Long-term fluorescent cellular tracing by the aggregates of AIE bioconjugates

Article abstract of DOI:10.1021/ja312581r

There is a great demand for long-term cellular tracers because of their great importance in monitoring biological processes, pathological pathways, therapeutic effects, etc. Herein we report a new type of fluorescence "turn-on" probe for tracing live cell

Full text of DOI:10.1021/ja312581r

Article
pubs.acs.org/JACS
Long-Term Fluorescent Cellular Tracing by the Aggregates of AIE
Bioconjugates
Zhengke Wang,†,‡,⊥,§ Sijie Chen,†,⊥,§ Jacky W. Y. Lam,†,⊥ Wei Qin,† Ryan T. K. Kwok,† Ni Xie,†
Qiaoling Hu,‡ and Ben Zhong Tang*,†,∥,⊥
†Department of Chemistry, Division of Biomedical Engineering, Institute for Advanced Study, and Institute of Molecular Functional
Materials, The Hong Kong University of Science and Technology (HKUST), Clear Water Bay, Kowloon, Hong Kong, China
‡Institute of Biomedical Macromolecules, MoE Key Laboratory of Macromolecular Synthesis and Functionalization, Zhejiang
University, Hangzhou 310027, China
∥Guangdong Innovative Research Team, SCUT-HKUST Joint Research Laboratory, State Key Laboratory of Luminescent Materials
and Devices, South China University of Technology (SCUT), Guangzhou 510640, China
⊥HKUST Shenzhen Research Institute, Nanshan, Shenzhen 518057, China
S
* Supporting Information
ABSTRACT: There is a great demand for long-term cellular tracers because of their great importance in monitoring biological
processes, pathological pathways, therapeutic effects, etc. Herein we report a new type of fluorescence “turn-on” probe for tracing
live cells over a long period of time. We synthesized the fluorogenic probe by attaching a large number of tetraphenylethene
(TPE) labels to a chitosan (CS) chain. The resultant TPE−CS bioconjugate shows a unique aggregation-induced emission (AIE)
behavior. It is nonfluorescent when dissolved but becomes highly emissive when its molecules are aggregated. The AIE
aggregates can be readily internalized by HeLa cells. The cellular staining by the TPE−CS aggregates is so indelible that it enables
cell tracing for as long as 15 passages. The internalized AIE aggregates are kept inside the cellular compartments and do not
contaminate other cell lines in the coculture systems, permitting the differentiation of specific cancerous cells from normal
healthy cells.
confluent cells are passaged, the internalized dyes can be
extruded out to the fresh growth media, due to the reverse
gradient in the fluorophore concentrations. Because of the poor
intracellular retention of the molecular species, most of the
conventional fluorophores are unsuitable for long-term cellular
tracing applications.2
Various approaches have been taken in order to keep the
dyes inside cells.2 A number of chemical reactions have been
utilized to fix fluorophoric units to the cellular components,
typical examples of which include thiol- and amine-mediated
reactions,2 with the azide−alkyne click reaction being the new
development in the area of research.4,5 However, the in situ
covalent attachment of the synthetic dyes to the intracellular
1. INTRODUCTION
The development of long-lasting cellular tracers is of great
scientific value and has important practical implications, for
such tracers may enable biological researchers and medicinal
practitioners to systematically and continuously follow cell
transplantation, migration, division, fusion, and lysis to study
cancer pathogenesis, invasion, and metastasis, to monitor flow
and circulation in capillaries, to examine neural network and
neuronal connectivity, to scrutinize cellular movements in
tissues and organisms, etc. over a large time span on site and in
time.1
Fluorescent molecules have been widely used as biological
imaging reagents.2,3 When the fluorophores are added into
culture media of live cells, the small molecules can penetrate
into cellular interiors, often due to the concentration gradient
in the extra- and intracellular fluorophores. However, when the
Received: December 27, 2012
© XXXX American Chemical Society
A
dx.doi.org/10.1021/ja312581r | J. Am. Chem. Soc. XXXX, XXX, XXX−XXX

Journal of the American Chemical Society
Article
biopolymers may alter their chain conformations and folding
patterns and thus exert undesired effects on the physiological
functions and metabolic processes of the cellular components.
The use of the Cu-based catalysts in the click reactions is an
issue of serious concern, owing to the severe cytotoxicity of the
metallic species.
Scheme 1. Synthesis of a Bioconjugate of Tetraphenylethene
(TPE) and Chitosan (CS)
a
Another approach taken by the researchers in the area is to
label biomacromolecules (e.g., proteins and dextrans) with
fluorophores through bioconjugation reactions outside cells
under abiotic conditions.2 Although this approach can avoid the
problems associated with the intracellular reactions, it has other
drawbacks. For example, the protein bioconjugates may be
readily decomposed by enzymatic proteolysis. The dextran
bioconjugates, on the other hand, are membrane-impermeant
and usually need to be loaded onto cells by harsh and invasive
techniques or processes, such as microinjection, electro-
poration, whole-cell patch clamping, perfusion, scrape loading,
microprojectile bombardment, and hypotonic shock, which
may harm cellular health and cause cellular infection or even
lysis.2
a
By the addition reaction of the isothiocyanate (ITC) group in TPE−
ITC with the amino group in CS.
