Water-soluble bioprobes with aggregation-induced emission characteristics for light-up sensing of heparin

Article abstract of DOI:10.1039/c4tb00367e

Two water-soluble cationic fluorene-based fluorescent probes for heparin detection are designed and synthesized. A slight change in the molecular design results in two probes with opposite optical properties in their solution and aggregation states as well as a response to heparin in buffer solution. The probe with a propeller-like conformation exhibits aggregation-induced emission (AIE) characteristics and shows a green fluorescence enhancement upon interaction with heparin; in contrast, the probe with a more planar conformation has a fluorescence quenching response. A comprehensive study on heparin detection using the two probes was conducted, which revealed that the AIE probe shows a better performance than the aggregation-caused quenching (ACQ) probe in terms of sensitivity. The AIE probe integrated with graphene oxide (GO) further improves the heparin detection sensitivity and selectivity. The solution of AIE probe/GO emits strong green fluorescence only in the presence of heparin, which allows for light-up visual discrimination of heparin from its analogues such as chondroitin-4-sulfate and hyaluronic acid. Moreover, the linear light-up response of AIE probe/GO enables heparin quantification in the range of 0-13.2 μM with a detection limit of 10 nM, which is of practical importance for heparin monitoring during surgery or therapy. This journal is the Partner Organisations 2014.

Full text of DOI:10.1039/c4tb00367e

Journal of
Materials Chemistry B
View Article Online
View Journal | View Issue
PAPER
Water-soluble bioprobes with aggregation-
induced emission characteristics for light-up
Cite this: J. Mater. Chem. B, 2014, 2,
sensing of heparin†
4134
Ryan T. K. Kwok,‡ab Junlong Geng,‡c Jacky W. Y. Lam,ab Engui Zhao,ab Guan Wang,c
c
c
abd
*
*
Ruoyu Zhan, Bin Liu and Ben Zhong Tang
Two water-soluble cationic fluorene-based fluorescent probes for heparin detection are designed and
synthesized. A slight change in the molecular design results in two probes with opposite optical
properties in their solution and aggregation states as well as a response to heparin in buffer solution. The
probe with a propeller-like conformation exhibits aggregation-induced emission (AIE) characteristics and
shows a green fluorescence enhancement upon interaction with heparin; in contrast, the probe with a
more planar conformation has a fluorescence quenching response. A comprehensive study on heparin
detection using the two probes was conducted, which revealed that the AIE probe shows a better
performance than the aggregation-caused quenching (ACQ) probe in terms of sensitivity. The AIE probe
integrated with graphene oxide (GO) further improves the heparin detection sensitivity and selectivity.
The solution of AIE probe/GO emits strong green fluorescence only in the presence of heparin, which
allows for light-up visual discrimination of heparin from its analogues such as chondroitin-4-sulfate and
hyaluronic acid. Moreover, the linear light-up response of AIE probe/GO enables heparin quantification
in the range of 0–13.2 mM with a detection limit of 10 nM, which is of practical importance for heparin
monitoring during surgery or therapy.
Received 5th March 2014
Accepted 22nd April 2014
DOI: 10.1039/c4tb00367e
www.rsc.org/MaterialsB
heparin rely on an indirect measurement, which monitors the
activated coagulation time or the activated partial thrombo-
plastin time,7–9 rather than direct measurement of the presence
of chemical species. These indirect assays are time-consuming,
unreliable and inaccurate because they lack specicity and have
potential interference from other factors.10
Introduction
Heparin, a highly sulfated, negatively charged polysaccharide,
plays an important role in the regulation of various physiolog-
ical processes such as cell growth and differentiation, and
inammatory processes.1 It has been widely used as an inject-
able anticoagulant drug to prevent thrombosis during surgery
or therapy.2,3 However, an overdose of heparin could induce
catastrophic complications such as hemorrhage or thrombo-
cytopenia.4–6 Heparin detection and quantication are thus of
critical importance. Current clinical laboratory assays for
Biosensors based on uorescent materials have attracted
much attention due to their superb sensitivity, selectivity and
rapidity. Many cationic materials have been developed for
heparin sensing through complexation aided by electrostatic
attractions.11–19 Among which, cationic uorophores, such as
phenylboronic acids,14 polycationic calyx[8]arenes,15 and chro-
mophore-tethered exible copolymers,16 have been used as
heparin receptors. However, the uorescence emissions of
these conventional uorophores are oen partially or
completely quenched upon complexation with heparin due to
formation of detrimental species such as excimers and exci-
plexes that generally relax nonradiatively.20 This aggregation-
caused quenching (ACQ) effect has been an intractable obstacle
to the development of uorescent sensors for sensitive quanti-
cation of heparin.
aHKUST Shenzhen Research Institute, No. 9 Yuexing 1st RD, South Area, Hi-tech Park,
Shenzhen, Nanshan, China 518057
bDepartment of Chemistry, Institute for Advanced Study, Division of Biomedical
Engineering, Division of Life Science, State Key Laboratory of Molecular
Neuroscience and Institute of Molecular Functional Materials, The Hong Kong
University of Science and Technology, Clear Water Bay, Kowloon, Hong Kong,
China. E-mail: tangbenz@ust.hk
cDepartment of Chemical and Biomolecular Engineering, National University of
Singapore, 4 Engineering Drive 4, 117585, Singapore. E-mail: cheliub@nus.edu.sg
dGuangdong Innovative Research Team, SCUT-HKUST Joint Research Laboratory,
State Key Laboratory of Luminescent Materials and Devices, South China University
of Technology, Guangzhou 510640, China
Recently, we have discovered that some propeller-like
molecules such as tetraphenylethene (TPE) and siloles display a
phenomenon that is exactly opposite to the ACQ effect and are
non-emissive when molecularly dissolved in solutions but are
induced to emit efficiently by aggregate formation.21–24 We have
† Electronic supplementary information (ESI) available: Intermediate synthesis;
NMR and high resolution mass spectra; size distributions of nanoaggregates;
uorescence photos and emission spectra. See DOI: 10.1039/c4tb00367e
‡ Ryan T. K. Kwok and Junlong Geng contributed equally to this work.
