Supramolecular Assembly of TPE-Based Glycoclusters with Dicyanomethylene-4H-pyran (DM) Fluorescent Probes Improve Their Properties for Peroxynitrite Sensing and Cell Imaging

Article abstract of DOI:10.1002/chem.202002772

Two red-emitting dicyanomethylene-4H-pyran (DM) based fluorescent probes were designed and used for peroxynitrite (ONOO^?) detection. Nevertheless, the aggregation-caused quenching effect diminished the fluorescence and restricted their further

Full text of DOI:10.1002/chem.202002772

Accepted Article
Accepted Article
Title: Supramolecular assembly of TPE-based glycoclusters with
dicyanomethylene-4H-pyran (DM) fluorescent probes improve
their properties for peroxynitrite sensing and cell imaging
Authors: Lei Dong, Mengqi Fu, Lifang Liu, Hai-Hao Han, Yi Zang, Guo-
Rong Chen, Jia Li, Xiao-Peng He, and Sébastien Vidal
This manuscript has been accepted after peer review and appears as an
Accepted Article online prior to editing, proofing, and formal publication
of the final Version of Record (VoR). This work is currently citable by
using the Digital Object Identifier (DOI) given below. The VoR will be
published online in Early View as soon as possible and may be different
to this Accepted Article as a result of editing. Readers should obtain
the VoR from the journal website shown below when it is published
to ensure accuracy of information. The authors are responsible for the
content of this Accepted Article.
To be cited as: Chem. Eur. J. 10.1002/chem.202002772
Link to VoR: https://doi.org/10.1002/chem.202002772
0
1/2020

Chemistry - A European Journal
10.1002/chem.202002772
FULL PAPER
Supramolecular assembly of TPE-based glycoclusters with
dicyanomethylene-4H-pyran (DM) fluorescent probes improve
their properties for peroxynitrite sensing and cell imaging
Lei Dong[a,b], Mengqi Fu , Lifang Liu , Hai-Hao Han , Yi Zang , Guo-Rong Chen , Jia Li* , Xiao-
[a]
[a]
[a]
[c]
[a]
[c]
[a]
[b]
Peng He* , and Sébastien Vidal*
Abstract: Two red-emitting dicyanomethylene-4H-pyran (DM) based
such as DNA, proteins, thiols and lipids are easily damaged
triggering cancer, septic shock, atherosclerosis and multiple
sclerosis.[2] However, peroxynitrite is difficult to detect in biological
systems due to its short half-life (~10-20 ms).[2] Despite this short
lifetime, the level of intracellular peroxynitrite is a key indicator to
monitor and predict diseases effectively.
-
fluorescent probes were designed and used for peroxynitrite (ONOO )
detection. Nevertheless, the aggregation-caused quenching effect
diminished the fluorescence and restricted their further applications.
To overcome this problem, tetraphenylethylene (TPE) based
glycoclusters were used to self-assemble with these DM probes to
obtain supramolecular water-soluble glyco-dots. This self-assembly
strategy enhanced the fluorescence intensity, leading to an enhanced
selectivity and activity of the resulting glyco-dot comparing to DM
probes alone in PBS buffer. The glyco-dots also exhibited better
Over the past decade, fluorescent probes for peroxynitrite
detection and imaging in cells with different mechanisms were
reported frequently. These probes displayed sensitivity, selectivity,
high spatial and temporal resolution for peroxynitrite detection in
vivo.[3] For the interference from background auto-fluorescence,
the development of long-wavelength fluorescent probes was
required to detect peroxynitrite in tissues. Sikora et al. reported
the reactivity of peroxynitrite with aromatic boronates was better
-
results during fluorescence sensing of intracellular ONOO than the
probes alone, thereby offering scope for the development of other
similar supramolecular glyco-systems for chemical biological studies.
-
than hypochlorous acid (HOCl/ClO ) and hydrogen peroxide
(
H
2
O
2
).[4] Since then, various boronate fluorescent probes were
Introduction
[5][6][7]
reported.
Besides, self-assembled nanostructures in relation
to therapeutic strategies for neurodegenerative disorders were
-
Peroxynitrite (ONOO ) is an important reactive oxidant and
reported.[8]
nitration agent in physiological and pathological processes. It has
been demonstrated to cause neuronal death in vivo such as brain
ischemia, amyotrophic lateral sclerosis, Alzheimer’s disease and
Parkinson’s disease.[1] Peroxynitrite can damage tissues by
releasing hydroxyl radical ( OH) and nitrogen dioxide (NO
In our previous studies, we developed fluorescent glycoprobes
based on organic dyes modified with carbohydrate moieties to
enhance not only their water solubility but also for targeting cancer
cells that over-express specific carbohydrate receptors. These
glycoprobes achieved the sensitive detection and targeted
imaging of ions or biomolecules.[9] Dicyanomethylene-4H-pyran
•
•-). Also,
2
due to its efficient nitration and oxidation properties, biomolecules
(
DM), due to its broad absorption band and relatively long
emission wavelength, was widely used for biosensing and
bioimaging.[10] However, the fluorescence properties of DM-based
[
a]
Dr L Dong, MQ Fu, LF Liu, HH Han, Prof GR Chen and Prof XP He
Key Laboratory for Advanced Materials and Joint International
Research Laboratory of Precision Chemistry and Molecular
Engineering, Feringa Nobel Prize Scientist Joint Research Center,
School of Chemistry and Molecular Engineering, Frontiers Center
for Materiobiology and Dynamic Chemistry, East China University of
Science and Technology
probes were generally influenced by the aggregation-caused
quenching (ACQ) effect from the low water-solubility in PBS buffer
thus restricting the sensitivity and further biological applications.[
Glycoclusters are water-soluble and typically display high affinity
for cell surface receptors and have been thus used for targeted
cell imaging and drug delivery.[ Recently, through the self-
assembly between perylenediimide (PDI)-based galactoclusters
11]
130 Meilong Rd., Shanghai 200237, PR China
12]
E-mail: xphe@ecust.edu.cn
Dr L Dong, Dr S Vidal
Institut de Chimie et Biochimie Moléculaires et Supramoléculaires,
Laboratoire de Chimie Organique 2-Glycochimie, UMR 5246, CNRS
and Université Claude Bernard Lyon 1, Université de Lyon
[
b]
c]
and
a
DM dye, we have developed
a
water-soluble
supramolecular “glyco-dot” system for the targeted fluorescence
imaging of liver cancer cells that over-express galactose
receptors.[ However, in that study DM was only used as an
imaging agent rather than a chemical probe for sensing
intracellular signalling molecules.
