA and

amoA sequence analysis: implications for molecular diversity

surveys. Appl Environ Microbiol 66: 5368–5382.

Rosello-Mora RA, Wagner M, Amann R & Schleifer K-H (1995)

The abundance of Zoogloea ramigera in sewage treatment

plants. Appl Environ Microbiol 61: 702–707.

Scholten E, Lukow T, Auling G, Kroppenstedt R, Rainey F &

Diekmann H (1999) Thauera mechernichensis sp. nov., an

aerobic denitrifier from a leachate treatment plant. Int J Syst

Bacteriol 49: 1045–1051.

Seviour RJ & Blackall LL (1999) The Microbiology of Activated

Sludge. Kluwer academic publisher, Dordrecht.

Seviour RJ, Mino T & Onuki M (2003) The microbiology of

biological phosphorus removal in activated sludge systems.

FEMS Microbiol Rev 27: 99–127.

Song B, Young L & Palleroni N (1998) Identification of denitrifier

strain T1 as Thauera aromatica and proposal for emendation

of the genus Thauera definition. Int J Syst Bacteriol 48:

889–894.

Song B, Palleroni NJ, Kerkhof LJ & Haggblom MM (2001)

Characterization of halobenzoate-degrading, denitrifying

Azoarcus and Thauera isolates and description of

Thauera chlorobenzoica sp. nov. Int J Syst Evol Microbiol 51:

589–602.

Thomsen TR, Nielsen JL, Ramsing NB & Nielsen PH (2004)

Micromanipulation and further identification of FISH-

labelled microcolonies of a dominant denitrifying bacterium

in activated sludge. Environ Microbiol 6: 470–479.

Thomsen TR, Kragelund C & Nielsen PH (2006) Abundance and

physiology of Aquaspirillum-related filamentous bacteria in

activated sludge. Wat Sci Technol 54: 237–245.

van Loosdrecht MCM, Smolder GJ, Kuba T & Heijnen JJ (1997)

Metabolism of micro-organisms responsible for enhanced

biological phosphorus removal from wastewater. Antonie van

Leeuwenhoek 71: 109–116.

Wagner M & Loy A (2002) Bacterial community composition and

function in sewage treatment systems. Current Opinion

Biotechnol 13: 218–227.

Wagner M, Loy A, Nogueira R, Purkhold U, Lee N & Daims H

(2002) Microbial community composition and function in

wastewater treatment plants. Antonie van Leeuwenhoek 81:

665–680.

Zeng RJ, Saunders AM, Yuan Z, Blackall LL & Keller J (2003)

Identification and comparison of aerobic and denitrifying

polyphosphate-accumulating organisms. Biotechnol Bioeng 83:

140–148.

Zita A & Hermansson M (1997) Effects of bacterial cell surface

structures and hydrophobicity on attachment to activated

sludge. Appl Environ Microbiol 63: 1168–1170.

FEMS Microbiol Ecol 60 (2007) 370–382c� 2007 Federation of European Microbiological SocietiesPublished by Blackwell Publishing Ltd. All rights reserved

382 T.R. Thomsen et al.