4134 | J. Mater. Chem. B, 2014, 2, 4134–4141
This journal is © The Royal Society of Chemistry 2014

View Article Online
Paper
Journal of Materials Chemistry B
coined this phenomenon as aggregation-induced emission bovine trachea and hyaluronic acid (HA) from Streptococcus equi
(AIE) and rationalized that the restriction of intramolecular were purchased from BioChemik and used as received.
rotations (RIR) in the aggregated state is the main cause for the Myoglobin, lysozyme, insulin, cytochrome c and human serum
AIE phenomenon.25–27 Such a phenomenon is not only of albumin (HSA) were purchased from Aldrich. Graphene oxide
academic value but also of practical implication as it permits (GO) was synthesized from graphite by a modied Hummers
the use of dye solutions with any concentration for bioassays method.51
and enables the development of “light-up” bioassays by taking
advantage of luminogenic aggregation.21 To date, a variety of
Instrumentation
AIE luminogens have been prepared and their biological
1H and 13C NMR spectra were measured on a Bruker ARX 400
applications in various elds such as cell imaging, and nucleic
spectrometer in CDCl3 or CD3OD using tetramethylsilane (TMS;
acids and protein assays have been explored.28–38 In a previous
d ¼ 0) as internal reference. UV spectra were measured on a
report, a cationic silole derivative has been utilized for uo-
Biochrom Libra S80PC double beam spectrometer. Photo-
rescence “turn-on” sensing of heparin. The probe, however,
luminescence spectra (PL) were recorded on a Perkin-Elmer LS
suffers from strong background noise and non-specic binding
55 spectrouorometer. High resolution mass spectra (HRMS)
to proteins to give false-positive signals.39 Thus, the develop-
were recorded on a GCT premier CAB048 mass spectrometer
ment of a “light-up” probe with high sensitivity, selectivity and
operating in MALDI-TOF mode. Particle size was determined
signal-to-noise ratio for heparin detection is highly appreciated.
using a ZetaPlus potential analyzer (Brookhaven instruments
Graphene oxide (GO) is a single carbon thin layer decorated
corporation, USA).
with epoxy and hydroxyl groups in its basal planes and carboxyl
groups at its periphery. The oxygen-rich functional groups
enable GO to interact with biomolecules through covalent,
Synthesis
noncovalent or electrostatic interactions. GO is also a well-
1,2-Bis[9,9-bis(6-bromohexyl)-2-uorenyl]-1,2-diphenylethene
known uorescence quencher owing to the strong p–p stacking (1). To a solution of 6 (2.98 g, 5 mmol), zinc dust (0.82 g, 12.5
interactions between GO and organic uorophores, which mmol) in dry distilled THF was added dropwise with titaniu-
ꢁ
induce energy transfer or nonradiative dipole–dipole coupling m(IV) chloride (1.3 mL, 12.5 mmol) under nitrogen at ꢀ78 C.
channels.40–43 Based on such a unique quenching property, The reaction mixture was warmed to room temperature and
many ultrasensitive GO-based biosensors have been success- then heated to reux for 12 h. Aer cooling to room tempera-
fully fabricated and applied to detect DNA and proteins.33,44–50 It, ture, the reaction was quenched by the addition of hydrochloric
therefore, would be nice if a water-soluble AIE probe could be acid solution. The mixture was then extracted with dichloro-
integrated with GO to develop a sensitive heparin assay.
methane several times. The organic layers were combined and
In this contribution, we report the design and synthesis of a washed with saturated brine solution and water, and dried over
pair of ACQ and AIE water-soluble cationic uorene-based anhydrous magnesium sulfate. Aer ltration and solvent
uorescent probes for heparin detection. Although the two evaporation, the product was puried by silica-gel column
probes have
distinctly different optical properties in solution and aggregated yield 1 as a green oil in 82% yield (2.38 g). H NMR (400 MHz,
states. The probe with propeller-shaped conformation CDCl3), d (TMS, ppm): 7.61–7.54 (m, 2H), 7.49–7.40 (m, 2H),
a similar molecular structure, they display chromatography using hexane–dichloromethane as eluent to
1
a
exhibits AIE characteristic and shows a green uorescence 7.29–7.22 (m, 6H), 7.13–7.00 (m, 14H), 3.36–3.22 (m, 8H), 1.90–
enhancement upon interaction with heparin; in contrast, the 1.49 (m, 16H), 1.26–1.08 (m, 8H), 1.06–0.85 (m, 8H), 0.57–0.43
probe with more planar structure shows
a
uorescence (m, 8H). 13C NMR (100 MHz, CDCl3), d (TMS, ppm): 150.5, 149.6,
quenching response. The AIE “light-up” probe enables visual 144.1, 143.3, 140.9, 140.8, 139.3, 131.5, 130.5, 127.6, 126.9,
detection of heparin and shows much higher sensitivity than 126.8, 126.4, 126.2, 122.6, 119.6, 118.9, 40.0, 34.0, 32.6, 29.0,
the ACQ “turn-off” probe. With the aid of GO the sensitivity and 27.8, 23.6. HRMS (MALDI-TOF): m/z 1160.6270 (M+, calcd
selectivity of the AIE probe towards heparin are further 1160.2327).
enhanced. This study highlights the excellent performance of
AIE luminophores in bioassays.
