Modified Galacto- or Fuco-Clusters Exploiting the Siderophore Pathway to Inhibit the LecA- or LecB-Associated Virulence of Pseudomonas aeruginosa

Article abstract of DOI:10.1002/cbic.202000490

Galacto- and fuco-clusters conjugated with one to three catechol or hydroxamate motifs were synthesised to target LecA and LecB lectins of Pseudomonas aeruginosa (PA) localised in the outer membrane and inside the bacterium. The resulting glycocluster–pseudosiderophore conjugates were evaluated as Trojan horses to cross the outer membrane of PA by iron transport. The data suggest that glycoclusters with catechol moieties are able to hijack the iron transport, whereas those with hydroxamates showed strong nonspecific interactions. Mono- and tricatechol galactoclusters (G1C and G3C) were evaluated as inhibitors of infection by PA in comparison with the free galactocluster (G0). All of them exhibited an inhibitory effect between 46 to 75 % at 100 μM, with a higher potency than G0. This result shows that LecA localised in the outer membrane of PA is involved in the infection mechanism.

Full text of DOI:10.1002/cbic.202000490

Combining Chemistry and Biology
Accepted Article
Title: Synthesis of modified galacto- or fuco-clusters exploiting
siderophore pathway to inhibit LecA or LecB associated virulence
of Pseudomonas aeruginosa
Authors: Mimouna Madaoui, Olivier Vidal, Albert Meyer, Mathieu Noël,
Jean-Marie Lacroix, Jean-Jacques Vasseur, Alberto Marra,
and Francois Morvan
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: ChemBioChem 10.1002/cbic.202000490
Link to VoR: https://doi.org/10.1002/cbic.202000490
01/2020

10.1002/cbic.202000490
ChemBioChem
FULL PAPER
Synthesis of modified galacto- or fuco-clusters exploiting siderophore
pathway to inhibit LecA or LecB associated virulence of Pseudomonas
aeruginosa
Mimouna Madaoui,[a] Olivier Vidal,[b] Albert Meyer,[a] Mathieu Noël,[a] Jean-Marie Lacroix,[b] Jean-
Jacques Vasseur,[a] Alberto Marra[a] and François Morvan*[a]
[a]
[b]
Dr M. Mimouna, A. Meyer, M. Noël, Dr J.-J. Vasseur, Prof Dr A. Marra, Dr F. Morvan
Institut des Biomolécules Max Mousseron (IBMM)
Université Montpellier, CNRS, ENSCM, Montpellier, France
E-mail: francois.morvan@umontpellier.fr
Dr O. Vidal, Prof. Dr J.-M. Lacroix
Unité de Glycobiologie Structurelle et Fonctionnelle (UGSF)
UMR 8576 CNRS, Université de Lille
Cité Scientifique, Avenue Mendeleiev, Bat C9, 59655 Villeneuve d'Ascq cedex, France
Supporting information for this article is given via a link at the end of the document.
Abstract: Galacto and fuco-clusters conjugated with one to three
catechol or hydroxamate motifs were synthesized to target LecA and
LecB lectins of Pseudomonas aeruginosa (PA) localized in the outer
membrane and in the bacterium. The resulting glycocluster
pseudosiderophore conjugates were evaluated as Trojan horse to
cross the outer membrane of PA thanks to iron transport. The data
suggests that glycoclusters with catechol moieties were able to hijack
the iron transport while those with hydroxamates showed strong non-
specific interactions. Mono- and tri-catechol galactoclusters (G1C and
G3C) were evaluated as inhibitors of the infection by PA in
comparison with the free galactocluster (G0). All of them exhibited an
inhibitory effect between 46 to 75% at 100 M with a higher potency
than G0. This result shows that LecA localized in the outer membrane
of PA is involved in the infection mechanism.
them demonstrated some in vivo activity against PA especially
antibiofilm property[8b, c, d , i, j, 9] and anti-bacterial adhesion.[8f, h, 10]
During the last few years, we have synthesized and evaluated, by
DNA-based microarray assays, the affinity of hundreds of
glycocluster-oligonucleotides bearing either
D-galactose or L-
fucose moieties and found some of them (Figure1, Gal4 and
Fuc4) with high affinity for those lectins.[11] The increase of affinity
of galactoclusters determined by the measure of Kd by ITC[11b] was
400-fold higher than methyl -D-galactoside, thanks to a cluster
effect by chelation while the increase of potency of the fucocluster
in comparison with monofucoside was lower (70-fold, IC50
determined by ELLA).[11a] Indeed, due to the shape of LecB, the
increase of affinity is rather due to an increase of the local
concentration of fucoside than to a chelation of two carbohydrate
recognition domains (CRDs). The activity against biofilm
formation has been established for two tetragalactoclusters
leading to a 40% reduction of biofilm.[8b] In contrast, fucoclusters
displaying high affinity for LecB were found unable to impair
biofilm formation (unpublished results).
Introduction
Bacterial infections with the appearance of antibiotics resistance
lead to
a
severe problem of public health.[1] Alternative
O
HO
OH
O
HO
HO
O
Gal4
approaches to antibiotics are to be developed. To this end,
glycoclusters exhibiting several epitopes recognized by the lectins
of bacteria are supposed to perturb biofilm formation and bacteria
cell recognition. They have been intensively synthesized.[2] These
glycoclusters are designed to interact with high affinity with the
lectins of bacteria thanks to the cluster effect.{Gonzalez-Cuesta,
2020 #147;Lundquist, 2002 #91} Pseudomonas aeruginosa (PA),
one of the most prevalent bacteria together with S. aureus and E.
coli, is a Gram-negative, motile, opportunistic bacterium involved
in nosocomial infections (10-30%).[4] This bacterium has two
soluble lectins, LecA and LecB, that specifically recognize D-
galactose and L-fucose, respectively, and are involved in its
virulence and biofilm formation.[5] Furthermore, LecA has been
shown to be involved in adhesion and intracellular uptake of the
bacterium[6] and LecB is involved in adhesion on airway epithelial
cells.[6a] Initially localized in the cytoplasm[7] both lectins were then
largely found in the outer membrane of the bacteria.[5f]
O
NH
HO
OH
O
O
N
N
O
HO
N
N
3
Fuc4
N
Me
HO
O
O
OH
N
HO
O
O
OH
O
O
O
O
P
N
2
2
P
-O
O
N
OH
O
P
O
N
N
O-
O
O
O
Cy3
O
2
O
O
-O
O
N
O
15-mer DNA tag
O
2
P
-O
O
N
O
O
N
HO
N
N
Figure 1. Structure of the Gal4 and Fuc4 glycocluster-oligonucleotides
endowed with high affinity toward LecA and LecB lectins, respectively.
Iron is a key nutrient in bacteria and for all living systems, but due
to the low solubility of iron(III) it is poorly available. To overcome
this limitation, bacteria have developed siderophore-dependent
iron acquisition systems.[12] Siderophores are low-molecular-
weight iron chelators synthesized by bacteria mainly constituted
of catechol or hydroxamate motifs.[12a, 13] The iron-siderophore
To date, several glycoclusters, presenting strong affinity for these
two targets, have been reported in the literature[8] and some of
1
This article is protected by copyright. All rights reserved.

