Tetraphenylethylene-based glycoclusters with aggregation-induced emission (AIE) properties as high-affinity ligands of bacterial lectins

Article abstract of DOI:10.1039/c8ob02035c

Tetraphenylethylene (TPE) is fluorescent through aggregation induced emission (AIE) in water. Herein, TPE was used as the core of glycoclusters that target the bacterial lectins LecA and LecB of Pseudomonas aeruginosa. Synthesis of these TPE-based glycocl

Full text of DOI:10.1039/c8ob02035c

Organic &
Biomolecular Chemistry
View Article Online
View Journal | View Issue
PAPER
Tetraphenylethylene-based glycoclusters with
aggregation-induced emission (AIE) properties as
high-affinity ligands of bacterial lectins†
Cite this: Org. Biomol. Chem., 2018,
16, 8804
Marion Donnier-Maréchal,a Shuay Abdullayev,a Marvin Bauduin,b Yoann Pascal,a
Meng-Qi Fu,c Xiao-Peng He,c Emilie Gillon,d Anne Imberty, d Eric Kipnis,b,e
a
Rodrigue Desseinb,f and Sébastien Vidal
*
Tetraphenylethylene (TPE) is fluorescent through aggregation induced emission (AIE) in water. Herein,
TPE was used as the core of glycoclusters that target the bacterial lectins LecA and LecB of
Pseudomonas aeruginosa. Synthesis of these TPE-based glycoclusters was accomplished by using
azide–alkyne “click” chemistry. The AIE properties of the resulting glycoclusters could be readily verified,
but imaging could not be pursued due to the overlap of the fluorescence signals from cells and bacteria.
Nonetheless, the glycoclusters displayed nanomolar affinities toward LecA and LecB. Further evaluation in
a cell-based anti-adhesive assay highlighted a limited decrease in adhesion (20%) for the fucosylated
glycocluster. This confirmed that these TPE-based glycoclusters are indeed LecA and LecB high-affinity
ligands. Nevertheless, the hypotheses involving their application in imaging or anti-adhesive therapy could
not be verified.
Received 20th August 2018,
Accepted 22nd October 2018
DOI: 10.1039/c8ob02035c
rsc.li/obc
ment is true for valency, controlling the exact tridimensional
shape of glycoclusters can sometimes be more difficult.
Introduction
Glycoclusters are composed of a multivalent core conjugated Glycoclusters are designed with a large series of aromatic
with a single copy of an oligosaccharide on each branching cores,1 among several inherently fluorescent scaffolds.2–12 We
point. This perfect control of the molecular structure allows recently applied perylenediimide (PDI) in an anti-adhesive
for an exact valency and provides a single and homogeneous strategy13 against Pseudomonas aeruginosa (PA) and evaluated
molecule, in contrast to glycopolymers or glyconanoparticles, successful PDI-based glyco-dots in the context of cancer cell
for which the valency is not fully controlled. While this state- detection.14 As a continuation of these investigations, we
report here tetraphenylethylene(TPE)-based glycoclusters as
multivalent lectin ligands. We also evaluated their anti-
adhesive properties against PA. The aggregation-induced emis-
aInstitut de Chimie et Biochimie Moléculaires et Supramoléculaires, Laboratoire de
sion (AIE) properties of TPE were investigated to capitalize on
Chimie Organique 2 – Glycochimie UMR 5246, CNRS - Université Claude Bernard
its fluorescence properties toward chemical biology appli-
cations and imaging of cells. Previous reports have used such
TPE-based glycoclusters functionalized with different carbo-
hydrates, such as mannose,15,16 glucosamine,17 and 6′-sialyl-
lactose,18 for the detection of lectins, alkaline phosphatase
and the influenza virus, respectively.
Multivalency is a major strategy used for the design of
lectin ligands with high affinity.19 The host–pathogen inter-
action of PA with the human host cell is mediated by LecA
and LecB. These two soluble lectins bind galactosides and
fucosides, respectively. While a large series of multivalent
ligands of these lectins has been designed,19 very few incor-
Lyon 1, 43 Boulevard du 11 Novembre 1918, F-69622 Villeurbanne, France.
E-mail: sebastien.vidal@univ-lyon1.fr
bUniversité de Lille, CHU Lille, Recherche translationnelle relations hôte-pathogènes
(EA 7366), France
cKey Laboratory for Advanced Materials & Institute of Fine Chemicals & Feringa
Nobel Prize Scientists Joint Research Center, School of Chemistry and Molecular
Engineering, East China University of Science and Technology, Shanghai 200237,
China
dUniv. Grenoble Alpes, CNRS, CERMAV, Centre de Recherche sur les Macromolécules
Végétales (CERMAV), 38000 Grenoble, France
eCHU Lille, Institut de Microbiologie, F-59000 Lille, France
fCHU Lille, Service de Réanimation Chirurgicale, F-59000 Lille, France
†Electronic supplementary information (ESI) available. See DOI: 10.1039/
c8ob02035c
8804 | Org. Biomol. Chem., 2018, 16, 8804–8809
This journal is © The Royal Society of Chemistry 2018

