Fibrous testing papers for fluorescence trace sensing and photodynamic destruction of antibiotic-resistant bacteria

Article abstract of DOI:10.1039/d0tb00002g

The increasing prevalence of antibiotic-resistant bacteria needs rapid identification and efficient destruction routes. This study proposes testing paper derived from electrospun fibrous mats and aggregation-induced emission (AIE) probes for trace sensing

Full text of DOI:10.1039/d0tb00002g

Journal of
Materials Chemistry B
View Article Online
View Journal
PAPER
Fibrous testing papers for fluorescence trace
sensing and photodynamic destruction of
antibiotic-resistant bacteria
Cite this:DOI: 10.1039/d0tb00002g
Long Zhao,ab Yuan Liu,a Zhanlin Zhang,a Jiaojun Wei,a Songzhi Xiea and
a
Xiaohong Li
*
The increasing prevalence of antibiotic-resistant bacteria needs rapid identification and efficient destruction
routes. This study proposes testing paper derived from electrospun fibrous mats and aggregation-induced
emission (AIE) probes for trace sensing and simultaneous destruction of antibiotic-resistant E. coli. Aptamers
are conjugated on fibers for selective capture of E. coli, and the capture capability can be regenerated via
rinsing with salt solution. Hydroxyl tetraphenylethene (TPE) is linked with two cephalosporin molecules to
construct TPE-Cep probes, and the fluorescence emission is turned on specifically in the presence of
b-lactamase, which is a critical marker for screening resistant bacteria. Fibrous mats are lit up only in the
presence of antibiotic-resistant bacteria, and the fluorescence intensity changes could be statistically fitted
into an equation for quantitative analysis. Fibrous strips display apparent color changes from blue to green
for a visual readout of bacterial levels, and the limit of detection (LOD) is much lower than those of previous
paper substrates. In addition, the TPE-Cep probes could produce reactive oxygen species (ROS) under room
light illumination to kill the captured bacteria. Thus, the integration of aptamer-grafted electrospun fibers
and functional AIE probes provides potential for selective capture, trace imaging and photodynamic
destruction of antibiotic-resistant bacteria.
Received 2nd January 2020,
Accepted 25th February 2020
DOI: 10.1039/d0tb00002g
rsc.li/materials-b
preventing cross-infection and providing guidance for treatment
protocols.2
Introduction
Infectious diseases have become a constant threat to human
A well-established gold standard for resistance detection is
health, and bacterial infections cause over 300 million cases of derived from bacterial growth in antibiotic-containing media.3
severe illness each year. Antibiotic treatment has made significant Conventional bacterial cultures usually take 2–3 days or even
contributions to infection control and remarkable reductions longer to enrich samples for isolation and counting of pure
in the morbidity and mortality of infected patients. However, colonies. These approaches are labor intensive, time consuming
the bacterial resistance to antibiotics has increased dramatically and less sensitive,4 and usually cannot recognize a low popula-
over recent years due to the overuse of antibiotics in humans, tion of antibiotic-resistant bacteria.5 Given such deficiencies,
veterinary medicine and the food industry.1 The widespread attempts have been made in recent years for visual assessment of
presence of antibiotic-resistant bacteria is regarded as one of antibiotic resistance, enabling a rapid and direct observation of
the most urgent challenges in human healthcare, animal drug-resistant strains.6 Among the resistance mechanisms, the
breeding and environmental control. Thus, efficient and rapid disruption of antibiotic activities by b-lactamases has become
identification of bacteria and drug-resistant strains plays the most significant and has an unavoidable consequence due to
important roles in monitoring epidemiology and directing the wide and extended uses of b-lactam antibiotics. For example,
effective risk management strategy. Especially in clinical practice, penicillin, cephalosporin, monobactam and carbapenem exhibit
a real-time and reliable analysis of pathogenic bacteria is vital for antibacterial activities via interactions with penicillin binding
proteins (PBPs) in the bacterial inner wall and then inhibiting
cell wall synthesis. The expression of b-lactamases could destroy
a Key Laboratory of Advanced Technologies of Materials, Ministry of Education,
b-lactam structural moieties and diminish the antibacterial
activities.7 Thus, b-lactamase has been utilized as a critical
School of Materials Science and Engineering, Southwest Jiaotong University,
Chengdu 610031, P. R. China. E-mail: xhli@swjtu.edu.cn; Fax: +8628-87634649;
marker for the detection and screening of resistant bacterial
populations. Li et al. developed non-fluorescent probes containing
a specific substrate of b-lactamase. The catalytic breakage of the
Tel: +8628-87634068
b School of Bioscience and Technology, Chengdu Medical College, Chengdu 610031,
P. R. China
J. Mater. Chem. B
This journal is © The Royal Society of Chemistry 2020

View Article Online
Paper
Journal of Materials Chemistry B
lactam ring caused the release of free fluorophores, and the aggregation state.18 The light-induced ROS generation from
fluorescence response was directly proportional to enzyme some AIE luminogens has been used to kill cancer cells and
concentration.8 However, the traditional imaging techniques bacteria, but challenges remain in imaging and distinguishing
face challenges in isolating specific strains and reporting a low antibiotic-resistant bacteria from susceptible bacteria.19
level of resistant bacteria.9
In the current study, electrospun fibrous strips were inte-
Another challenge is the development of new antibacterial grated with capture ligands and AIE probes for selective sensing
agents and alternative therapeutics to treat antibiotic-resistant and photodynamic destruction of antibiotic-resistant bacteria.