10.1002/cbic.202000490
ChemBioChem
FULL PAPER
complex is taken up by ferric-chelate specific TonB-dependent
transporters allowing the transport of iron into bacteria.[14]
Thus, it has been demonstrated that it is possible to hijack the
transport of iron into bacteria for the uptake of antibiotics into
bacteria leading to an increase of antibiotic potency (Trojan horse
strategy).[15]
compulsory to protect the catechol hydroxyls since they are able
to chelate the copper leading to an inefficient CuAAC and to some
degradations.
O
O
H2N
BOP, DIEA
CH2Cl
N3
Ac2O
N
N3
OH
H
Pyridine
r.t., 2 h
OH
O
OH
2, r.t., 3 h
50%
To our knowledge, this strategy has been only reported recently
with the synthesis of calixarene-based glycoclusters against PA
exhibiting four hydroxamate motifs.[9c] Most bacteria produce their
own siderophores to catch iron from the medium, but they can
also use xenosiderophores made by others microorganisms.
Along this line PA, which synthesizes two major siderophores,
pyoverdine (PVD) and pyochelin (PCH), is also able to use lot of
xenosiderophores (siderophore piracy) like enterobactin,
cepabactin, mycobactin and carboxymycobactin, desferrichrysin,
desferricrocin, coprogen, vibriobactin, aerobactin, fungal
siderophores and deferrioxamines (for a review see Cornelis, P.,
and Dingemans[16]). Since LecA and LecB lectins are mostly
cytoplasmic and only around 5% membrane bound, we
hypothesized that targeting lectins where they are produced could
improve the efficiency of their inhibition. To do so, we developed
OH
OH
54%
1
2
N
N3
H
OAc
OAc
3
Scheme 1. Synthesis of N-(3-azidopropyl)-2,3-diacetoxy-benzamide (3).
Protected azidobutyl-hydroxamate 7 was obtained in three steps
(Scheme 2). Acetohydroxamic acid 4 was first acetylated yielding
5[18] which was secondly N-alkylated with 1,4-dibromobutane to
give 6. The third step was a substitution of bromine atom with
azide by treatment with tetramethylguanidinium azide (TMG-N3)
affording 7.
a
Trojan horse strategy based on galacto/fucocluster
siderophores to help the inhibitors to reach the largest possible
amount of lectin targets.
Br
H
H
Ac
Br
2O, NaOH
N
N
HO
AcO
Cs2CO
2, DMF
100 °C, 2 h, MW
CH2Cl2
r.t., 2 h
O
O
4
5
35%
75%
Results and Discussion
Br
N3
Since catechol and hydroxamate are the most represented motifs
in siderophore, we decided to introduce them one, two or three
times and evaluate their effect on the targeting of PA. The
different neosiderophores were conjugated to galacto and fuco-
clusters and labelled with a fluorophore (Cy3) to visualize and
quantify their interaction with the bacteria. As a control, the Cy3-
glycocluster was also synthesized without a siderophore moiety.
The resulting conjugates were evaluated on mutants to confirm or
infirm the pathway through the siderophore active transport.
The syntheses of these bioconjugates were mainly performed on
solid support using phosphoramidite chemistry and copper-
catalyzed alkyne azide cycloaddition (CuAAC). Indeed, solid
supported synthesis allowed a rapid synthesis of complex
structures in low amount (< mg). This scale of synthesis produces
enough material for a screening by fluorescence monitoring.
Basically the glycocluster was synthesized and Cy3-labelled on
solid support[17] then alkynes functions were introduced in the
scaffold for a last conjugation performed in solution with a
catechol or a hydroxamate azide. Indeed, the catechol motifs
were introduced in solution at the last step since they showed
some instabilities under the ammonia treatment required to
cleave the conjugate from the solid support. To this end, three
new building blocks were synthesized: the di-acetyl
catecholamide propylazide 3 (Scheme 1), the O-acetyl N-
butylazide hydroxamate 7 (Scheme 2) and the O-DMTr-O’-
levulinyl tris-hydroxylmethyl ethane (THME) cyanoethyl
diisopropyl phosphoramidite 10 (Scheme 3).
TMGN3
N
N
AcO
AcO
CH3CN
80 °C, MW
60%
O
O
6
7
Scheme 2. Synthesis of N-(4-azidobutyl)-N-acetoxyacetamide (7).
In order to introduce a phosphoramidite derivative on a lateral
chain
of
the
scaffold,
we
synthesized
the
tris-
hydroxylmethylethane (THME) phosphoramidite 10 protected on
one hydroxyl with an acid labile dimethoxytrityl (DMTr) group and
on a second hydroxyl with a hydrazine labile levulinyl (lev) group
(Scheme 3). DMTr and Lev groups are orthogonal and so can be
selectively removed allowing a selective reaction on a hydroxyl or
on another with a subsequent phosphoramidite derivative. To this
end, DMTr-THME 8[19] was protected with levulinic acid by
dicyclohexyl carbodiimide (DCC) activation and then the resulting
compound
9
was phosphitylated by 2-cyanoethyl-N,N-
diisopropylchloro phosphoramidite in presence of DIEA affording
10 (85%).
HO
O
DCC, DMAP
DMTr
O
OH
HO
+
CH2Cl2
0 °C, 3 h
72%
O
8
O
-Pr) N
(i
2
Cne
O
P
HO
O
O
O
CneOPN(i-Pr)2Cl
Di-acetyl catechol propylazide 3 was synthesized in two steps
starting from 2,3-dihydroxybenzoic acid on which 3-azido
propylamine was coupled through an amide linkage using
BOP/DIEA (50%). Then the hydroxyls were protected by
treatment with acetic anhydride (54%) (Scheme 1). It is
DMTrO
O
DMTrO
DIEA, CH2Cl2
0 °C to r.t., 1 h
O
O
9
10
85%
Scheme 3. Synthesis of the protected phosphoramidite 10.
2
This article is protected by copyright. All rights reserved.

10.1002/cbic.202000490
ChemBioChem
FULL PAPER
A portion of 18 was treated with TCA to remove the MMTr of Cy3
and finally with aqueous ammonia to give the control Cy3 labelled
galactocluster (cy3G0) (Scheme 5). The other portion of 18 was
coupled with tris-propargyl pentaerythritol 19[23] by solid phase
phosphoramidite chemistry (SPPC) followed by deprotection with
TCA and then aqueous ammonia giving rise to Cy3-galactocluster
20 with three alkynes in solution. A last coupling with catechol
azide 3 or hydroxamates azide 7, followed by mild deacetylation
afforded the galactocluster-tricatechol conjugate (cy3G3C) or the
galactocluster-trihydroxamate conjugate (cy3G3H) respectively.
The same strategy was applied for the synthesis of Cy3F0, Cy3F3G
and cy3F3H (Schemes S1-S2).
For the preparation of the control glycocluster without
catechol/hydroxamate and the glycocluster with three
catechol/hydroxamate the synthesis was identical for the first
steps leading to 18 (Scheme 4). The propargyl α-D-mannoside
12[20] was immobilized on azide solid support 11[21] by CuAAc
affording
13[17]
and
the
propargyl-diethyleneglycol
phosphoramidite 14[11a] was introduced on each hydroxyl to form
15 after oxidation of intermediate phosphite linkages. The DMTr
group was removed by treatment with trichloroacetic acid (TCA)
and the THME derivative protected with a levulinyl group 10 was
introduced by solid phase phosphoramidite chemistry (SPPC)
followed by a Cy3 phosphoramidite keeping its MMTr group.