Fluorescence output of a bioconjugate is often limited by the
degree of labeling (DL) or number of fluorescent labels that are
attached to a biopolymer chain. A high DL frequently does not
lead to an enhanced but to a weakened emission, due to the
notorious aggregation-caused quenching (ACQ) effect on the
fluorescence process.6 In a heavily labeled bioconjugate, the
aromatic fluorophores tend to aggregate in aqueous media,
generating such detrimental species as excimers. The excimer
formation prompts nonradiative decays of the excited states and
quenches fluorescence of the bioconjugate. For example,
fluorescence of heavily labeled casein conjugates is quenched
to a level of typically <3% of that of their corresponding free
fluorophores.2 Although more fluorophores can be loaded onto
microspheres of synthetic polymers (e.g., polystyrene beads)
through elaborate control, the ACQ problem still persists,
hampering the fabrication of highly emissive polymer beads.2
We have observed a unique photophysical process termed
aggregation-induced emission (AIE),7−10 which is opposite to
the common ACQ effect discussed above.6 An AIE fluorogen is
virtually nonemissive as an isolated molecule but becomes
highly fluorescent in the aggregate state.7−13 Tetraphenylsilole
(TPS)9 and tetraphenylethene (TPE)10 are two archetypical
AIE fluorogens. In our previous work, we synthesized a TPS
derivative.14 Whereas conventional dyes (e.g., MitoTracker
Green FM) could stain HeLa cells for only one passage, the
aggregates of the TPS derivative could trace the cells for four
passages.14 In another work, we fabricated nanoparticles of a
TPE-decorated fumaronitrile using lipid−poly(ethylene glycol)
complexes as the coencapsulation matrixes. The nanoparticle
surfaces were further functionalized by an HIV-1 transactivator
of transcription (Tat) and the resultant Tat−AIE nanoparticles
could trace live cells for 10−12 passages in vitro.15
highly emissive bioconjugates can enter living cells in a
noninvasive manner. While the nanoparticles in our previous
work were fabricated by multistep procedures,15 the TPE−CS
bioconjugates in this work spontaneously cluster into micro-
particles inside live cells. The internalized microparticles do not
leak out from the intracellular compartments and can trace the
living cells for as long as 15 passages.
2. RESULTS AND DISCUSSION
Synthesis of TPE−ITC Adduct. Since our first report on
the AIE behavior of TPE,16 many TPE derivatives have been
prepared.10 In this work, we prepared a TPE−ITC adduct, i.e.,
1-[4-(isothiocyanatomethyl)phenyl]-1,2,2-triphenylethene, by
the synthetic route shown in Scheme S1 in the Supporting
Information (SI).17 The product was characterized by standard
spectroscopic techniques, from which satisfactory analysis data
corresponding to its expected molecular structure were
obtained. The product, for example, shows peaks at δ 4.63
1
and 48.75 in its H (Figure 1a) and 13C (Figure S1a in the SI)
NMR spectra, respectively, due to the resonance of the
methylene unit between the TPE and ITC units, while the peak
at m/z = 403.1386 in the HRMS spectrum (Figure S2, SI)
perfectly matches the theoretical mass of the adduct (m/z =
403.1395).
TPE−ITC is AIE active, like many other TPE derivatives
previously prepared in our groups.10 As seen from Figure 2a,
practically no photoluminescence (PL) signals are collected
from the dilute THF solution of TPE−ITC. Its phenyl rings
undergo active intramolecular rotations in the THF solution,
which nonradiatively annihilates its excited states and renders it
nonemissive in the solution state.7,8 However, when a large
amount of water (fw > 70 vol %) is admixed with THF, the
fluorogen starts to radiate. In the THF/water mixture with an
fw of 90%, its PL intensity (I) becomes ∼36,200-fold stronger
than that in the THF solution (I0). In the aqueous mixtures
with high fw values, the fluorogenic molecules aggregate. In the
aggregates, intramolecular rotations of the aromatic rotors of
the fluorogens are physically hindered. This restriction of the
intramolecular rotations (RIR) blocks nonradiative decay
channels and populates radiative transitions, thus making the
aggregates highly fluorescent.8−10
Fluorescent cellular tracing for even longer periods of time is
desirable and useful for many biomedical applications (vide
supra). In this work, we functionalized TPE with a reactive
isothiocyanate (ITC) group. The obtained TPE−ITC adduct
was used to label chitosan (CS), a cytocompatible biopolymer
derived from shells of crabs, shrimps, lobsters, etc. (Scheme 1).
The AIE attribute of TPE−ITC enabled fabrication of highly
fluorescent TPE−CS bioconjugates through the attachment of
a large number of AIE labels to a CS chain. The fluorescence
output of the bioconjugate can be enhanced to a great extent
(up to 2 orders of magnitude) by simply increasing its DL. The
B
dx.doi.org/10.1021/ja312581r | J. Am. Chem. Soc. XXXX, XXX, XXX−XXX

Journal of the American Chemical Society
Article
chains were also labeled by fluorescein (F), an ACQ unit, by
the same synthetic route (Scheme S2 in the SI). The analysis
data of the resultant FCS bioconjugates are in nice agreement
with those reported in the literatures,18,19 with the examples of
NMR spectra of a TPE−CS bioconjugate shown in Figures 1
and S1 (SI). The resonance peaks of the protons of the phenyl
rings of the TPE−CS bioconjugate and the carbons of its
phenyl rings/olefin double bond appear at δ 7.50−6.70 and
132.1−128.8 in the 1H and 13C NMR spectra, respectively. The
appearance of these new peaks confirms the successful labeling
of the CS chains by the TPE units.
The efficiencies of labeling by the bioconjugation reactions
were evaluated by UV, PL and NMR analyses.18,19 At a feed
ratio (Rf = TPE/CS) of 1 mol %, the DL of the TPE−CS
bioconjugate determined by UV analysis is 0.88 mol % (Table
S1 in the SI). The DL is increased with an increase in Rf. When
the Rf is increased to 20 mol %, the DL reaches 7.20 mol %.
The DLs of the TPE−CS bioconjugates calculated from the
NMR spectral data are in good agreement with those obtained
from the UV data. The DL of the FCS bioconjugate prepared at
an Rf of 1 mol % is too low to be determined by the UV and
NMR analyses. Using the highly sensitive PL technique, the DL
of the bioconjugate is estimated to be 0.09 mol %. Even when
the Rf is increased to as high as 10 mol %, the DL of the FCS
bioconjugate is still as low as 0.52 mol %.
All the DLs of the TPE−CS bioconjugates are much higher
than those of their FCS congeners (Table S1, SI), indicating
that TPE−ITC is much more reactive than FITC. This can be
readily comprehended from the difference in their structures. In
TPE−ITC, there is a methylene spacer between the phenyl ring
and the ITC group. This saturated spacer shuts down the
electronic interaction between the ITC and phenyl units and
mitigates the steric hindrance effect of the phenyl ring, thus
rendering the ITC group very reactive. In FITC, however, the
ITC group is directly linked to the phenyl ring. The electronic
and steric effects collectively passivate the ITC group, making it
difficult to label the CS chain. Indeed, other researchers have
reported the similarly low DLs in their systems of CS labeling
by FITC.18
1
Figure 1. H NMR spectra of (a) TPE−ITC, (b) CS, and (c) TPE−
CS measured in (a) chloroform-d and (b, c) an acetic acid-d4/water-d2
mixture at room temperature. The solvent peaks are marked with
asterisks.