infections. In contrast to target limitation and resistance generation E. coli, as a biomarker of environmental contamination, is one
for common antibiotics, photodynamic therapy (PDT) has become type of commensal bacteria in the gastrointestinal tract of
a promising alternative to deal with bacteria and prevent the humans and many animals. Resistant E. coli makes up the
generation of resistant bacteria.10 Typically, bacteria are disrupted majority of bacterial pathogens commonly associated with
by reactive oxygen species (ROS) generated after light illumination antimicrobial resistance and survives almost any individual
of photosensitizers.11 Jia et al. introduced a protoporphyrin photo- antibiotic treatment.20 Electrospun fibrous mats have high
sensitizer and quaternary ammonium salt groups into cellulose, specific surface areas and facilitate high sensitivity and rapid
and ROS was produced under light irradiation for the effective response for sensing applications due to the easy accessibility
destruction of drug-resistant bacterial strains.12 Lin et al. developed of functional moieties and the increase in the reaction rates.21
silica nanoparticles with loaded chlorin e6, and singlet oxygen was In our previous study, TPE and mannose were conjugated on
continuously generated under light illumination to inhibit bacterial electrospun fibers, and the mannose grafts established specific
growth.13 Thus, the integration of diagnostic and therapeutic interactions with FimH proteins from the fimbriae of E. coli.
capabilities within activatable probes should provide great potential The AIE feature of TPE grafts led to an obvious fluorescence
in the simultaneous imaging and photodynamic ablation of ‘‘turn-on’’ of the fibrous mats.22 In the current study, aptamer
bacteria.14 But poor photostability and low bacterial affinity of ligands are designed for specific recognition of E. coli and
conventional fluorescent photosensitizers hinder their successful exhibit advantages over the commonly used antibodies due to
application in the real-time detection and simultaneous treatment their low cost, few batch-to-batch variations and high stability
of bacterial infections.15 Furthermore, most of the fluorescent in a wide range of temperatures and pH values.23 As shown in
probes and photosensitizers are extremely hydrophobic and have Scheme 1, electrospun blend fibers of poly(styrene-co-maleic
strong p–p stacking interactions, resulting in aggregation-induced anhydride) and polystyrene (PSMA/PS) with grafted E. coli-specific
quenching (ACQ). The quenched emission and reduced ROS aptamers (Apt-g-PSMA/PS) were constructed for bacterial capture,
generation decrease the labeling and antibacterial capabilities, followed by utilization of AIE probes for recognizing and destroying
but little progress has been made in circumventing the ACQ profiles antibiotic-resistant strains. The AIE probes consisted of TPE cores
of photosensitizers.16,17 Alternatively, probes with aggregation- and two terminal cephalosporin molecules (TPE-Cep) to promote
induced emission (AIE) characteristics, such as derivatives water solubility. After capture by E. coli, the TPE-Cep probes were
from propeller-like tetraphenylethene (TPE) molecules, have only cleaved by endogenous b-lactamase in penicillin-resistant
shown high fluorescence emission and ROS generation in an bacteria to yield TPE with two hydroxyl groups (TPE-OH), which
Scheme 1 (a) Schematic drawing of bacterial capture and destruction on fibrous mats. Aptamers are grafted on PSMA/PS fibers (1), followed by specific
capture of E. coli/pUC19 and E. coli (2). TPE-Cep probes are used to light up E. coli/pUC19 only (3), followed by destruction of E. coli/pUC19 under room
light illumination (4). (b) Schematic drawing of bacterial sensing and destroying mechanism. TPE-Cep probes are recognized and cleaved into fluorescent
TPE-OH by endogenous b-lactamase in antibiotic-resistant bacteria, followed by light irradiation to produce ROS for bacterial destruction.
J. Mater. Chem. B
This journal is © The Royal Society of Chemistry 2020

View Article Online
Journal of Materials Chemistry B
Paper
formed aggregates and emitted fluorescence immediately. Owing ATG AGT GTT GTG AAA TGT TGG GAC ACT AGG TGG CAT AGA
to the AIE activity and excellent water dispersion of TPE-Cep GCC GCA AAA GTG CAC GCT ACT TTG CTA A-30, which was
probes, the imaging process of resistant bacteria can be simplified obtained from Sangon Biotech Co., Ltd (Shanghai, China).
without washing. As a photosensitizer, the released TPE-OH could S. epidermidis (ATCC 12228) and P. aeruginosa (ATCC 35032)
produce ROS under light illumination to destroy bacteria. Thus, the were purchased from the American Type Culture Collection
TPE-Cep probes demonstrate multiple functions toward exactly (Rockville, MD). Escherichia coli JM109 (E. coli) and penicillin-
imaging and image-guided PDT of antibiotic-resistant bacteria.
resistant E. coli (E. coli/pUC19) were donated by Xiamen University
(Xiamen, China). All other chemicals were of analytical grade and
obtained from Changzheng Regents Company (Chengdu, China),
unless otherwise indicated.
