790 © IWA Publishing 2016 Water Science & Technology | 73.

4 | 2016

Factors affecting population of filamentous bacteria in

wastewater treatment plants with nutrients removal
Aleksandra Miłobędzka, Anna Witeska and Adam Muszyński

ABSTRACT
Aleksandra Miłobędzka (corresponding author)
Filamentous population in activated sludge and key operational parameters of full-scale municipal
Adam Muszyński
wastewater treatment plants (WWTPs) with bulking problems representative for Poland were Department of Biology, Faculty of Environmental
Engineering,
investigated with quantitative fluorescence in situ hybridization. Statistical analyses revealed few Warsaw University of Technology,
00-653 Warsaw,
relationships between operational parameters and biovolume of filamentous bacteria. Sludge age Poland
E-mail: aleksandra_milobedzka@is.pw.edu.pl
was not only positively correlated with abundance of Chloroflexi (parametric correlation and
Anna Witeska
principal component analysis (PCA)), but also differentiated Microthrix population (analysis of
Department of Informatics and Environmental
variance (ANOVA)). Phylum Chloroflexi and pH presented a negative relation during the study (PCA). Quality Research, Faculty of Environmental
Engineering,
ANOVA showed that pH of influent and sludge volume index (SVI) differentiated abundance of types Warsaw University of Technology,
00-653 Warsaw,
0803 and 1851 of Chloroflexi and candidate division TM7. SVI increased along with higher abundance Poland
of Microthrix (positive parametric and non-parametric correlations and positive relation in PCA).
Biovolumes of morphotypes 0803 and 1851 of Chloroflexi were differentiated by organic matter in
influent, also by nutrients in the case of Chloroflexi type 1851. Chemical and biological oxygen
demands (COD and BOD5, respectively) were negatively correlated with Microthrix. COD also
differentiated the abundance of Haliscomenobacter hydrossis. Results of the study can be used to
prevent WWTPs from excessive proliferation of filamentous bacteria and operational problems
caused by them – bulking and foaming of activated sludge.
Key words | activated sludge, bulking, foaming, quantitative fluorescence in situ hybridization