1,2-Bis{9,9-bis[6-(N,N,N-trimethylammonium)hexyl]-2-uorenyl}-
1,2-diphenylethene tetrabromide (2). To a solution of 1 (25 mg,
0.02 mmol) in THF (5 mL) trimethylamine (2 mL) was added
ꢁ
dropwise at ꢀ78 C. The mixture was stirred for 12 h and then
Experimental section
Materials
allowed to warm to room temperature. The precipitate was
redissolved by addition of methanol (5 mL). Aer the mixture
All chemicals and reagents were purchased from Aldrich and was cooled to ꢀ78 ꢁC, additional trimethylamine (2 mL) was
used as received without further purication unless otherwise added and the mixture was stirred at room temperature for 24 h.
noted. Tetrahydrofuran (THF) was distilled from sodium Aer solvent removal, water was added to redissolve all the
benzophenone ketyl under nitrogen immediately prior to use. precipitates and the aqueous solution was extracted with
Methanol and ethanol were dried and distilled from calcium dichloromethane. The aqueous layer was freeze-dried to yield 2
oxide. Heparin from bovine intestinal mucosa (Fluka) has 170 U as a greenish powder in 83% yield (25 mg). 1H NMR (400 MHz,
mgꢀ1. The molecular weight of heparin was determined by CD3OD), d (TMS, ppm): 7.69–7.64 (m, 2H), 7.54–7.50 (m, 2H),
disaccharide (644.2 g molꢀ1). Chondroitin 4-sulfate (ChS) from 7.39–7.34 (m, 2H), 7.33–7.26 (m, 4H), 7.16–6.95 (m, 14H), 3.32–
This journal is © The Royal Society of Chemistry 2014
J. Mater. Chem. B, 2014, 2, 4134–4141 | 4135

View Article Online
Journal of Materials Chemistry B
Paper
3.22 (m, 8H), 3.11 (s, 36H), 1.93–1.69 (m, 8H), 1.63–1.52 (m, 8H), water. UV, PL and particle size analyses of the resulting mixtures
1.16–1.03 (m, 16H), 0.59–0.45 (m, 8H). 13C NMR (100 MHz, were carried out immediately.
CD3OD), d (TMS, ppm): 152.1, 151.0, 145.5, 144.7, 142.5, 142.3,
140.9, 132.5, 131.6, 128.9, 128.1, 127.7, 124.2, 121.0, 120.4, 67.6,
Titration of heparin, ChS and HA
60.5, 55.9, 53.7, 41.1, 30.4, 26.9, 23.9. HRMS (MALDI-TOF): m/z
PBS solutions (150 mM, pH ¼ 7.4) of 2 and 4 with a concen-
tration of 12 mM were prepared. Into a cuvette, heparin solution
(2 mM) was added dropwise at an interval of 3.3 mL into the PBS
solutions of the dye (3 mL). Upon each addition, the mixture
was gently mixed using a pipette before the PL measurement.
The PL spectra were measured in the wavelength range of 380–
700 nm at an excitation wavelength of 365 nm. Titration of ChS
and HA was conducted using the same conditions.
+
ꢁ
1317.6105 [(M ꢀ Br) , calcd 1317.6083]. Melting point: 195 C.
1,2-Bis[9,9-bis(6-bromohexyl)-2-uorenyl]ethene (3). To a
solution of 8 (0.40 g, 0.75 mmol), zinc dust (0.12 g, 1.88 mmol)
in dry THF was added dropwise with titanium(IV) chloride (0.2
mL, 1.88 mmol) under nitrogen at ꢀ78 ꢁC. The reaction mixture
was warmed to room temperature and then heated to reux for
12 h. Aer cooling to room temperature, the reaction was
terminated by the addition of hydrochloric acid solution. The
mixture was then extracted with dichloromethane several times.
The organic layers were combined and washed with saturated
brine solution and water, and dried over anhydrous magnesium
sulfate. Aer ltration and solvent evaporation, the product was
puried by silica–gel column chromatography using hexane–
dichloromethane as eluent to yield 3 as a colourless oil in 79%
Optimization of GO concentration for heparin detection
Into a cuvette, heparin solution (16.5 mL, 2 mM) was added into
the PBS solution of 2 (3 mL, 12 mM) to furnish a mixture with a
heparin concentration of 11 mM. The GO solution (2.5 mg mLꢀ1
)
was subsequently added dropwise at an interval of 4.8 mL into
the cuvette. The PL spectra were measured in the wavelength
range of 380–700 nm at an excitation wavelength of 365 nm. The
same experiments were conducted for ChS and HA.
1
yield (0.30 g). H NMR (400 MHz, CDCl3), d (TMS, ppm): 7.70–
7.68 (d, 4H), 7.56–7.52 (m, 4H), 7.36–7.30 (m, 6H), 7.29–7.28 (d,
2H), 3.29–3.24 (t, 8H), 2.03–1.99 (m, 8H), 1.69–1.62 (m, 8H),
1.24–1.16 (m, 8H), 1.12–1.05 (m, 8H), 0.70–0.58 (m, 8H). 13C
NMR (100 MHz, CDCl3), d (TMS, ppm): 150.8, 150.5, 140.8,
136.5, 128.6, 127.1, 126.9, 125.7, 122.7, 120.5, 120.0, 119.7, 54.9,
40.3, 34.0, 32.6, 29.0, 27.7, 23.5. HRMS (MALDI-TOF): m/z
1008.1726 (M+, calcd 1008.1701).