Experimental section
Materials
Synthesis of TPE-Cep
PSMA (Mw: 170 kDa, maleic anhydride content: 14.8 wt%) and
PS (Mw: 400 kDa) were obtained from Shanghai Zhaocheng Fig. 1a shows the preparation process of TPE-Cep probes by
Scientific Development Corp (Shanghai, China). Penicillin, conjugation of TPE-OH and GCLE, followed by the removal of
4-methoxybenzyl 3-chloromethyl-7-(2-phenylacetamido)-3-cephem- p-methoxybenzyl protection groups. TPE-OH was synthesized
4-carboxylate (GCLE), bovine serum albumin (BSA), trypsin, dichloro- from 4-hydroxybenzophenone via the McMurry reaction as
fluorescein diacetate (DCFH-DA), b-lactamase, lysozyme, esterase, described previously.24 Briefly, 1.64 g of Zn powder (25.0 mmol)
hydroxybenzophenone and potassium clavulanate (PCA) were and 1.56 mL of TiCl4 (14.0 mmol) were mixed in 50 mL of dry
procured from Sigma (St Louis, MO). Sodium iodide, anisole, tetrahydrofuran (THF) and then refluxed for 2 h. After the addition
trifluoroacetic acid, and toluidine blue O (TBO) were purchased of dry THF (30 mL) containing 1.98 g of 4-hydroxybenzophenone
from Tianjin Heowns Biochemical Technology Co., Ltd (Tianjin, (10.0 mmol), the mixture was refluxed for 4 h and incubated with
China). The sequences for 30-amino-modified E. coli-specific 10% aqueous K2CO3 solution. The reaction mixture was vacuum-
aptamer were as follows: 50-GCA ATG GTA CGG TAC TTC CCC filtered to remove insoluble materials, and the aqueous layer was
Fig. 1 (a) Synthetic route for TPE-Cep. (b) 1H NMR spectra of TPE-GCLE and (c) TPE-Cep. (d) Fluorescence spectra of TPE-GCLE in DMSO/water
mixtures of different water fractions up to 99%. (e) Fluorescence intensity change of TPE-GCLE at 466 nm after incubation in DMSO/water mixtures with
different water fractions (n = 3).
J. Mater. Chem. B
This journal is © The Royal Society of Chemistry 2020

View Article Online
Paper
Journal of Materials Chemistry B
extracted with ethyl acetate. The organic fractions were water- and 20 mM) and TPE-Cep probes (2 mM) were mixed in aqueous
washed, Na2SO4-dried, and then vacuum-evaporated, followed by solutions for 20 min, followed by measuring the fluorescence
separation using silica gel to give TPE-OH. A yield of 56% was intensities as above.26 The specificity of the TPE-Cep probes was
1
obtained. The H NMR spectrum was recorded using a Bruker also investigated in the presence of other substances, such as
AM 400 apparatus, using tetramethylsilane (TMS) as the internal proteins (BSA, trypsin, lysozyme and esterase), inorganic salts
reference. 1H NMR (CDCl3, ppm): d 7.06 (d, 8H), 6.88 (d, 2H), and (CaCl2 and CuSO4), and organic molecules (glucose).
6.57 (d, 8H).
Preparation and modification of electrospun fibers
GCLE was conjugated on TPE-OH via esterification as
described previously.8 Briefly, GCLE (1.46 g, 3.0 mmol), sodium Blend fibers of PSMA and PS were prepared by electrospinning,
iodide (0.43 g, 3.0 mmol), and K2CO3 (0.40 g, 1.3 mmol) were and aptamers were grafted on the fibers via amidation reaction
mixed in 8 mL of acetonitrile for 15 min, followed by dropwise (Scheme 1a). Briefly, PSMA and PS were mixed with different
addition of TPE-OH (0.36 g, 1.0 mmol) in 4 mL of acetonitrile. ratios of THF/acetone/dimethylformamide (5 : 3 : 2, v/v/v). Electro-
After vigorous stirring for 1 h, the solvent was removed by spinning was conducted under a fixed electric field of 20 kV/15 cm,
vacuum evaporation, and the crude product was dissolved in a and the flow rate of polymer solution was 10 mL hꢀ1 27
The
.
mixture of water (10 mL) and ethyl acetate (15 mL). The organic collected PSMA/PS fibers were immersed in Tris–HCl buffer
fractions were washed with 10% aqueous Na2S2O3 solution and (10 mM, pH 7.4) containing 30-amino-modified E. coli-specific
dried with Na2SO4, followed by separation using silica gel to aptamer (1 mM), followed by incubation at room temperature for
1
give TPE-GCLE (0.37 g, 30% yield) as a yellow solid. H NMR 3 h to obtain Apt-g-PSMA/PS fibers. The reaction media were
(CDCl3, ppm): d 7.33–7.46 (m, 14H), 7.06–7.17 (m, 18H), 6.90–6.96 retrieved and precipitated with ethanol, and the aptamer concen-
(m, 4H), 5.80 (m, 1H), 5.11 (d, 2H), 5.05 (d, 1H), 4.86 (d, 1H), 4.77 tration was measured using an ultraviolet-visible (UV-Vis) spectro-
(d, 1H), 3.75 (s, 3H), 3.57 (d, 2H), 3.45 (d, 1H), and 3.27 (d, 1H) photometer (UV-2550, Shimadzu, Japan) to determine the grafting
(Fig. 1b).
content on the fibers.28 The resulting fibers were treated with
TPE-GCLE (148 mg, 0.12 mmol) was dissolved in 4 mL of 100 mM ethanolamine for 1 h to block unoccupied anhydride
dichloromethane at 0 1C under argon, followed by adding groups, followed by washing with Tris–HCl buffer and drying under
0.10 mL of anisole and 0.50 mL of trifluoroacetic acid. After a N2 stream.28
vigorous stirring for 1 h, the solvent was removed by vacuum
evaporation, and the residue was separated using silica gel to
Characterization of electrospun fibers
give TPE-Cep (52 mg, 42% yield) as a yellow solid. 1H NMR Scanning electron microscopy (SEM; JSM-7800F, JEOL, Japan)
(CDCl3, ppm): d 9.19 (d, 1H), 7.41–7.55 (m, 14H), 6.98–7.13 was used to observe the morphologies of the electrospun fibers,
(m, 14H), 5.80 (m, 1H), 5.09 (d, 1H), 4.98 (d, 1H), 4.73 (d, 1H), and the fiber size, pore size and the porosity of fibrous mats
3.62 (d, 2H), and 3.49 (d, 2H) (Fig. 1c).
