Labeling of Microthrix parvicella in situ: A novel FRET probe based on bisoctyl rhodamine B

Article abstract of DOI:10.1016/j.saa.2019.01.057

Filamentous bacteria, particularly Microthrix parvicella, are mainly responsible for bulking or foaming of activated sludge. Based on the affinity of M. parvicella to the hydrophobic characteristics of long-chain fatty acids, a novel bisoctyl rhodamine B (BORB) and a novel fluorescence resonance energy transfer (FRET) complex probe were prepared herein to study their properties. When the FRET probe was used in in situ activated sludge, M. parvicella was clearly labeled at 20 nmol/L, which was a reduction of 50 times compared to that of the BORB (1 μmol/L) alone and 500 times compared to the carbazole–quinoline probe reported previously. Compared with fluorescence in situ hybridization, M. parvicella could be clearly labeled using BORB and the FRET probe in situ without requiring complicated pretreatments (i.e., shock and broken process, fixed sample, digestion, and lysozyme treatment). This study discusses the facile approach developed for labeling M. parvicella in early warning expansion, thereby inhibiting and controlling sludge bulking in situ.

Full text of DOI:10.1016/j.saa.2019.01.057

Accepted Manuscript
Labeling of Microthrix parvicella in situ: A novel FRET probe
based on bisoctyl rhodamine B
Yingchun Gu, Dayong Lin, Xuening Fei, Yongqiang Chen,
Cuihong Wang, Qi Yang, Yalin Tang
PII:
S1386-1425(19)30063-0
https://doi.org/10.1016/j.saa.2019.01.057
SAA 16743
DOI:
Reference:
Spectrochimica Acta Part A: Molecular and Biomolecular
Spectroscopy
To appear in:
Received date: 8 July 2018
Revised date:
Accepted
date:
7 January 2019
1
5 January 2019
Please cite this article as: Yingchun Gu, Dayong Lin, Xuening Fei, Yongqiang Chen,
Cuihong Wang, Qi Yang, Yalin Tang , Labeling of Microthrix parvicella in situ: A
novel FRET probe based on bisoctyl rhodamine B. Saa (2019), https://doi.org/10.1016/
j.saa.2019.01.057
This is a PDF file of an unedited manuscript that has been accepted for publication. As
a service to our customers we are providing this early version of the manuscript. The
manuscript will undergo copyediting, typesetting, and review of the resulting proof before
it is published in its final form. Please note that during the production process errors may
be discovered which could affect the content, and all legal disclaimers that apply to the
journal pertain.

ACCEPTED MANUSCRIPT
Labeling of Microthrix parvicella In Situ: A Novel FRET Probe Based on Bisoctyl
Rhodamine B
1
#
#2
1,2
1
1
Yingchun Gu , Dayong Lin , Xuening Fei *, Yongqiang Chen , Cuihong Wang , Qi
1
3
Yang , Yalin Tang *
1School of Science, Tianjin Engineering Technology Center of Chemical Wastewater
Source Reduction and Recycling, Tianjin Chengjian University, Tianjin, 300384, China;
jvgugugu@126.com
2School of Environmental Science and Engineering, Tianjin University, Tianjin, 300072,
China; xueningfei@126.com
3Beijing National Laboratory for Molecular Sciences, Institute of Chemistry, Chinese
Academy of Sciences (ICCAS), Beijing, 100190, China; tangyl@iccas.ac.cn
*
Correspondence: xueningfei@126.com (Fei), tangyl@iccas.ac.cn (Tang)
#These authors contributed equally to this work and should be considered co-first authors.
Abstract
Filamentous bacteria, particularly Microthrix parvicella, are mainly responsible for bulking
or foaming of activated sludge. Based on the affinity of M. parvicella to the hydrophobic
characteristics of long-chain fatty acids, a novel bisoctyl rhodamine B (BORB) and a novel
fluorescence resonance energy transfer (FRET) complex probe were prepared herein to
study their properties. When the FRET probe was used in in situ activated sludge, M.
parvicella was clearly labeled at 20 nmol/L, which was a reduction of 50 times compared
to that of the BORB (1 μmol/L) alone and 500 times compared to the carbazole–quinoline
probe reported previously. Compared with fluorescence in situ hybridization, M. parvicella
could be clearly labeled using BORB and the FRET probe in situ without requiring
complicated pretreatments (i.e., shock and broken process, fixed sample, digestion, and
lysozyme treatment). This study discusses the facile approach developed for labeling M.
parvicella in early warning expansion, thereby inhibiting and controlling sludge bulking in
situ.

ACCEPTED MANUSCRIPT
Keywords Bisoctyl rhodamine B, FRET probe, Microthrix parvicella, properties, labeling
1
. Introduction
Sludge bulking occurs frequently in wastewater treatment plants (WWTPs)[1, 2], thereby
decreasing pollutant treatment efficiency, producing plenty of bubbles, and causing some
serious operational problems. Sludge bulking is mainly related to the excessive proliferation
of filamentous bacteria[3-5], of which Microthrix parvicella is one of the most frequently
responsible factors for these problems, as indicated by global surveys[6-8]
.
Effective identification of M. parvicella is a precondition to perceive and prevent its
proliferation in the early stages of sludge bulking[9, 10]. Fluorescence in situ hybridization
(
FISH) and polymerase chain reaction (PCR)[10-12] are widely used to identify M. parvicella.
However, these methods are time consuming as well as costly and have low cell
permeability, low fluorescent pretreatment, narrow application scope, and complicated
processes[13-15]. Based on the surface affinity of M. parvicella to the hydrophobic
characteristics of long-chain fatty acids (LCFAs), carbazole dye containing a long-chain
alkane is suitable for labeling M. parvicella in in situ sludge samples[13]. Quantum dots
(
QDs) with hydrophobic long-chain alkanes and high working concentrations (0.010
mmol/L) were also studied, which provided a high detection limit and low sensitivity to
effectively identify M. parvicella[16]
.
Fluorescence resonance energy transfer (FRET)
imaging offers location and
quantitative information at the molecular level and is used to research biologic mechanisms
with high selectivity and high sensitivity. This method has attracted considerable attention
in the application of nucleic acid structures, cancer recognition, and drug delivery[17, 18]
.
However, FRET probes have not been applied for labeling or detecting M. parvicella in situ.
Due to good solubility, excellent photostability, high extinction coefficient, and high
fluorescence quantum yields, rhodamine B (RhB) has been widely used as a fluorophore
FRET system to detect DNA, metal ions, small molecules, and other analytes[19-23]. However,
RhB or its derivative has not been applied for labeling or detecting M. parvicella in situ. In
this study, based on the affinity of M. parvicella to LCFAs, a novel rhodamine B derivative,
bisoctyl rhodamine B (BORB), was synthesized and modified using the hydrophobic bisoctyl
group. Using BORB as the energy acceptor and QDs as the energy donor, a novel FRET
probe was developed for labeling M. parvicella in situ.