INTRODUCTION

Bulking and foaming of activated sludge occurs during
excessive proliferation of a specific group of activated
sludge biocenosis–filamentous bacteria. Those microorgan-
isms impair settling and flocculating properties of sludge
in secondary clarifiers, which results in the presence of
solid particles in effluent (Kragelund et al. , ;
Miura et al. ). Poor settling ability of activated sludge
and deterioration of the floc structure limit the effectiveness
of dewatering processes (Eikelboom ). It also increases
costs of wastewater treatment and lowers the quality of final
effluent. The problems listed above arise mostly during
changes in operating conditions, mainly in terms of organic
load, nutrients and oxygen supply, and are reported by
Jenkins et al. () and Wanner & Jobbágy () around
the world. However, these problems are not easy to investi-
gate due to difficulties with identification of filamentous
bacteria according to morphology and chemical staining
alone (Nielsen et al. a; Wanner et al. ). For a
doi: 10.2166/wst.2015.541
long time Sphaerotilus was considered to be the main causa-
tive agent in this phenomenon, until Eikelboom ()
conducted several researches and isolated many filamentous
organisms, and wrote an identification key for filaments’
classification. Another important step in understanding fila-
mentous bulking and foaming was made in the late 1990s by
Blackall et al. (), Seviour et al. (, ) in Australia
and Tandoi et al. () in Italy, who isolated other key
players in filamentous population of activated sludge – Gor-
donia and Microthrix parvicella. Those studies together with
pure culture research with different substrates by Chiesa &
Irvine () and Jenkins et al. () provided the base for
comprehensive understanding of the ecophysiology of fila-
mentous bacteria in activated sludge. What is more,
thanks to that knowledge, control of filamentous bacteria
growth became possible and strategies developed on meta-
bolic and kinetic selection of microorganisms were used
(Wanner ).
 791 A. Miłobędzka et al. | Factors affecting population of filamentous bacteria in WWTPs with nutrient removal
Expansion of molecular methods targeting nucleic acids
resulted in expanding knowledge of identity of filaments in
WWTPs. Fluorescence in situ hybridization (FISH) allows
visualization of microorganisms with epifluorescence or
confocal laser scanning microscopy. It uses DNA oligonu-
cleotides with fluorochromes which bind to active cells
with specific abundant ribosomal ribonucleic acid (rRNA).
This method was optimized by Amann et al. (), and
allowed not only for identification, but also visualization
and quantification of bacteria without previous isolation.
That was a revolution in comprehension of microbial eco-
system functioning (Nielsen & McMahon ). More
reliable identification can be performed by culture-indepen-
dent molecular methods, but only a few surveys have been
carried out based on FISH analysis, as reviewed by Nielsen
et al. (a), Wanner et al. () and Wanner & Jobbágy
(). Furthermore, determination of the impact of techno-
logical parameters on the populations of filamentous
bacteria in activated sludge is also problematic (Eikelboom
; Mielczarek et al. ).
Miłobe˛ dzka & Muszyński (2015) showed that filamen-
tous populations in five Polish WWTPs were rather similar
to each other and only the abundance of Microthrix was dif-
ferentiated by season. However, studies providing larger
numbers of data, appropriate for statistical analyses, are
needed to reveal factors that affect filamentous bacteria
populations.
AIM
A 3-year study on population dynamics of filamentous bac-
teria in five Polish full-scale municipal WWTPs with
nutrients removal was carried out. Quantitative fluor-
escence in situ hybridization (qFISH) was applied to
evaluate the abundance of 11 groups of filamentous bac-
teria. Factors affecting the microbial population were
searched for using statistical analyses.
This study was conducted to find relations between the
filamentous bacteria in activated sludge and key parameters
in five full-scale municipal plants with bulking problems
representative for Poland. WWTPs differed in the configur-
ation of reactors and main technological parameters of
wastewater treatment process. During the study, chemical
and operational data like biological and chemical oxygen
demands in the process tank, total nitrogen and total phos-
phorus in influent, sludge age, sludge volume index (SVI)
were measured to assess their influence on the microbial
Water Science & Technology | 73.4 | 2016
community. qFISH was used to define abundance of fila-
mentous bacteria.
MATERIALS AND METHODS
Sampling
The study was carried out for 3 years, starting from 2011.
Five full-scale municipal WWTPs, differing in treated waste-
water and operational parameters, were selected for
research. Activated sludge samples were collected after
winter at the beginning of March, and after summer at the
end of September, following the procedure recommended
by Nielsen et al. (b).
FISH identification
FISH procedures were performed according to Nielsen et al.
(b). The majority of the bacterial community was targeted
with the 6-Fam labelled universal probe EUBmix (equimolar
mixture of EUB338, EUB338II and EUB338III). Filamentous
bacteria were identified with a wide selection of 11 oligop-
robes (labelled with Cy3): CFXmix (equimolar concentration
of GNSB-941 and CFX1223, targeting phylum Chloroflexi);
T0803 (Chloroflexi type 0803); Chl1851 (Chloroflexi type
1851); MPAmix (equimolar concentration of MPA645,
MPA223 and MPA60, targeting morphotype Microthrix);
G123T (Thiothrix/type 021N); Myc657 (Mycolata group);
Spin1449 (Skermania piniformis); Gor596 (family Gordonia);