1, Rue Victor Grignard, F-69622 Villeurbanne, France
13]
E-mail: sebastien.vidal@univ-lyon1.fr
Prof Y Zang and Prof J Li
National Center for Drug Screening, State Key Laboratory of Drug
Research, Shanghai Institute of Materia Medica, Chinese Academy
of Sciences
[
Tetraphenylethene (TPE) is a fluorogen with aggregation-
1
89, Guo Shoujing Rd., Shanghai 201203, P. R. China
induced emission properties, and TPE-based glycoclusters have
been developed for lectin sensing.[14] The amphiphilic TPE-based
Email: jli@simm.ac.cn
Supporting information for this article is given via a link at the end of
the document.
glycoclusters in PBS buffer, might well self-assemble with the
hydrophobic DM probes. The corresponding glyco-dots with
This article is protected by copyright. All rights reserved.

Chemistry - A European Journal
10.1002/chem.202002772
FULL PAPER
satisfying water-solubility were monitored for their response
peroxynitrite in PBS buffer. Therefore, we have designed two DM-
boronate fluorescent probes (DMB1 and DMB2) for peroxynitrite
detection. Their molecular design with two different moieties
between the DM core and the phenylboronic acid will be used to
determine the influence on the fluorescence properties. The
resulting probes could release strong fluorescence emission upon
exposure to peroxynitrite in PBS buffer mixed with 20% MeCN,
but exhibited low fluorescence changes due to the ACQ effect in
pure PBS buffer. To overcome this problem, TPE-based
glycoclusters bearing two or four carbohydrate residues (TPE-
For their excellent fluorescence properties, we have introduced
boronic moieties on the DM fluorogen for efficient peroxynitrite
detection. We conjugated different phenylboronates moieties with
the known 4-(dicyanometheylene)-2-(tert-butyl)-6-methyl-4H-
pyran 5[ (Scheme 2). DMB1 was obtained by condensation of
4-formylphenylboronic acid pinacol ester with compound 5 in the
presence of catalytic piperidine. The electron-poor boronate ester
could block the ICT effect to quench fluorescence. DM-OH,
having strong fluorescence emission, was synthesized similarly
15]
from compound
5
and 4-hydroxybenzaldehyde. Further
4-bromomethylbenzeneboronic acid
functionalization with
a
Gly
2
and TPE-Gly
4
) were designed and synthesized through
pinacol ester provided probe DMB2. The intramolecular D-π-A
system was influenced by the PET effect and thereby quenched
the fluorescence.
Meanwhile, several TPE-based glycoclusters were synthesized.
Dimethoxy-TPE[16] 2 was demethylated then propargylated and
conjugated with azido-functionalized glycosides (β-galactose, β-
glucose, α-mannose and α-fucose) through azide-alkyne click
CuAAC cycloaddition to improve the water-solubility of the DM
probes. The DM-probes could readily self-assemble in the
hydrophobic core of the TPE-based glycoclusters to obtain the
supramolecular
glyco-dots
system
with
fluorescence
enhancement or FRET phenomena (Scheme 1). The water-
soluble glyco-dots were therefore investigated for their response
to peroxynitrite in terms of sensitivity, efficiency and selectivity in
comparison to the DM probes alone in pure PBS buffer. These
glyco-dots were also studied for their imaging properties of
exogenous and endogenous peroxynitrite in living cells.
chemistry (Scheme 2). After deprotection under MeOH/H
solution, divalent TPE-based glycoclusters (TPE-Gly
obtained, while the tetravalent TPE-based glycoclusters (TPE-
Gly
) were previously reported by ourselves.[14]
2
O/Et
3
N
2
) were
4
Scheme 1. Self-assembly process of glyco-dots for peroxynitrite detection and
structures of DMB1, DMB2, DM-OH and TPE-based glycoclusters.
Results and Discussion
Scheme 2. Synthesis of DM-based fluorescent probes (DMB1, DMB2 and DM-
OH) and the divalent TPE-based glycoclusters (TPE-Gly ). See Supporting
Information for details.
2
Synthesis of DM-based fluorescent probes.
This article is protected by copyright. All rights reserved.