The appearances of solid powders of the TPE−CS and FCS
bioconjugates are whitened and darkened, respectively, with
increases in their DLs (Figure S3 in the SI). When the solid
powder of the TPE−CS bioconjugate with a DL of 0.83 mol %
is excited by a UV light, it emits a greenish-blue light (Figure
3a). The fluorescence of the bioconjugate is intensified with an
increase in its DL: the more the TPE labels, the stronger the
Figure 2. (a) PL spectra of TPE−ITC in THF/water mixtures with
different fractions of water (fw); c = 50 μM, λex = 338 nm. (Inset) Plot
of the relative PL intensity (I/I0) of TPE−ITC at 479 nm vs the
composition of the THF/water mixture (fw). I0 = PL intensity of
TPE−ITC in THF solution. (b) Photographs of solid powders of
TPE−ITC and fluorescein isothiocyanate (FITC) taken under room
lighting and UV illumination.
The solid powders of TPE−ITC are gray in appearance
under normal room lighting and emit a strong greenish-blue
light upon UV illumination (Figure 2b, upper panel). This is
very different from the fluorescence behaviors of the conven-
tional ACQ fluorophores. FITC, for example, is a fluorescein
(F) dye functionalized by an isothiocyanate (ITC) unit that is
reactive toward an amino group in a biopolymer (such as
protein). It has been utilized as a fluorescent dye in an array of
biological applications (e.g., flow cytometry). Though FITC is
emissive in dilute solutions, its solid powders are non-
fluorescent under UV excitation, as can be seen from the
photograph shown in the lower panel of Figure 2b, due to the
severe self-quenching effect in the dye aggregates.
Figure 3. Fluorescent images of (a−c) TPE−CS and (d−f) FCS
bioconjugates with degrees of labeling (DLs; mol %) of (a) 0.83, (b)
2.77, (c) 7.86, (d) 0.09, (e) 0.11, and (f) 0.52. The photographs were
taken under UV illumination (365 nm).
Labeling CS Chains with TPE Fluorogens. CS chains
were labeled by the AIE-active TPE units via the synthetic route
shown in Scheme 1.18 For the purpose of comparison, the CS
C
dx.doi.org/10.1021/ja312581r | J. Am. Chem. Soc. XXXX, XXX, XXX−XXX

Journal of the American Chemical Society
Article
light emission (cf. panels a−c). This suggests that the TPE−CS
conjugates are AIE active. A diametrically opposed effect is
observed in the case of the FCS bioconjugate. While the FCS
conjugate with a DL of 0.09 mol % emits a bright-yellow light,
the emission is drastically weakened with an increase in its DL
(panels d−f). The FCS bioconjugate with a DL of 0.52 mol %
becomes totally nonfluorescent, due to the ACQ effect of the
fluorescein aggregates in the “heavily” labeled FCS system.
AIE Behaviors of TPE−CS Bioconjugates. The AIE effect
of the TPE−CS bioconjugates suggested by the fluorescent
images shown in panels a−c of Figure 3 was further studied
spectroscopically. The PL spectra of the dilute solutions of TPE
and CS run nearly parallel to the abscissa, verifying their
nonemissive nature in the solution state.10 The TPE−CS
bioconjugates are soluble in acidic media (aqueous acetic acid
with pH 5 being used as the solvent in this work). However,
different from those of their parent forms (TPE and CS), the
dilute solutions of the TPE−CS bioconjugates are fluorescent
(Figure 4a). When a TPE group is bound to a CS backbone,
is changed to a neutral one, which is immiscible with the
aqueous medium. The polymer chains therefore contract in
conformation and form aggregates that wrap the TPE labels.
This prompts the RIR process and brightens the emission of
the TPE−CS bioconjugate.
Live Cell Imaging. Being a biomacromolecule from the
natural resources, CS is cytophilic.20 The novel AIE effect of the
TPE−CS bioconjugates encouraged us to utilize them as
bioprobes for cellular imaging applications. HeLa cells were
cultured in the aqueous buffers containing the bioconjugates at
pH 6.2 using a standard cell-staining protocol. As can be seen
from Figure 5, the living HeLa cells cultured in the presence of
Figure 5. (a) Bright-field and (b) fluorescent images of the HeLa cells
cultured in the presence of the TPE−CS bioconjugates with different
DLs. [TPE−CS] = 0.1 mg/mL, pH = 6.2. Photographs taken under
(a) room lighting and (b) UV irradiation.
the TPE−CS bioconjugate with a DL of 0.83 mol % emits a
weak light, in accordance with its weak PL spectrum shown in
Figure 4a. When the DL of the bioconjugate is increased to
2.77 mol %, the stained cells become emissive. Very bright light
is emitted from the living HeLa cells stained by the
bioconjugate with a DL of 7.86 mol %.
In our previous work, a commercially available dye named
CellTracker Green CMFDA (or 5-chloromethylfluorescein
diacetate)2 was used to stain HeLa cells.21 The fluorescence
process of the commercial dye was activated by the esterase
hydrolysis with the involvement of an addition reaction of its
benzyl chloride moiety to the intracellular thiol unit. No
chemical reactions are involved in our cellular staining system.
The light emissions of the TPE−CS bioconjugates are triggered
by a rapid physical process of spontaneous aggregation of the
internalized AIE fluorogen molecules inside the cellular
compartments.
The intracellular imaging processes of our AIE-active TPE−
CS bioconjugates were further scrutinized by the careful 3D
analysis using a confocal laser scanning microscope.22 The
extracellular bioconjugate molecules dissolved in the culture
media (pH = 6.2) had been thoroughly washed away before the
image was taken. As can be seen from the projections of the 3D
image onto the x−z and y−z planes shown in Figure 6a, the PL
signals are from the TPE−CS bioconjugates internalized by the
HeLa cells. The bioconjugate molecules engulfed by the live
cells form aggregates, as the TPE-labeled CS chains are
insoluble in the cytosols at the intracellular pH (7.2−7.4;
Figure S4). The intracellular aggregates emit brightly, thanks to
the AIE nature of the TPE labels.