View Article Online
Organic & Biomolecular Chemistry
Paper
porate
a fluorescent core for further chemical biology ITC binding studies
applications.
The affinity of the designed TPE-based glycoclusters was
measured by isothermal titration microcalorimetry (ITC)
towards LecA or LecB with the corresponding galactosylated
or fucosylated glycoclusters (Table 1). The glucosylated
glycocluster 4a was used as a negative control of affinity
toward LecA and LecB. The high affinity of the galactosy-
lated TPE-based glycocluster 4b was evidenced by an appar-
ent dissociation constant (Kd) of 80 nM toward LecA.
Moreover, the relative potency (β = 875) is among the best
reported for such LecA ligands.19 The stoichiometry of the
complex (N = 0.275–1/4) implies that each of the four
galactoside epitopes is involved in a binding interaction
with the lectin. The fucosylated glycocluster 4c displayed a
nanomolar apparent Kd value, so binding properties some-
what similar to the monovalent methyl fucoside used as a
reference with relative potency (β = 5). Nevertheless, the
cluster effect for multivalent LecB ligands is much more
difficult to obtain given the tetrahedral geometry of the
multimeric protein.19 Again, all four fucoside epitopes
could interact simultaneously with a LecB binding site (N
= 0.197–1/4 or even 1/5).
Results and discussion
Synthesis of the TPE-based glycoclusters
The synthesis of the known tetrapropargylated tetraphenyl-
ethylene (TPE) core 1 was achieved through the McMurry reac-
tion of 4,4′-dimethoxybenzophenone,20–22 followed by de-
methylation with boron tribromide to the tetra-phenol,20,23
and then propargylation to the desired core 1.20,23 Four azido-
functionalized carbohydrates 2a–d 24–27 were then conjugated
to the tetra-alkynylated TPE core 1 under Meldal’s con-
ditions28,29 (CuI, iPr2NEt) using microwave activation
(Scheme 1). The resulting acetylated glycoclusters 3a–d were
then deacetylated to afford the desired TPE-based glycoclusters
4a–d in good isolated yields.
Both glycoclusters 4b and 4c displaying nanomolar
affinities for their lectins therefore appear as very promising
candidates for an anti-adhesive strategy against PA.
Photophysical properties
UV-vis spectra of the TPE-based glycoclusters (Fig. 1a) dis-
played an absorbance peak at ca. 340 nm, which is character-
istic of TPE. Upon excitation at 340 nm, while we observed
minimal fluorescence in organic solvents, including DMF,
DMSO, MeOH and EtOH, the glycoclusters exhibited a much
stronger emission in pure water and phosphate buffered saline
(0.05 M, pH 7.4). This is in agreement with their typical AIE
properties,32 because the amphiphilic TPE-based glycoclusters
could form aggregates in water, as was demonstrated by
increasing the water fraction of a DMSO solution of the TPE-
based glycoclusters 4a–d (Fig. 1c and d). The difference in the
fluorescence enhancement of the glycoclusters might be due
Scheme 1 Synthesis of the TPE-based glycoclusters.
Table 1 ITC measurements for the binding of TPE-based glycoclusters 4a–c to LecA or LecB
Ligand
Lectin
Na
−ΔH (kJ mol−1
)
−TΔS (kJ mol−1
)
Kd b (nM)
βc
TPE-(EG3Glc)4 (4a)
TPE-(EG3Gal)4 (4b)
TPE-(EG3Glc)4 (4a)
TPE-(EG3Fuc)4 (4c)
LecA
LecA
LecB
LecB
No binding observed (see ESI Fig. S3)
0.275 0.005 111.0 0.5
No binding observed (see ESI Fig. S4)
0.197 0.001 119
70.5
79
80
84
5
5
875
5
2
a Stoichiometry of binding defined as the number of glycoclusters per monomer of LecB. b The present dissociation constants are presented for
possible aggregates of the TPE-based glycoclusters and therefore must be considered as apparent Kd values. c Relative potency: calculated using
methyl β-D-galactopyranoside (β-GalOMe) as a monovalent reference with Kd = 70 µM for LecA,30 or methyl α-L-fucopyranoside (α-FucOMe) as a
monovalent reference with Kd = 0.43 µM for LecB.31
This journal is © The Royal Society of Chemistry 2018
Org. Biomol. Chem., 2018, 16, 8804–8809 | 8805