Levulinyl group was removed by treatment with a solution of
hydrazinium acetate[22] leading to 16. Then azide phenyl tetra-
acetyl-galactoside 17a[11a] was conjugated four times by CuAAC
affording the solid-supported glycocluster 18.
OH
Cy3
O
R1
O
R1
O
O
O
O
P
O
O
P
O
P
O
P
O
O
O
O
O
R1
1) 3% TCA
CH
O
P
2Cl
2, r.t., 60 s
OH, r.t., 16 h
O
O
O
O
2) NH4
18
O
O
O
OH
OH
O
P
O
6
O
O
DMTrO
O
OH
O
O
O
DMTrO
N
HO
HO
O
R1
N
N
G0
O
12
, Na Ascorbate
O
O
OH
OH
O
N
6
O
O
CuSO4
HO
HO
OCne
O
THPTA, TEAAc
r.t., 2 h
P
i
-Pr) N
O
O
1) SPPC with
(
2
O
N
N
O
11
13
N3
O
19
O
O
Cl
2, r.t., 60 s
2) 3% TCA, CH2
3) NH4
O
OCne
DMTrO
O
OH, r.t., 16 h
P
R
O
i
-Pr) N
O
1)
(
2
R
2
OCne
14
O
O
O
P
O
P
OCne
O
O
BMT, CH3CN
O
N
R
O
6
O
O
O
O
O
P
P
O
, H
2O, Pyr, THF
2) I2
O
P
O
O
CneO
O
O
N
N
Cy3
O
OCne
O
O
R1
O
O
R
R1
R =
O
O
P
O
O
P
O
15
O
O
P
O
O
O
P
O
O
O
O
R1
or
O
7
3
CneO
O
O
1)
Cu(0) nanoparticles, THPTA
P
O
O
P
OCne
O
P
ODMTr
H
O
O
O
(MMTr)Cy3
HO
OH
2O-Dioxane, r.t., 3 h
O
P
O
O
6
i
-Pr) N
(
O
O
1) SPPC with
2
N
N-MeOH-H2O
O
2) Et3
r.t., 3 h
R1
OLev
10
R
R
O
OCne
N
N
CneO
O
O
OCne
20
P
P
O
O
P
O
O
2) Cy3(MMTr) Phosphoramidite
3) 2.2% Hydrazine, AcOH-Pyr.
O
R2
R
O
O
O
O
N
N
r.t., 30 min
P
O
N
N
O
O
CneO
O
N
N
P
O
O
6
OCne
R2
O
O
O
O
O
N
N
O
N
R
N
N
R2
O
O
N
R =
P
O
16
O
O
CneO
Cy3
O
O
O
P
O
R1
O
O
(MMTr)Cy3
HO
R1
O
O
O
O
O
P
O
O
P
O
P
N3
O
OAc
P
AcO
AcO
O
O
O
O
O
O
R1
R
O
R
O
CneO
O
NH
O
O
OCne
O
O
O
P
P
P
OAc
P
O
O
OCne
O
O
O
O
O
17a
OH
O
R
O
O
P
O
O
O
6
O
O
CuSO
4, Na Ascorbate
P
O
N
O
O
THPTA, 60 °C, 2 h, vortex
O
R1
CneO
O
N
N
P
O
O
6
N
OCne
O
O
N
N
N
O
O
R
O
N
N
OH
HO
HO
=
O
R1
N
NH
O
O
N
O
OH
HO
O
N
OAc
AcO
AcO
OH
O
R =
O
18
NH
R2
=
R2 =
cy3G3H
O
N
N
cy3G3C
HO
H
OAc
O
Scheme 4. Synthesis of solid-supported Cy3-glycocluster 18. SPPC: i) 3% TCA
in CH2Cl2; ii) amidite + benzylthiotetrazole (BMT), dry CH3CN; iii) Capping: Ac2O,
N-methylimidazole, pyridine, THF; iv) 0.1M I2, H2O, THF, pyridine.
Scheme 5. Synthesis of Cy3-galactocluster without catechol (cy3G0) and with
three catechol (cy3G3C) or three hydroxamates units (cy3G3H).
3
This article is protected by copyright. All rights reserved.

10.1002/cbic.202000490
ChemBioChem
FULL PAPER
For the synthesis of the mono- and di-catechol conjugates, the
intermediate 15 was conjugated with fully protected galactoside
azide 17a, then the mono-propargyl 22a[19] or the di-propargyl
22b[24] phosphoramidite derivative was coupled by SPPC followed
by Cy3 phosphoramidite coupling to give, after ammonia
deprotection, 23a and 23b in solution, respectively (Scheme 6).
Finally, di-acetyl catechol propyl azide 4 or di-acetyl hydroxamate
butyl azide 7 were conjugated by CuAAC affording after
deacetylation the Cy3-galactoclusters mono- cy3G1C and di-
catechol cy3G2C, and the corresponding Cy3-galactocluster
mono- cy3G1H and di-hydroxamate cy3G2H.
For the synthesis of the fucoclusters, the same protocol was
applied starting from 15 on which the fully protected fucoside
azide 17b was introduced yielding in fine the Cy3-fucoclusters
mono- cy3F1C and di-catechol cy3F2C and the corresponding Cy3-
fucocluster mono- cy3F1H and di-hydroxamate cy3F2H (Scheme
S3).
R
R
O
OCne
O
P
ODMTr
P
O
OCne
O
O
R
O
O
15
+ 17
, Na Ascorbate
O
O
CuSO4
P
O
O
O
THPTA, H2O-dioxane
60 °C, 2 h, vortex
CneO
O
O
P
O
O
6
OCne
N
O
R
N
N
N
O
N
The resulting 28 glycoclusters were characterized by C18 reverse
phase HPLC and MALDI-TOF mass spectrometry. For the
glycoclusters with catechols, HPLC profiles of acetylated
derivative showed a thin peak while the fully deprotected
compounds were eluted as a broad peak with and increasing
complexity with the number of catechol motifs. This phenomenon
could be explained by hydrogen bounding of the catechol with the
stationary phase. In contrast the HPLC of glycoclusters with
hydroxamates showed nice profiles. MALDI-TOF MS spectra
showed only the [M-H]- ion corresponding to the expected
structures.
O
O
N
OAc
AcO
R =
O
21
NH
O
AcO
OAc
OCne
ODMTr
O
O
O
P
P
P
i
-Pr) N
2
O
O
(
Cy3
O
1) SPPC with
R
R = CH
R =
22a
22b
R2
3
O
O
R1
O
O
R1
O
O
Phosphoramidite
O
P
O
P
2) Cy3
3) NH4OH, r.t., 16 h
O
O
O
O
R1
O
O
O
O
O
P
O
O
P
O
O
OH
O
O
O
6
O
N
O
Evaluation of labelling efficiency of Pseudomonas
aeruginosa by galactocluster-siderophore conjugates.