were determined from the SEM images.29 The hydrophilicity of
fibrous mats was determined from the water contact angles
(WCAs) using a Kruss GmbH DSA100 Mk 2 goniometer (Hamburg,
Characterization of TPE-Cep
The AIE properties of TPE-GCLE were determined from the changes Germany). The chemical compositions of fibers were analysed
in fluorescence emissions in dimethyl sulfoxide (DMSO)/water using X-ray photoelectron spectroscopy (XPS, XSAM800, Kratos
mixtures.25 Briefly, TPE-GCLE was dissolved in DMSO at a concen- Ltd, UK), which was operated between 50 and 1300 eV with an
tration of 1 mM, and water was added dropwise with fractions from energy step of 0.5 eV using a pass energy of 300 eV. The density of
0 to 99% under vigorous stirring. The emission spectra of the the anhydride groups on the fiber surface was measured via the
resulting solutions or suspensions were measured using a fluores- TBO binding method after reaction with ethanolamine and con-
cence spectrophotometer (Hitachi F-7000, Japan) at the excitation/ version into carboxyl groups.30 Briefly, the fibers were immersed in
emission wavelengths of 365/466 nm.
100 mM ethanolamine at 25 1C for 1 h, and then incubated in TBO
The sensitivity of TPE-Cep probes was detected from the solution (0.50 mM), which was adjusted to pH 10 with 0.10 mM
fluorescence emission change in the presence of b-lactamase. NaOH. After constant agitation for 5 h at room temperature, the
Briefly, a TPE-Cep solution (3 mM) in phosphate buffered saline retrieved fibers were intensively washed with 0.1 mM NaOH
(PBS, pH 7.4) was mixed with b-lactamase solution (3.0 U mLꢀ1). solution to remove the noncomplexed TBO. The resulting fibers
After incubation at 37 1C for 20 min, the reaction mixtures were were mixed in acetic acid solution (50%) for 10 min to desorb TBO,
measured using a fluorescence spectrophotometer as above and which was detected using a UV-Vis spectrophotometer at 633 nm.
photographed under a 365 nm UV lamp. The inhibition assay of
enzyme activity was performed in the presence of PCA.26 Briefly,
Bacterial capture on fibrous mats
PCA solution (1, 5, and 10 mM) was premixed with b-lactamase The bacterial capture on fibrous mats was examined using
(5.0 U mLꢀ1) for 10 min at 37 1C, followed by addition of E. coli/pUC19 as a drug-resistant bacterial model, and using
TPE-Cep probes (2 mM). After vigorous shaking for 20 min, the E. coli, S. epidermidis and P. aeruginosa as controls. Bacteria
fluorescence intensity of the reaction mixtures was measured were cultured in Luria-Bertani (LB) media for 12 h and diluted with
immediately using a fluorescence spectrophotometer as above. PBS. Fibrous mats were cut into round discs with a diameter of
To investigate the specific interactions between b-lactamase 10 mm, followed by incubation with 1 mL of bacterial suspension
and b-lactam ring, b-lactamase (5.0 U mLꢀ1), penicillin (5, 10, in PBS (105 CFU mLꢀ1). Fibrous mats were intensively rinsed with
J. Mater. Chem. B
This journal is © The Royal Society of Chemistry 2020

View Article Online
Journal of Materials Chemistry B
Paper
PBS, and the bacterial concentration in the washing solution was increased by around 17 fold, indicating the maintenance of AIE
determined using a UV-Vis spectrophotometer at 600 nm (OD600) activities.
to calculate the amount of bacteria captured. The bacteria-loaded
Fluorescence response of TPE-Cep probes to b-lactamase
fibrous mats were observed by SEM as above after fixation in 4%
paraformaldehyde and lyophilization.
The b-lactam of cephalosporin was subjected to specific hydrolysis
by b-lactamase, and cephalosporin could be used as a substrate for
b-lactamase recognition.8 Fig. 2a shows the fluorescence intensities
of TPE-Cep probes upon the addition of b-lactamase in PBS. The
cephalosporin structure in the TPE-Cep probes could be recognized
and hydrolyzed by b-lactamase and subsequently release TPE-OH,
Bacterial detection and killing efficiency
The bacteria-captured fibers were incubated with TPE-Cep
probes at a concentration of 1.0 mg mLꢀ1 in the dark for
20 min. After PBS washing, the fibrous mats were detected
using a fluorescence spectrophotometer at the excitation wave-
lengths of 365 nm and photographed under a 365 nm UV lamp.
In addition, the bacteria-captured fibers were treated with
penicillin solution (1 mM) for 20 min before incubation with
TPE-Cep probes. To regenerate the aptamer sensors, the
bacteria-captured fibrous mats were rinsed using 2 M NaCl at
60 1C.31 The regenerated fibrous mats were used to capture and
detect bacteria, and the fluorescence spectra were recorded as
above. The detection and regeneration processes were repeated
for 5 cycles. The ROS levels in the bacterial cells were measured
after incubation with DCFH-DA as described previously with
some modifications.32 Briefly, TPE-Cep probe-stained fibers
with captured bacteria were resuspended in PBS containing
30 mg mLꢀ1 DCFH-DA at 37 1C for 30 min. The fibers were
collected and exposed to room light for 10 min, while the control
groups were kept in the dark. The fluorescence intensity was
measured at the excitation/emission wavelengths of 485/520 nm.
The morphology of bacteria was observed by SEM as above after
fixation in 4% paraformaldehyde and lyophilization.
Statistical analysis
The statistical significance of the data obtained was analyzed by
Student’s t-test. Data are expressed as mean ꢁ standard devia-
tion (S.D.). Probability values of p o 0.05 were interpreted as
denoting statistical significance.