View Article Online
Paper
Organic & Biomolecular Chemistry
Fig. 1 Photophysical properties of 4a–c. (a) Stacked UV-vis absorbance spectra of 4a–d (60 μM) in different organic solvents (DMSO, DMF, EtOH,
MeOH), pure water and phosphate buffered saline (0.05 M, pH 7.4). The UV-vis absorbance measurements were carried out at room temperature on
a Varian Cary 60 UV-vis spectrophotometer. (b) Stacked fluorescence spectra of 4a–d (60 μM) in different organic solvents (DMSO, DMF, EtOH,
MeOH), pure water and phosphate buffered saline (0.05 M, pH 7.4) with an excitation wavelength of 340 nm. The fluorescence measurements were
carried out at room temperature on an Agilent Cary Eclipse fluorescence spectrophotometer. (c) Plot of the fluorescence intensity of 4a–d (60 μM)
in DMSO as a function of water fraction ( fw%); the original intensities of the glycoclusters are normalized as 1.0. (d) AIE of 4a–d in DMSO with
increasing water fraction ( fw%) photographed with a camera.
to the structural difference of the carbohydrate epitopes conju- also confirmed the formation of larger aggregates upon
gated on the TPE core.
increasing concentrations with typically 20 nm aggregates at
The galactosylated glycocluster 4b was investigated using 20 µM, 40 nm aggregates at 60 µM, 60 nm aggregates at
dynamic light scattering (DLS) and high-resolution trans- 180 µM and 150 nm aggregates at 540 µM (Fig. 2b). These
mission electron microscopy (HR-TEM) to better understand observations could be further confirmed by HR-TEM images
the aggregation of such TPE-based glycoclusters (Fig. 2). The (Fig. 2c), in which large aggregates can be observed at 180 or
fluorescence intensity of the glycocluster increased drastically 540 µM.
from 10 to 90 µM due to the AIE effect (Fig. 1a). Then fluo-
PDI-based glycoclusters have previously been shown to
rescence quenching occurred above these concentrations provide weak fluorescence in cell adhesion assays, but
(from 120 to 600 µM) due to the aggregation-caused quench- their fluorescence is not strong enough for them to be
ing (ACQ) effect, bringing the intensity back almost to its used as imaging agents.13 TPE-based glycoclusters could
initial point (Fig. 2a). The DLS data collected for compound 4b also not be applied in this context, since their fluore-
8806 | Org. Biomol. Chem., 2018, 16, 8804–8809
This journal is © The Royal Society of Chemistry 2018

View Article Online
Organic & Biomolecular Chemistry
Paper
Fig. 2 Concentration-dependent aggregation of the TPE-based glyco-
cluster 4b. (a) Plot of the fluorescence intensity (at 488 nm) of 4b as a
function of its increasing concentration in deionized water. (b) Dynamic
light scattering (DLS) of 4b at different concentrations in deionized
water. (c) Transmission electron microscopy images of 4b at different
concentrations in deionized water. The black arrows indicate 4b aggre-
gates at higher concentrations.
scence in the blue range overlaps with the intrinsic fluo-
rescence of PAO1.
Fig. 3 Bacterial adhesion to A549 cells (3 h infection, MOI 10) in the
presence or absence of different TPE-based glycoclusters at increasing
concentrations (µM) after washing off (5 times) excess non-adherent
bacteria with PBS. (a) Inhibition of adhesion of wild-type PAO1. All
results are compared to the first column without adhesion inhibitors
(PAO1, positive control). (b) Inhibition of adhesion of ΔlecA strain. All
results are compared to the second column without adhesion inhibitors
(PAO1ΔlecA), and the first column showing the PAO1 wild-type strain
remains as a control. (c) Inhibition of adhesion of ΔlecB strain. All results
are compared to the second column without adhesion inhibitors
(PAO1ΔlecB), and the first column showing the PAO1 wild-type strain
remains as a control. All experiments were performed in triplicate.
Results are mean SEM. (*p < 0.05); ns: non-significant.
Cell adhesion assays
The involvement of PA lectins, namely LecA and LecB, in
binding TPE-based glycoclusters to host cells was further
investigated in an in vitro cell culture model using A549 lung
epithelial cells. Inhibition of adhesion reaching 20% with
fucosylated glycocluster 4d was clearly observed while a non-
significant inhibition of adhesion was demonstrated with
galactosylated glycocluster 4b (Fig. 3a). Adhesion assays per-
formed with both lecA or lecB mutant strains (Fig. 3b and c)
confirmed the lack of inhibition of the galactosylated glyco-
cluster 4b. The fucosylated glycocluster 4c had no inhibitory
effect on the adhesion of lecA or lecB mutant strains, highlight-
Conclusions
ing that its inhibition is independent of both LecA and LecB. TPE was incorporated as the core of glycoclusters to take
These results suggest a lack of specificity in the interaction of advantage of its intrinsic fluorescence properties caused by
fucose with LecA, probably due to non-specific interactions the AIE effect. The synthesis of TPE-based glycoclusters could
with LecA.
be performed by azide–alkyne “click” chemistry and the result-
This journal is © The Royal Society of Chemistry 2018
Org. Biomol. Chem., 2018, 16, 8804–8809 | 8807