Figure 4. (a) PL spectra of TPE−CS bioconjugates with different DLs
measured in 0.1 M aqueous acetic acid at [TPE−CS] = 0.1 mg/mL.
(b) Plot of the relative PL intensity (I/I0) of TPE−CS vs its DL. I0 =
PL intensity in 0.1 M acetic acid solution. Inset: Fluorescent images of
the TPE−CS solutions taken at pH = 5 and pH = 8 under UV
illumination.
the intramolecular rotations of its phenyl rings are not as easy
as in the unbound, “free” state. This activates the RIR process,
thus making the TPE−CS bioconjugate fluorescent even in the
solution state.8−10
The fluorescence intensity of the TPE−CS bioconjugate is
increased with an increase in its DL in a nonlinear fashion
(Figure 4b). The bioconjugate with a high DL of 7.86 mol % is
much more fluorescent than its counterpart with a low DL of
0.83 mol %: the light emission of the former is >35-fold more
intense than that of the latter. The heavy labeling of the CS
chain by the hydrophobic TPE fluorogens may have decreased
the miscibility of the macromolecular chain with the aqueous
medium. The polymer chain may have entangled or wrapped
the TPE fluorogen and strengthened the RIR process, thereby
nonlinearly boosting its fluorescence.
When the pH of the aqueous medium is changed from 5 to
8, the PL intensity of the TPE−CS bioconjugate is jumped by
more than 2 times, as clearly shown by the intense blue light
emitted from the sample vial in Figure 4b. The dissolution of
the TPE−CS bioconjugate in the acidic medium is realized by
the ionization or quaternization of the amino units in the CS
chain. When the acidic medium (pH 5) is changed to a basic
one (pH 8), the quaternized or ionized macromolecular chain
Leakage-Free Staining. Conventional fluorophores can
hardly be retained inside live cells for a long period of time
under physiological conditions. It is very difficult for them to
survive the passage processes, during which they are easily
D
dx.doi.org/10.1021/ja312581r | J. Am. Chem. Soc. XXXX, XXX, XXX−XXX

Journal of the American Chemical Society
Article
In the case of bioimaging using small-molecule fluorophores,
fluorescence of the stained cells is often drastically weakened
with an increase in the number of passages. The TPE−CS
bioconjugate is a big polymer and behaves differently from its
small-molecule counterpart. The fluorescence intensity of the
stained cells does not change much, while the relative number
of the stained cells decreases rapidly with increasing number of
passages. This implies that the aggregates of the TPE−CS
bioconjugate are not uniformly split during the process of cell
division (Figure S5, SI).
The persistence in the fluorescence signals from the stained
live cells monitored in the first five days was evaluated by a
confocal microscope (Figure S6, SI). The experimental data
further confirm that the stained live cells can retain the high
fluorescence intensities as the original ones in a long period of
time. To our delight, the stained cells still fluoresce brightly
even for as long as 15 passages (Figure 7b). It is worth noting
that CellTracker Green CMFDA, the best-known cellular
tracer, can track the living cells for no more than three
passages.24
In our previous study, MitoTracker Green FM (MTG) and
an aminated silole, both being small molecules, were used for
cell tracing studies.14 Virtually no fluorescence signals were
detected in the MTG-stained cells after the first passage, owing
to the low working concentration of MTG (in the low
nanomolar range to avoid the nonspecific staining and inherent
cytotoxicity).2 The aminated silole performed better and could
trace the live cells for four passages, due to its high working
concentration (in the range of mM) and the slow release of its
internalized nanoparticles to the fresh cell growth media.14 The
cellular imaging performance of the TPE−CS bioconjugate is
even better, thanks to its polymeric nature and to the big sizes
of its aggregates, which makes it difficult for the internalized
microparticles to escape from the cellular enclosures and to
break down in the cell division processes.
In a conventional system, even if a zero rate of dye efflux is
assumed, 15 times of cell divisions will lead to a 215- or 32,768-
fold dilution of the dye signal, which makes the labeled cells
undistinguishable. In our system, large AIE aggregates are
preserved in one cell, instead of being averaged to two daughter
cells in every cell division cycle. The long retention duration of
the microparticles of the TPE−CS bioconjugates in a specific
preloaded cell line may enable their use in a variety of unique
biomedical applications as long-term cellular tracers for
monitoring biological processes, pathological pathways, ther-
apeutic effects, and so on.
Figure 6. (a) Optical section (x−y axis) captured by confocal laser
scanning microscope, with projections on x−z and y−z planes, of the
HeLa cells incubated in the presence of a solution (0.1 mg/mL) of a
TPE−CS bioconjugate with a DL of 7.86 mol %. (b) Bright-field,
fluorescence, and overlay images of the TPE−CS-stained HeLa cells
cocultured with unstained 3T3 cells.
extruded back to the growth media. To examine if the TPE−CS
bioconjugates will diffuse from the intracellular compartments
to the extracellular media, the HeLa cells stained by the TPE−
CS bioconjugates were cocultured with unstained 3T3 cells in
Petri dishes. These two types of cell lines are chosen in this
study because of their distinct cell morphologies.
After overnight coculture, the aggregates of the TPE−CS
bioconjugates remain inside the living HeLa cells, as verified by
the intense PL signals collected from these cells (Figure 6b).
On the other hand, the 3T3 cells in the same Petri dish remain
nonfluorescent, revealing that the TPE−CS bioconjugates have
neither escaped from the HeLa cells nor penetrated into the
3T3 cells in the coculture media. As discussed above, the
conventional dyes usually function as single molecular species
in the cellular imaging systems, which can be easily extruded
out from the cells, unless they are bound by the covalent bonds
generated by in situ chemical reactions with the intracellular
reactive species such as thiol.2,23 The TPE−CS bioconjugates
function as supramolecular aggregates, which are difficult to
extrude out once they are internalized,14,21 which accounts for
their remarkable leakage-free staining behaviors.
Long-Term Tracing. The leakage-free retention of the
TPE−CS bioconjugates in the stained cells suggests that they
can serve as fluorescent bioprobes for long-term cell tracing.
This is indeed the case. The living cells stained by the heavily
labeled CS chains emit intensely at the first passage (Figure 7).