R1
N
N
N
N
O
The Cy3 fluorescence intensity of each glycocluster was
measured. Indeed, it is known that catechol reduces the
fluorescence intensity of fluorescent molecules.[25] While the
fluorescence intensity of cy3G1C was similar to that of cy3G0, those
of cy3G2C and cy3G3C were dramatically reduced by 78% and
96%, respectively (Figure S4). One can imagine that such
decrease of fluorescence is due to strong π,π interactions
between the catechols and the indoles of Cy3 leading to non-
radiative energy loss. For the fucoclusters series with catechols
the decrease of fluorescent was lower with only 50% for cy3F3C
(Fig; S5). In contrast, the Cy3 fluorescence intensity of
glycoclusters with hydroxamates motifs was increased by ~20%
for cy3G1H and cy3G2H and by 15% for cy3G3H with respect to
cy3G0 (Figure S6) while similar intensity was observed for cy3F0,
for cy3G1H and cy3G2H and slightly higher for cy3F3H (Figure S7).
O
N
OH
HO
HO
=
O
R1
NH
O
R
2 = CH
OH
23a
23b
3
R2
=
O
or
7
3
1)
Cu(0) nanoparticles, THPTA
N
N
Cy3
O
H
R3
O
N
O
2O-Dioxane, r.t., 2 h
N-MeOH-H2O
2) Et3
P
O
r.t., 2 h
R1
R2
O
O
R1
O
O
O
P
O
P
O
P
O
O
O
O
R1
O
O
O
O
O
P
O
O
O
O
OH
P
O
O
6
O
N
N
O
R1
N
N
O
N
O
O
N
OH
HO
HO
=
O
R1
Fluorescence quantification of bacterial labelling by cy3-
galacto/fucoclusters.
NH
O
OH
R2
Trojan horse strategy is based on the use of the siderophore
pathway to help entrance of inhibitors of virulence in the bacterial
envelope to reach their specific target. We believed that such a
strategy could help preventing LecA or LecB-dependant virulence
of the bacteria by targeting the lectin before its exposure on the
bacterial surface. Our first design of molecules was to simply add
1 to 3 catechol or hydroxamate residues on the galactocluster G0
or fucocluster F0 structure and to evaluate the inhibitory potential
of the modified molecules. Synthesis of these molecules is
expensive and time consuming. Consequently, only small amount
of 1, 2 or 3 catechols-galacto/fucoclusters and 1, 2 or 3
hydroxamates-galacto/fucoclusters have been produced first and
fluorescently labelled with cyanine (Cy3) to assess by
fluorescence quantification their possible association with the
bacteria.
N
N
R3
O
N
CH3
OH
O
HO
HO
N
cy3G1C
cy3G2C
H
R3
N
cy3G1H
cy3G2H
O
Scheme 6. Synthesis of the Cy3-galactoclusters with one (cy3G1C) or two
catechol (cy3G2C) motifs and with one (cy3G1H) or two (cy3G2H) hydroxamate
motifs.
4
This article is protected by copyright. All rights reserved.

10.1002/cbic.202000490
ChemBioChem
FULL PAPER
The Cy3-galactocluster/fucocluster-siderophores were incubated
with different PA strains: the wild type PAO1 and three isogenic
mutants, fpvA, exbB1 and lecA (galactoclusters) or lecB
(Figure 2). The cy3G1C and cy3G2C exhibited a 5-fold increase of
labelling and the cy3G3C showed a strong increase of 20-fold.
Thus, the addition of
1 to 3 catechol residues on the
(fucoclusters).
The
Cy3-galactocluster-siderophores
are
galactoclusters structure increase its association with the WT
bacterial envelope.
expected to target LecA lectin (and fucoclusters, LecB)
associated with the bacterial cell surface as well as
periplasmic/cytoplasmic LecA (LecB). lecA (lecB) mutant, that
doesn't express the LecA (LecB) lectin, will then allow
discrimination of non-LecA (LecB) specific labeling of the bacteria
and should be considered as background. The fpvA mutant
doesn't express the pyoverdine specific receptor (PVD-R) that
recognizes pyoverdine and allows the active transport of iron-
pyoverdine complex into the bacterium. Iron-pyoverdine
recognition by FpvA is essentially due to association of the
receptor with the chromophore(catecholate)/Fe/hydroxamates
When the galactoclusters were incubated with lecA mutant we
observed a very low labelling for cy3G0, cy3G1C and cy3G2C
showing no interaction of the galactocluster with the mutant
devoid of LecA synthesis. Initial works of Glick and Garber have
located the soluble LecA lectin mainly in the cytoplasm of the
bacteria with a small percentage (3-9%) membrane bound or
located in the periplasmic space.[7] Membrane localisation of LecA
was confirmed by one other group.[30] However, how did the lectin
come in the envelope of the bacteria still remains unknown since
the lecA gene doesn’t contain any specific signal peptide
encoding sequence.[30] Difference of labelling observed here
between PAO1 and the lecA mutant must signify specific
association of galactoclusters with or without catechols with the
target LecA and its location in the bacterial envelope.
complex.[26]
The
Cy3-galacto(fuco)cluster-siderophores
described herein don’t share common structures with pyoverdine
but have been designed to include 1 to 3 catechols or 1 to 3
hydroxamates residues. It has been described in the literature that
antibiotics modified by the addition of catechols, cifedorocol[27] or
hydroxamates, albomycins-like,[28] display better minimum
inhibitory concentration (MIC) than their non-modified counterpart
demonstrating entrance of the antibacterial compounds via the
siderophores uptake pathway. If in the experiments described
here the fpvA mutant doesn’t display any difference of Cy3-
galacto(fuco)cluster-siderophores uptake than does the WT
PAO1, no conclusions will be made. But if differences are
observed they had to be attributed to entrance of the compounds
via the FpvA receptor. The exbB1 mutant is used herein for the
same purpose. Siderophores-Fe entrance in the periplasm of the
bacteria is dependent on energy transfer by the TonB/ExbB/ExbD
cluster.[29] One can expect then that inactivation of the
TonB/ExbB/ExbD machinery in Pseudomonas aeruginosa will
lead as well as to a reduction of galacto(fuco)cluster-siderophores
entrance in the bacterial envelope.
When the mutants fpva and exbB1 were incubated with Cy3-
galactocluster catechol conjugates and cy3G0, we observed the
same level of labelling for cy3G0 and a strong decrease of labelling
for the cy3G1C, cy3G2C and cy3G3C showing that the high labelling
on PAO1 was due to some interaction between Cy3-
galactocluster catechol conjugates and the iron transport
mechanism. Increase of fluorescence labelling of the bacteria with
catechols-Cy3-galactoclusters compared to Cy3-galactoclusters
is not alone a direct proof of the entrance of the molecules in the
bacterial envelope via the siderophore pathway. One may argue
that it could be the result of unspecific interaction of the catechols
with the bacterial envelope. As example, cy3G3C also displayed a
high labelling of the lecA, fpvA and exbB1 mutants even if it is 1,5
to 3-fold less than for wild-type PAO1. This phenomenon would
be due to some non-specific adsorption of cy3G3C on the surface
of bacteria rather than some internalization of it thanks to iron
transport due to the recognition of the three-catechol motif by the
bacteria. But, since lecA, fpvA and exbB1 mutants are isogenics
of PAO1 strain, the decrease of labelling observed is obviously
due to the absence of membrane associated LecA for the lecA
mutant and absence of a functioning siderophore pathway for the
two latter. Consequently, as expected, our data show that addition
of 1 to 3 catechol residues allows entrance of the Cy3-
galactocluster in the bacterial envelope via the siderophore
pathway. Then, higher amount of the molecules has been
produced to assess its inhibitory potential against PA virulence.