View Article Online
Paper
Organic & Biomolecular Chemistry
ing glycoclusters displayed the expected AIE effect. However,
the blue fluorescence from cells and bacteria overlapped, pre-
venting the application of the TPE-based glycoclusters as
imaging agents. ITC binding studies with LecA and LecB as
the two proteins involved in the adhesion of Pseudomonas
aeruginosa to host cells provided strong binding properties
towards both lectins with Kd values in the nanomolar range.
Further evaluation of the TPE-based glycoclusters as anti-
adhesive agents proved less encouraging, since only a very
6 Y. Huang, J. Wang and Z. Wei, Chem. Commun., 2014, 50,
8343–8345.
7 X. Wang, O. Ramström and M. Yan, Anal. Chem., 2010, 82,
9082–9089.
8 D. Grünstein, M. Maglinao, R. Kikkeri, M. Collot,
K. Barylyuk, B. Lepenies, F. Kamena, R. Zenobi and
P. H. Seeberger, J. Am. Chem. Soc., 2011, 133, 13957–13966.
9 Y. Wang, J. C. Gildersleeve, A. Basu and M. B. Zimmt,
J. Phys. Chem. B, 2010, 114, 14487–14494.
limited decrease of adhesion could be observed (20%) for 10 K. Petkau, A. Kaeser, I. Fischer, L. Brunsveld and A. P. H.
the fucosylated glycocluster. These TPE-based glycoclusters J. Schenning, J. Am. Chem. Soc., 2011, 133, 17063–17071.
therefore appear as high-affinity ligands of the LecA and 11 K.-R. Wang, H.-W. An, F. Qian, Y.-Q. Wang, J.-C. Zhang and
LecB bacterial lectins, yet their AIE fluorescence properties X.-L. Li, RSC Adv., 2013, 3, 23190–23196.
could not be exploited for further biomedical applications. 12 M. L. Lepage, A. Mirloup, M. Ripoll, F. Stauffert,
Nevertheless, improvement of the biological properties of
such glycoclusters can be readily performed through modify-
A. Bodlenner, R. Ziessel and P. Compain, Beilstein J. Org.
Chem., 2015, 11, 659–667.
ing the valency to increase the affinity and anti-adhesive pro- 13 M. Donnier-Maréchal, N. Galanos, T. Grandjean, Y. Pascal,
perties. Improvements can also be achieved by modifying
the functional groups on the TPE scaffold to provide fluo-
rescent probes with different emission properties that do
D.-K. Ji, L. Dong, E. Gillon, X.-P. He, A. Imberty, E. Kipnis,
R. Dessein and S. Vidal, Org. Biomol. Chem., 2017, 15,
10037–10043.
not overlap with biological samples through a red-shifting 14 Y. Liu, D.-K. Ji, L. Dong, N. Galanos, Y. Zang, J. Li, S. Vidal
strategy.
and X.-P. He, Chem. Commun., 2017, 53, 11937–11940.
15 T. Sanji, K. Shiraishi, M. Nakamura and M. Tanaka, Chem.
– Asian J., 2010, 5, 817–824.
16 J.-X. Wang, Q. Chen, N. Bian, F. Yang, J. Sun, A.-D. Qi, C.-G. Yan
and B.-H. Han, Org. Biomol. Chem., 2011, 9, 2219–2226.
17 Q. Chen, N. Bian, C. Cao, X.-L. Qiu, A.-D. Qi and B.-H. Han,
Chem. Commun., 2010, 46, 4067–4069.
Conflicts of interest
There are no conflicts to declare.
18 T. Kato, A. Kawaguchi, K. Nagata and K. Hatanaka,
Biochem. Biophys. Res. Commun., 2010, 394, 200–204.
19 S. Cecioni, A. Imberty and S. Vidal, Chem. Rev., 2015, 115,