Results and discussion
Characterization of TPE-Cep probes
TPE-GCLE was prepared by coupling GCLE on TPE-OH, which
was synthesized via McMurry reaction between two molecules
1
of 4-hydroxybenzophenone (Fig. 1a). As shown in the H NMR
spectra, the disappearance of peaks at 5.11 (d, 2H) and 3.75 ppm
(s, 3H) indicated the removal of p-methoxybenzyl groups from
TPE-GCLE (Fig. 1b), while the peak at 9.19 ppm (d, 1H) belonged
to the carboxyl groups of the resulting TPE-Cep (Fig. 1c). The AIE
properties of TPE-GCLE were measured from fluorescence changes
in DMSO after the addition of water.25 Fig. 1d shows the fluores-
cence spectra of TPE-GCLE, indicating significant increases in the
fluorescence intensities with increasing water fractions. DMSO is a
good solvent for TPE-GCLE, and the addition of water induced
aggregation and then fluorescence emission. As shown in Fig. 1e,
the fluorescence emission of TPE-GCLE became visible when the
water fractions were over 40%. When the water fractions increased
from 40% to 99%, the fluorescence intensities of TPE-GCLE
Fig. 2 (a) Fluorescence spectra of TPE-Cep in PBS and in the presence of
b-lactamase. Insets are photographs of the solutions under illumination
obtained by using a hand-held ultraviolet lamp. (b) Fluorescence intensities
of TPE-Cep after incubation with b-lactamase in the presence of PCA and
penicillin at different concentrations (n = 3). (c) Fluorescence intensities of
TPE-Cep probes in the presence of b-lactamase, BSA, trypsin, lysozyme,
esterase, glucose, CaCl2 and CuSO4 (n = 3).
J. Mater. Chem. B
This journal is © The Royal Society of Chemistry 2020

View Article Online
Paper
Journal of Materials Chemistry B
which formed aggregates immediately in PBS and emitted fluores-
cence at 468 nm. As shown in the inset of Fig. 2a, under UV
illumination, the fluorescence emission of TPE-Cep was observed
in the presence of b-lactamase.
To verify that the fluorescence turn-on resulted from the
hydrolysis of cephalosporin, PCA was added as an inhibitor of
b-lactamase. As shown in Fig. 2b, there were decreases of
around 63%, 76% and 84% in the fluorescence intensities after
the addition of 1, 5 and 10 mM PCA, respectively, due to the
irreversible binding with the active site of b-lactamase.33 In
addition, penicillin could interrupt the specific interaction
between b-lactamase and TPE-Cep, due to the competitive
binding with b-lactamase.26 As expected, the fluorescence emis-
sion of the probes was much weaker with the increase of penicillin
concentrations (Fig. 2b). The above data indicated that the fluores-
cence turn-on was mainly attributed to the specific recognition of
TPE-Cep and hydrolysis by b-lactamase into its highly fluorescent
counterpart TPE-OH. The specificity of TPE-Cep probes to
b-lactamase was evaluated from the fluorescence response in the
presence of various interferences. As shown in Fig. 2c, there was no
significant increase in the fluorescence intensities of TPE-Cep
probes after incubation with proteins BSA, enzymes (trypsin,
lysozyme and esterase), inorganic salts (CaCl2 and CuSO4), and
organic molecules (glucose). Thus, TPE-Cep could be an effective
and specific probe for the detection of b-lactamase from antibiotic-
resistant bacteria.
Characterization of Apt-g-PSMA/PS fibrous mats
Fig. 3a shows typical SEM images of PSMA/PS and Apt-g-PSMA/
PS fibers containing 50% PSMA. The aptamer grafting maintained
the integral fibrous morphology and randomly arrayed feature. The
average sizes of pristine and aptamer-grafted fibers were 1.38 and
1.84 mm, respectively, providing high specific surface areas for
bacterial binding. PSMA/PS fibers showed a hydrophobic surface
with a WCA of around 1261, and the aptamer grafts on the fiber
surface significantly increased the hydrophilicity (381). The increase
in the hydrophilicity of fibers led to water swelling of the fiber
Fig. 3 (a) Typical SEM images and (b) XPS spectra of electrospun PSMA/PS
and Apt-g-PSMA/PS fibers. (c) Densities of anhydride groups and aptamers
on the surface of fibers with different PSMA fractions (n = 3).
Bacterial capture on fibers
matrix and a significant increase in size. In addition, the aptamer Fig. 4a summarizes the amount of bacteria captured on fibrous
grafting led to a decrease in the average pore sizes from 13.5 to mats after incubation with E. coli/pUC19 suspensions. There
7.7 mm and porosities from 93% to 82%. Fig. 3b shows the was a rapid increase in the bacterial amount during the early
typical XPS profiles of the fibers. Additional N1s (400 eV) and incubation stage and no more increase after incubation for
P2p peaks (135 eV) appeared, due to amidation between term- 50 min, which was set as the incubation time during the
inal amino groups of aptamer and anhydride groups on the following tests. Fig. 4b shows the amount of E. coli/pUC19
fiber surface.
captured by fibers, indicating significantly higher capture
PSMA was mixed with PS in different fractions to adjust the efficiency by fibers with higher aptamer levels (p o 0.05). The
densities of anhydride groups and aptamer grafts on the fiber amount of captured bacteria indicated no significant increase
surface. Fig. 3c shows the densities of anhydride groups on the when the PSMA fractions in fibers increased to over 50%. It was
fiber surface, determined after transformation into carboxyl indicated that the reaction between bacteria and aptamer grafts
groups. The increase in PSMA fractions from 40% to 60% led to reached saturation when the graft density of aptamers increased to
an almost linear increase in anhydride group densities from 0.9 nmol mgꢀ1, which was used in the following investigation.