ACCEPTED MANUSCRIPT
2
2
2
. Materials and Methods
.1. Materials and instruments
.1.1. Chemical and biochemical reagents
Sodium borohydride (≥99%), tellurium powder (AR), cadmium chloride (AR),
3
-dimethylaminophenol (≥98%), and 3-aminophenol (≥98%) were supplied by
Shanghai Macklin Biochemical Co., Ltd (China) FISH reagents were supplied by Takara.
Other commonly used reagents and solvents were purchased from local companies.
2
.1.2. Instruments
Nuclear magnetic resonance (NMR) spectra were obtained in CDCl using a Bruker
3
AVANCE III 400 MHz spectrometer. Mass spectrometric analysis (MS) was conducted
using an electrospray ionization mass spectrometer. High-performance liquid
chromatography (HPLC) was used to analyze the chemical purity on a Lab Alliance
PC4006A with the mobile phase formed by 90% methanol and 10% water and a column of
2
reversed phase C18 ODS (5 μm, 4.6 × 250 mm ). UV–Vis and fluorescence spectra were
obtained using Perkin Elmer UV–Vis Spectrometer (TU-1900) and HITACHIF-7000
fluorescence spectrophotometer, respectively, with excitation and emission slits both at 5 nm
and a PTM voltage of 400 V.
2
.2. Synthesis of quantum dots
Te + 4NaBH + 7H O → 2NaHTe + Na
Sodium borohydride (37.8 mg) and tellurium powder (25.5 mg) were added to ultrapure
2
4
2
2
B
4
O
7
+ 14H
2
water (5 mL) and heated to 40°C–50°C for 30 min under nitrogen protection to obtain the
NaHTe precursor based on the following chemical reactions:
2
+
+
Cd + TGA → (Cd-TGA)
+
2+
(
Cd-TGA) + TGA → (Cd-TGA)
2
+
(
Cd-TGA) + NaHTe → (CdTe)-TGA
Cadmium chloride (25.5 mg) and mercaptoacetic acid, also known as thioglycolic acid
(TGA), (37.8 mg) were added to ultrapure water (50 mL), and 1 mol/L of NaOH under

ACCEPTED MANUSCRIPT
magnetic agitation was used to adjust the solution pH to ~9–10. Then, under nitrogen
protection, the NaHTe precursor (5 mL) was added and heated to 90°C for 5–10 min in a
water bath to obtain the QD solution. The final nitrogen-protected QD solution was cooled to
room temperature and stored at 4°C.
2
.3. Synthesis of bisoctyl rhodamine B
BORB was synthesized using 3-(dimethylamino)phenol and isobenzofuran-1,3-dione as
starting materials via the Friedel–Crafts acylation, dehydration reaction, and condensation
reaction (Scheme 1).
Scheme 1 Synthetic route of bisoctyl rhodamine B
2
3
.3.1. 2-(4-(dimethylamino)-2-hydroxybenzoyl)benzoic acid
To
obtain
2-(4-(dimethylamino)-2-hydroxybenzoyl)benzoic
acid
mixture,
-(dimethylamino)phenol (2.74 g, 20 mmol) and isobenzofuran-1,3-dione (3.00 g, 21 mmol)
were dissolved and heated with toluene (20 mL) under reflux for 4 h. After cooling to
0°C–60°C, the mixture was poured into an NaOH solution (20 mL, 35%). The mixture was
filtered, and the solids were washed with water (3 × 50 mL) before CH OH recrystallization
5
3
and dried under vacuum. 2-(4-(dimethylamino)-2-hydroxybenzoyl)benzoic acid was
obtained as a brown solid (4.75 g). Yield: 73.2%, melting point: 137°C–140°C.
2
.3.2. 3-(dioctylamino)phenol
3-(dioctylamino)phenol was obtained by dissolving 3-aminophenol (2.00 g, 18.2 mmol)
and bromooctane (8.78 g, 45.5 mmol) in ethanol (EtOH) (15 mL) and heated under reflux for