Heparin quantication
A PBS solution of 2 (3 mL, 12 mM) was transferred into cuvettes.
Varying concentrations of heparin solutions (0–13.2 mM) were
prepared by adding the stock heparin solution (2 mM) at an
interval of 3.3 mL. The GO solution (57.6 mL, 2.5 mg mLꢀ1) was
subsequently added into the cuvette. The PL spectra were
measured from 380–700 nm at an excitation wavelength of 365
nm.
1,2-Bis{9,9-bis[6-(N,N,N-trimethylammonium)hexyl]-2-uorenyl}-
ethene tetrabromide (4). To a solution of 3 (50 mg, 0.05 mmol)
in THF (10 mL) trimethylamine (4 mL) was added dropwise at
ꢁ
ꢀ78 C. The mixture was stirred for 12 h and then allowed to
warm to room temperature. The precipitate was redissolved by
the addition of methanol (5 mL). Aer the mixture was cooled to
Results and discussion
Design and synthesis
ꢁ
ꢀ78 C, additional trimethylamine (2 mL) was added and the
mixture was stirred at room temperature for 24 h. Aer solvent
removal, water was added to redissolve all the precipitate and TPE is an archetypal AIE luminophore and emits intense light in
the aqueous solution was extracted with dichloromethane. The the aggregated state despite its solution is almost non-uores-
aqueous layer was freeze-dried to yield 4 as a pale yellow powder cent. Several uorescent probes have been developed from TPE-
in 85% yield (53 mg). 1H NMR (400 MHz, CD3OD), d (TMS, based water-soluble AIE luminogens for sensing of bio-
ppm): 7.76–7.71 (m, 6H), 7.63–7.61 (d, 2H), 7.43–7.39 (m, 4H), macromolecules such as nucleic acids and proteins.30–32,38 Most
7.36–7.30 (m, 4H), 3.22–3.18 (m, 8H), 3.03 (s, 36H), 2.17–2.10 of these probes emit blue light. However, for biological appli-
(m, 8H), 1.59–1.55 (m, 8H), 1.16–1.08 (m, 16H), 0.75–0.52 (m, cations, it is more desirable to have uorophores with longer-
8H). 13C NMR (100 MHz, CD3OD), d (TMS, ppm): 152.3, 151.9, wavelength emission as they suffer little interference from
142.5, 138.2, 129.7, 128.4, 128.3, 127.0, 124.1, 121.9, 121.1, optical self-absorption and auto-uorescence from the back-
120.8, 67.7, 56.3, 53.5, 41.2, 30.2, 26.8, 24.7, 23.6. HRMS ground. Increasing p-conjugation and introducing donor and
(MALDI-TOF): m/z 1197.8608 [(M ꢀ Br + Cl)+, calcd 1196.5187]. acceptor groups in the luminogenic structure are the common
ꢁ
Melting point: 202 C.
ways to red-shi the emission. In this work, the AIE probe 2 has
high p-conjugation and hence green emission has been
obtained by replacing the two phenyl rings of TPE by uorene
units. Meanwhile, a control probe 4 with stilbene-like structure
Preparation of aggregates
Stock THF solutions of 1 and 3 with a concentration of 1 mM is also developed by changing the phenyl rings of stilbene into
were prepared. An aliquot (0.1 mL) of this stock solution was uorene moieties. The synthetic routes to probes 2 and 4 are
transferred to a 10 mL volumetric ask. Aer adding an depicted in Scheme 1. Compound 5 was synthesized by Friedel–
appropriate amount of THF, water was added dropwise under Cras acylation of uorene and benzoyl chloride in the pres-
vigorous stirring to furnish 10 mM THF–water mixtures with ence of AlCl3 as catalyst. It was then coupled with 1,6-dibro-
water fractions (fw) of 0–99 vol%. Ethanol and hexane were used mohexane in basic solution to afford 6. Compound 1 was
to prepare the nanoaggregates of 2 and 4 instead of THF and subsequently synthesized by McMurry coupling of 6 catalyzed
4136 | J. Mater. Chem. B, 2014, 2, 4134–4141
This journal is © The Royal Society of Chemistry 2014

View Article Online
Paper
Journal of Materials Chemistry B
absorption band recorded at 367 nm in the spectrum of 3.
Although 1 has more p-conjugated units, it shows a blue-shi
spectrum, revealing that it adopts a more twisted conformation
caused by the steric repulsion between the aryl rings. To verify
this assumption, theoretical calculations were carried out. The
optimized molecular structures and HOMO and LUMO energy
levels of 1 and 3 are shown in Fig. 2. The HOMO and LUMO of 1
are contributed by both orbitals of the phenyl and uorene
rings. Such an X-shaped orbital distribution has somewhat
shortened the effective conjugation length. On the other hand,
the uorene rings in 3 lie on the same plane of the vinyl core,
thus contributing largely to the energy levels of the luminogen.
This orbital distribution results in a linear and extended
conjugation in 3. The calculated energy band gap for 1 is 4.05
eV, which is wider than that of 3 (3.28 eV). Thus, the theoretical
study nicely explains the blue-shi in the absorption of 1 from
that of 3.
The photoluminescence (PL) properties of 1 and 3 in the
solution and aggregated states are then investigated. In pure
THF solution, 1 is non-orescent. The emission remains weak
when up to 70% of water is added to the THF solution (Fig. 3).