525–561.
Acknowledgements
The authors thank the Université Claude Bernard Lyon 1,
Université Grenoble Alpes and the CNRS for financial support. 20 X.-M. Hu, Q. Chen, J.-X. Wang, Q.-Y. Cheng, C.-G. Yan,
Vaincre la Mucoviscidose is gratefully acknowledged for
financial support under grant number RF20160501652. A. I.
J. Cao, Y.-J. He and B.-H. Han, Chem. – Asian J., 2011, 6,
2376–2381.
acknowledges support from the ANR projects Glyco@Alps 21 Q. Qi, Y. Liu, X. Fang, Y. Zhang, P. Chen, Y. Wang, B. Yang,
(ANR-15-IDEX-02) and Labex ARCANE (ANR-11-LABX-003). Y. P.
is grateful to the Région Auvergne-Rhône-Alpes (ARC 1 Santé).
B. Xu, W. Tian and S. X.-A. Zhang, RSC Adv., 2013, 3, 7996–
8002.
Dr A. Berlioz-Barbier and C. Duchamp are gratefully acknowl- 22 C. Li, X. Luo, W. Zhao, C. Li, Z. Liu, Z. Bo, Y. Dong,
edged for mass spectrometry analyses. M.-Q. F. and Y. Q. Dong and B. Z. Tang, New J. Chem., 2013, 37, 1696–1699.
X.-P. H. are grateful to the National Natural Science 23 J. Wu, S. Sun, X. Feng, J. Shi, X.-Y. Hu and L. Wang, Chem.
Foundation of China (Grant No. 21722801).
Commun., 2014, 50, 9122–9125.
24 S. Wang, N. Galanos, A. Rousset, K. Buffet, S. Cecioni,
D. Lafont, S. P. Vincent and S. Vidal, Carbohydr. Res., 2014,
395, 15–18.
25 S. Cecioni, D. Goyard, J.-P. Praly and S. Vidal, Methods Mol.
Biol., 2012, 808, 57–68.
26 S. Cecioni, M. Almant, J.-P. Praly and S. Vidal, Synthesis of
Azido-Functionalized Carbohydrates for the Design of
Glycoconjugates, in Carbohydrate Chemistry: Proven
Synthetic Methods, ed. P. Kováč, CRC Press, Boca Raton,
2012, vol. 1, pp. 175–180.
Notes and references
1 Y. M. Chabre and R. Roy, Chem. Soc. Rev., 2013, 42, 4657–
4708.
2 K.-R. Wang, H.-W. An, Y.-Q. Wang, J.-C. Zhang and X.-L. Li,
Org. Biomol. Chem., 2013, 11, 1007–1012.
3 C. Xue, S. P. Jog, P. Murthy and H. Liu, Biomacromolecules,
2006, 7, 2470–2474.
4 K.-R. Wang, Y.-Q. Wang, H.-W. An, J.-C. Zhang and X.-L. Li, 27 J. L. Xue, S. Cecioni, L. He, S. Vidal and J.-P. Praly,
Chem. – Eur. J., 2013, 19, 2903–2909. Carbohydr. Res., 2009, 344, 1646–1653.
5 R. L. Phillips, I.-B. Kim, L. M. Tolbert and U. H. F. Bunz, 28 M. Meldal and C. W. Tornøe, Chem. Rev., 2008, 108, 2952–
J. Am. Chem. Soc., 2008, 130, 6952–6954.
3015.
8808 | Org. Biomol. Chem., 2018, 16, 8804–8809
This journal is © The Royal Society of Chemistry 2018

View Article Online
Organic & Biomolecular Chemistry
Paper
29 C. W. Tornøe, C. Christensen and M. Meldal, J. Org. Chem., 31 C. Sabin, E. P. Mitchell, M. Pokorná, C. Gautier, J.-P. Utille,
2002, 67, 3057–3064.
M. Wimmerová and A. Imberty, FEBS Lett., 2006, 580, 982–
30 S. Cecioni, R. Lalor, B. Blanchard, J.-P. Praly, A. Imberty,
987.
S. E. Matthews and S. Vidal, Chem. – Eur. J., 2009, 15, 32 R. Hu, N. L. C. Leung and B. Z. Tang, Chem. Soc. Rev., 2014,
13232–13240.
43, 4494–4562.
This journal is © The Royal Society of Chemistry 2018
Org. Biomol. Chem., 2018, 16, 8804–8809 | 8809