ACCEPTED MANUSCRIPT
1
2 h. After cooling and distilling under reduced pressure, the residue was purified via silica
gel chromatography, yielding a brown oil (3.01 g). Yield: 50%.
.3.3.
2
N-(9-(2-carboxyphenyl)-6-(dimethylamino)-3H-xanthen-3-ylidene)-N-octyloctan-1-aminiu
m (BORB)
2-(4-(dimethylamino)-2-hydroxybenzoyl)benzoic acid (0.285 g, 1.00 mmol) and
3
-(dioctylamino)phenol (0.41 g, 1.25 mmol) were dissolved in methyl sulfonic acid (5 mL)
and heated to 80°C–85°C for 14 h under nitrogen protection. Then, the mixture was naturally
cooled and poured into ice water and neutralized using sodium carbonate. The solid was
filtered and chromatographed over silica gel to yield the final product, i.e., BORB. Yield:
1
7
5.1%, melting point: 120°C–123°C, purity: 99.5% (HPLC). H NMR (400 MHz, CDCl
3
) δ
ppm: 0.89–0.92 (t, J = 7.2 Hz, 4H), 1.30–1.34 (m, 20H), 1.62–1.62 (m, 4H), 3.07 (s, 6H),
3
7
2
1
.30–3.34 (t, 4H), 6.41–6.57 (m, 4H), 6.71–6.77 (m, 2H), 7.21–7.23 (d, J = 7.2Hz, 1H),
1
3
.60–7.68 (m, 2H), and 8.10–8.12 (d, J = 7.6 Hz, 1H); C NMR (400 MHz, CDCl ) δ ppm:
3
2.7, 27.1, 27.2, 29.3, 29.5, 31.8, 40.4, 51.4, 97.3, 98.1, 107.2, 108.4, 109.3, 109.6, 125.1,
26.2, 129.4, 129.5, 133.8, 151.1, 153.0, 154.0, 154.2, and 169.2. MS (m/z): calcd: 583.4
+
+
[
M] , found: 583.3[M] (Figure S1–4).
2
.4. Activated sludge sample
Sludge samples containing abundant M. parvicella were collected from an activated
sludge WWTP in Tianjin (China) and stored at 4°C. The sludge concentration was 3–5 g
suspended solid/L (SS/L) and diluted using water to 1 g SS/L before use.
2
.5. Fluorescence quantum yields
Fluorescence quantum yields (Φ) were calculated (Eq. 1) using RhB in EtOH as a
standard with Φ value of 0.89[24, 25]
.
2
F A  n 
x
s
x
x s 


,
(1)




F A
ns
s
x

ACCEPTED MANUSCRIPT
where s and x represent the reference and sample, respectively; Φ represents the
fluorescence quantum yields; F represents the integral areas of fluorescence spectrum peaks;
A represents the absorbance; and n represents the refractive indexes.
2
.6. Spectra properties of BORB
Fluorescent compounds often show different spectra in different solvents. Herein,
BORB was dissolved in different solvents to yield 1 μmol/L solutions to measure the
UV–Vis absorption and fluorescent spectra. The photostability of organic dyes played an
essential role in their application as biological fluorescent labeling regents. An
iodine–tungsten lamp (500 W) was used as the light source, and 50.0 g/L of sodium nitrite
aqueous solution was used as the cold trap to decrease the effect of heat from illumination
and UV radiation shorter than 400 nm. The photostability experiment was performed in
EtOH using 1 μmol/L of BORB at room temperature in the dark[26]. Absorbance was
measured at different illuminating time points to evaluate the photostability of BORB. The
thermostability of BORB (1 μmol/L) in EtOH was measured at 80°C in an oil bath.
Absorbance was measured at different illuminating time points to evaluate the
thermostability of BORB.
2
.7. Labeling of M. parvicella BORB
The effects of BORB concentration and its duration on the labeling of M. parvicella
were investigated. To investigate the effect of BORB concentration on the labeling of M.
parvicella, different doses of BORB (500 μmol/L in EtOH) were added to 1 mL of sludge
samples (1 g SS/L). The samples were vigorously stirred and examined using an inverted
fluorescence microscope . To avoid biased images, 15–20 images were randomly selected
instead of selecting them according to the photo quality or labeling effect during the imaging
process[27]. Image analysis was performed using the Image Pro Plus 6 (IPP 6) software.
2
.8. Fabrication of the FRET system and FRET labeling of M. parvicella
2
.8.1. Fabrication of the FRET system

ACCEPTED MANUSCRIPT
QD solutions (118 μL) (42.6 μmol/L, calculated by empirical formula[28]) were added
into 10 individual test tubes containing 0, 5, 10, 15, 20, 25, 35, 50, 70, and 125 μL of BORB
solution in EtOH (5 μmol/L). The solution in each test tube was diluted with 100 μmol/L of
hexadecyl trimethyl ammonium bromide (CTAB) or phosphate-buffered saline (PBS) to
achieve the final volume of 5 mL with QD concentrations of 1.0 μmol/L. All samples were
mixed using a vortex mixer for 10 min to ensure sufficient mixing. The fluorescence
emission spectra were recorded at the excitation wavelength of 470 nm with both the
excitation and emission slits at 5 nm.
2
.8.2. FRET labeling of M. parvicella
The FRET systems of BORB (500 nmol/L) and QDs (1.0 μmol/L) were added to 1 mL
(100 μmol/L of CTAB) of sludge (1 g SS/L) to obtain samples with final BORB
concentrations of 5, 10, 20, 40, and 160 nmol/L. The samples were vigorously stirred for
several minutes and examined using an inverted fluorescence microscope.
2
.8.3. Transfer efficiency of the FRET system
Fluorescence resonance energy transfer efficiency of the FRET system is calculated
using Eq. 2[20]
:
E=1-F /F =1- /
,
D
(2)
DA
D
DA
where subscript D and DA represent the F or τ of donor alone and F or τ of donor in the
present of receptor, respectively; F is the fluorescence emission intensity; and τ is the
fluorescence lifetime.
3
. Results
3
.1. Properties of BORB
BORB has excellent properties, including long fluorescence emission wavelength, good
photostability, good thermostability (Figure 1), and high fluorescence quantum yields (Φ)
Table 1).
(

ACCEPTED MANUSCRIPT
2
2
1
1
50
00
50
00
b
EtOH
a
MeOH
EtOH
0
0
0
.2
.1
.0
MeOH
CH CN
3
H O
H O
2
2
5
0
0
DMSO
DMF
DMF
CH CN
3
DMSO
5
50
600
650
5
00
550
600
Wavelength(nm)
Wavelength (nm)
0
.3
.2
.1
.0
0.3
0.2
0.1
0.0
c
d
0
0
0
0
2
4
6
8
0
2
4
6
8
Time (h)
Time (h)
Figure 1. (a) UV–Vis and (b) fluorescent spectra of bisoctyl rhodamine B (BORB; 1 μmol/L)
in different solvents; (c) photostability curve and (d) thermostability curve of BORB (1
μmol/L) in CH OH.
3
The spectra of BORB in different solvents are in accordance with those of RhB or other RhB
derivatives reported[29]
absorption maximum wavelength ( abs) at ~555 nm, and
fluorescence emission ( em) band at ~580 nm in all solvents. This suggests that the
,
λ
λ
bisoctyl group on N does not change the conjugated ring system or the spectra wavelength.
In protic solvents, the solvent effect was observed in cases wherein the λabs of BORB
was red-shifted with increase in the dipole moment and solvent polarity. Thus, the excited
state was more stable than the ground state because of wavelength shifts due to hydrogen