In contrast to the Cy3-galactoclusters-catechol conjugates
Cy3G1C and Cy3G2C, the Cy3-fucoclusters-catechol conjugates
Cy3F1C and Cy3F2C showed a low increase of fluorescence for the
control strain while cy3F3C showed a large increase (16-fold)
(Figure S8). Surprisingly, the percentage of labelling observed for
Cy3F0 are similar for the WT and for the mutant lecB suggesting
no specific interaction with the membrane bound lectin. In vitro
experiments, has shown that F0 is highly affine for LecB34 but no
interaction with membrane bound LecB has been demonstrated
to date. Tielker and coworkers11 have demonstrated localization
of LecB in the PA outer membrane where its presence helps
biofilm development. Thus, our result indicates that the
fucoclusters developed herein don’t interact with membrane
bound LecB in vivo.
Figure 2. Percentage of labelling of PA strains with 1 μM Cy3-galactocluster
with 0 to 3 catechols. Data were normalized to 100% for the Cy3-galactocluster
with 0 catechols (G0) added to the wild type strain PAO1. Grey “*” are statistical
comparison between G0 and 1, 2, or 3 catechols for the PAO1. Black “*” are
statistical comparison of each mutant, lecA, exbB1 or fpvA labelled with Cy3-
galactocluster with 0, 1, 2 or 3 catechols respectively to the corresponding
PAO1 labelled with Cy3-galactocluster with the same number of catechols (0 to
3 catechols). With *, p<0.05; **, p<0.01; ***, p<0.001; ****, p<0.0001.
The data were normalized with a percentage of labelling fixed at
100% for cy3G0 (1 μM) when incubated with wild type PAO1
5
This article is protected by copyright. All rights reserved.

10.1002/cbic.202000490
ChemBioChem
FULL PAPER
For the WT strain and the fpvA mutant we observed a higher
labelling for cy3F3C whereas the others mutants, exbB1 and lecB,
afforded similar low levels of fluorescent labelling. Difference
between labelling of the fpvA and exbB1 mutant suggest
interaction of cy3F3C with the siderophore pathway independent
of the pyoverdin uptake. Additionally, difference of cy3F3C
labelling between WT and the lecB mutant suggests that
interaction with the siderophore pathway helps the fucocluster to
reach soluble LecB to interact with confirming that three catechol
residues enhance association of the glycocluster with siderophore
pathway. Nevertheless, since the control molecule cy3F0 doesn’t
show any difference of labelling between WT and lecB mutant no
further experiments were conducted with fucoclusters-catechol
conjugates.
All the data suggested that there are some non-specific
interactions of the Cy3-galactocluster or fucocluster hydroxamate
conjugates with the bacteria and that the hydroxamate motifs
reported in this study are not recognized by the bacteria as a
siderophore.
To summarize this first study, the data show that catechol-
galactoclusters were internalized by iron-assisted transport while
hydroxamate glycoclusters were not. For the fucocluster-catechol
conjugates, the increase of labelling was not really significant and
it seems that the fucocluster was not recognized by LecB while
fucocluster-hydroxamate conjugates showed high non-specific
interaction. Hence, for the evaluation of the anti-infectious
properties of glycoclusters on bacteria, we only selected the
monocatechol galactocluster, since there is a similar behaviour
between G1C and G2C, and the tri-catechol galactocluster G3C
to evaluate the effect of the number of catechol on the activity.
Solution phase synthesis of mono- and tri-catechol
glycoclusters G1C and G3C
The syntheses were performed in solution to obtain few hundred
milligrams of each compound. Since we observed some instability
of acetyl groups on catechol, the more stable benzoyl group was
selected. 2,3-Dihydroxybenzoic acid was first protected by
treatment with benzoic anhydride to give 24. The carboxylic
function was activated as an anhydride by treatment with ethyl
chloroformate and eventually 3-azido-propylamine was added to
form the amide linkage affording the N-(3-azidopropyl)-2,3-
dibenzoxybenzamide 25 in 62% overall yield (Scheme 7).
Figure 3. Percentage of labelling of PA strains with 1 μM Cy3-galactocluster
with 0 to 3 hydroxamate. Data were normalized to 100% for the Cy3-
galactocluster with 0 hydroxamate (G0) added to the wild type strain PAO1.
Grey “*” are statistical comparison between G0 and 1, 2, or 3 hydroxamate for
the PAO1. Black “*” are statistical comparison of each mutant, lecA or exbB1
labelled with Cy3-galactocluster with 0, 1, 2 or 3 hydroxamate respectively to
the corresponding PAO1 labelled with Cy3-galactocluster with the same number
of hydroxamate (0 to 3 hydroxamate). With *, p<0.05 ; **, p<0.01 ; ***, p<0.001 ;
****.
O
O
O
O
Bz2O
OH
O
O
Cl
O
1
DIEA
CH2Cl2
0 °C, 15 min
Pyridine
r.t., 16 h
93%
OBz
OBz
OBz
OBz
24
O
N
N3
H2N
N3
H
CH2Cl2
0 °C to r.t., 1 h
67%
OBz
OBz
The Cy3-galactocluster hydroxamate conjugates cy3G1H, cy3G2H
and cy3G3H as well as cy3G0 were incubated with PAO1, exbB1
and lecA (Figure 3). The labelling of PAO1 increased by 3-fold for
cy3G1H and by almost 6-fold for cy3G2H and cy3G3H with respect
to cy3G0. As for the catechol series, the highest substituted
hydroxamate glycoconjugate exhibited the highest labelling.
However, surprisingly, for the exbB1 mutant the increase of
labelling was similar showing that the increase of labelling should
not be due to some interaction of the galactoclusters with the iron-
transport mechanism. Hence exbB1 is not interacting with its
hydroxamate glycoconjugate. Since exbB1 is a more general
partner of the siderophore uptake than FpvA (restricted to the
pyoverdin-like molecules uptake) the fpvA mutant has not been
tested. Finally, the labelling of the lecA mutant was also found to
be increased but to a lower extent.
25
Scheme 7. Synthesis of N-(3-azidopropyl)-2,3-dibenzoxy-benzamide (25).
The monocatechol-glycocluster G1C was synthesized in four
steps (Scheme 8). Protected catechol derivative 25 was
conjugated by CuAAc to propargyldiethyleneglycol α-mannoside
26[31] and the free hydroxyls were phosphorylated by means of
propargyldiethyleneglycol phosphoramidite 14 followed by
-
oxidation with solid-supported A26 IO4 reagent to give 28. The
galactoside units were finally introduced by CuAAC to afford G1C
after deprotection.
The same trend was observed for the fucocluster-hydroxamate
conjugates with a high non-specific interaction when the number
of hydroxamate motifs increased (Figure S9) but also an absence
of interaction with the siderophore pathway.
6
This article is protected by copyright. All rights reserved.