Chemistry - A European Journal
10.1002/chem.202002772
FULL PAPER
Fluorescence properties of the DM probes.
use our previously developed TPE-based glycoclusters (TPE-
We initially investigated the UV-Vis and fluorescence properties
of DMB1 and DMB2 in PBS buffer mixed with 20% MeCN to
improve the solubility (Figure S1). The two probes had wide
absorption areas (300-450 nm) originally without the addition of
peroxynitrite. Compared with the low fluorescence intensity of
DMB2 (Ф < 0.01) influenced by PET effect, the fluorescence of
DMB1 (Ф < 0.01) was quenched completely by the electron-poor
boronate blocking the ICT effect. In the presence of peroxynitrite
2 4
Gly and TPE-Gly ) to self-assemble with the DM probes in order
to enhance their water dispersibility.
Glyco-dots self-assembly from DM-based probes and TPE-
based glycoclusters.
To improve the water-solubility of the DM fluorescent dye, we
have investigated its self-assembly with TPE-glycoclusters in
PBS buffer. The hydrophobic cavity of TPE-glycoclusters may
accommodate DM probes inside to avoid the ACQ restriction. We
employed DLS and TEM to measure the size and morphology of
the resulting glyco-dots. Next, the fluorescence variation of glyco-
dots were measured at 365 nm or 470 nm excitation under
different concentration conditions.
(
100 μM), two novel absorption peaks were observed at 320 nm
and 470 nm, along with a colorimetric response from pale yellow
to orange (Figure S2). Meanwhile, an obvious fluorescence
enhancement was observed at 625 nm, which displayed a large
Stokes shift (145 nm). The fluorescence intensities of the
solutions containing DMB1 (Ф = 0.11) and DMB2 (Ф = 0.10)
increased after addition of peroxynitrite based on the
fluorescence properties of DM-OH (Table S1).
Dynamic light scattering (DLS) and transmission electron
microscope (TEM) were used to measure the size distribution and
morphology, respectively, of DMB1, DMB2, TPE-Gal
and the corresponding glyco-dots. We observed that the size of
TPE-Gal (~60 nm) was larger than that of TPE-Gal (~20 nm) by
2 4
, TPE-Gal ,
Plausible sensing mechanisms of the two DM probes for
peroxynitrite can be proposed. For probe DMB1, the electron-
poor boron atom was electrophilic enough to allow condensation
with the anionic and nucleophilic oxygen atom on peroxynitrite.
Rearrangement to the borate with loss of nitrite, cleavage of the
C-B bond and creation of the C-O bond, and further hydrolysis
afforded the desired DM-OH with strong fluorescence emission,
followed by the ICT effect recovery (Scheme S1A). The response
process of DMB2 would include one additional rearrangement
step in comparison to DMB1, which is the spontaneous
elimination of quinone methide to afford the desired phenol DM-
OH (Scheme S1B). This two-step response mechanism would
probably influence the chemical reactivity and fluorescence
intensity of DMB2 after responding to peroxynitrite in vitro or in
vivo. Mass spectrometry (HRMS) and chromatography (HPLC)
analyses further confirmed the response mechanism for both DM
probes in PBS buffer with 20% MeCN (Figure 1). Both probes
incubated with peroxynitrite yielded the same new compound
4
2
DLS (Figure 2). This suggests that the tetra-substitution pattern
of TPE represents a larger molecular volume than the di-
substituted system in their aggregated status. The diameter of
DMB2 (~90 nm) was determined to be larger than that of DMB1
(~20 nm), which illustrated that DMB2 had a lower water-solubility
leading to aggregation in PBS solution. After incubating DM
probes with TPE-glycoclusters, the size distribution of the
supramolecular glyco-dots was narrowed comparing to the DM
probes alone, suggesting the complete self-assembly between
the two partners (Figure 2).
(
DM-OH, m/z = 339.1, Figure S3). Also, the HPLC analysis of the
starting materials and incubated probes clearly indicated the
transformation of both DMB1 (14.1 min) and DMB2 (15.3 min) into
the fluorescent DM-OH (10.5 min).
Figure 2. Aggregate size variation of DM probes, TPE-based glycoclusters and
their glyco-dots after self-assembly obtained by DLS.
Figure 1. HPLC analysis for the response mechanism of (A) DMB1 (10 μM) and
(B) DMB2 (10 μM) in PBS buffer (pH 7.4, 0.01 M) with 20% MeCN.
In their representative TEM images (Figure 3), sheet-like and
filament-like structures were observed for DMB1 and DMB2,
respectively. The TPE-Gal
4
aggregates appeared to be larger
However, the presence of a large amount of MeCN (20%) in
than those of TPE-Gal , which is in agreement with the DLS
2
PBS is toxic to cells, thus limiting the biological application of the
probes. Indeed, we determined that DMB1 and DMB2 when
reacted with peroxynitrite also suffered from ACQ effect in full
PBS buffer (Figure S4). To overcome this problem, we sought to
results. Interestingly, after self-assembly, structurally uniform
glyco-dots formed between the TPE glycoclusters and DM probes
(
Figure 6). The DLS and TEM data well corroborated that the DM
This article is protected by copyright. All rights reserved.

Chemistry - A European Journal
10.1002/chem.202002772
FULL PAPER
probes could self-assemble into the hydrophobic core of TPE-
glycoclusters to construct the well-dispersible supramolecular
glyco-dots in aqueous solution.
Figure 4. Fluorescence changes of TPE-Gal
PBS solution (pH 7.4, 0.01 M, λex = 320 nm) at 25°C with addition of DMB1 (0-
2 4
(20 µM) or TPE-Gal (20 µM) in
Figure 3. Aggregation of DM probes, TPE-based glycoclusters and their glyco-
dots after self-assembly obtained by TEM.
30 µM) and DMB2 (0-80 µM).