3. CONCLUSIONS
In this work, we have designed and synthesized a TPE−ITC
adduct with AIE characteristics. The insertion of a saturated
flexible methylene spacer between the phenyl ring and the ITC
group has helped alleviate the detrimental effects caused by the
electronic interaction and steric hindrance. This renders TPE−
ITC more reactive than FITC, a widely used fluorescent
labeling reagent for bioimaging. The high reactivity of TPE−
ITC has enabled more TPE labels to be attached to a CS chain.
Whereas the fluorescence of the FCS bioconjugate is quickly
quenched when its DL is slightly increased due to the ACQ
effect commonly observed in the conventional fluorophore
systems, the emission of the TPE−CS bioconjugate is boosted
with an increase in its DL, thanks to the unique AIE effect of
the TPE label.
Figure 7. Fluorescent images of the HeLa cells stained by the
aggregates of the TPE−CS bioconjugate with a DL of 7.86 mol % at
different passages taken at magnifications of (a) ×100 and (b) ×1000.
Numbers of passages are denoted by the Arabic numerals (1−15) on
the right sides of the images.
E
dx.doi.org/10.1021/ja312581r | J. Am. Chem. Soc. XXXX, XXX, XXX−XXX

Journal of the American Chemical Society
Article
Some of the TPE labels attached to the CS backbones may
be wrapped by the polymer chains. After being strangled by the
polymer chains, the phenyl rotors of TPE become difficult to
undergo intramolecular rotations. This RIR process hampers
the excited states of TPE from nonradiatively decaying. In the
TPE−CS bioconjugate with a DL of 7.86 mol %, the heavily
labeled CS chain becomes more hydrophobic and less miscible
with the aqueous medium, which further strengthens the RIR
process, thereby rendering the TPE−CS bioconjugate highly
fluorescent. The bioconjugate becomes more emissive when the
pH of the medium is increased (from acidic to alkalescent),
owing to the decrease in its solubility in the medium and the
aggregate formation of its molecules at high pH.
The AIE activity and pH sensitivity of the bioconjugates have
enabled them to work as fluorescent light-up bioprobes for
intracellular imaging applications. When internalized, the TPE-
labeled CS chains become insoluble and spontaneously form
microparticles at intracellular pH. This triggers the AIE
processes of the bioconjugates and makes the stained live
cells highly emissive. The internalized microparticles do not
leak out in the coculture system, allowing visual differentiation
of one specific cell line from the other unstained cell lines. The
outstanding intracellular retention of the TPE−CS bioconju-
gate microparticles permits the stained cells to be traced for as
long as 15 passages. This makes the TPE−CS conjugates
promising candidate materials for applications in biomedical
areas as long-term cellular tracers which are under high demand
in such important areas as cancer metastasis, neuron
networking, embryo development, and stem cell differentiation.
to 65 °C. Hydrogen peroxide (6 mL) was added into the CS
solution, and the mixture was stirred at 65 °C for 6 h. The thus-
treated CS was precipitated with a 10% sodium hydroxide
solution, filtered, and washed with distilled water until the pH
became neutral. The collected precipitate was lyophilized at
−50 °C for three days. The deacetylation degree of CS was
93.64%, as determined by conductometric titration using a
conductivity meter (DDS-307).25 The Mη value of the CS was
decreased by the degradation reaction to 5.60 × 104, as
estimated by the molecular weight measurement with an
Ubbelohde viscometer at 25 0.5 °C using a mixture of 0.1 M
acetic acid and 0.2 M sodium acetate as the solvent.
Labeling of CS by TPE−ITC. The degraded CS (0.1 g, 0.61
mmol) was added into a two-necked flask, evacuated under
vacuum and flushed with dry nitrogen three times. DMSO (10
mL) was added into the flask, and the mixture was stirred at 60
°C for 24 h. A specific amount of TPE−ITC (with Rf in the
range of 1−20 mol %) was added into the flask, and the
resultant mixture was stirred for 24 h. The product was washed
with distilled water five times and acetone three times and then
was dissolved in an aqueous mixture of acetic acid with an equal
volume of acetone. The solution was precipitated with a 10%
sodium hydroxide solution, filtered, and washed with distilled
water until the pH became neutral. The product was dried
under vacuum at 60 °C. The labeling experiments of CS by
FITC (Rf = 1−10 mol %) were conducted as a control by
employing the published experimental procedures.18 The
labeling efficiencies or DLs of the TPE−CS and FCS
bioconjugates were determined by UV, PL, and NMR
analyses,18,26 and the results are summarized in Table S1 in
the SI.
4. EXPERIMENTAL SECTION
General Information. All the chemicals and reagents were
purchased from Sigma and Aldrich and used without further
purification, unless specified otherwise. THF, toluene, and
dichloromethane were purified by simple distillation prior to
use. 1-[4-(Azidomethyl)phenyl]-1,2,2-triphenylethylene (6 in
the SI) was prepared using previously reported procedures.17
1H and 13C NMR spectra were measured on a Bruker ARX 400
spectrometer. HRMS spectra were taken on a Finnigan TSQ
7000 triple quadrupole spectrometer operating in the MALDI-
TOF mode. UV−vis absorption spectra were recorded on a
Milton Ray Spectronic 3000 array spectrophotometer, and PL
spectra were measured on a Perkin-Elmer LS 50B spectro-
fluorometer with a xenon discharge lamp excitation.
Cellular Imaging. HeLa cells were grown overnight on a
cover slide in a 35-mm Petri dish. The living cells were stained
with a solution of a TPE−CS bioconjugate (100 μg/mL) and
were further incubated for another 4 h. After careful washing,
fluorescence images of the stained cells were taken on a
fluorescence microscope (Olympus BX41) with a long-pass
emission filter using the following parameters: excitation
wavelength = 330−380 nm and dichroic mirror = 400 nm.
Coculture of HeLa and 3T3 Cells. HeLa cells were stained
with a TPE−CS solution (100 μg/mL) for 4 h. After washing
five times with PBS, the stained HeLa cells and the unstained
3T3 cells were detached from their respective culture dishes by
treating them with trypsin−EDTA solution. The stained HeLa
cells (1.5 × 103) and the unstained 3T3 cells (2.5 × 103) were
resuspended with 2 mL of DMEM and then transferred to a
new 25-mm Petri dish. After incubating at 37 °C for 12 h, phase
contrast and fluorescence images of the cocultured cells were
taken from the same spot using a Zeiss laser scanning confocal
microscope (LSM7 DUO; excitation: 405 nm, filter: 449−520
nm).