10.1002/cbic.202000490
ChemBioChem
FULL PAPER
HO
N
OH
O
N
O
O
25
HO
OR
O
HO
O
OH
N
OAc
O
AcO
AcO
O
Cu(0) nanoparticles
H2O-Dioxane
17a
O
O
NH
O
Cu(0) nanoparticles
H2O-Dioxane
55 °C, 16 h
OAc
O
R = H
55 °C, 16 h
26
87%
29
i
82%
CneOPN( -Pr)2Cl, DIEA
, 0 °C to r.t., 1 h
CH2Cl2
14
OH
R = (
1)
OH
O
30
i
-Pr)
Tetrazole, CH
78%
R1
2N-P(CneO)
OBz
BzO
3CN
HO
HO
r.t., 4 h
N
N
-
26
1)
R1
R2
O
BMT, CH
H
2) Amberlyst A26 IO4
r.t., 2 h
O
P
O
O
3CN
N
N
r.t., 4 h
P
R2
O
R2
O
O
R1
O
O
-
IO
O
100%
O
O
26
2) Amberlyst A
r.t., 2 h
4
O
27
P
O
O
P
O
R2
O
O
R
R
O
O
O
OCne
O
R1
O
P
17a
O
P OCne
O
O
N
R
O
O
O
N
O
Cu(0) nanoparticles
H2O-Dioxane
OBz
O
O
P
OR3
O
O
N
R3O
R3O
O
BzO
CneO
=
N
O
N
N
55 °C, 16 h
R1
P
H
O
NH
O
O
O
N
OCne
OR3
O
R
O
O
R =
R
R
3 = Ac,
R2 = O-Cne
31
28
Et3N-MeOH-H2O
r.t., 16 h
O
-
3 = H, R2 = O
54%
32
R
R
O
O
O
P
O
P O
O
O
O
Scheme 9. Synthesis of the alkynyl-galactocluster EG2.
O
R
O
P
Et3N-MeOH-H2O
r.t., 16 h
OH
O
O
O
HO
N
O
N
P
H
O
O
O
N
N
O
i
N( -Pr)2
O
R
P
O
O
O
O
O
1)
OCne
TsO
O
2
N
34
HO
O
N
O
Tetrazole, CH3CN, r.t., 2 h
-, r.t., 2 h
O
N
O
OH
HO
HO
R =
O
2) Amberlyst A26 IO4
60%
G1C
NH
O
33
74%
OH
25
O
P
O
Cu(0) nanoparticles
TEAAc
O
O
Scheme 8. Synthesis of glycoclusters G1C in solution.
O
O
TsO
H2O-Dioxane
55 °C, 16 h
65%
2
O
O
Cne
35
R
R
P
O
O
O
O
P O
P
O
O
O
P
O
Et3N-MeOH-H2O
r.t., 16 h
O
O
R
O
O
O
N
N
N
N
O
O
O
O
TsO
H
O
2
O
O
BzO
Cne
P
O
OCH3
O
OBz
3
O
36
N
R
O
N
O
, DMF
1) TMGN3
55 °C, 16 h
O
N
O
P
OH
O
O
HO
HO
R =
G0
O
2) BzCl, Pyridine
0 °C, 1 h
N
N
H
BzO
O
NH
O
O
O
N3
2
N
N
O
OH
29% (3 steps)
OBz
3
37
Figure 4. Structure of glycocluster G0.
Scheme 10. Synthesis of the azidated tricatechol derivative 37.
The synthesis of G3C was carried out according to a convergent
strategy. The galactocluster 32 was synthesized with a propargyl
diethyleneglycol arm on the anomeric carbon of mannoside
(Scheme 9) and the tricatechol 37 was synthesized with an azide
diethyleneglycol arm (Scheme 10) allowing a final conjugation of
both units 32 and 37 (Scheme 11).
Since the mannoside 26 exhibited an alkyne function, it was not
possible to first introduce propargyl diethyleneglycol
phosphoramidite on it and then the azide galactosides. Hence the
azide tetraacetylgalactoside 17a was conjugated by CuAAC to
propargyl diethyleneglycol and then converted to its
phosphoramidite derivative 30 which was coupled with 26
affording the alkynyl-galactogluster 31 which was finally
deprotected to give 32.
For the siderophore synthesis, the tripropargyl-pentaerythritol 33
was coupled to tosyltriethyleneglycol phosphoramidite 34 and
after oxidation of phosphite triesters into phosphotriesters the
three catechols were introduced by CuAAC affording 35 (Scheme
10). Since during azidation partial debenzoylation occurred, 35
was first deprotected and azidation was performed by treatment
with TMG azide. Finally, the hydroxyls were reprotected by
treatment with benzoyl chloride affording the azide-sidererophore
37. The perbenzoylation using benzoyl anhydride was inefficient,
on the other hand, the use of benzoyl chloride led to the expected
compound with a side-product corresponding to the loss of a
catechol carboxy acid and a benzoylation of the intermediate
amine. This side-reaction is surprising since amides of aliphatic
amines are usually very stable.
7
This article is protected by copyright. All rights reserved.

10.1002/cbic.202000490
ChemBioChem
FULL PAPER
OH
OH
O
HO
O
OH
OH
OH
OH
OH
HN OH
O
O
HO
HO
HN
O
NH
O
O
OH
N N
N
N
N
N
NH
N
N
N
O
OH
OH
O
N
O
O
N
O
OH
HO
N
HO
O
N
N
N
O
1) Cu(0) nanoparticles
NH
O
O
H
H
OH
-
O
O
O
N
2O-Dioxane
55 °C, 6 days
O
O
O
N
N
N
O
P
P
-
O
O
N-MeOH-H2O
2) Et3
r.t., 16 h
O
O
O
-
OH
O
OH
P
O
+
O
32
37
O
O
O
O
O
-
O
HO
38%
P
O
O
N
O
NH
O
N
N
O
-
OH
P
O
O
O
O
O
O
N
O
N
O
N
O
O
G3C
Scheme 11. Synthesis of the galactocluter-tricatechol G3C.
A last CuAAC conjugation allowed the formation of benzoylated
G3C (Scheme 11). Surprisingly the reaction was very sluggish
requiring 6 days. A reduced accessibility of the azide and alkyne
functions with the two quite large units could explain the slowness
of the click reaction. After chromatography on C18 reverse phase
and deprotection G3C was afforded.
As expected, compared to PAO1, lecA mutant displays a lower
infection percentage with only 28±14 % of the WT infection
percentage. Concentrations of galactocluster G0 ranging from
100 to 500 μM reduce up to 70% the percentage of infection of
PAO1. Thus 100 μM of G0 is already sufficient to reduce the
whole LecA dependant percentage of infection for the PAO1
strain. Increase of the G0 concentration in the medium doesn’t
increase the level of inhibition since probably all the LecA
exposed on the surface of the bacterial cell are already masked
by 100 μM of the galactocluster. Lower concentration of the
galactoclutser G0 decreases the efficiency of inhibition since 10
μM only reduce to 25% (76±21 %) the percentage of infection of
PAO1 (data not shown).