Subsequently, DM probes (10 µM) and TPE-glycoclusters (150
µM) were incubated in PBS buffer (pH 7.4, 0.01 M) leading to
supramolecular glyco-dots. The fluorescence of TPE (λem = 490
nm) was gradually decreased with an increasing concentration of
DM probes (λex = 470 nm) as excited at 320 nm (Figure 4). This
implies a Förster resonance energy transfer (FRET) between the
two closely associated fluorophores. Meanwhile, the fluorescence
intensity of DMB2 (acceptor) was increased (λem = 625 nm) with
TPE-glycoclusters (donor), which agrees with a typical FRET
Then, we measured the color and fluorescence change of
glyco-dots upon addition of peroxynitrite in PBS buffer. We
observed a clear change in solution color of the self-assembled
TPE-glycodots (Figure S2) after being exposed to peroxynitrite in
PBS buffer. Besides, we measured the fluorescence ratiometric
change of the glyco-dots in the presence of peroxynitrite in pure
PBS buffer. Similarly, we observed a gradually quenched TPE
fluorescence upon titration with both DM probes. However, unlike
DMB1 alone, the reaction product of DMB1 and peroxynitrite (i.e.
DM-OH) exhibited a fluorescence enhancement, indicative of a
FRET process (ФET = 89%) from TPE to DM-OH (Figure 5).
Interestingly, the fluorescence intensity at 625 nm of DM-OH with
2 4
process (ФET = 68% from TPE-Gal , ФET = 75% from TPE-Gal ).
Nevertheless, the fluorescence intensity of DMB1 remained
unchanged (Figure 4). The PET effect from the arylboronate
moiety might be influenced by the assembly with TPE-
glycocluster, leading to a partial recovery of the fluorescence
emission of DMB2. On the contrary, while the ICT effect from the
boronate moiety still quenches its fluorescence, DMB1 has also
proven to accept the energy but not released through irradiative
process, thus only quenching the fluorescence of TPE-
glycoclusters.
TPE-Gal
2 4
was much stronger than that with TPE-Gal (ФET =
5
9%) , which suggests that, in the glyco-dot system, the distance
between the di-substituted TPE and DM-OH was closer for
achieving a higher FRET efficiency than the tetra-substituted
counterpart. A similar phenomenon was observed for the DMB2-
based glyco-dots. The single fluorescence peak and intense
emission at 625 nm indicated of high energy transfer (ФET = 90%)
bewteen TPE-Gal
However, the characteristic double emission peaks of DMB2 at
90 nm overlapped with the emission of DM-OH at 625 nm in
TPE-Gal participation (Figure 5), thereby was similar to the
results without peroxynitrite (Figure 4). The self-assembly affinity
of TPE-Gal with DM probes in the presence of peroxynitrite is still
unknown and worthy of investigation.
2
and DM-OH in the presence of peroxynitrite.
5
4
4
This article is protected by copyright. All rights reserved.

Chemistry - A European Journal
10.1002/chem.202002772
FULL PAPER
Figure 6. Fluorescence changes of DMB1 (10 µM) and DMB2 (10 µM) pre-
mixed with peroxynitrite (100 µM) in PBS buffer (pH 7.4, 0.01 M, λex = 470 nm)
at 25°C, followed by addition of TPE-Gal (0-200 µM) and TPE-Gal (0-200 µM).
2 4
Fluorescence properties of the glyco-dots.
To evaluate the response of glyco-dots to ONOO , some
fundamental parameters were measured in full PBS buffer
-
(
without additional 20% of MeCN). We firstly analyzed the time-
dependent fluorescence changes of glyco-dots in presence of
peroxynitrite (150 μM). Both probes exhibited a fluorescence
“
turn-on” response to peroxynitrite in PBS buffer (Figure 7A,
Figure S6). An excess of peroxynitrite was used to ensure that all
probe molecules react producing the fluorescent species DM-OH.
An obvious fluorescence enhancement was observed for the
probes in the presence of peroxynitrite and this change was time
dependent. DMB1-based glyco-dots reacted faster than the
DMB2-based one. This could be due to the assumption that
DMB1 produces DM-OH by a single “synthetic” step while DMB2
requires a two-step reaction for the elimination of quinone methide.
2 4
Figure 5. Fluorescence changes of TPE-Gal (20 µM) or TPE-Gal (20 µM) in
PBS solution (pH 7.4, 0.01 M, λex = 320 nm) at 25°C pre-mixed with excess
peroxynitrite (100 µM), followed by addition of DMB1 (0-30 µM) and DMB2 (0-
The fluorescence enhancement capacity of TPE-Gal
4
based
2
based
50 µM).
glyco-dots seemed to be stronger than that of the TPE-Gal
counterparts.