ACCEPTED MANUSCRIPT
bonds
.
The absorbance increased due to the decreased energy of the bonding and
antibonding orbitals with increased solvent polarity.
λem red-shifted with increase in the solvent dipole moment. In protic solvents,
fluorescence intensities evidently decreased with increase in the solvent polarity and reached
the lowest in water due to the highest solvent polarity. When internal charge transfer
occurred due to molecule excitation, the excited state showed higher polarity than the
ground state. The increase in solvent polarity caused a stronger stable effect of the solvent
in the excited state than in the ground state, resulting in stronger stability of the excited
state and a red shift in
λem. Notably, CH
3
OH and EtOH have similar polarities due to which
λem and fluorescence intensity of BORB were different.
Table 1. Solvent properties and UV–Vis characteristics of BORB
Solvent
nD20 a
Methanol Water
EtOH
1.003
CH CN DMF DMSO
3
1.33
1.69
32.7
545
572
1.33
1.82
1.334
3.92
36.6
1.43
3.82
38.3
1.48
3.96
47.2
b

5.64
24.5
εc
8
0.1
λ
max (nm)_
em (nm)
556
572
17
544
572
28
558
576
21
556
580
25
555
582
27
λ
Stokes shift 27
d
Φ × 100
78.9
6.9
110
91
10.2
15
a
b
c
d
Refractive index. Dipole moment. Dielectric constant. Fluorescence quantum
yield.

ACCEPTED MANUSCRIPT
The decrease in solvent polarity enhanced Φ value because the excited state exhibited
stronger stability than the ground state. Φ was the lowest in water, presumably because the
bonds between the xanthene and benzene rings were rotated, thereby increasing the twist
angles of water molecules, destructing the coplanarity, increasing the nonradiative decay,
and causing obvious fluorescence quenching.
The photostability and thermostability of BORB were also studied herein. The
absorption intensity did not show any obvious changes at different exposure times, even after
continuous irradiation for 7 h. High photostability and thermostability are beneficial for the
wide application of BORB.
3
.2. Determination of M. parvicella
Before testing, dominant filamentous bacteria in activated sludge were first identified
using the FISH probe and Gram staining under OLYMPUS-IX71 fluorescent microscope
(Figure 2).
Gram staining was performed to classify bacterial species into two groups, namely gram
positive and gram negative. The chemical and physical properties of the cell walls of
bacterial species were determined by staining the peptidoglycan. Gram staining is not an
infallible tool for identification or phylogeny and is of extremely limited used in
environmental microbiology[30-32]. Upon Gram staining, filamentous bacteria can be
morphologically distinguished.
The total bacteria were labeled with the fluorescent EUBmix probe containing EUB338
(
(
(
Cy
3
-GCTGCCTCCCGTAGGAGT, targeting for most bacteria); EUB338-II
-GCAGCCACCCGTAGGTGT, targeting for Planctomycetales); and EUB338-III
-GCTGCCACCCGTAGGTGT, targeting for Verrucomicrobiales) in 1:1:1 ratio. M.
cy
3
cy
3
parvicella were labeled with the fluorescent MPAmix probe containing MPA60
(
(
(
cy
cy
cy
3
-GGATGGCCGCGTTCGACT, targeting for most M. parvicella); MPA223
-GCCGCGAGACCCTCCTAG, targeting for most M. parvicella); and MPA645
3
-CCGGACTCTAGTCAGAGC, targeting for most M. parvicella) in 1:1:1 ratio[33]
.
3

ACCEPTED MANUSCRIPT
a
a
b
b
c
Figure 2. (a) Images of activated sludge samples using Gram staining (10 × 40), (b) EUBmix
FISH staining (10 × 100), and (c) MPAmix FISH staining (10 × 100).
Filamentous bacteria are characterized by their filaments. These bacteria were identified
using strong Gram-positive (Figure 2a); previously reported M. parvicella showed the
following morphological characteristics[31, 34]: long, curly filaments without branches of
~0.7–0.9 μm diameter and 200–400 μm length. FISH with targeted oligonucleotide probes
[
4]
was used to identify and classify filamentous bacteria . The fragments emitting green
fluorescence (EUBmix, Figure 2b) represented the totle bacteria, whereas the fragments
emitting red fluorescence (MPAmix, Figure 2c) represented M. parvicella. Both Gram
staining and FISH analysis revealed extensive and dominant existence of M. parvicella in the
activated sludge samples.
3
.3. Labeling of M. parvicella by BORB
The concentration, duration, and corresponding mean fluorescence intensity of BORB
are discussed to understand the factors affecting labeling efficiency.
BORB concentration can affect the labeling efficiency. Different concentrations of
BORB were used to label M. parvicella (fixed sludge concentration of 1 g/L). Mean
fluorescence intensity was used to roughly evaluate the expression of M. parvicella.
Integrated optical density (IOD) represents the total fluorescence intensity of the entire
bacteria. The areas of fluorescence spectrum peaks and IOD were obtained using the IPP 6
imaging software by “spatial filtering,” “count and measure objects,” “irregular AOI,”
“arithmetic and logical operations,” and “count and measure objects.” The mean
fluorescence intensity can be calculated using Eq. 3.
Mean fluorescence intensity = IOD/Area
(3).