HHY654 (Haliscomenobacter hydrossis); Curvi997 (types
1701 and 0041/0675); and TM7905 (type 0041/0675).
Detailed information about the oligoprobes can be found in
probeBase (Loy et al. ).
Quantification of probe-defined bacteria
Quantification procedures were performed similarly to
Mielczarek et al. (). Twenty separate images for each
specific probe and another 20 corresponding images with
EUBmix were captured with a Nikon Eclipse 50i micro-
scope. ImageJ software (Collins ) was used to
determine the biovolume of bacteria. The microbial abun-
dance was quantified as the percentage of pixel area
fluorescing with a specific probe (Cy3-labelled) in relation
to the area fluorescing with the EUBmix probe (6-Fam-
labelled) and it was then calculated as a mean of 20 separate
measurements (expressed as % of the EUBmix probe).
 792 A. Miłobędzka et al. | Factors affecting population of filamentous bacteria in WWTPs with nutrient removal
Statistical analyses
Standard statistical analyses, correlation analyses (para-
metric with Pearson product-moment correlation
coefficient and non-parametric with Spearman’s rank corre-
lation coefficient), analysis of variance (ANOVA) with
significance level 0.05, principal component analysis
(PCA) and cluster analysis were carried out in order to
find differentiating factors and the strength of relationships
between quantified bacteria populations and wastewater
and operational data. Cluster analysis was performed, the
Ward’s method, using variance analysis to calculate the dis-
tance between the clusters. For cluster formation, the
Euclidean distance was chosen. All analyses were
performed in STATISTICA™ from StatSoft®.
RESULTS
Investigated wastewater treatment plants
All tested plants reported operational problems connected
with settling of activated sludge, while foam was observed
on the tanks only in WWTP IV. All five full-scale municipal
investigated WWTPs had biological N-removal (nitrification
and denitrification) and four of them (except WWTP II) also
the enhanced biological phosphorus removal (EBPR) step.
The plants ranged in size: 18,000–110,000 population equiva-
lents (PE); also the fraction of industrial contribution to the
organic matter in the influent depended on the plant, and it
came mainly from the food industry and constituted 0–50%.
All of the plants, except WWTP V dosed iron-based coagulants
(PIX) to improve phosphorus elimination. WWTP V occasion-
ally used polyaluminium chloride compounds (PAX), mostly
during autumn and winter, to improve the settling properties
and to control excessive growth of filamentous bacteria. The
main operational parameters of selected plants are described
in Table 1; more details can be found elsewhere (Miłobe˛ dzka
& Muszyński 2015).
Identity and abundance of filamentous bacteria
Filamentous bacteria were abundant in all tested plants and
constituted on average 23 ± 2% of all bacteria identified by
the universal probe EUBmix. The most abundant bacterial
groups belonged to phylum Chloroflexi (CFXmix probe;
12% of all bacteria), genus Microthrix (MPAmix probe;
7%), and species Haliscomenobacter hydrossis (HHY654
probe; 2%). Chloroflexi type 0803, Chloroflexi type 1851
Water Science & Technology | 73.4 | 2016
and candidate division TM7 accounted for minor fractions
and usually did not exceed 2%. Mycolata and Skermania
piniformis occurred just once, whereas Curvibacter, Thio-
thrix/021N and Gordonia were not found in any of the
tested samples. More details about the abundance of specific
bacteria in filamentous populations in Polish WWTPs can
be found in a study focused on dynamics of filamentous bac-
teria in Polish plants by Miłobe˛ dzka & Muszyński (2015).
Correlation analyses
Only medium strength correlations were found between the
abundance of specific filamentous bacteria and chemical or
operational parameters (Table 2). Abundance of morpho-
types 1851 and 0803 of phylum Chloroflexi was
differentiated by several factors described below. On the
other hand, Chloroflexi hybridizing with the broad phylum
probe CFXmix were connected only with sludge age (posi-
tive parametric correlation; r ¼ 0.453 with a significance
level alpha ¼ 0.012). What is more, Chloroflexi that were
not further identified (not classified to morphotypes 0803
and 1851) were also in positive but non-parametric corre-
lation to sludge age (r ¼ 0.412, alpha ¼ 0.024).
Chemical oxygen demand (COD; non-parametric
r ¼ 0.442, alpha ¼ 0.015) and biological oxygen demand
(BOD5; parametric r ¼ 0.378, alpha ¼ 0.040; non-parametric
r ¼ 0.443, alpha ¼ 0.014) were positively correlated with
abundance of morphotype Chloroflexi type 1851 (Table 2).
Microthrix (targeted by MPAmix) contributed to the SVI
increase, which was confirmed by positive parametric
(r ¼ 0.474, alpha ¼ 0.008) and non-parametric (r ¼ 0.446, alpha
¼ 0.010) correlations. This genus was in negative relation to
organic matter in influent (expressed as COD; parametric and
non-parametric r ¼ 0.372, alpha ¼ 0.043) and BOD5 (non-
parametric r ¼ 0.511, alpha ¼ 0.004) and total N (non-para-
metric r ¼ 0.465, alpha ¼ 0.010), as shown by negative
correlations. Positive connection was found between Micro-
thrix and total P (positive parametric: r ¼ 0.574, alpha ¼ 0.001
and non-parametric: r ¼ 0.431, alpha ¼ 0.017 correlations).
Candidate division TM7 was in positive parametric cor-
relation (r ¼ 0.387, alpha ¼ 0.035) with pH in influent
(Table 2).
Analysis of variance
ANOVA showed that abundance of Chloroflexi type 0803
was differentiated by COD, BOD5, pH and SVI. Those par-
ameters, together with total P and total N also differentiated
Chloroflexi type 1851.
 793 A. Miłobędzka et al. | Factors affecting population of filamentous bacteria in WWTPs with nutrient removal Water Science & Technology | 73.4 | 2016