Aerwards, the PL starts to increase swily. The higher the
water fraction, the stronger is the light emission. At 90% water
content, the PL intensity at 502 nm is 120-fold higher than that
in pure THF solution. The uorescence enhancement of 1 is
attributed to the formation of nano-aggregates as suggested by
the particle size analysis (Fig. S9†). Clearly, 1 is AIE-active. In
contrast, 3 displays an opposite luminescence behaviour. In
pure THF solution, 3 uoresces strongly at 420 nm. Addition of
water into the THF solution, however, has quenched the light
emission progressively. At 99% aqueous mixture, only a very
weak signal is recorded, suggesting that 3 is a typical ACQ
luminogen. The uorescence photos of THF–water mixtures of
1 and 3 with different water fractions taken under UV illumi-
nation also demonstrate the distinct opposite PL properties of 1
from that of 3 (Fig. S10†).
Scheme
1 Synthetic routes to water-soluble cationic fluorene-
substituted ethenes 2 and 4. R ¼ n-hexyl.
by TiCl4 and Zn. Treatment of 1 with trimethylamine nally
furnished the desirable product 2 in 83% yield. On the other
hand, luminogen 4 was synthesized from 2-bromouorene.
Reaction of 2-bromouorene with 1,6-dibromohexane in the
presence of tetrabutylammonium bromide in basic medium
gave 7. Lithiation of 7 with subsequent addition of dime-
thylforamide afforded formyl uorene 8. McMurry coupling of 8
followed by amination with trimethylamine generated 4.
Detailed experimental procedures can be found in the experi-
mental section or ESI.† Flurophores 1–4 were characterized by
NMR and MS spectroscopies and gave satisfactory data corre-
sponding to their molecular structures (Fig. S1–8†).
Optical properties
Aer investigating the PL of the precursors 1 and 3, we
further examined the optical properties of 2 and 4 to see
whether amination would affect the light emission process. Due
to their cationic character, 2 and 4 are soluble in water and
We rst studied the optical properties of precursors 1 and 3.
Compounds 1 and 3 are hydrophobic luminogens and dissolve
readily in tetrahydrofuran (THF). As shown in Fig. 1A, the UV
spectra of 1 in THF solution exhibit two absorption bands at 282
and 348 nm associated with the p–p* transition of the phenyl
and uorene rings, respectively. In contrast, there is only one
Fig. 1 UV spectra of (A) 1 and 3 in THF and (B) 2 and 4 in ethanol.
Concentration: 10 mM.
Fig. 2 Molecular amplitude plots HOMO and LUMO of 1 and 3.
J. Mater. Chem. B, 2014, 2, 4134–4141 | 4137
This journal is © The Royal Society of Chemistry 2014

View Article Online
Journal of Materials Chemistry B
Paper
detection and differentiation of its analogues such as chon-
droitin-4-sulfate (ChS) and hyaluronic acid (HA) (Scheme 2),
which oen exist as contaminants in heparin samples.
The principle of uorescence assay of heparin and its
analogues using 2 and 4 is illustrated in Scheme 3. Addition of
negatively-charged analytes induces the formation of the probe-
analyte complex through electrostatic interactions, which trig-
gers uorescence change. Due to the different affinity of the
analytes to the dye molecules, different extents of PL changes
are expected.
Titration of 2 and 4 with heparin and its analogues (0–11 mM)
was conducted in PBS buffer (150 mM, pH ¼ 7.4) and the cor-
responding PL spectra are shown in Fig. 4 and Fig. S15–16.†
Luminogen 2 shows a slight stronger PL in PBS buffer than in
water, presumably due to the slight aggregation of the uo-
rophore in this aqueous medium. The emission becomes
stronger gradually when an increasing amount of heparin is
added to the buffer solution. The complexation of 2 with
heparin has activated the RIR process of the uorophore, which
reduces the energy loss via nonradiative relaxation channels
and hence turns on the PL of 2. In the presence of 11 mM of
heparin, the PL intensity increases by ꢂ3-fold. Such a “light-up”
phenomenon was also observed when ChS, instead of heparin,
was used but the maximum emission enhancement was merely
1.4-fold. The lower charge density of ChS compared to heparin
maybe the major reason to account for the smaller magnitude
of uorescence enhancement. The PL of 2 changes little in the
presence of HA, which is indicative of no or very weak electro-
static interactions between 2 and HA. On the other hand, since 4
is an ACQ dye, it is highly emissive in PBS buffer. Addition of 11
mM of heparin and ChS, respectively, to the buffer solution, has
quenched the light emission by 50 and 80%. Similar to 2, HA
causes no observable change on the PL of 4. Such results show
that the AIE probe 2 has a higher sensitivity than the ACQ probe
Fig. 3 PL spectra of (A) 1 and (C) 3 in THF–water mixtures with
different water fractions (fw). Concentration: 10 mM; excitation wave-
length: 365 nm. (B and D) Plots of relative PL intensity (I/I0) of (B) 1 at
502 nm and (D) 3 at 420 nm versus the composition of THF–water
mixtures of 1 and 3. I0 ¼ PL intensity in pure THF solution (fw ¼ 0). Inset:
fluorescence photos of THF–water mixtures of (upper) 1 and (lower) 3
at fw ¼ 0 and 99 vol% taken under 365 nm UV illumination from a
hand-held UV lamp.
ethanol but insoluble in hexane. Thus, ethanol or PBS buffer
and ethanol–hexane mixture are chosen for the UV measure-
ment and PL analysis in the solution and aggregated state,
respectively. The UV spectra of 2 and 4 in ethanol (Fig. 1B) and
in PBS buffer (Fig. S11†) resemble those of the precursors 1 and
3 in THF. While 2 absorbs at 280 and 350 nm, the absorption
maximum of 4 occurs at a longer wavelength of 368 nm, which
is in correlation with the fact that its precursor 3 is more
conjugated than 1. Similar to 1, 3 is AIE-active: its ethanol
solution gives almost no light upon photoexcitation but its
nano-aggregates in the ethanol–hexane mixture with 80%
hexane content are strong green emitters (Fig. S12–S13†).