ACCEPTED MANUSCRIPT
The results were shown in Figure 3.
a
b
e
c
f
g
1
0 μm
10 μm
1
0 μm
1
60
y=19.27109+111.27971x
2
R =0.99
d
140
1
20
1
0 μm
10 μm 100
0
1
0 μm
1
2
Concentration (μmol/L)
Figure 3. Images (10 × 40) of M. parvicella labeled by BORB (a. 0.05 μmol/L, b. 0.2
μmol/L, c. 0.4 μmol/L, d. 1.0 μmol/L, e. 1.5 μmol/L, f. 2.0 μmol/L) and (g) The relation of
mean fluorescence intensities and concentration of BORB.
As shown in Figure 3, the increase in BORB concentration gradually enhanced the mean
fluorescence intensity with good linear relation.
When BORB concentration was <0.2 μmol/L, the mean fluorescence intensity was <120
and all the filaments and zooglea showed faint and weak fluorescence intensities with
indistinct boundaries, which suggests that it hardly labeled M. parvicella. When the
concentration was added to 0.4–1.0 μmol/L, the mean fluorescence intensity increased to
~120–130. Meanwhile, moderate fluorescence intensities could be observed, and the
filaments and zooglea were clearly distinguished. The fluorescence intensity of filaments
was stronger than that of the zooglea. Thus, BORB could obviously distinguish M. parvicella
from zooglea and could be used as a labeling probe to label M. parvicella. However, when
the probe concentration was increased to 1.5 μm/L, the fluorescence intensity increased up to
~140. Both the filaments and zooglea showed high fluorescence intensities and so could not
be distinguished. Therefore, 0.4–1.0 μmol/L was determined to be the appropriate
concentration range for labeling.

ACCEPTED MANUSCRIPT
Labeling stability is important for biological efficiency. Images of M. parvicella at different
times were labeled by BORB (1 μmol/L) in natural light without any other illumination. The
results of the mean fluorescence intensity (Figure 4) showed no remarkable changes for
different duration times. Even after 24 h, the filaments and zooglea were clearly observed
and easily distinguished. These results showed that results of labeling M. parvicella by
BORB were stable and that morphologies of M. parvicella were maintained in the duration
time.
a
a
b
e
1
0 μm
1
0 μm
c
c
d
Time (h)
1
0 μm
10 μm
Figure 4. Images (10 × 40) of different duration times (a: 0 h, b: 4 h, c: 12 h, d: 24 h; 400×)
labeled by BORB (1 μmol/L) and (e) the relation of mean fluorescence intensities and
duration times.
3
.4. Fabrication of the FRET system
Fluorescence quantum yields (Φ) of QDs were calculated to be 1.44 using Eq. 1 and
fluorescence quantum yield (Φ) of BORB was 1.11. The higher fluorescence quantum yield
of QDs was beneficial for the labeling of M. parvicella and made the FRET system efficient.
To discuss the ionic species causing the occurrence of the FRET phenomenon, surfactants of
PBS and CTAB were used to study the energy exchange (Figure 5).

ACCEPTED MANUSCRIPT
1
1
200
000
a
1
1
200
000
b
1
:0.01 mmol/L(BORB)
2:0.5 mmol/L(QDs)+0.01 mmol/L(BORB)
:1.0 mmol/L(QDs)+0.01 mmol/L(BORB)
4:1.5 mmol/L(QDs)+0.01 mmol/L(BORB)
:2.0 mmol/L(QDs)+0.01 mmol/L(BORB)
:2.0 mmol/L(QDs)
1
:0.04 mmol/L(BORB)
2:0.01 mmol/L(BORB)+1.0 mmol/L(QDs
:0.02 mmol/L(BORB)+1.0 mmol/L(QDs
4:0.03 mmol/L(BORB)+1.0 mmol/L(QDs
:0.04 mmol/L(BORB)+1.0 mmol/L(QDs
3
3
8
6
4
2
00
00
00
00
0
8
00
5
6
5
600
400
6:1.0 mmol/L(QDs)
5
5
4
3
4
3
2
2
2
00
0
1
6
1
6
5
00
550
600
650
5
00
550
600
650
Wavelength (nm)
Wavelength (nm)
Figure 5. Fluorescent spectra (E
x
= 470 nm, slit = 5 nm, voltage = 700 eV) of (a) BORB
(0.01 μmol/L) with addition of QDs in CTAB (100 μmol/L, pH = 7.4). (b) QDs (1 μmol/L)
with addition of BORB in CTAB (100 μmol/L, pH = 7.4).
When BORB was fixed at 0.01 μmol/L in PBS, the fluorescence intensity of QDs
increased with the addition of QDs, and the intensity at 572 nm increased until the
fluorescence spectra of BORB were enclosed by the QD spectra (Figures not shown). The
spectra of QDs and BORB at 572 nm overlapped so it was not conclusive an ambiguous
energy exchange. When QDs were fixed at 1.0 μmol/L in PBS, there was a weak increase in
the fluorescence intensity of BORB, whereas that of QDs decreased, which suggests a weak
energy exchange between them. When BORB was fixed at 0.01 μmol/L in CTAB, with the
addition of QDs, the fluorescence intensity of QDs increased, whereas that of BORB
obviously increased. It cannot be concluded that there was an FRET phenomenon. When
QDs were fixed at 1.0 μmol/L in CTAB, with the addition of BORB, the fluorescence
intensity of BORB increased, whereas that of QDs greatly decreased. It showed a strong
energy exchange between them; therefore, the FRET phenomenon occurred. Thus, we will
be using CTAB as the surfactant in the following FRET experiments.
The FRET phenomenon mainly relies on some crucial factors, such as the spectra of
donor and acceptor, distance and fluorescence efficiency. To further understand the FRET
process, its underlying mechanisms were studied (Figure 6).
There was an overlap at about 555 nm between the absorption spectrum of BORB and the
fluorescence emission spectrum of QDs, which meets the prerequisite of the FRET process
(
Figure 6a)[35]. Based on BORB and QDs, a FRET system was fabricated and the properties
were discussed. As shown in Figure 6b, QDs (1 μmol/L) showed strong fluorescence