Table 1 | Influent and operational parameters (mean values of data used for statistical analyses) of the WWTPs tested in this survey

Parameter WWTP I WWTP II WWTP III WWTP IV WWTP V

Size designed (PE) 55,400 83,000 163,500 53,040 22,500

Size actual (PE) 73,400 99,000 110,000 76,000 18,000
Reactor type A2O AO Anaerobic tank þ anoxic/ UCT A2O
aerobic oxidation ditch
(oxidation-reduction
potential controller)
Predenitrification Yes No No No Yes
Presettling Yes Yes Yes Yes No
Fermenter Yes No No No No
Aeration Fine bubble Surface (vertical type) Surface (horizontal rotors) Fine bubble diffusers Fine bubble
diffusers diffusers
Wastewater type Domestic Domestic; industrial Domestic; industrial Domestic; industrial Domestic
(% of overall (20–25%): (5–10%): fruit and (30-50%): fruit and
BOD5) slaughterhouse, dairy vegetables processing, vegetables processing,
breweries, landfill leachate sugar refining
SVI [mL/g] Summer 218 178 223 89 139
Winter 233 169 194 82 158
COD [mg/L] 428 670 954 1,333 1,457
BOD5 [mg/L] 268 311 352 927 431
N total [mg/L] 66 69.4 83.3 72.8 87.2
P total [mg/L] 10.6 10.8 17.1 15.5 20
pH 7.7 7.6 7.7 7.4 8.4
MLSS [g/L] 3.8 4.5 5.6 7.7 4.0
SRT [d] 25 37 39 37 18

MLSS, mixed liquor suspended solids; SRT, sludge retention time.

Process configurations: A2O, anaerobic–anoxic–aerobic; AO, anaerobic–aerobic.

Sludge age influenced the population of Microthrix;
usually biovolume of this bacteria decreased when sludge
age increased.
Abundance of filamentous candidate division TM was
influenced by influent pH (both parameters slightly increased,
but no visible link was noticed), total N (both increased) and
SVI (higher abundance of bacteria when SVI decreased).
COD was the only factor differentiating in a statistically
significant way the biovolume of H. hydrossis, and no
pattern in this relation was observed.
Principal component analysis
The data were standardized in order to perform a transform-
ation that would facilitate the comparison of values from
numerous variables, irrespective of their distribution and
the units they had been measured in. The analysis of
factor coordinates points to a high correlation between
BOD5, COD, Ntotal and Ptotal in the data set, within the
first component. A high correlation within the second com-
ponent involves pH and H. hydrossis (Table 3).
On the basis of the scree plot (not shown) and the infor-
mation about the variance accounted for by particular
components, two principal components, which explained
57% of variance, were identified. The biplot of the variables
projection onto the factor plane is presented in Figure 1.
Connection between phylum Chloroflexi and sludge age
shown by parametric correlation was also found with PCA.
The analysis also revealed negative relation between phylum
Chloroflexi and pH (Figure 1). Another relation which was
confirmed by PCA was found for genus Microthrix and the
SVI (Figure 1).
Cluster analysis
Connection between Microthrix and SVI and an obvious
link between Chloroflexi and the group named in this
study – ‘other Chloroflexi’ (further not identified fraction,
 794 A. Miłobędzka et al. | Factors affecting population of filamentous bacteria in WWTPs with nutrient removal Water Science & Technology | 73.4 | 2016