Synthesis of TPE−ITC. The azido-functionalized TPE (6;
0.330 g, 0.852 mmol) and triphenylphosphine (0.112 g, 0.426
mmol) were added into a two-necked flask, which was
evacuated under vacuum and flushed with dry nitrogen three
times. Carbon disulfide (0.55 g, 7.242 mmol) and distilled
dichloromethane (50 mL) were added into the flask under
stirring. The resultant reaction mixture was refluxed overnight,
followed by the removal of the solvent under a reduced
pressure. The crude product was precipitated with cold ether
(250 mL), and the precipitate was filtered and washed with
cold ether (30 mL) three times. The product was dried under
Long-Term Cell Tracing. Living HeLa cells (3 × 104) were
cultured overnight and then stained in a Petri dish at 50%
confluence with a TPE−CS solution (100 μg/mL). After an
image was taken at the end of 24 h of incubation (referred to as
the end of the first passage), 25% of the cells in the completely
filled Petri dish were transferred to a new dish with fresh
growth medium. Another image was taken after 24 h in the
then half-filled Petri dish, i.e., the end of the second passage.
The living cells were further incubated for another 24 h to the
end of the third passage. This process was iterated to proceed
to the 15th passage.
1
vacuum to give a white solid in 85.2% yield. H NMR (400
MHz, CDCl3) δ (ppm): 7.00−7.13 (m, 19H), 4.63 (s, 2H).
HRMS (MALDI-TOF) m/e: 403.1386 ([M]+, calcd:
403.1395).
Degradation of CS. Six (6) grams of CS with a viscosity-
average molecular weight (Mη) of 1.16 × 106 was added into a
mixture of distilled water (190 mL) and acetic acid (4 mL),
which was stirred for 1 h at room temperature and then heated
F
dx.doi.org/10.1021/ja312581r | J. Am. Chem. Soc. XXXX, XXX, XXX−XXX

Journal of the American Chemical Society
Article
39, 2522−2544. (c) Qin, A.; Lam, J. W. Y.; Tang, B. Z. Macromolecules
2010, 43, 8693−8702.
(6) (a) Nadeau, J. Introduction to Experimental Biophysics; CRC Press:
Boca Raton, FL, 2011. (b) Orellana, G., Moreno-Bondi, M. C., Eds.
Frontiers in Chemical Sensors: Novel Principles and Techniques; Springer:
Berlin, 2010. (c) Birks, J. B. Photophysics of Aromatic Molecules; Wiley:
London, 1970.
(7) Luo, J.; Xie, Z.; Lam, J. W. Y.; Cheng, L.; Chen, H.; Qiu, C.;
Kwok, H. S.; Zhan, X.; Liu, Y. Q.; Zhu, D. B.; Tang, B. Z. Chem.
Commun. 2001, 1740−1741.
(8) (a) Hong, Y.; Lam, J. W. Y.; Tang, B. Z. Chem. Soc. Rev. 2011, 40,
5361−5388. (b) Wang, M.; Zhang, G.; Zhang, D.; Zhu, D.; Tang, B. Z.
J. Mater. Chem. 2010, 20, 1858−1867. (c) Hong, Y.; Lam, J. W. Y.;
Tang, B. Z. Chem. Commun. 2009, 4332−4353. (d) Liu, J.; Lam, J. W.
Y.; Tang, B. Z. Chem. Rev. 2009, 109, 5799−5867.
(9) For a review, see: Liu, J.; Lam, J. W. Y.; Tang, B. Z. J. Inorg.
Organomet. Polym. Mater. 2009, 19, 249−285.
(10) (a) Zhao, Z.; Lam, J. W. Y.; Tang, B. Z. Curr. Org. Chem. 2010,
14, 2109−2131. (b) Zhao, Z.; Lam, J. W. Y.; Tang, B. Z. J. Mater.
Chem. 2012, 22, 23726−23740.
ASSOCIATED CONTENT
■
S
* Supporting Information
Schemes depicting the synthetic routes to TPE−ITC and FCS;
a table listing the DLs of the TPE−CS and FCS bioconjugates;
and figures showing the 13C NMR spectra of TPE−ITC, CS
and TPE−CS, the HRMS spectrum of TPE−ITC, the
photographs of the TPE−CS and FCS bioconjugates with
different DLs, dynamic uptake process of TPE−CS by HeLa
cell, dynamic distribution of TPE−CS during the cell division,
intensities of the fluorescent signals from the stained cells, and
cytotoxicity data. This material is available free of charge via the
Internet at http://pubs.acs.org.
AUTHOR INFORMATION
■
Corresponding Author
tangbenz@ust.hk
Author Contributions
§Z.W. and S.C. contributed equally to this work.
(11) For reviews, see: (a) Chi, Z.; Zhang, X.; Xu, B.; Zhou, X.; Ma,
C.; Zhang, Y.; Liu, S.; Xu, J. Chem. Soc. Rev. 2012, 41, 3878−3896.
(b) Salinas, Y.; Martinez-Manez, R.; Marcos, M. D.; Sancenon, F.;
Costero, A. M.; Parra, M.; Gil, S. Chem. Soc. Rev. 2012, 41, 1261−
1296. (c) Tanaka, K.; Chujo, Y. Macromol. Rapid Commun. 2012, 33,
1235−1255. (d) Anthony, S. P. ChemPlusChem 2012, 77, 518−531.
(e) Akiyama, S. J. Synth. Org. Chem. Jpn. 2012, 70, 465−472.
(f) Hatano, K.; Matsuoka, K.; Terunuma, D. Trends Glycosci.
Glycotechnol. 2012, 24, 78−94. (g) Yang, G.; Li, S.; Wang, S.; Li, Y.
C. R. Chim. 2011, 14, 789−798. (h) Yuan, C.; Xin, Q.; Liu, H.; Wang,
L.; Jiang, M.; Tao, X. Sci. China, Ser. B 2011, 54, 587−595.
(i) Strassert, C. A.; Mauro, M.; De Cola, L. Adv. Inorg. Chem. 2011,
63, 47−103. (j) Corey, J. Y. Adv. Organomet. Chem. 2011, 59, 1−180.