Luminogen 4 inherits the ACQ characteristics from 3. While it
emits intense blue light in ethanol but it becomes non-emissive
when its molecules aggregate in the ethanol–hexane mixture
with 99% hexane content. Visual observation of the uores-
cence photos of 2 and 4 in the ethanol–hexane mixtures with
different hexane fractions also demonstrates the AIE and ACQ
features of 2 and 4, respectively (Fig. S14†). These results indi-
cate that the amination reaction only enhances the water
solubility of the uorophores but exerts no effect on their
optical properties.
Scheme 2 Chemical structures of heparin (Hep), chondroitin 4-
sulfate (ChS) and hyaluronic acid (HA).
Heparin detection
The obvious difference in emission behaviours of 2 and 4
Scheme 3 Schematic illustration of detection of heparin and its
promotes us to compare their sensing performance in heparin analogues by 2 and 4 in PBS buffer.
4138 | J. Mater. Chem. B, 2014, 2, 4134–4141
This journal is © The Royal Society of Chemistry 2014

View Article Online
Paper
Journal of Materials Chemistry B
weakens the light emission. In contrast, heparin is highly
negatively-charged and thus shows stronger affinity to bind with
2 than GO. With such regard, the strong uorescence from the
2/Hep complex can be preserved. To prove such speculation, we
rst introduced GO into 2 in PBS buffer. The intrinsic uores-
cence of 2 in PBS buffer (12 mM) becomes weaker upon addition
of GO and is quenched completely at a concentration of 10 mg
mLꢀ1 GO. This implies the strong binding of 2 to GO and GO is
really an efficient PL quencher. The uorescence change of
buffer solutions of dye/biopolymer complexes upon addition of
different concentrations of GO are then studied (Fig. 5A). With
an increase in the GO concentration, the emission from all the
solutions decreased gradually. However, the decay rate for the 2/
Hep solution is much slower than those for 2/ChS and 2/HA
solutions, thanks to the tight complex formed by 2 and heparin.
The largest emission difference between 2/Hep/GO, 2/ChS/GO
and 2/HA/GO is found at [GO] ¼ 48 mg mLꢀ1, which is most
likely to be the optimum GO concentration for the heparin
quantication study. Indeed, as shown in Fig. 5B, the PL of 2/
Hep/GO is 13 and 146-fold higher than those of 2/ChS/GO and
2/HA/GO, respectively at 520 nm. To further address the selec-
tivity of the 2/GO system towards heparin, a control experiment
with other proteins including myoglobin, lysozyme, insulin,
cytochrome c, and human serum albumin (HSA) is performed
under the same experimental conditions. As shown in Fig. S17,†
only the uorescence of the 2/Hep complex is preserved in the
presence of GO. While in the case of other protein complexes
with 2, most of the uorescence are quenched once GO is
introduced. Obviously, introducing GO into 2 in the heparin
assay has greatly improved the assay sensitivity and selectivity.
Fig. 4 Titration of (A) 2 and (C) 4 with different concentrations of
heparin in PBS buffer (150 mM, pH ¼ 7.4). Dye concentration: 12 mM;
excitation wavelength: 365 nm. (B and D) Plots of relative PL intensity
(I/I0) of (B) 2 at 520 nm and (D) 4 at 424 nm versus Hep, ChS or HA
concentrations in PBS buffer.
4 in heparin detection. However, selectivity still needs to be
improved in order to discriminate heparin from ChS.
Dye/GO complex for heparin sensing
Heparin quantication
To further improve the sensitivity and selectivity of the AIE
probe 2 in heparin sensing, GO is introduced into the assay
system. The schematic illustration of the uorescence assay for
heparin using 2 integrated with GO is shown in Scheme 4.
Addition of GO into the buffer solution of the 2/analyte
complex may disassemble the complex due to the competition
of GO with the analyte for 2, the extent of which depends on the
charge density of the analyte. For analytes such as ChS and HA
with low charge density, GO wins the competition. This disas-
sembles the dye/biopolymer complex and subsequently
Heparin quantication is conducted in PBS buffer (150 mM, pH
¼ 7.4) with an optimum dye GO concentration of 12 mM and 48
mg mLꢀ1, respectively. While, no uorescence is detected when
2 is integrated with GO, the buffer solution starts to emit when
heparin is added (Fig. 6A). The plot of (I/I0 ꢀ 1) versus the
heparin concentration gives a linear curve in the range of 0–13.2
mM (Fig. 6B). Such a detection range is suitable for heparin
Fig. 5 (A) Change in the relative PL intensity (I/I0) of 2/Hep, 2/ChS or 2/
HA in PBS buffer with GO concentration, where I and I0 are the PL
intensity of 2/Hep, 2/ChS or 2/HA at 520 nm in the presence or
absence of GO, respectively. (B) PL spectra of 2/Hep, 2/ChS, 2/HA and
Scheme
4
Schematic illustration of heparin and its analogues 2 with 48 mg mLꢀ1 GO in PBS buffer. Concentration (mM): 12 (2) and 11
detection with 1/GO.