Table 2 | Results of statistical analyses performed to find differentiating factors and relationships between quantified bacterial populations and chemical/operational data

Probe-defined bacteria Relationship Sludge age BOD5 COD N P SVI pH

Chloroflexi type 0803 Parametric correlation

Non-parametric correlation
ANOVA þ þ þ þ
PCA
Chloroflexi type 1851 Parametric correlation þ
Non-parametric correlation þ þ
ANOVA þ þ þ þ þ þ
PCA
Chloroflexi Parametric correlation þ
Non-parametric correlation
ANOVA
PCA þ 
Other Chloroflexi Parametric correlation
Non-parametric correlation þ
ANOVA
PCA
H. hydrossis Parametric correlation
Non-parametric correlation
ANOVA þ
PCA
Microthrix Parametric correlation   þ þ
Non-parametric correlation   þ þ
ANOVA þ
PCA þ
TM7 Parametric correlation þ
Non-parametric correlation
ANOVA þ þ þ
PCA

BOD, biological oxygen demand; COD, chemical oxygen demand; N, total nitrogen; P, total phosphorus; SVI, sludge volume index; ‘ þ ’, positive relations, ‘  ’ negative relations. All pre-
sented correlations had medium strength.

Table 3 | Variable factor coordinates based on correlation result of morphotypes conducted from bacteria hybridized
with CFXmix) can be seen in Figure 2. Different distinctions
F1 F2
can be noticed for the three groups: (1) Chloroflexi type
Chloroflexi type 0803 0.27 0.19 0803, H. hydrossis, pH, Microthrix and SVI; (2) Chloroflexi
Other Chloroflexi 0.24 0.83 and sludge age; and (3) with wastewater parameters:
Chloroflexi 0.12 0.86 BOD5, COD, Ptotal and Ntotal constituting a different set.
H. hydrossis 0.21 0.40
Microthrix 0.50 0.35
Sludge age 0.11 0.72 DISCUSSION
BOD5 0.87 0.09
COD 0.95 0.20 Full understanding of the bulking mechanism has still not
Ntotal 0.86 0.10 been achieved. Reasons for growth of filaments are very
Ptotal 0.90 0.09 complex; however, some factors can be designated from a
SVI 0.68 0.30
practical approach: high levels of sulfide, long solids reten-
tion time (SRT), low levels of food-to-microorganisms
pH 0.07 0.56
(F/M), nutrients, dissolved oxygen (DO) and pH (Wanner
 795 A. Miłobędzka et al. | Factors affecting population of filamentous bacteria in WWTPs with nutrient removal
Figure 1 | Principal component analysis showing differences and similarities in abun-
dance of filamentous bacteria and chemical and operational parameters.
Variables plotted onto the two-factor plane.
Figure 2 | Cluster analysis of investigated bacteria and parameter of wastewater.
; Jenkins et al. ; Grady et al. ). The conducted
study allowed the connection of abundance of filamentous
bacteria with other factors.
Factors affecting population structure
Sludge age was positively correlated with abundance of
Chloroflexi (parametric correlation). A similar observation
was made for a group of Chloroflexi which was not investi-
gated in this study – named ‘Other Chloroflexi’. An
Water Science & Technology | 73.4 | 2016
example of the same connection was noticed for bacteria,
which could be classified in this group and may be abundant
in Polish WWTPs – morphotype 0914 hybridizing with
probe CFX67. Abundance of type 0914 has been examined
in Australian WWTPs by Speirs et al. () and was
higher in biological nutrient removal plants with long
sludge age plants.
It was also revealed by ANOVA that sludge age differen-
tiates Microthrix population; which is not surprising,
because high sludge age is required for growth of Candida-
tus M. parvicella (Rossetti et al. ). For the same
reason – slow growth rate and difficulties in maintenance
in pure culture it is still named just ‘Candidatus’ and full
characterization is not possible (Blackall et al. ; Rossetti