ACCEPTED MANUSCRIPT
intensity (Figure 6b, line 1), while BORB (0.125 μmol/L) solution alone hardly showed any
fluorescence (Figure 6b, line 0). However, when 0–125 nmol/L of BORB was added to QDs
(1.0 μmol/L), the fluorescence intensities of QDs at 513 nm significantly decreased with a
little blue-shift while the intensities of BORB gradually increased at 572 nm with a little red
shift. The reason for λem shift may be the interaction between QDs and BORB, changing the
environment and causing quantum confinement effects on QDs and electrostatic effect on
BORB. This phenomenon indicates that BORB absorbed the energy of QDs via the FRET
process. Thus, the FRET phenomenon occurred and the FRET system was fabricated. The
working curve was established by plotting the ratio of the emission fluorescence intensities at
5
72 and 513 nm versus BORB concentration (0–0.125 μmol/L) (Figure 6c). It shows an
2
excellent linear relationship (R = 0.99) between the concentration and I572/I513
.
1
μmol/LQDs +0.12 5μ mol/LBORB
1μmol/LQDs +0.07 5μ mol/LBORB
μmol/LQDs +0.0 5μ mol/LBORB
1μmol/LQDs +0.037 5μ mol/LBORB
b
1
4
0
0
0
1
1
1
1
μmol/LQDs +0.02 5μ mol/LBORB
μmol/LQDs +0.02 0μ mol/LBORB
μmol/LQDs +0.01 5μ mol/LBORB
μmol/LQDs +0.01 0μ mol/LBORB
2
1μmol/LQDs +0.00 5μ mol/LBORB
1
0
μÌmol/LQDs
.125μmol/LBORB
500
600
700
Wavelength(nm)
3
2
1
0
d
c
80
40
0
(
-x/32.13)
y=-98.15e
+84.54
Y=0.0217X+0.1171
2
2
R =09972
R =0.9934
0
30
60
90
120
0
30
60
90
120
Concentration (nmol/L)
Concentration of BOR B( nmol/L)
Figure 6. (a) Absorption and emission spectra of QDs and BORB. (b) Fluorescent spectra of
the FRET system of QDs (1.0 μmol/L) with different concentration of BORB (λex = 470 nm,
slit = 5 nm, voltage = 400 eV). (c) The linear relation between I572/I513 and the concentration
of BORB. (d) FRET transfer efficiency.

ACCEPTED MANUSCRIPT
According to Eq. (2), the FRET efficiency data (Figure 6d) were calculated as 2.3%,
.4%, 24.5%, 34.8%, 37.5%, 54.6%, 62.1%, 75.3%, and 83.4%, respectively. Increase in the
7
concentration of BORB tends to increase the efficiency of fluorescence energy transfer with
a good relationship.
The distance between QDs and BORB can be calculated by the Förster’s nonradiative
energy transfer theory. The Förster radius R
donor and the acceptor in Eq. 4 (4).
0
represents the critical distance between the
1
/6
2
4
R  0.2108K  n  J()
(4).
0
D
R


2
K represents the orientation factor, and the value was dominated by the relative orientation
of QDs and BORB, n
R
represents the solvent refractive index, and Φ represents Φ of the
D
donor[36]. In these experiments, K = 2/3, n
2
= 1.33 (water), and Φ = 0.323. J(λ) was used for
R
D
calculating the overlap integral between the acceptor’s absorption spectrum and the donor’s
fluorescence emission spectrum, which is represented in Eq. (5)

4
J()  F () () d
(5).

D
A
0
In Eq. (5), λ represents the wavelength, F
donor, and ε
D
(λ) represents the fluorescence intensity of the
A
(λ) represents the extinction coefficient of the acceptor. J(λ) was calculated as
8
−1
−1
4
7
.4014 × 10 M cm nm . As we know, FRET can occur in the distance range of 1–10 nm
between the donor and acceptor, and the degree is strongly associated with the distance. Here
was calculated to be 5.92 nm. Thus, the FRET process occurred between QDs and the
BORB acceptor.
R
0
The possible mechanism of FRET processes is shown in Figure 6a. The positive charge
end of the CTAB molecule interacted with QDs to form QD–CTAB complexes[37]
.
Meanwhile, the alkyl groups in CTAB molecules can be combined with RhB molecules via
intermolecular forces. Such a combination shortens the distance between CdTe QDs and
BORB and facilitating the resonance energy transfer between QDs and BORB. Thus, based
on QDs and the novel organic BORB probe, a novel labeling FRET system for M.
parvicella was fabricated.

ACCEPTED MANUSCRIPT
3
.5. Labeling of M. parvicella by FRET
Different concentrations of the FRET probe were used to label M. parvicella. The results
are shown in Figure 7.
1
1
1
1
40
30
20
10
f
0
40
80
120
160
Concentration of BORB (nmol/L)
Figure 7. Images(10 × 40) of M. parvicella labeled by the FRET probe (QDs, 1.0 μmol/L)
with different concentration of BORB (a: 160 nmol/L, b: 40 nmol/L, c: 20 nmol/L, d: 10
nmol/L, e: 5 nmol/L, 400×) and (f) the relation of mean fluorescence intensities and
concentration of BORB in the FRET system.
In the FRET system, energy is transferred from QDs (1.0 μmol/L) to BORB, so the
optimum concentration of labeling M. parvicella of FRET was substantially reduced. The
labeling images show that when the concentration was <10 nmol/L, the mean fluorescence
intensity was <120, and all the filaments and zooglea showed faint and weak fluorescence
intensities with indistinct boundaries. When the concentration increased to 10–30 nmol/L,
the mean fluorescence intensity increased to ~120–130. Meanwhile, moderate fluorescence
intensities could be observed and clear boundaries among the filaments and zooglea were