Infection protection assay with G0, G1C and G3C
The monocatechol G1C and tricatechol G3C galactoclusters were
tested as inhibitors of infection in comparison with the glycocluster
G0 (Figure 4).[8b] NCI-H292 cells were incubated with bacteria in
presence of galactoclusters with or without catechols during the
2h of infection. Then, all the bacteria stayed on the outside of NCI-
H292 cells were killed by gentamicin treatment. Finally, cells were
washed to remove antibiotic and dead bacteria and lysed. Serial
dilution of cell lysates was then prepared and plated on to LB agar
to quantify the rate of infection by comparison with the control,
untreated PAO1. Various concentrations of each inhibitor were
tested in order to demonstrate dose dependant inhibition of
infection (Figure 5).
In a previous work, we have demonstrated that galactoclusters
(G0), targeting the soluble LecA lectin of Pseudomonas
aeruginosa (PA), when added in a concentration as low as 10 μM,
can reduce considerably (up to 40%) the biofilm production by the
bacteria.[8b] Several similar molecules were developed by other
8i, j, 9]
groups showing equivalent results[8b-d,
demonstrating the
importance of the lectin during the biofilm building even if it is still
not very clear how the lectin can help its development. LecA
involvement in PA virulence is a lot more complex since the lectin
was shown to display on his own cytotoxic effect on respiratory
epithelial cells[32] and acts, associated to the bacteria, as an
adhesin[5f]/invasin[6a] to host tissue by directly binding to the
globotriosyl receptor (Gb3)[6a] promoting cell infection by the
bacteria.
c
c
o
o
n
ntrol (PAO1)
AO1+G0
AO1+G1C
AO1+G3C
cA
100
100
P
PA
PA
P
PPA
le
lec
****
****
5500
**
**
****
****
*
*
ns
ns
****
****
In the present work, we show that 100 μM of galactoclusters (G0)
is sufficient to reduce up to 70% the PA infection in an ex-vivo
model of infection using the human pulmonary cell line NCI-H292.
Increased concentration of the inhibitor doesn’t increase the
inhibition efficiency suggesting that the maximum inhibitory
potential was reached. Infection done with the lecA mutant strain
has also shown a 70% reduction compared to the WT which mean
that 30% of the PA infection potential is independent of LecA.
Our previous publication has shown that 130 μM of the
galactocluster G0 decreases the adhesion force existing between
PA and fixed cells on an AFM tip.[10b] The combined results
****
ns
****
****
****
****
0
M
M
M
M
M
M
M
M
M
µ
µ
µ
µ
µ
µ
µ
µ
µ
lecA
0 µM
100
200
500
100
200
500
100
200
500
Figure 5. Inhibition of PAO1 infection using G0, G1C, and G3C at concentration
ranging from 100 to 500 µM compared to the lecA mutant. Data were normalized
to 100% for the WT strain PAO1. Black “*” are statistical comparison between
PAO1 and G0, and lecA wile grey “*” were statistical comparison between G0
and G1C and G3C. With *, p<0,05 ; **, p<0,01 ; ***, p<0,001 ; ****, p<0,0001.
8
This article is protected by copyright. All rights reserved.

10.1002/cbic.202000490
ChemBioChem
FULL PAPER
suggest that galactoclusters prevent host cell Gb3 LecA
dependent recognition by the bacteria diminishing the infection
ratio.
Such a 70% inhibition of PA ability to colonize pulmonary cell ex-
vivo using β-D-galactopyranoside-presenting glycoclusters was
also observed by Malinovská and co-workers using higher
concentrations of inhibitor (up to 2 mM).[10c] The galactoclusters
developed herein seem, up to date, being more efficient to reduce
PA infection on pulmonary cells and should be good candidates
to further in vivo protection assay in an animal model.
although the protection assays with G0 at micromolar
concentration led to very interesting results, the assays with
catechol-galactoclusters were disappointing, because no gain
compared to G0 and even a reduction of the inhibitor efficiency
was observed. Nevertheless, evidences have been gained that
addition of 1 to 3 catechols promotes entrance of chemical
compounds into PA envelope via the siderophore pathway and
can be suitable for other inhibitors.
Finally, this work shows that the Trojan horse strategy targeting
LecA is not helpful against PA virulence. On other hand, we
proved that 100 μM of G0 added in the culture medium were
sufficient to reduce PA virulence to a same extent as the lecA
mutant. If the virulence of the bacteria is in fact promoted by non-
membrane bound LecA, soluble galactoclusters, such as G0,
recruiting soluble, bacterial free, LecA in the tissue
neighbourhood and preventing their further association either with
bacterial or cell membrane can certainly help to protect against
PA infection.
Unfortunately, addition of one or three catechols on the
galactocluster structure (G1C and G3C) doesn’t increase the
inhibition potential of the molecule as compared to G0 with even
a percentage of infection observed for 100 μM of G1C (54±20 %)
or G3C (38±5 %) higher than that observed for G0 (25±5 %).
Since no direct proof of entrance of the molecules in the
cytoplasm of the bacteria was demonstrated in our work, we can’t
rule out the fact that simply the inhibitor doesn’t reach intracellular
LecA to increase the inhibition efficiency. One may argue that the
presence of catechols could decrease affinity of the galactocluster
for membrane bound LecA or the stability of LecA/galactocluster
complex. It is possible that the ability of catechols to interact with
divalent cations may have competed with the affinity of LecA with
Ca2+ necessary for the LecA/galactose interaction.
Our experiments show that the addition of one or three catechols
promotes entrance of the galactoclusters in the bacterial envelope
via the siderophore pathway. Thus, one other interpretation will
be that the catechol-galactoclusters reach its intracellular LecA
target but that this has no influence on the bacterial virulence
since during the two hours of infection only the membrane bound
LecA is used by the bacteria during the infection process.
According to Diggle and coworkers, membrane bounds LecA may
have come from other bacterial cell lysis, liberating soluble lectin
which then link to the bacterial surface and help virulence.[30] Then,
LecA binding on Gb3 receptor exposed on host cell surface will
allow engulfment of the bacteria and cell infection.13 This model
can explain why only a few percentage of LecA is membrane
bound, how it came here, and why bacterial free LecA displays
cytotoxic effect on host tissue. In addition, this may also explain
why, as this work shows, the force entry of the galactocluster
targeting LecA into the bacterial envelope by Trojan horse
strategy, did not succeed in improving its inhibition potential.
Experimental Section
All reagents for synthesis were commercial and used without purification.