2.1 to 11.4 mmol mgꢀ1 fibers. The grafting amount of aptamers Fig. 4c shows a typical SEM image of the bacteria-captured
on fibers increased with the PSMA contents and the increase of fibers, displaying the abundant existence of bacteria on fibers.
aptamer densities was consistent with that of the anhydride In order to confirm the capture specificity of aptamers to E. coli,
groups on the fiber surface (Fig. 3c), suggesting the sufficient S. epidermidis and P. aeruginosa were used as controls.34 As shown
conjugation of aptamers on fibers.
in Fig. 4d, compared with the efficient capture of E. coli/pUC19
J. Mater. Chem. B
This journal is © The Royal Society of Chemistry 2020

View Article Online
Journal of Materials Chemistry B
Paper
Fig. 4 (a) Bacterial capture amount after incubation for up to 60 min with E. coli/pUC19 at 105 CFU mLꢀ1 (n = 3). (b) Bacterial capture amount by fibers
with different PSMA fractions after incubation with E. coli/pUC19 at 105 CFU mLꢀ1 for 50 min (n = 3). (c) Typical SEM morphology of bacteria-captured
fibrous mats. Arrows indicate the captured bacteria. (d) Bacterial capture amount after incubation with E. coli/pUC19, E. coli, S. epidermidis and
P. aeruginosa at 105 CFU mLꢀ1 for 50 min (n = 3).
and E. coli, there was no apparent capture of S. epidermidis and target peptidoglycan synthesis in bacteria via inactivating the
P. aeruginosa on the fiber surface.
transpeptidase activity of PBPs.36 Then, the cephalosporin moieties
in the TPE-Cep probes could recognize PBPs on the periplasmic
surface of the cytoplasmic membrane of bacteria,35 and the specific
binding between them led to weak fluorescence due to the AIE
effect. The addition of penicillin would lead to the occupation of
the PBP sites, leading to no significant fluorescence emission after
incubation with the TPE-Cep probes. Thus, in the case of bacterial
mixtures, the determination of E. coli/pUC19 could be performed
after treatment with both TPE-Cep probes and penicillin, and the
fluorescence difference with and without penicillin addition could
reflect the presence of E. coli.
Fig. 5c shows typical fluorescence spectra of fibers in response to
bacteria levels up to 106 CFU mLꢀ1. As summarized in Fig. 5d,
the fluorescence intensity of the bacteria-captured fibers progres-
sively increased as the bacterial concentrations increased up to
105 CFU mLꢀ1. The fluorescence intensity of fibrous mats was
drafted against E. coli/pUC19 levels and expressed by the equation:
Bacteria detection with TPE-Cep probes
The expression of b-lactamase is responsible for the resistance
of E. coli/pUC19 to b-lactam antibiotics.35 As shown in Fig. 2, the
TPE-Cep probes showed high sensitivity to b-lactamase and strong
fluorescence emission. We then investigated their applicability in
the fluorescence detection of E. coli/pUC19. Fig. 5a shows the
fluorescence intensities of captured bacteria on fibers after incuba-
tion with the TPE-Cep probes. As expected, fibers with captured
E. coli/pUC19 showed high fluorescence intensities, while almost no
fluorescence emission was observed for fibers after incubation with
S. epidermidis and P. aeruginosa. Although the antibiotic-susceptible
E. coli strain could not express b-lactamase, fibers with captured
E. coli showed a weak fluorescence intensity after treatment with
the TPE-Cep probes. In order to clarify the fluorescence emission,
fibers with captured E. coli strain were treated with free penicillin
before incubation with the TPE-Cep probes. As shown in Fig. 5b,
the treatment with penicillin led to no remarkable change in the
Y = 26X0.28 (R2 = 0.992)
emission intensities of fibers after capture of E. coli/pUC19 at 104 where Y represents the fluorescence intensity and X is the bacterial
and 105 CFU mLꢀ1 (p 4 0.05), which was attributed to the high- concentration (0–105 CFU mLꢀ1). Thus, the bacterial levels can be
level expression of b-lactamase in this strain. However, significantly determined from the fluorescence emissions according to the
lower fluorescence intensities were observed for E. coli-captured above equation. On the basis of the signal-to-noise ratio of 3,37
fibers after penicillin treatment (p o 0.05). As is known, b-lactams the limit of detection (LOD) of the current assay was 60 CFU mLꢀ1
,
J. Mater. Chem. B
This journal is © The Royal Society of Chemistry 2020

View Article Online
Paper
Journal of Materials Chemistry B
Fig. 5 (a) Fluorescence intensities of E. coli/pUC19, E. coli, S. epidermidis and P. aeruginosa captured on fibers after incubation with TPE-Cep probes (n = 3).
(b) Fluorescence intensities of captured E. coli/pUC19 and E. coli (104 and 105 CFU mLꢀ1) after incubation with TPE-Cep probes in the absence and presence of
penicillin (n = 3). (c) Fluorescence spectra and (d) intensities of fibrous mats after incubation with E. coli/pUC19 at different concentrations, followed by treatment
with TPE-Cep probes (n = 3). (e) Color strips of fibrous mats in the presence of E. coli/pUC19 at different concentrations taken at 365 nm under illumination of a
hand-held UV lamp. (f) Fluorescence intensity of bacteria captured on fibers after TPE-Cep treatment for different cycles (n = 3).
Fig. 6 (a) Fluorescence spectra of DCFH-DA in the presence of captured E. coli/pUC19, TPE-Cep probes and the mixtures in the dark and under light
illumination. (b) Fluorescence intensity of DCFH-DA in the presence of captured bacteria, TPE-Cep probes and the mixtures in the dark and under room
light illumination after incubation with E. coli/pUC19 of 104 and 105 CFU mLꢀ1 (n = 3). (c) SEM images of E. coli/pUC19 captured after incubation with
TPE-Cep probes in the dark and (d) under light irradiation.
J. Mater. Chem. B
This journal is © The Royal Society of Chemistry 2020

View Article Online
Journal of Materials Chemistry B
Paper
which was lower than that of previous assays on paper substrates. results in the release of TPE derivatives and fluorescence turn-on.