Then, we tested the selectivity of the four glyco-dots for
peroxynitrite in the presence of several potential interfering ROS
and RNS (Figure 7B, Figure S7). All glyco-dots exhibited a
strong emission intensity enhancement at 625 nm upon addition
of peroxynitrite, whereas minimal fluorescence changes were
observed in the presence of other ROS and RNS. The association
of the DM probes with TPE-glycoclusters did not compromise the
selectivity of the sensing system. We also measured the
fluorescence intensity changes of the glyco-dots with increasing
peroxynitrite in PBS (Figure 7C, Figure S8). All glyco-dots
exhibited an obvious fluorescence enhancement with addition of
peroxynitrite (0-80 μM) in pure PBS buffer. Among them, the
glyco-dots composed of DMB1 showed a higher fluorescence
increase than those composed of DMB2, which could be similarly
ascribed to the single “synthetic” step process of the former. TPE-
To further examine that TPE-glycocluster could assemble with
DM-OH in order to block its ACQ effect, TPE-Gal or TPE-Gal
2 4
was added into the PBS solution of DMB1 or DMB2 with
excessive peroxynitrite. Using the DM excitation of 470 nm, we
observed a gradually enhanced fluorescence of both DM probes
in pure PBS solution (Figure 6). After self-assembly with TPE-
Gal
2 4
or TPE-Gal , DMB1 respectively displayed 11.5-fold (Ф 0.02-
0
.23) or 15.5-fold (Ф 0.02 to 0.31) fluoresence increase in the
presence of peroxynitrite. Slightly weaker fluorescence signal
were increased by DMB2-based glyco-dots self-assembled with
TPE-Gal
.24). Meanwhile, other TPE-based glycoclusters (Glc, Man, Fuc)
could increase the fluorescence intensity of DM probes after self-
assemby, similar to TPE-Gal or TPE-Gal (Figure S5). These
2 4
(10.5-fold, Ф 0.02-0.21) or TPE-Gal (12-fold, Ф 0.02 to
0
2
4
results clearly suggest that the self-assembly with TPE-
glycoclusters largely enhance the fluorescence intensity of the
relatively hydrophobic DM probes for peroxynitrite in pure PBS
buffer.
Gal
4
better improved the fluorescence properties of the glyco-dots
system, thus suggesting its ability to
comparing to the TPE-Gal
2
better overcome the ACQ problem of DM. Sigmoidal titration
curves were obtained for the glyco-dots with increasing
peroxynitrite; however DM probes alone showed minimal
fluorescent response to the highest concentration of peroxynitrite
in PBS buffer.
This article is protected by copyright. All rights reserved.

Chemistry - A European Journal
10.1002/chem.202002772
FULL PAPER
Figure 8. Fluorescence imaging and quantification of HepG2 cells incubated
with glyco-dots self-assembled from DM probes (10 µM, yellow) and TPE-based
glycoclusters (150 µM) without (-) or with (+) a subsequent addition of SIN-1
(150 µM, a peroxynitrite inducer). Excitation and emission wavelengths for
glyco-dots are 470-490 nm and 580-650 nm, respectively. The cell nuclei were
stained by Hoechst 33342 (blue). Error bars represent the s.d. *p < 0.05, **p <
0.01, ***p < 0.001, ****p < 0.0001. n = 3.
Figure 7. (A) Time-dependent fluorescence enhancement of these glyco-dots
for ONOO response. (B) Selectivity of two glyco-dots self-assembled from DM
-
To test whether the enhanced fluorescence imaging effect
probes and TPE-Gal
emission enhancements of two glyco-dots self-assembled with DM probes and
TPE-Gal with addition of peroxynitrite concentration (0-80 µM). The glyco-dots
were self-assembled from DM probes (10 µM) and TPE-based glycoclusters
4
for several ROS, RNS and peroxynitrite. (C) Fluorescence
would be the result of galactose-ASGPr interaction facilitating
active endocytosis of the material, the glyco-dots formed between
DMB1 and TPE-based glycoclusters modified with other
carbohydrate epitopes (Glc, Man and Fuc) were used. To our
surprise, all these glycoclusters enhanced the fluorescence
imaging effect of DMB1 in HepG2 cells (Figure S9). This
observation differs from our previous study using PDI-based
glycoclusters for targeted delivery of a DM-based imaging agent
to cancer cells that highly express specific carbohydrate
receptors.[ This previous study indicated that PDI-based glyco-
4
(150 µM), and then detect the ROS, RNS or peroxynitrite (150 µM) in PBS buffer
(pH 7.4, 0.01 M, λex = 470 nm) at 25°C.
Glyco-dots imaging of peroxynitrite in living cells.
Human hepatoma cell line (HepG2) highly expresses the
asialoglycoprotein receptor (ASPGr) which binds preferentially to
galactosides. Hence, the glyco-dots self-assembly obtained from
DM probes and galactose-based glycoclusters were incubated
with HepG2 cells and employed to detect the exogenous and
endogenous ONOO-.
We first determined that DMB1 showed a two-fold higher
fluorescence intensity in cells treated with SIN-1 (a peroxynitrite
inducer) than DMB2 alone (Figure 8). In contrast, the
fluorescence intensity of the glyco-dots produced in the cells was
observed to be much stronger than that of the probes alone. We
13]
dots could target selectively HepG2 cells based on
a
carbohydrate-receptor interaction at the cell surface. We
preliminarily hypothesized that, unlike PDI-based glyco-dots[13]
that predominantly facilitate multivalent carbohydrate-receptor
interactions, the TPE-based glyco-dots might adopt a different
mode of interaction with the cell membrane. The good water-
solubility of TPE-based glyco-dots may improve the dispersion of
DM probes in aqueous solution that is beneficial for cell
endocytosis and (b) TPE-based glyco-dots may increase the
fluidity of the cell membrane, thus improving the penetration of the
DM probes into cells. However, more detailed biophysical assays
are needed to confirm this hypothesis, which is not the main focus
of the present study that develops a sensitivity-enhanced system
by means of the simple supramolecular self-assembly between
fluorogenic probes and glycoclusters.
observed that 1) TPE-Gal
enhancement for both DM probes than TPE-Gal
4
led to
a
better fluorescence
, and that 2)
2
DMB1-based glyco-dots performed better than DMB2-based
glyco-dots, similar to the observation in the solution-state
fluorescence studies.