(Hep, ChS or HA); excitation wavelength: 365 nm.
This journal is © The Royal Society of Chemistry 2014
J. Mater. Chem. B, 2014, 2, 4134–4141 | 4139

View Article Online
Journal of Materials Chemistry B
Paper
4 T. E. Warkentin, M. N. Levine, J. Hirsh, P. Horsewood,
R. S. Roberts, M. Gent and J. G. Kelton, N. Engl. J. Med.,
1995, 332, 1330–1335.
5 B. Girolami and A. Girolami, Semin. Thromb. Hemostasis,
2006, 32, 803–809.
6 G. J. Despotis, G. Gravlee, K. Filos and J. Levy, Anesthesiology,
1999, 1122–1151.
7 P. D. Raymond, M. J. Ray, S. N. Callen and N. A. Marsh,
Perfusion, 2003, 269–277.
8 D. J. Murray, W. J. Brosnahan, B. Pennell, D. Kapalanski,
J. M. Weiler and J. Olson, J. Cardiothor. Vasc. An., 1997, 11,
24–28.
9 B. Boneul and P. de Moerloose, Semin. Thromb. Hemostasis,
2001, 27, 519–522.
10 M. N. Levine, J. Hirsh, M. Gent, A. G. Turpie,
M. Cruickshank, J. Weitz, D. Anderson and D. Johnson,
Arch. Intern. Med., 1994, 154, 49–56.
11 K.-Y. Pu and B. Liu, Adv. Funct. Mater., 2009, 19, 277–284.
12 R. Zhan, Z. Fang and B. Liu, Anal. Chem., 2010, 82, 1326–
1333.
Fig. 6 (A) Fluorescence photos of 2/GO in PBS buffer in the absence
and presence of HA, ChS or Hep taken under 365 nm UV illumination
using a hand-held UV lamp. Concentration: 12 mM (2), 11 mM (HA and
ChS), 2.2–11 mM (Hep), 48 mg mLꢀ1 (GO). (B) Calibration curve for Hep
quantitation. Concentration: 12 mM (2); 48 mg mLꢀ1 (GO); excitation
wavelength: 365 nm.
monitoring during surgery or therapy (2–11 mM).50 The lower
detection limit for the heparin assay is determined to be 10 nM,
based on 3 ꢃ S0/S, where S0 is the standard deviation of back-
ground and S is the sensitivity.
´
´˜
13 E. Climent, P. Calero, M. D. Marcos, R. Martınez-Manez,
´
F. Sancenon and J. Soto, Chem.–Eur. J., 2009, 15, 1816–
1820.
Conclusions
14 A. T. Wright, Z. Zhong and E. V Anslyn, Angew. Chem., Int.
Ed., 2005, 44, 5679–5682.
15 T. Mecca, G. M. L. Consoli, C. Geraci, R. La Spina and
F. Cunsolo, Org. Biomol. Chem., 2006, 4, 3763–3768.
16 W. Sun, H. Bandmann and T. Schrader, Chem.–Eur. J., 2007,
13, 7701–7707.
17 J. C. Sauceda, R. M. Duke and M. Nitz, ChemBioChem, 2007,
8, 391–394.
18 R. B. C. Jagt, R. F. Gomez-Biagi and M. Nitz, Angew. Chem.,
Two cationic uorene-based uorescent probes have been
synthesized for heparin detection. The AIE probe performs
better than the ACQ probe in terms of sensitivity and selectivity.
The AIE probe integrated with GO enables “light-up” heparin
quantication in a concentration range of 0–13.2 mM with a
detection limit of 10 nM, which is of practical use for heparin
monitoring during surgery or therapy. This study not only
provides an assay for rapid quantication of heparin, but also
demonstrates that the water-soluble AIE luminogens are
promising biosensors by taking advantage of the strong uo-
rescence of its complexes with biomacromolecules.
´
Int. Ed., 2009, 48, 1995–1997.
19 R. S. Dey and C. R. Raj, Chem.–Asian J., 2012, 7, 417–424.
20 J. B. Birks, Photophysics of Aromatic Molecules, Wiley,
London, New York, 1970.
21 Y. Hong, J. W. Y. Lam and B. Z. Tang, Chem. Commun., 2009,
4332–4353.
22 M. Wang, G. Zhang, D. Zhang, D. Zhu and B. Z. Tang, J.
Mater. Chem., 2010, 20, 1858.
23 Y. Hong, J. W. Y. Lam and B. Z. Tang, Chem. Soc. Rev., 2011,
40, 5361–5388.
24 J. Luo, Z. Xie, J. W. Y. Lam, L. Cheng, B. Z. Tang, H. Chen,
C. Qiu, H. S. Kwok, X. Zhan, Y. Liu and D. Zhu, Chem.
Commun., 2001, 381, 1740–1741.
Acknowledgements
This work was partially supported by the National Basic
Research Program of China (973 Program; 2013CB834701), the
Research Grants Council of Hong Kong (604711, 604913,
HKUST2/CRF/10 and N_HKUST620/11), Innovation and Tech-
nology Commission (ITCPD/17-9), the University Grants
Committee of Hong Kong (AoE/P-03/08) and the Singapore
Ministry of Defense (R279-000-340-232). B. Z. T. is grateful for
the support from the Guangdong Innovative Research Team
Program of China (201101C0105067115).
25 J. Chen, C. C. W. Law, J. W. Y. Lam, Y. Dong, S. M. F. Lo,
I. D. Williams, D. Zhu and B. Z. Tang, Chem. Mater., 2003,
79, 1535–1546.