Naik et al. integrated thermal lysis and DNA amplification from E. coli-specific aptamers are grafted on fibers for selective capture of
bacteria into a single reaction step for E. coli assay, indicating a E. coli, and the interaction with bacteria reaches saturation when
LOD of 100 CFU mL .
ꢀ1 38 In addition, a color strip was constructed to the graft density of aptamers increases to 0.9 nmol mgꢀ1. In contrast
achieve a visual readout of bacterial levels. As shown in Fig. 5e, the to S. epidermidis, P. aeruginosa and the antibiotic-susceptible E. coli
fluorescence emissions of fibrous strips turned from blue to green strain, antibiotic-resistant E. coli/pUC19C captured on fibers showed
after incubation with E. coli/pUC19 from 102 to 105 CFU mLꢀ1. Thus, strong fluorescence emissions after incubation with the TPE-Cep
the fibrous strips indicated potential in the real-time detection of probes. The fluorescence intensity changes could be statistically
antibiotic-resistant bacteria, which is useful in healthcare or social fitted to an equation for the quantitative analysis of bacteria. Color
settings without complex experimental procedures.
strips from blue to green are established for a visual readout of
Aptamers usually bind target substances via non-covalent E. coli/pUC19C levels, and the LOD is much lower than those of
interactions, and the interactions could be disturbed by envir- previous paper substrates. In addition, the antibiotic-resistant
onmental pH, temperature and ionic strengths.39 In the current bacteria are efficiently destroyed by ROS produced under room
study, the bacteria-captured fibers were regenerated after incu- light illumination of AIE probes. Thus, this study demonstrates a
bation in salt solutions, which could unfold the aptamer strategy for selective capture, specific imaging, trace sensing,
structure without causing damage.31 Fig. 5f shows the fluores- and photodynamic destruction of antibiotic-resistant bacteria.
cence intensities of bacteria captured by the regenerated fibers
in the presence of TPE-Cep probes. The aptamer-grafted fibers
exhibited bacterial capture ability even after hybridization– Conflicts of interest
denaturation–hybridization for 5 cycles, showing high regen-
There are no conflicts of interest to declare.
eration ability of the fibrous biosensor.
ROS generation and bacterial killing by TPE-Cep probes
Acknowledgements
ROS generation involves the photoexcitation of photosensitizers
from the ground state to the singlet excited state and then inter-
system crossing to the triplet state.16 Chen et al. developed a
tetraphenylethene-based pyridinium salt bioprobe, and the AIE
feature selectively lit up the cell membrane and produced ROS
under room light illumination. ROS could oxidize the substances
in the plasma membrane and destroy the integrity.40 In the current
study, the ROS production in the captured bacteria was examined
after incubation with nonemissive DCFH-DA, whose fluorescence
was turned on at 520 nm after oxidation by ROS.32 As shown in
Fig. 6a, there was no fluorescence emission of DCFH-DA in the
presence of bacteria or TPE-Cep probes alone. In addition, the
incubation of DCFH-DA in the dark with the bacteria-captured fibers
and TPE-Cep probes did not affect the fluorescence emission at
520 nm. However, under room light illumination, strong fluores-
cence was produced after incubation with the bacteria-captured
fibers, indicating efficient ROS production by TPE-Cep probes
in the captured bacteria. Similar light-induced ROS-generation
properties were confirmed after incubation with different bacterial
concentrations (Fig. 6b). The morphologies of bacteria captured on
fibers were detected after treatment with the TPE-Cep probes and
room light illumination. As shown in Fig. 6c, E. coli/pUC19 captured
on fibers exhibited regular smooth bodies and a well-defined border
for overlapped bacteria, indicating a healthy state. After light illumi-
nation, shrinkage and fusion of bacteria were observed (Fig. 6d),
demonstrating the significant destruction of antibiotic-resistant
bacteria.41
This work was supported by the National Natural Science
Foundation of China (31771034), the Key Research and Devel-
opment Program of Sichuan Province (2018SZ0348), and the
Analytical and Testing Center of Southwest Jiaotong University
for SEM analysis.
Notes and references
1 R. A. Blaustein, D. R. Shelton, J. A. S. Van Kessel, J. S. Karns,
M. D. Stocker and Y. A. Pachepsky, Environ. Monit. Assess.,
2016, 188, 56.
2 L. Varadi, J. L. Luo, D. E. Hibbs, J. D. Perry, R. J. Anderson,
S. Orenga and P. W. Groundwater, Chem. Soc. Rev., 2017, 46,
4818–4832.
´
3 N. Drabinska, B. de Lacy Costello, K. Hewett, A. Smart and
N. Ratcliffe, TrAC, Trends Anal. Chem., 2019, 115, 1–12.
4 C. R. Nemr, S. J. Smith, W. Liu, A. H. Mepham, R. M.
Mohamadi, M. Labib and S. O. Kelley, Anal. Chem., 2019,
91, 2847–2853.
5 T. Huang, Y. Zheng, Y. Yan, L. Yang, Y. Yao, J. Zheng, L. Wu,
X. Wang, Y. Chen and J. Xing, Biosens. Bioelectron., 2016, 80,
323–330.
6 B. R. Choi, J. Cho, J. Kim, J. S. Kim, K. Kim and H. Lee, Adv.
Mater. Technol., 2019, 4, 1800375.
7 J. M. Blair, M. A. Webber, A. J. Baylay, D. O. Ogbolu and
L. J. Piddock, Nat. Rev. Microbiol., 2015, 13, 42–51.
8 L. Li, Z. Li, W. Shi, X. Li and H. Ma, Anal. Chem., 2014, 86,
6115–6120.
Conclusions
9 Q. Shao, Y. Zheng, X. Dong, K. Tang, X. Yan and B. Xing,
Chem. – Eur. J., 2013, 19, 10903–10910.