ACCEPTED MANUSCRIPT
clearly distinguished. Moreover, the fluorescence of filaments was brighter than that of
zooglea. Thus, BORB could distinguish M. parvicella from zooglea and could be used as a
labeling probe for M. parvicella. However, when the concentration was >40 nmol/L, the
fluorescence intensity increased up to ~140. Both the filaments and zooglea showed strong
fluorescence intensity and so could not be distinguished. Therefore, 20 nmol/L of BORB is
the appropriate concentration for labeling M. parvicella, which is 500 times less than the
appropriate concentration for the carbazole–quinoline probe previously reported by our
group[13, 38]
.
The FISH method requires M. parvicella to be pretreated with enzymes by shock and
broken process, fixed sample, digestion, and lysozyme treatment, which is time consuming
and has several disadvantages, such as the complicated processing, low cell permeability,
and high expense. On the other hand, using our probes, M. parvicella can be quickly
detected, with a simple process, good stability, and no destructiveness.
4
. Conclusions
A novel fluorescent BORB modified by long-chain alkane was synthesized and
characterized for labeling M. parvicella, which led to the novel development of the FRET
probe, which was used to label M. parvicella in activated sludge. Compared with FISH, M.
parvicella could be clearly identified using BORB and the FRET probe in situ without shock
and broken process, fixed sample, digestion, or lysozyme treatment. The appropriate
identification concentration of labeling M. parvicella in situ was only 20 nmol/L, which was
a reduction of 50 times compared to that of the BORB (1 μmol/L) alone and 500 times
compared to the carbazole–quinoline probe reported previously by our group. It provided a
sensitive potential detection method for labeling M. parvicella in the early warning of M.
parvicella expansion in activated sludge bulking in situ.
Acknowledgments: The present study was financially supported by the National Natural
Science Foundation of China (grant no. 51578360, 51778398, 21502137, 11604238), the key
technologies R & D program of Tianjin (15ZCZDSF00150).

ACCEPTED MANUSCRIPT
References
[
1] A.J. Schuler, D. Jassby, Filament content threshold for activated sludge bulking: Artifact
or reality? Water Res. 41 (2007) 4349-4356.
[
2] E. Jiao, C. Gao, R. Li, T. Ye, Y. Peng, Energy saving control strategies for
Haliscomenobacter hydrossis filamentous sludge bulking in the A/O process treating real
low carbon/nitrogen domestic wastewater, Environ. Technol. 39 (2018) 2117-2127.
[
3] L.A.D. Santos, V. Ferreira, M.M. Neto, M.A. Pereira, M. Mota, A. Nicolau, Study of 16
Portuguese activated sludge systems based on filamentous bacteria populations and their
relationships with environmental parameters, Appl. Microbiol. Biot. 99 (2015) 1-10.
[
4] J. Guo, Y. Peng, S. Wang, X. Yang, Z. Yuan, Filamentous and non-filamentous bulking
of activated sludge encountered under nutrients limitation or deficiency conditions, Chem.
Eng. J. 255 (2014) 453-461.
[
5] L. Vanysacker, C. Denis, J. Roels, K. Verhaeghe, I.F.J. Vankelecom, Development and
evaluation of a TaqMan duplex real-time PCR quantification method for reliable
enumeration of Candidatus Microthrix, J. Microbiol. Methods 97 (2014) 6-14.
[
6] J. Wang, R. Qi, M. Liu, Q. Li, H. Bao, Y. Li, S. Wang, V. Tandoi, M. Yang, The potential
role of 'Candidatus Microthrix parvicella' in phosphorus removal during sludge bulking in
two full-scale enhanced biological phosphorus removal plants, Water Sci. Technol. 70
(2014) 367-375.
[
7] A. Pajdak-Stós, W. Kocerba-Soroka, J. Fyda, M. Sobczyk, E. Fiałkowska,
Foam-forming bacteria in activated sludge effectively reduced by rotifers in laboratory- and
real-scale wastewater treatment plant experiments, Environ. Sci. Pollut. Rese. 24 (2017)
1
-8.
[
8] D. Mamais, G. Nikitopoulos, E. Andronikou, E. Gavalakis, A. Andreadakis, K.
Noutsopoulos, C. Giotakis, G. Tsimarakis, Influence of the presence of long chain fatty
acids (LCFAs) in the sewage on the growth of M. Parvicella in activated sludge wastewater

ACCEPTED MANUSCRIPT
treatment plants, Global Nest J. 8 (2006) 82-88.
9] L. Speirs, T. Nittami, S. Mcilroy, S. Schroeder, R.J. Seviour, Filamentous bacterium
[
Eikelboom type 0092 in activated sludge plants in Australia is a member of the phylum
Chloroflexi, Appl. Environ. Microbiol. 75 (2009) 2446-2452.
[
10] D.S.L. Araújo, V. Ferreira, M.M. Neto, M.A. Pereira, M. Mota, A. Nicolau, Study of
6 Portuguese activated sludge systems based on filamentous bacteria populations and their
relationships with environmental parameters, Appl. Microbiol. Biot. 99 (2015) 5307-5316.
1
[
11] R. Erhart, D. Bradford, R.J. Seviour, R. Amann, L.L. Blackall, Development and use
of fluorescent in situ hybridization probes for the detection and identification of “
Microthrix parvicella ” in activated sludge, Syst. Appl. Microbiol. 20 (1997) 310-318.
[
12] P. Wang, Z. Yu, R. Qi, H. Zhang, Detailed comparison of bacterial communities during
seasonal sludge bulking in a municipal wastewater treatment plant, Water Res. 105 (2016)
57-566.
1
[
13] S. Li, X. Fei, X. Jiao, D. Lin, B. Zhang, L. Cao, Synthesis and characterization of the
fluorescent probes for the labeling of Microthrix parvicella, Appl. Microbiol. Biot. 100
(2016) 2883-2894.
[
14] T. Hug, W. Gujer, H. Siegrist, Rapid quantification of bacteria in activated sludge
using fluorescence in situ hybridization and epifluorescence microscopy, Water Res.39
(2005) 3837-3848.
[
15] D.P. Mesquita, A.L. Amaral, E.C. Ferreira, Activated sludge characterization through
microscopy: A review on quantitative image analysis and chemometric techniques, Anal.
Chim. Acta 802 (2013) 14-28.
[
16] X. Fei, W. Sun, L. Cao, X. Jiao, D. Lin, G. Jia, Design and preparation of quantum dots
fluorescent probes for in situ identification of Microthrix parvicella in bulking sludge, Appl.
Microbiol. Biot. 100 (2016) 961-968.