et al. ; McIlroy et al. ).
pH of influent and SVI differentiated abundance of
types 0803 and 1851 of Chloroflexi and candidate division
TM7 (ANOVA). Although SVI depends on an abundance
of filamentous bacteria, its influence on listed, quantified
microorganisms can be the result of a different relation. A
very strong correlation with other factors could have con-
cealed the causative agent. Probably, a relation with
Microthrix (described later) and proliferation of this bacteria
could have decreased abundance of types 0803 and 1851 of
Chloroflexi and candidate division TM7.
Phylum Chloroflexi and pH presented different behav-
iour during the study (PCA). The effect of those changes in
pH is hard to interpret because only slight changes in this
parameter were noticed during the whole period of study.
Examples of pH influence on the abundance of Chloroflexi
can be found in the literature (Miura et al. ), but bacteria
were investigated in pilot-scale submerged membrane bio-
reactors and did not hybridize with probes for
morphotypes found in Polish WWTPs. However, Miura
et al. () stated that utilization of glucose and N-acetyl
glucosamine by bacteria affiliated with the Chloroflexi
subphylum 1 was low at low pH.
SVI increased with higher abundance of Microthrix the
dependence in the study was so strong that it was confirmed
by all analyses: positive parametric and non-parametric corre-
lations, similar relation in PCA, and cluster analysis. Similar
results were presented by Westlund et al. (). Abundant
Microthrix with filaments protruding from flocs corresponded
to high SVI. On the other hand, a different study by Knoop &
Kunst () showed no negative effect on settleability of acti-
vated sludge with M. parvicella – bacteria did not impair the
structure of flocs, because filaments from this genus were short.
Haliscomenobacter hydrossis, Chloroflexi and Microthrix
belong to specialized filamentous bacteria involved in
 796 A. Miłobędzka et al. | Factors affecting population of filamentous bacteria in WWTPs with nutrient removal
degradation of complex matter, but they utilize different sub-
strates. Multiparticulates are decomposed with egzoenzymes
like glucuronidase, phosphatase and chitinase, esterase by
filamentous members of Bacteroidetes and Chloroflexi
(Nielsen et al. ). In the study, the abundance of bacteria
from Bacteroidetes H. hydrossis was differentiated by COD
of wastewater, while morphotypes 0803 and 1851 of Chloro-
flexi were also affected by BOD5. Phylum presents activity of
egzoenzymes like galactosidase and protease.
Although Chloroflexi type 0803 has been shown by
Kragelund et al. () as able to store and to accumulate
intracellular polyphosphate (polyphosphate granules were
observed), in the presented study the second investigated
Chloroflexi morphotype’s biovolume – type 1851 was differ-
entiated by concentration of nutrients in influent.
Candidatus M. parvicella can secrete lipase and esterase
and assimilate long chain fatty acids under all e-acceptor
conditions. Microthrix uptake different substrate – lipids,
in various conditions (Nielsen et al. ), and that is why
it can be in opposite relation to bacteria described earlier.
COD and BOD5 were negatively correlated with biovolume
of Microthrix.
In future it is advisable to use more advanced, modern
techniques like DNA extraction, sequencing library prep-
aration and sequencing or ribosomal intergenic spacer
analysis and denaturing gradient gel electrophoresis to con-
firm results of identification of filamentous bacteria in