To explore where the probes located in cells, the colocalization
experiments was performed to confirm the subcellular distribution
of glyco-dots (DMB1+TPE-Gal ) in HepG2 cells (Figure 9). The
4
fluorescence of DMB1 almost completely overlapped with the
signal from a commercial MitoTracker Red for mitochondrial
target. However, the fluorescence of DMB1 did not coincide with
the fluorescence of LysoTracker, a commercial lysosome marker.
This article is protected by copyright. All rights reserved.

Chemistry - A European Journal
10.1002/chem.202002772
FULL PAPER
The fluorescence intensity of the linear region of interest (ROIs)
of two channels across a selected cell altered close synchrony,
and the Pearson’s colocalization coefficient were determined to
be 0.92 for mitochondria and 0.45 for lysosome, respectively.
Therefore, the glyco-dots have ability to accumulate and image in
-
mitochondria for ONOO generation and functions.
Figure 10. Fluorescence imaging and quantification of RAW 264.7 incubated
with glyco-dots self-assembled from DM probes (10 µM, yellow) and TPE-based
glycoclusters (150 µM) without or with
a
subsequent addition of
lipopolysaccharide (LPS, 1 mg/mL) or SIN-1 (150 µM). Excitation and emission
wavelengths for glyco-dots are 470-490 nm and 580-650 nm, respectively. The
cell nuclei were stained by Hoechst 33342 (blue). Error bars represent the s.d.
*
p < 0.05, **p < 0.01, ***p < 0.001. n = 3
Conclusion
Two red-emitting (625 nm) phenylboronate fluorescent probes
based on dicyanomethylene-4H-pyran (DMB1 and DMB2) have
been designed and synthesized. Their sensing mechanism for
peroxynitrite was studied by MS and HPLC. However, an
insufficient fluorescence response was observed after exposure
to peroxynitrite in full PBS buffer, which we ascribed to the ACQ
effect. The poor water solubility and ACQ effect observed
prompted our attempt of the supramolecular self-assembly
between water soluble TPE-based glycoclusters and DM probes.
The resulting glyco-dots displayed a significantly improved
sensitivity (~60-fold) for peroxynitrite in pure PBS buffer. The
glyco-dots also exhibited a better fluorescence imaging effect for
intracellular peroxynitrite (exogenously added and LPS-
stimulated) than the probes alone, thereby offering scope for the
development of other similar self-assembling systems for
chemical biology studies.
Figure 9. Fluorescence co-localization of DMB1+TPE-Gal
MitoTracker Red (a mitochondria tracker, 500 nM) and (B) LysoTracker Red (a
lysosome tracker, 500 nM) in HepG2 cells. (C) Fluorescence quantification of
selected cellular regions (white solid lines in panel A) of DMB1+TPE-Gal
MitoTracker Red (500 nM) in HepG2 cells. (D) Fluorescence quantification of
selected cellular regions (white solid lines in panel B) of DMB1+TPE-Gal and
LysoTracker Deep Red (500 nM) in HepG2 cells. SIN-1 (500 µM) was then
added to activate fluorescent. Excitation wavelengths for DMB1+TPE-Gal and
subcellular organelle dye (LysoTracker Red or MitoTracker Red) are 488 nm
and 559 nm, respectively. Emission wavelengths for DMB1+TPE-Gal and
4
with (A)
4
and
4
4
4
subcellular organelle dye (LysoTracker Red or MitoTracker Red) are 580–650
nm and 570-590 nm, respectively.
Experimental Section
We also tested the response of the glyco-dots for endogeneous
peroxynitrite produced by lipopolysaccharide (LPS) in another
macrophage cell line (RAW264.7) (Figure 10). Similarly, we
observed an enhanced fluorescence intensity for the glyco-dots
after reacting with endogenous peroxynitrite. This suggests that
our self-assembly strategy was generally applicable to enhancing
the imaging capacity of otherwise cell impermeable fluorogenic
probes in different cells. The glyco-dots were also shown to be
not toxic to HepG2 cells with concentrations higher than that use
for fluorescence imaging (Figure S10).
Synthesis of probes. Detailed descriptions of the syntheses can be found
in the Supporting information.
UV-vis absorbance spectroscopy. The UV-vis absorbance
measurements were carried out at room temperature using a Varian Cary
60 UV-vis spectrophotometer. All spectra were corrected for background
intensities by subtracting the spectra of pure solvent measured under
identical conditions. UV-vis-NIR absorption spectra (0-800 nm) of probes.
Fluorescence spectroscopy. The fluorescence measurements were
carried out at room temperature using an Agilent Cary Eclipse
fluorescence spectrophotometer.
Fluorescence quantum yields. DMB1 or DMB2 (with
a final
concentration of 30 μM) was dissolved in aqueous 0.01 M PBS (pH 7.4).
DMB1 and DMB2 (with a final concentration of 10 μM) with addition of
peroxynitrite (with a final concentration of 100 μM) dissolved in aqueous
0.01 M PBS (pH 7.4). Rhodamine B (with a final concentration of 5 μM)
This article is protected by copyright. All rights reserved.

Chemistry - A European Journal
10.1002/chem.202002772
FULL PAPER
was dissolved in aqueous 0.01 M PBS (pH 7.4). The UV-vis absorbance
measurements were carried out at room temperature using a Varian Cary
microplate reader (Molecular Device, USA). The optical density of the
result in MTS assay was directly proportional to the number of viable cells.