¨
26 Z. Li, Y. Dong, B. Mi, Y. Tang, M. Haussler, H. Tong, Y. Dong,
J. W. Y. Lam, Y. Ren, H. H. Y. Sung, K. S. Wong, P. Gao,
I. D. Williams, H. S. Kwok and B. Z. Tang, J. Phys. Chem. B,
2005, 109, 10061–10066.
Notes and references
1 I. Capila and R. J. Linhardt, Angew. Chem., Int. Ed., 2002, 41,
391–412.
2 J. Fareed, D. A. Hoppensteadt and R. L. Bick, Semin. Thromb.
Hemostasis, 2000, 26, 5–21.
27 Z. Zhao, Z. Wang, P. Lu, C. Y. K. Chan, D. Liu, J. W. Y. Lam,
H. H. Y. Sung, I. D. Williams, Y. Ma and B. Z. Tang, Angew.
Chem., Int. Ed., 2009, 48, 7608–7611.
3 R. J. Linhardt and N. S. Gunay, Semin. Thromb. Hemostasis, 28 C. W. T. Leung, Y. Hong, S. Chen, E. Zhao, J. W. Y. Lam and
1999, 25, 5–16.
B. Z. Tang, J. Am. Chem. Soc., 2013, 135, 62–65.
4140 | J. Mater. Chem. B, 2014, 2, 4134–4141
This journal is © The Royal Society of Chemistry 2014

View Article Online
Paper
Journal of Materials Chemistry B
29 W. Qin, D. Ding, J. Liu, W. Z. Yuan, Y. Hu, B. Liu and 40 X. Qi, H. Li, J. W. Y. Lam, X. Yuan, J. Wei, B. Z. Tang and
B. Z. Tang, Adv. Funct. Mater., 2012, 22, 771–779. H. Zhang, Adv. Mater., 2012, 24, 4191–4195.
30 Y. Hong, H. Xiong, J. W. Y. Lam, M. Haussler, J. Liu, Y. Yu, 41 L. Cai, R. Zhan, K. Pu, X. Qi, H. Zhang, W. Huang and B. Liu,
¨
Y. Zhong, H. H. Y. Sung, I. D. Williams, K. S. Wong and
B. Z. Tang, Chem.–Eur. J., 2010, 16, 1232–1245.
31 Y. Hong, L. Meng, S. Chen, C. W. T. Leung, L.-T. Da,
Anal. Chem., 2011, 7849–7855.
42 X. Huang, X. Qi, F. Boey and H. Zhang, Chem. Soc. Rev., 2012,
41, 666–686.
M. Faisal, D.-A. Silva, J. Liu, J. W. Y. Lam, X. Huang and 43 X. Gu, G. Yang, G. Zhang, D. Zhang and D. Zhu, ACS Appl.
B. Z. Tang, J. Am. Chem. Soc., 2012, 134, 1680–1689. Mater. Interfaces, 2011, 3, 1175–1179.
32 H. Shi, R. T. K. Kwok, J. Liu, B. Xing, B. Z. Tang and B. Liu, J. 44 C.-H. Lu, J. Li, J.-J. Liu, H.-H. Yang, X. Chen and G.-N. Chen,
Am. Chem. Soc., 2012, 134, 17972–17981. Chem.–Eur. J., 2010, 16, 4889–4894.
33 X. Xu, J. Li, Q. Li, J. Huang, Y. Dong, Y. Hong, J. Yan, J. Qin, 45 F. Liu, J. Y. Choi and T. S. Seo, Biosens. Bioelectron., 2010, 25,
Z. Li and B. Z. Tang, Chem.–Eur. J., 2012, 18, 7278–7286. 2361–2365.
34 M. Nakamura, T. Sanji and M. Tanaka, Chem.–Eur. J., 2011, 46 H. Dong, W. Gao, F. Yan, H. Ji and H. Ju, Anal. Chem., 2010,
17, 5344–5349. 82, 5511–5517.
35 T. Noguchi, T. Shiraki, A. Dawn, Y. Tsuchiya, L. T. N. Lien, 47 S. He, B. Song, D. Li, C. Zhu, W. Qi, Y. Wen, L. Wang, S. Song,
T. Yamamoto and S. Shinkai, Chem. Commun., 2012, 48,
8090–8092.
36 X. Wang, A. R. Morales, T. Urakami, L. Zhang, M. V Bondar,
H. Fang and C. Fan, Adv. Funct. Mater., 2010, 20, 453–459.
48 Y. Wang, Z. Li, J. Wang, J. Li and Y. Lin, Trends Biotechnol.,
2011, 29, 205–212.
M. Komatsu and K. D. Beleld, Bioconjugate Chem., 2011, 22, 49 X. Xu, J. Huang, J. Li, J. Yan and Z. Li, Chem. Commun., 2011,
1438–1450.
12385–12387.
37 D. A. Links, J. Mater. Chem., 2011, 3170–3177.
50 D. L. Rabenstein, Nat. Prod. Rep., 2002, 19, 312–331.
38 M. Wang, D. Zhang, G. Zhang, Y. Tang, S. Wang and D. Zhu, 51 Y. Xu, H. Bai, G. Lu, C. Li and G. Shi, J. Am. Chem. Soc., 2008,
Anal. Chem., 2008, 80, 6443–6448.
39 M. Wang, D. Zhang, G. Zhang and D. Zhu, Chem. Commun.,
2008, 4469–4471.
130, 5856–5857.
This journal is © The Royal Society of Chemistry 2014
J. Mater. Chem. B, 2014, 2, 4134–4141 | 4141