(k) Kim, H. N.; Guo, Z.; Zhu, W.; Yoon, J.; Tian, H. Chem. Soc. Rev.
2011, 40, 79−93. (l) Wu, J.; Liu, W.; Ge, J.; Zhang, H.; Wang, P.
Chem. Soc. Rev. 2011, 40, 3483−3495. (m) Accetta, A.; Corradini, R.;
Marchelli, R. Top. Curr. Chem. 2011, 300, 175−216. (n) Shimizu, M.;
Hiyama, T. Chem.Asian J. 2010, 5, 1516−1531. (o) Chan, C. P.-Y.
Bioanalysis 2009, 1, 115−133. (p) Zhao, Y. S.; Fu, H.; Peng, A.; Ma,
Y.; Xiao, D.; Yao, J. Adv. Mater. 2008, 20, 2859−2876. (q) Yang, B.;
Ma, Y.-G.; Shen, J.-C. Chem. J. Chin. Univ. 2008, 29, 2643−2658.
(r) Chen, J.; Cao, Y. Macromol. Rapid Commun. 2007, 28, 1714−1742.
(s) Itami, K.; Ohashi, Y.; Yoshida, J. J. Org. Chem. 2005, 70, 2778−
2792.
Notes
The authors declare no competing financial interest.
ACKNOWLEDGMENTS
■
The work reported in this paper was partly supported by
National Basic Research Program of the Ministry of Science
and Technology of China (973 Program; 2013CB834701), the
Research Grants Council of Hong Kong (HKUST2/CRF/10
and N_HKUST620/11), the University Grants Committee of
Hong Kong (AoE/P-03/08), and the Guangdong Innovative
Research Team Program of China (201101C0105067115).
REFERENCES
■
(1) For selected examples of reviews, see: (a) Woolf, N. J. Prog.
Neurobiol. 1998, 55, 59−77. (b) Nemoto, T. Mol. Cells 2008, 26, 113−
120. (c) Mattoussi, H.; Palui, G.; Na, H. B. Adv. Drug Delivery Rev.
2012, 64, 138−166. (d) Taylor, A.; Wilson, K. M.; Murray, P.; Fernig,
D. G.; Levy, R. Chem. Soc. Rev. 2012, 41, 2707−2717. (e) Dreaden, E.
C.; Alkilany, A. M.; Huang, X.; Murphy, C. J.; El-Sayed, M. A. Chem.
Soc. Rev. 2012, 41, 2740−2779.
(2) (a) Johnson, I., Spence, M. T. Z., Eds. The Molecular Probes
Handbook: A Guide to Fluorescent Probes and Labeling Technologies,
11th ed.; Life Technologies Corp.: New York, 2010. (b) Herold, K. E.,
Rasooly, A., Eds. Biosensors and Molecular Technologies for Cancer
Diagnostics; Taylor & Francis: Oxford, 2012. (c) Thompson, R. B., Ed.
Fluorescence Sensors and Biosensors; CRC Press: Boca Raton, FL, 2005.
(3) For reviews, see: (a) Yao, S.; Belfield, K. D. Eur. J. Org. Chem.
2012, 17, 3199−3217. (b) Kim, W. H.; Lee, J.; Jung, D.-W.; Williams,
D. R. Sensors 2012, 12, 5005−5027. (c) Coto-Garcia, A. M.; Sotelo-
Gonzalez, E.; Fernandez-Arguelles, M. T.; Pereiro, R.; Costa-
Fernandez, J. M.; Sanz-Medel, A. Anal. Bioanal. Chem. 2011, 399,
29−42. (d) Tanury, P.; Malhotra, A.; Byrne, L. M.; Santra, S. Adv.
Drug Delivery Rev. 2010, 62, 424−437. (e) Escobedo, J.; Rusin, O.;
Lim, S.; Strongin, R. M. Curr. Opin. Chem. Biol. 2010, 14, 64−70.
(f) Meech, S. R. Chem. Soc. Rev. 2009, 38, 2922−2934. (g) Terai, T.;
Nagano, T. Curr. Opin. Chem. Biol. 2008, 12, 515−521.
(12) (a) Shustova, N. B.; Ong, T.-C.; Cozzolino, A. F.; Michaelis, V.
K.; Griffin, R. G.; Dinca, M. J. Am. Chem. Soc. 2012, 134, 15061−
̆
15070. (b) Tian, D.; Jiang, J.; Hu, H.; Zhang, J.; Cui, C. J. Am. Chem.
Soc. 2012, 134, 14666−14669. (c) Xu, Y.; Chen, L.; Guo, Z.; Nagai, A.;
Jiang, D. J. Am. Chem. Soc. 2011, 133, 17622−17625. (d) Ren, Y.; Kan,
W. H.; Henderson, M. A.; Bomben, P. G.; Berlinguette, C. P.;
Thangadurai, V.; Baumgartner, T. J. Am. Chem. Soc. 2011, 133,
17014−17026. (e) Perez, A.; Luis Serrano, J.; Sierra, T.; Ballesteros,
A.; de Saa, D.; Barluenga, J. J. Am. Chem. Soc. 2011, 133, 8110−8113.
(f) Procopio, E. Q.; Mauro, M.; Panigati, M.; Donghi, D.; Mercandelli,
P.; Sironi, A.; D’Alfonso, G.; De Cola, L. J. Am. Chem. Soc. 2010, 132,
14397−14399. (g) Dong, J.; Solntsev, K. M.; Tolbert, L. M. J. Am.
Chem. Soc. 2009, 131, 662−670. (h) Peng, Q.; Yi, Y.; Shuai, Z.; Shao, J.
J. Am. Chem. Soc. 2007, 129, 9333−9339. (i) Yu, G.; Yin, S.; Liu, Y.;
Chen, J.; Xu, X.; Sun, X.; Ma, D.; Zhan, X.; Peng, Q.; Shuai, Z.; Tang,
B. Z.; Zhu, D.; Fang, W.; Luo, Y. J. Am. Chem. Soc. 2005, 127, 6335−
6346. (j) Luo, Z.; Yuan, X.; Yu, Y.; Zhang, Q.; Leong, D. T.; Lee, J. Y.;
Xie, J. J. Am. Chem. Soc. 2012, 134, 16662−16670.
(4) (a) Yang, P.-Y.; Wang, M.; Li, L.; Wu, H.; He, C. Y.; Yao, S. Q.