ACCEPTED MANUSCRIPT
[
17] S. Hamdghadareh, A. Salimi, F. Fathi, S. Bahrami, An amplified comparative
fluorescence resonance energy transfer immunosensing of CA125 tumor marker and
ovarian cancer cells using green and economic carbon dots for bio-applications in labeling,
imaging and sensing, Biosens. Bioelectron. 96 (2017) 308-316.
[
18] D. Li, W. Zhou, R. Yuan, X. Yun, A DNA-fueled and catalytic molecule machine lights
up trace under-expressed MicroRNAs in living cells, Anal. Chem. 89 (2017) 9934-9940.
[
19] J. Li, F. Mei, W.Y. Li, X.W. He, Y.K. Zhang, Study on the fluorescence resonance
energy transfer between CdTe QDs and butyl-rhodamine B in the presence of CTMAB and
its application on the detection of Hg(II), Spectrochim. Acta A 70 (2008) 811-817.
[
20] A.R. Clapp, I.L. Medintz, J.M. Mauro, B.R. Fisher, M.G. Bawendi, H. Mattoussi,
Fluorescence resonance energy transfer between quantum dot donors and dye-labeled
protein acceptors, J. Am. Chem. Soc. 126 (2004) 301-310.
[
21] H.H. Cai, H. Wang, J. Wang, W. Wei, P.H. Yang, J. Cai, Naked eye detection of
glutathione in living cells using rhodamine B-functionalized gold nanoparticles coupled
with FRET, Dyes Pigments 92 (2012) 778-782.
[
22] Y. Ge, A. Liu, R. Ji, S. Shen, X. Cao, Detection of Hg 2+ by a FRET ratiometric
fluorescent probe based on a novel pyrido[1,2-a]benzimidazole-rhodamine system, Sensor.
Actuat. B-Chem. 251 (2017) 410-415.
[
23] M. Suresh, S. Mishra, S.K. Mishra, E. Suresh, A.K. Mandal, A. Shrivastav, A. Das,
2
+
Resonance energytransfer approach and a new ratiometric probe for Hg in aqueous media
and living organism, Org. Lett. 11 (2009) 2740-2743.
[
24] D. Magde, R. Wong, P.G. Seybold, Fluorescence quantum yields and their relation to
lifetimes of rhodamine 6G and fluorescein in nine solvents: improved absolute standards
for quantum yields, Photochem. Photobiol. 75 (2002) 327-334.
[
25] R.T. Myers, Simplified forms of the Myers-London model for intermolecular forces in

ACCEPTED MANUSCRIPT
liquids, J. Phys. Chem. 84 (1980) 1871-1871.
[
26] B. Song, Q. Zhang, W.H. Ma, X.J. Peng, X.M. Fu, B.S. Wang, The synthesis and
photostability of novel squarylium indocyanine dyes, Dyes Pigments 82 (2009) 396-400.
[
27] H. Daims, S. Lücker, M. Wagner, daime, a novel image analysis program for microbial
ecology and biofilm research, Environ. Microbiol. 8 (2006) 200-213.
[
28] C. Dong, H. Qian, N. Fang, J. Ren, Study of fluorescence quenching and dialysis
process of CdTe quantum dots, using ensemble techniques and fluorescence correlation
spectroscopy, J. Phys. Chem. B 110 (2006) 11069-11075.
[
29] M. Takahashi, M. Ui, T. Inuzuka, T. Sengoku, H. Yoda, Synthesis and spectral
properties of perylene–rhodamine dyads with lipophilic dendritic auxiliaries, Tetrahedron
0 (2014) 9175-9182.
7
[
30] D. Jenkins, M.G. Richard, G.T. Daigger, Manual on the causes and control of activated
sludge bulking, foaming, and other solids separation problems, 3rd Edition, Crc Press
(2003).
[
31] D.H. Eikelboom, H.J.J.V. Buijsen, Microscopic sludge investigation
manual, IMG-TNO Report A 94A, 2nd edn. Delft, The Netherlands (1981).
[
32] S. Rossetti, C. Christensson, L.L. Blackall, V. Tandoi, Phenotypic and phylogenetic
description of an Italian isolate of "Microthrix parvicella", J. Appl. Microbiol. 82 (1997)
05-410.
4
[
33] J. Pernthaler, F.O. Glöckner, W. Schönhuber, R. Amann, Fluorescence in situ
hybridization (FISH) with rRNA-targeted oligonucleotide probes, Method in Microbiol. 30
(2001) 207-210.
[
34] P. H. Nielsen, H. Daims, H. Lemmer. FISH Handbook for Biological Wastewater
Treatment. IWA Publishing, London, UK.(2009)
35] R. Zhang, F. Yan, Y. Huang, D. Kong, Q. Ye, J. Xu, L. Chen, Rhodamine-based
[

ACCEPTED MANUSCRIPT
ratiometric fluorescent probes based on excitation energy transfer mechanisms:
construction and applications in ratiometric sensing, Rsc. Adv. 6 (2016) 50732-50760.
[
36] Y. Zou, F. Yan, L. Dai, Y. Luo, Y. Fu, N. Yang, J. Wun, L. Chen, High
photoluminescent carbon nanodots and quercetin-Al 3+ construct a ratiometric fluorescent
sensing system, Carbon 77 (2014) 1148-1156.
[
37] Z. Wang, J. Li, B. Liu, J. Hu, X. Yao, J. Li, Chemiluminescence of CdTe nanocrystals
induced by direct chemical oxidation and its size-dependent and surfactant-sensitized
effect, J. Phys. Chem. B 109 (2005) 23304-23311.
[
38] X. Jiao, X. Fei, S. Li, D. Lin, H. Ma, B. Zhang. Design mechanism and property of the
novel fluorescent probes for the identification of Microthrix Parvicella in situ. Materials 10
2017) 804-819.
(

ACCEPTED MANUSCRIPT
Graphical abstracts
FRET process between BORB and QDs, and images of labeling for M. parvicella.

ACCEPTED MANUSCRIPT
Highlights
1
2
3
4
.A novel bisoctyl Rhodamine B for labeling M. parvicella was synthesized.
.A novel FRET probe labeling for M. parvicella was prepared.
.The labeling procession occured in situ without broken or lysozyme pretreatment.
.It provides an potential method for warning expansion in activated sludge bulking in situ.