Poland. Moreover, the development of new FISH probes
for unidentified fraction Chloroflexi gives new possibilities
to expand the scope of research and precisely classify
which group of Chloroflexi studied factors had the most
influence. Presented results of the study could be verified
in a controlled laboratory-scale reactor. Another interesting
aspect of such research would be knowing not only which
factors have an influence on bacteria, but to what extent
and what causes the connection.
CONCLUSIONS
Populations of filamentous bacteria in WWTPs with nutri-
ents removal were affected by several factors. FISH
investigation and statistical analyses revealed the following:
• Sludge age was positively correlated with abundance of
Chloroflexi and differentiated abundance of Microthrix.
• pH of influent and SVI differentiated abundance of types
0803 and 1851 of Chloroflexi and candidate division
Water Science & Technology | 73.4 | 2016
TM7. Phylum Chloroflexi and pH presented different
behaviour during the study.
• SVI increased with higher abundance of Microthrix (posi-
tive parametric and non-parametric correlations).
• Organic matter in influent (expressed as COD and
BOD5) differentiated abundance of morphotypes 0803
and 1851 of Chloroflexi, and was also negatively corre-
lated with Microthrix. COD of wastewater differentiated
abundance of H. hydrossis.
• Nitrogen and phosphorus in influent differentiated bio-
volume of Chloroflexi type 1851.
REFERENCES
Amann, R. I., Binder, B. J., Olson, R. J., Chisholm, S. W.,
Devereux, R. & Stahl, D. A.  Combination of 16S rRNA-
targeted oligonucleotide probe with flow cytometry for
analyzing mixed microbial populations. Appl. Environ.
Microbiol. 56, 1919–1925.
Blackall, L. L., Seviour, E. M., Cunningham, M. A., Seviour, R. J.
& Hugenholtz, P.  ‘Microthrix parvicella’ is a novel, deep
branching member of the actinomycetes subphylum. Syst.
Appl. Microbiol. 17, 513–518.
Chiesa, S. C. & Irvine, R. L.  Growth and control of
filamentous microbes in activated sludge: an integrated
hypothesis. Water Res. 19, 471–479.
Collins, T. J.  Imagej for microscopy. Biotechniques 43, 25–30.
Eikelboom, D. H.  Filamentous organisms observed in
activated sludge. Water Res. 9, 365–388.
Eikelboom, D. H.  Process Control of Activated Sludge Plants
by Microscopic Investigation. IWA Publishing, London, UK.
Grady Jr., C. P. L., Daigger, G. T., Love, N. G. & Filipe, C. 
Biological Wastewater Treatment, 3rd edn. CRC Press,
New York, USA.
Jenkins, D., Richard, M. C. & Daigger, G. T.  Manual on the
Causes and Control of Activated Sludge Bulking and
Foaming. Lewis Publishers, New York, NY, USA.
Jenkins, D., Richard, M. & Daigger, G.  Manual on the
Causes and Control of Activated Sludge Bulking, Foaming
and Other Solids Separation Problems. IWA Publishing,
CRC Press, London, UK.
Knoop, S. & Kunst, S.  Influence of temperature and sludge
loading on activated sludge settling, especially on Microthrix
parvicella. Water Sci. Technol. 37 (4–5), 27–35.
Kragelund, C., Levantesi, C., Borger, A., Thelen, K., Eikelboom,
D., Tandoi, V., Kong, Y., van der Waarde, J., Krooneman, J.,
Rossetti, S., Thomsen, T. R. & Nielsen, P. H.  Identity,
abundance and ecophysiology of filamentous Chloroflexi
species present in activated sludge treatment plants. FEMS
Microbiol. Ecol. 59 (3), 671–682.
Kragelund, C., Thomsen, T. R., Mielczarek, A. T. & Nielsen, P. H.
 Eikelboom’s morphotype 0803 in activated sludge
belongs to the genus Caldilinea in the phylum Chloroflexi.
FEMS Microbiol. Ecol. 76 (3), 451–462.
 797 A. Miłobędzka et al. | Factors affecting population of filamentous bacteria in WWTPs with nutrient removal