60 UV-vis spectrophotometer. UV absorption values were collected at 470
nm. The fluorescence measurements were carried out at room
temperature by an Agilent Cary Eclipse fluorescence spectrophotometer
with an excitation wavelength of 480 nm. Integrated fluorescence curve
between 500 nm and 850 nm. Finally, the fluorescence quantum yield was
calculated according to equation (1).
Acknowledgements
The authors thank the financial support from Université Claude
Bernard Lyon 1 and CNRS, the Natural Science Foundation of
China (nos. 21788102, 91853201, 21722801 and 21776078), the
Shanghai Municipal Science and Technology Major Project (No.
2
퐹ꢀ 퐴ꢁ 푛ꢀ
푌 = 푌 ·
·
·
2
（1）
푢
푟
퐹ꢁ 퐴ꢀ 푛ꢁ
u = probe，r = rhodamine B, Y = fluorescence quantum yield, F =
integrated fluorescence intensity, A = absorbance, n = refractive index.
2
018SHZDZX03), the National Key Sci-Tech Special Projects of
Infection Diseases of China (2018ZX10732202), and the
international cooperation program of Shanghai Science and
Technology Committee (17520750100). Lei Dong is grateful to
the China Scholarship Council for a PhD stipend (201606740066).
Supramolecular self-assembly of glyco-dots. DM probe (1 mM, DMSO)
was added to a solution of TPE-based glycoclusters (15 mM, PBS buffer).
The resulting mixture was stirred during 30 min to produce the
supramolecular glyco-dots for subsequent experiments.
Keywords: glycocluster • Self-assembly • Peroxynitrite detection
• tetraphenylethylene • Cell imaging • dicyanomethylene-4H-
pyran
Preparation of ROS/RNS. Details descriptions of the ROS and RNS
preparation can be found in the Supporting information.
Transmission electron microscope (TEM). A droplet of probes (DMB1
and DMB2 = 10 μM, TPE-Gal
2
1
Tridiem energy filter operating at 200 kV was used for TEM, and data were
processed using Image J software.
2
and TPE-Gal
00 mesh holey carbon copper grids for TEM characterizations. JEOL
400 equipped with a Gatan Orius charged-coupled device camera and
4
= 150 μM) was dropped onto
References
[
1] a) K.-T. Lin, J.-Y. Xue, M. C. Lin, E. G. Spokas, F. F. Sun, P. Y.-K. Wong,
Am. J. Physiol. Cell Physiol. 1998, 274, C855-C860; b) K. B. Li, L. Dong,
S. Zhang, W. Shi, W. P. Jia, D. M. Han, Talanta 2017, 165, 593-597.
2] P. Pacher, J. S. Beckman, L. Liaudet, Physiol. Rev. 2007, 87, 315-424.
3] a) J. Zhang, X. Zhen, J. Zeng, K. Pu, Anal. Chem. 2018, 90, 9301-9307; b)
X. Xie, F. Tang, G. Liu, Y. Li, X. Su, X. Jiao, X. Wang, B. Tang, Anal. Chem.
[
[
Dynamic light scattering. Dynamic light scattering carried out using a
Horiba LB-550 Dynamic Light Scattering Nano-Analyzer.
2
018, 90, 11629-11635; c) Y. Wu, A. Shi, Y. Li, H. Zeng, X. Chen, J. Wu,
Statistical analysis. Error bars represent s.d. Data analysis was
performed with GraphPad PRISM (GraphPad Software, Inc.) using two-
tailed unpaired Student’s t-test. P-values <0.05 were considered
statistically significant. All experiments were repeated at least three times
with representative data shown.
X. Fan, Analyst 2018, 143, 5512-5519; d) L. Li, M. Zan, X. Qie, J. Yue, P.
Miao, M. Ge, Z. Chang, Z. Wang, F. Q. Bai, H. X. Zhang, J. K. Ferri, W. F.
Dong, J. Phys. Chem. A 2018, 122, 217-223; e) Y. Li, X. Xie, X. Yang, M.
Li, X. Jiao, Y. Sun, X. Wang, B. Tang, Chem. Sci. 2017, 8, 4006-4011; f) K.
Debowska, D. Debski, B. Michalowski, A. Dybala-Defratyka, T. Wojcik, R.
Michalski, M. Jakubowska, A. Selmi, R. Smulik, L. Piotrowski, J. Adamus,
A. Marcinek, S. Chlopicki, A. Sikora, Chem. Res. Toxicol. 2016, 29, 735-
Cell culture. HepG2 and RAW264.7 cells were maintained in a Dulbecco’s
Modified Eagle’s Medium (Invitrogen, Carlsbad, CA, USA) supplemented
with 10% fetal bovine serum (Gibco, Gland Island, NY, USA) in a
746.
[
[
4] A. Sikora, J. Zielonka, M. Lopez, J. Joseph, B. Kalyanaraman, Free Radic.
Biol. Med. 2009, 47, 1401-1407.
humidified atmosphere of 5% CO
cells reached 90% confluency.
2
and 95% air at 37°C and split when the
5] a) M. Weber, A. B. Mackenzie, S. D. Bull, T. D. James, Anal. Chem. 2018,
90, 10621-10627; b) J. Zhou, Y. Li, J. Shen, Q. Li, R. Wang, Y. Xu, X. Qian,
Fluorescence imaging of cells. Cells were seeded on a black 96-well
microplate with optically clear bottom (Greiner bio-one, Germany)
overnight. To produce peroxynitrite exogenously, the cells were first
incubated with DM probe / TPE-glycocluster (10 μM / 150 μM, 1% DMSO
in PBS, pH 7.4) for 30 min, followed by incubation with SIN-1 (150 μM) for
RSC Adv. 2014, 4, 51589-51592; c) J. Kim, J. Park, H. Lee, Y. Choi, Y. Kim,
Chem. Commun. 2014, 50, 9353-9356; d) F. Yu, P. Song, P. Li, B. Wang,
K. Han, Analyst 2012, 137, 3740-3749.
[6] A. C. Sedgwick, H. H. Han, J. E. Gardiner, S. D. Bull, X. P. He, T. D. James,
Chem. Commun.2017, 53, 12822-12825.