Chem.Eur. J. 2012, 18, 6528−6541. (b) Wang, G.; Geng, J.; Zhang,
X.; Cai, L.; Ding, D.; Li, K.; Wang, L.; Lai, Y.-H.; Liu, B. Polym. Chem.
2012, 3, 2464−2470. (c) Gasser, G.; Pinto, A.; Neumann, S.; Sosniak,
A. M.; Seitz, M.; Merz, K.; Heumann, R.; Metzler-Nolte, N. Dalton
Trans. 2012, 41, 2304−2313. (d) Singh, S.; Aggarwal, A.; Thompson,
S.; Tome, J. P. C.; Zhu, X.; Samaroo, D.; Vinodu, M.; Gao, R.; Drain,
C. M. Bioconjugate Chem. 2010, 21, 2136−2146.
(13) (a) Wee, K.-R.; Han, W.-S.; Cho, D. W.; Kwon, S.; Pac, C.;
Kang, S. O. Angew. Chem., Int. Ed. 2012, 51, 2677−2680. (b) Li, C.;
Wu, T.; Hong, C.; Zhang, G.; Liu, S. Angew. Chem., Int. Ed. 2012, 51,
455−459. (c) Jin, X.-H.; Wang, J.; Sun, J.-K.; Zhang, H.-X.; Zhang, J.
Angew. Chem., Int. Ed. 2011, 50, 1149−1153. (d) Feng, J.; Tian, K.;
Hu, D.; Wang, S.; Li, S.; Zeng, Y.; Li, Y.; Yang, G. Angew. Chem., Int.
Ed. 2011, 50, 8072−8076. (e) Zhang, Z.; Xu, B.; Su, J.; Shen, L.; Xie,
(5) (a) Qin, A.; Lam, J. W. Y.; Tang, B. Z. Prog. Polym. Sci. 2012, 37,
182−209. (b) Qin, A.; Lam, J. W. Y.; Tang, B. Z. Chem. Soc. Rev. 2010,
G
dx.doi.org/10.1021/ja312581r | J. Am. Chem. Soc. XXXX, XXX, XXX−XXX

Journal of the American Chemical Society
Article
Y.; Tian, H. Angew. Chem., Int. Ed. 2011, 50, 11654−11657.
(f) Shimizu, M.; Takeda, Y.; Higashi, M.; Hiyama, T. Angew. Chem.,
Int. Ed. 2009, 48, 3653−3656. (g) Mutai, T.; Tomoda, H.; Ohkawa,
T.; Yabe, Y.; Araki, K. Angew. Chem., Int. Ed. 2008, 47, 9522−9524.
(h) Wu, Y.-T.; Kuo, M.-Y.; Chang, Y.-T.; Shin, C.-C.; Wu, T.-C.; Tai,
C.-C.; Cheng, T.-H.; Liu, W.-S. Angew. Chem., Int. Ed. 2008, 47, 9891−
9894. (i) Bhongale, C. J.; Hsu, C. S. Angew. Chem., Int. Ed. 2006, 45,
1404−1408.
(14) Yu, Y.; Feng, C.; Hong, Y.; Liu, J.; Chen, S.; Ng, K. M.; Luo, K.
Q.; Tang, B. Z. Adv. Mater. 2011, 23, 3298−3302.
(15) Li, K.; Qin, W.; Ding, D.; Geng, J.; Liu, J.; Zhang, X.; Liu, H.;
Liu, B.; Tang, B. Z. Sci. Rep. 2013, 3, 1150.
(16) Dong, Y. Q.; Lam, J. W. Y.; Qin, A.; Sun, J. X.; Liu, J.; Li, Z.;
Zhang, S.; Sun, J. Z.; Kwok, H. S.; Tang, B. Z. Appl. Phys. Lett. 2007,
91, 011111-1−011111-3.
(17) (a) Banerjee, M.; Emond, S. J.; Lindeman, S. V.; Rathore, R. J.
Org. Chem. 2007, 72, 8054−8061. (b) Tang, L.; Jin, J. K.; Qin, A. J.;
Yuan, W. Z.; Mao, Y.; Mei, J.; Sun, J. Z.; Tang, B. Z. Chem. Commun.
2009, 4974−4976.
(18) (a) Huang, M.; Ma, Z. S.; Khor, E.; Lim, L. Y. Pharm. Res. 2002,
19, 1488−1494. (b) Jia, X. Y.; Chen, X.; Xu, Y. L.; Han, X. Y.; Xu, Z.
R. Carbohydr. Polym. 2009, 78, 323−329.
(19) Jiang, G. B.; Quan, D. O.; Liao, K. R.; Wang, H. H. Mol.
Pharmaceutics 2006, 3, 152−160.
(20) Huang, M.; Khor, E.; Lim, L. Y. Pharm. Res. 2004, 21, 344−353.
(21) Yu, Y.; Hong, Y.; Feng, C.; Liu, J.; Lam, J. W. Y.; Faisal, M.; Ng,
K. M.; Luo, K. Q.; Tang, B. Z. Sci. China, Ser. B 2009, 52, 15−19.
(22) (a) Wang, J.; Mei, J.; Hu, R.; Sun, J. Z.; Qin, A.; Tang, B. Z. J.
Am. Chem. Soc. 2012, 134, 9956−9966. (b) Shi, H.; Kwok, R. T. K.;
Liu, J.; Xing, B.; Tang, B. Z.; Liu, B. J. Am. Chem. Soc. 2012, 134,
17972−17981.
(23) Liu, Y.; Yu, Y.; Lam, J. W. Y.; Hong, Y.; Faisal, M.; Yuan, W. Z.;
Tang, B. Z. Chem.Eur. J. 2010, 16, 8433−8438.
(24) Christian, W.; Jonhson, T. S.; Gill, T. J. J. Biomed. Sci. Eng. 2008,
1, 163−169.
(25) Raymond, L.; Morin, F. G.; Marchessault, R. H. Carbohydr. Res.
1993, 246, 331−336.
(26) Kolb, H. C.; Finn, M. G.; Sharpless, K. B. Angew. Chem., Int. Ed.
2001, 40, 2004−2021.
H
dx.doi.org/10.1021/ja312581r | J. Am. Chem. Soc. XXXX, XXX, XXX−XXX