TPE-Cep probes are constructed by linking TPE-OH with cephalo-
sporin molecules, and the hydrolysis of TPE-Cep by b-lactamase 10 L. Zhai and K.-W. Yang, Dyes Pigm., 2015, 120, 228–238.
J. Mater. Chem. B
This journal is © The Royal Society of Chemistry 2020

View Article Online
Paper
Journal of Materials Chemistry B
´
11 A. Almeida, M. A. Faustino and J. P. Tome, Future Med. 26 J. Lv, C. Chen, C. Qian, W. Chen, Y. Xiang and J. Zhang,
Chem., 2015, 7, 1221–1224. Sens. Actuators, B, 2016, 230, 653–657.
12 R. Jia, W. Tian, H. Bai, J. Zhang, S. Wang and J. Zhang, Adv. 27 L. Zhao, T. Wang, Q. Wu, Y. Liu, Z. Chen and X. Li, ACS Appl.
Healthcare Mater., 2019, 8, 1801591. Mater. Interfaces, 2017, 9, 3400–3410.
13 J. F. Lin, J. Li, A. Gopal, T. Munshi, Y. W. Chu, J. X. Wang, 28 S. J. Lee, R. Tatavarty and M. B. Gu, Biosens. Bioelectron.,
T. T. Liu, B. Shi, X. Chen and L. Yan, Chem. Commun., 2019,
55, 2656–2659.
14 H. Jin, T. Zhu, X. Huang, M. Sun, H. Li, X. Zhu, M. Liu,
2012, 38, 302–307.
29 B. Zou, Y. Liu, X. Luo, F. Chen, X. Guo and X. Li, Acta
Biomater., 2012, 8, 1576–1585.
Y. Xie, W. Huang and D. Yan, Biomaterials, 2019, 211, 30 L. Ying, C. Yin, R. Zhuo, K. Leong, H. Mao, E. Kang and
68–80. K. Neoh, Biomacromolecules, 2003, 4, 157–165.
15 Y. Zhang, X. Zhao, Y. Li, X. Wang, Q. Wang, H. Lu and 31 L. Guo and D.-H. Kim, Biosens. Bioelectron., 2012, 31, 567–570.
L. Zhu, Dyes Pigm., 2019, 165, 53–57.
32 D. Y. Lyon, L. Brunet, G. W. Hinkal, M. R. Wiesner and
16 E. Zhao, Y. Chen, H. Wang, S. Chen, J. W. Lam, C. W. Leung,
P. J. Alvarez, Nano Lett., 2008, 8, 1539–1543.
Y. Hong and B. Z. Tang, ACS Appl. Mater. Interfaces, 2015, 7, 33 H. C. Neu and K. P. Fu, Antimicrob. Agents Chemother., 1978,
7180–7188. 14, 650–655.
17 M. C. DeRosa and R. J. Crutchley, Coord. Chem. Rev., 2002, 34 Y. S. Kim, M. Y. Song, J. Jurng and B. C. Kim, Anal. Biochem.,
233, 351–371. 2013, 436, 22–28.
18 Y. Yuan, G. Feng, W. Qin, B. Z. Tang and B. Liu, Chem. 35 A. B. Shapiro, Methods Appl. Fluoresc., 2016, 4, 024002.
Commun., 2014, 50, 8757–8760.
19 F. Hu, S. Xu and B. Liu, Adv. Mater., 2018, 30, 1801350.
36 A. M. King, D. T. King, S. French, E. Brouillette, A. Asli,
J. A. N. Alexander, M. Vuckovic, S. N. Maiti, T. R. Parr Jr and
E. D. Brown, ACS Chem. Biol., 2016, 11, 864–868.
37 C. Shen, X. Li, A. Rasooly, L. Guo, K. Zhang and M. Yang,
Biosens. Bioelectron., 2016, 85, 220–225.
38 P. Naik, S. Jaitpal, P. Shetty and D. Paul, Sens. Actuators, B,
2019, 291, 74–80.
39 J.-H. Park, J.-Y. Byun, W.-B. Shim, S. U. Kim and M.-G. Kim,
Biosens. Bioelectron., 2015, 73, 26–31.
¨
20 L. George, A. Mu¨ller, B. Roder, V. Santala and A. Efimov,
Dyes Pigm., 2017, 147, 334–342.
21 S.-X. Wang, C. C. Yap, J. He, C. Chen, S. Y. Wong and X. Li,
Nanotechnol. Rev., 2016, 5, 51–73.
22 L. Zhao, Y. Chen, J. Yuan, M. Chen, H. Zhang and X. Li, ACS
Appl. Mater. Interfaces, 2015, 7, 5177–5186.
23 J. Chung, J. S. Kang, J. S. Jurng, J. H. Jung and B. C. Kim,
Biosens. Bioelectron., 2015, 67, 303–308.
40 W. Zhang, Y. Huang, Y. Chen, E. Zhao, Y. Hong, S. Chen,
J. W. Lam, Y. Chen, J. Hou and B. Z. Tang, ACS Appl. Mater.
Interfaces, 2019, 11, 10567–10577.
24 T. Kato, A. Kawaguchi, K. Nagata and K. Hatanaka, Biochem.
Biophys. Res. Commun., 2010, 394, 200–204.
25 R. Hu, R. Ye, J. W. Lam, M. Li, C. W. Leung and B. Z. Tang, 41 X. You, H. Ma, Y. Wang, G. Zhang, Q. Peng, L. Liu, S. Wang
Chem. – Asian J., 2013, 8, 2436–2445.
and D. Zhang, Chem. – Asian J, 2017, 12, 1013–1019.
J. Mater. Chem. B
This journal is © The Royal Society of Chemistry 2020