3
0 min. The cell nuclei were stained with Hoechst 33342 (5 μg/mL) at 37°C
[7] X. Sun, Q. Xu, G. Kim, S. E. Flower, J. P. Lowe, J. Yoon, J. S. Fossey, X.
Qian, S. D. Bull, T. D. James, Chem. Sci. 2014, 5, 3368.
in a humidified atmosphere of 5% CO in air for 5 min. Then, cells were
2
washed with PBS (phosphate buffered saline) three times. The
fluorescence images were recorded using an Operetta high-content
imaging system (Perkin Elmer, US) and quantified and plotted by
Columbus analysis system (Perkin Elmer, US).
[8] a) A. A. Rajkiewicz, K. Skowerski, B. Trzaskowski, A. Kajetanowicz, K.
Grela, ACS Omega 2019, 4, 1831-1837; b) A. Angelova, M. Drechsler, V.
M. Garamus, B. Angelov, Acs Omega 2018, 3, 3235-3247.
[9] a) X.-P. He, R.-H. Li, S. Maisonneuve, Y. Ruan, G.-R. Chen, J. Xie, Chem.
Commun. 2014, 50, 14141-14144; b) K.-B. Li, D. Zhou, X.-P. He, G.-R.
Chen, Dyes Pigm. 2015, 116, 52-57; c) L. Dong, Y. Zang, D. Zhou, X. P.
He, G. R. Chen, T. D. James, J. Li, Chem. Commun. 2015, 51, 11852-
Cell viability assay. Cells were plated overnight on 96-well plates in
growth medium. After seeding, cells were treated with glycodots of
different concentrations for 48 hours. Then, 10 μL per well of MTS/PMS
11855; d) K. B. Li, N. Li, Y. Zang, G. R. Chen, J. Li, T. D. James, X. P. He,
H. Tian, Chem. Sci. 2016, 7, 6325-6329.
(
20:1, Promega Corp) solution was added to each well containing 100 μL
of growth medium. After incubation at 37°C under 5% CO for 2 h, the
absorbance of the solutions was measured at 470/480 nm, using an M5
[
10] a) D. K. Ji, G. R. Chen, X. P. He, H. Tian, Adv. Funct. Mater. 2015, 25,
2
3483-3487; b) D. K. Ji, Y. Zhang, Y. Zang, J. Li, G. R. Chen, X. P. He, H.
Tian, Adv. Mater. 2016, 28, 9356-9363.
This article is protected by copyright. All rights reserved.

Chemistry - A European Journal
10.1002/chem.202002772
FULL PAPER
[
11] a) K. Gu, W. Qiu, Z. Guo, C. Yan, S. Zhu, D. Yao, P. Shi, H. Tian, W. H.
Zhu, Chem. Sci. 2019, 10, 398-405; b) K. Gu, Y. Xu, H. Li, Z. Guo, S. Zhu,
S. Zhu, P. Shi, T. D. James, H. Tian, W. H. Zhu, J. Am. Chem. Soc. 2016,
[14] M. Donnier-Marechal, S. Abdullayev, M. Bauduin, Y. Pascal, M. Q. Fu, X.
P. He, E. Gillon, A. Imberty, E. Kipnis, R. Dessein, S. Vidal, Org. Biomol.
Chem. 2018, 16, 8804-8809.
1
38, 5334-5340.
[15] G. Zhao, Y. Zhu, S. Guang, F. Ke, H. Xu, New J. Chem. 2018, 42, 555-563.
[16] H.-T. Feng, J.-B. Xiong, Y.-S. Zheng, B. Pan, C. Zhang, L. Wang, Y. Xie,
Chem. Mater. 2015, 27, 7812-7819.
[
[
12] M. Donnier-Marechal, N. Galanos, T. Grandjean, Y. Pascal, D. K. Ji, L.
Dong, E. Gillon, X. P. He, A. Imberty, E. Kipnis, R. Dessein, S. Vidal, Org.
Biomol. Chem. 2017, 15, 10037-10043.
13] Y. Liu, D. K. Ji, L. Dong, N. Galanos, Y. Zang, J. Li, S. Vidal, X. P. He,
Chem. Commun. 2017, 53, 11937-11940.
This article is protected by copyright. All rights reserved.

Chemistry - A European Journal
10.1002/chem.202002772
FULL PAPER
Entry for the Table of Contents
FULL PAPER
Self-assembly of TPE-based
glycoclusters with fluorogenic probes
allowed for the detection of
Lei Dong, Mengqi Fu, Lifang Liu, Haihao
Han, Yi Zang, Guo-Rong Chen, Jia Li*,
Xiao-Peng He* and Sébastien Vidal*
peroxynitrite in vitro and in cell assays
through the recovery of proper water
solubility and fluorescence emission.
Page 1 – Page 8
Supramolecular assembly of TPE-
based glycoclusters with
dicyanomethylene-4H-pyran (DM)
fluorescent probes improve their
properties for peroxynitrite sensing
and cell imaging
This article is protected by copyright. All rights reserved.