Loy, A., Horn, M. & Wagner, M.  Probebase: an online
resource for rRNA-targeted oligonucleotide probes. Nucleic
Acids Res. 31, 514–516.
McIlroy, S. J., Kristiansen, R., Albertsen, M., Karst, S. M., Rossetti,
S., Nielsen, J. L. & Nielsen, P. H.  Metabolic model
for the filamentous ‘Candidatus Microthrix parvicella’ based
on genomic and metagenomic analyses. ISME J. 7 (6),
1161–1172.
Mielczarek, A. T., Kragelund, C., Eriksen, P. S. & Nielsen, P. H.
 Population dynamics of filamentous bacteria in Danish
wastewater treatment plants with nutrient removal. Water
Res. 46 (12), 3781–3795.
Miłobe˛ dzka, A. & Muszyński, A.  Population dynamics of
filamentous bacteria identified in Polish full-scale wastewater
treatment plants with nutrients removal. Water Sci. Technol.
71 (5), 675–684.
Miura, Y., Watanabe, Y. & Okabe, S.  Significance of
Chloroflexi in performance of submerged membrane
bioreactors (MBR) treating municipal wastewater. Environ.
Sci. Technol. 41, 7787–7794.
Nielsen, P. H. & McMahon, K. D.  Microbiology and
microbial ecology of the activated sludge process. In:
Activated Sludge–100 Years and Counting (D. Jenkins & J.
Wanner, eds). IWA Publishing, London, UK, pp. 171–194.
Nielsen, P. H., Kragelund, C., Seviour, R. J. & Nielsen, J. L. a
Identity and ecophysiology of filamentous bacteria in
activated sludge. FEMS Microbiol. Rev. 33, 969–998.
Nielsen, P. H., Daims, H. & Lemmer, H. (eds) b FISH
Handbook for Biological Wastewater Treatment. IWA
Publishing, London, UK.
Nielsen, P. H., Mielczarek, A. T., Kragelund, C., Nielsen, J. L.,
Saunders, A. M., Kong, Y., Hansen, A. A. & Vollertsen, J. 
A conceptual ecosystem model of microbial communities in
First received 29 July 2015; accepted in revised form 7 October 2015. Available online 27 October 2015
Water Science & Technology | 73.4 | 2016
enhanced biological phosphorus removal plants. Water Res.
44 (17), 5070–5088.
Rossetti, S., Tomei, M. C., Nielsen, P. H. & Tandoi, V. 
‘Microthrix parvicella’, a filamentous bacterium causing
bulking and foaming in activated sludge systems: a review of
current knowledge. FEMS Microbiol. Rev. 29 (1), 49–64.
Seviour, E. M., Williams, C. J., Seviour, R. J., Soddell, J. A. &
Lindrea, K. C.  A survey of filamentous bacterial
populations from foaming activated sludge plants in eastern
states of Australia. Water Res. 24, 493–498.
Seviour, E. M., Williams, C., DeGrey, B., Soddell, J. A., Seviour, R. J.
& Lindrea, K. C.  Studies on filamentous bacteria from
Australian activated sludge plants. Water Res. 28, 2335–2342.
Speirs, L. B. M., McIlroy, S. J., Petrovski, S. & Seviour, R. J. 
The activated sludge bulking filament Eikelboom
morphotype 0914 is a member of the Chloroflexi. Environ.
Microbiol. Rep. 3 (2), 159–165.
Tandoi, V., Rossetti, S., Blackall, L. L. & Majone, M.  Some
physiological properties of an Italian isolate of ‘Microthrix
parvicella’. Water Sci. Technol. 37 (1–2), 1–8.
Wanner, J.  Activated Sludge Bulking and Foaming Control.
Technomic Publishing Company, Lancaster, PA, USA.
Wanner, J. & Jobbágy, A.  Activated sludge solids separation. In:
Activated Sludge – 100 Years and Counting (D. Jenkins & J.
Wanner, eds). IWA Publishing, London, UK, pp. 171–194.
Wanner, J., Kragelund, C. & Nielsen, P. H.  Microbiology of
bulking. In: Microbial Ecology of Activated Sludge (R. J.
Seviour & P. H. Nielsen, eds). IWA Publishing, London, UK,
pp. 191–214.
Westlund, Å. D., Hagland, E. & Rothman, M.  Bulking and
foaming caused by Microthrix parvicella at three large sewage
treatment plants in the Greater Stockholm area. Water Sci.
Technol. 34, 281–287.