Dry solvents and reagents CH3CN, pyridine, DIEA and NEt3 were distilled
over CaH2 and CH2Cl2 was distilled over P2O5, others solvents were
commercial and used without distillation. Sensitive reactions were
performed under argon atmosphere. Reactions under microwaves were
achieved on Monowave 300 Anton Paar. The reactions were monitored by
TLC using silica gel 60 F254 precoated plates. TLC plates were analysed
by UV light (λ = 254 nm) and revealed by treatment with KMnO4,
Ninhydrine in EtOH, 10% H2SO4 in EtOH/H2O (1:1 v/v), phosphomolybdic
acid 20 wt% solution in EtOH or molibdenum blue according to Dittmer and
Lester followed by heating. Products were purified on column
chromatography using silica gel Si 60 (40-63 µm) or cartridge of silica gel
35-45 µm. Reverse phase purification was executed with C18 flash
chromatography (40 µm). Reverse phase C18 HPLC analyses were
performed with Dionex Ultimate 3000 instrument equipped with an
automatic injector and a photometer DAD 3000 with Nucleodur® 100 Å, 3
µm C18ec, 75 mm DI 4.6 mm, Macherey-Nagel column (flow 1 mL/min
using linear gradient of CH3CN in 0.05M aqueous TEAAc pH 7). NMR
analyses were performed at 298 K using a 200 MHz, 400 MHz, 500 MHz
or 600 MHz spectrometer (Bruker) using deuterated solvent. Observed
multiplicities are labelled with following abbreviation: s (singlet), d (doublet),
dd (doublet of doublet), t (triplet), tt (triplet of triplet), q (quartet), p (quintet),
m (multiplet). Shifts (δ) were referenced relative to deuterated solvent and
expressed in part per million (ppm), coupling constants were expressed in
hertz (Hz). High resolution (HR-ESI-QTOF) mass spectra were achieved
with Q-Tof Micromass spectrometer. MALDI-TOF analysis were performed
on a Shimadzu Assurance equipped with 337 nm nitrogen laser. Spectra
were recorded, in negative or positive mode, using THAP with 10% of
ammonium citrate as a matrix in water CH3CN (1:1 v/v). Liquid samples
were mixed with the matrix as 1:5 v/v ratio and 1 µL was deposited on the
stainless-steel plate for drying.
Conclusion
The purpose of the present work was to develop inhibitors of PA
virulence by targeting the lectins LecA and LecB involved in
biofilm formation and host tissue infection. Since the two lectins
are mostly cytoplasmic and only around 5% membrane bound,
Trojan horse strategy based on catechol- or hydroxamate-
modified galacto/fucoclusters mimicking siderophores has been
developed to help the inhibitors to reach the largest possible
amount of lectin targets. Our results show that only galactocluster-
catechol conjugates were able to penetrate the bacterial envelope
via the siderophore pathway. Protection assays of human
pulmonary cell culture against PA infection using galactoscluster
G0 or its catechols (G1C and G3C) associated counterparts have
been successful in this work comforting us in the efficiency of
galactoclusters as inhibitor of PA virulence. As discussed above,
N-(3-azidopropyl)-2,3-dihydroxybenzamide 2: 2,3-Dihydroxybenzoic
acid (894 mg, 5.8 mmol) was dissolved in anhydrous CH2Cl2 (17 mL), 3-
azidopropylamine (701 mg, 7 mmol) and anhydrous DIEA (2 mL, 11.6
mmol) were added and the mixture was dried over molecular sieve 4Å for
1h under argon and cooled down to 0 °C. BOP (2.6 g, 5.8 mmol) was
added and the solution was stirred for 3h at room temperature. The crude
was directly applied on a silica gel column for flash chromatography using
cyclohexane with AcOEt (from 20% to 80%). Product was lyophilized from
dioxane to afford compound 2 (716 mg, 50%) as a colorless oil. TLC Rf:
0.53 Cyclohexane: AcOEt 4:6 v/v. 1H NMR (400 MHz, CDCl3) δ 12.65 (s,
1H, OH), 7.05 (dd, J = 7.8, 1.2 Hz, 1H, Ar), 6.89 (dd, J = 8.1, 1.1 Hz, 1H,
Ar), 6.76 (t, J = 8.0 Hz, 1H, Ar), 6.66 (s, 1H, NH), 5.86 (s, 1H, OH), 3.56
9
This article is protected by copyright. All rights reserved.

10.1002/cbic.202000490
ChemBioChem
FULL PAPER
(q, J = 6.4 Hz, 2H, NCH2CH2), 3.47 (t, J = 6.4 Hz, 2H, CH2CH2N3), 1.91 (p,
J = 6.4 Hz, 2H, CH2CH2N3). 13C NMR (101 MHz, CDCl3) δ 170.3, 149.2,
146.1, 118.9, 118.3, 116.0, 114.3, 49.8, 37.7, 28.6. HR-ESI-QToF MS
(positive mode): m/z calcd. for C10H13N4O3 [M+H]+ 237.0988, found
diol 8[19] (DMTrTHME) (850 mg, 2.0 mmol) was dissolved into 10 mL of
anhydrous dichloromethane. Then, 1,3-dicyclohexylcarbodiimide (413 mg,
2.0 mmol) and 4-(dimethy1amino)pyridine (25 mg, 0.2 mmol) were added.
After cooling to 0 °C, levulinic acid (205 µL, 2.0 mmol) was added. The
mixture was stirred at 0 °C for 3h and 5 mL methanol was added, followed
by addition of hexane (10 mL). After filtration of DCU, the solution was
concentrated. The residue was purified by chromatography on silica gel
using ethyl acetate, cyclohexane, triethylamine from 2:7:1 to 5:5:1 v/v/v
and compound 9 was obtained as viscous syrup (689 mg, 66%). TLC Rf:
0.20 cyclohexane/EtOAc/Et3N (5:4:1 v/v/v). 1H NMR (400 MHz, CDCl3) δ
7.49 (m, 2H, Ar), 7.38 (m, 4H, Ar), 7.34 (m, 2H, Ar), 7.26 (t, J = 7.4 Hz, 1H,
Ar), 6.90 (d, J = 8.8 Hz, 4H, Ar), 4.24 (s, 2H, COOCH2), 3.83 (s, 6H, CH3O),
3.53 (s, 2H, DMTrOCH2), 3.13 (q, J = 9.2 Hz, 2H, CCH2), 2.76 (t, J = 6.4
Hz, 2H, CH3COCH2), 2.58 (t, J = 6.6 Hz, 2H, CH2COO), 2.21 (s, 3H,
OCCH3), 0.98 (s, 3H, Me). 13C NMR (100 MHz, CDCl3) δ 206.6, 172.9,
158.4, 144.8, 135.8, 130.0, 128.0, 127.8, 126.8, 113.1, 86.1, 66.8, 66.5,
66.0, 55.1, 40.6, 37.9, 29.7, 27.9, 17.4. HR-ESI-QToF MS (positive mode):
m/z calcd. for C31H36O7Na [M+Na]+ 543.2359, found 543.2357.2-[(4,4'-
Dimethoxytrityl)oxymethyl]-2-methylpropan-3-ol-yl Levulinate (2-
cyanoethyl-N,N-diisopropyl) phoshoramidite 10: To a solution of 2-
[(4,4'-dimethoxytrityl)oxymethyl]-2-methylpropan-3-ol-yl levulinate 9 (265
mg, 0.5 mmol) and DIEA (131 µL, 0.75 mmol) in anhydrous CH2Cl2 (6 mL)
was added 2-cyanoethyl-N,N-diisopropylchloro phosphoramidite (123 µL,
0.55 mmol). The resulting mixture was stirred for 1.5h at room temperature.
Water (1 mL) was added and the solution was diluted with CH2Cl2 (15 mL),
and washed with a saturated solution of NaHCO3 (15 mL) then with brine
(15 mL). The organic layer was dried over Na2SO4, filtered and
concentrated. The crude product was purified by silica gel column
chromatography using cyclohexane, AcOEt and triethylamine from 2:7:1
to 4:5:1, v/v/v affording 10 (307 mg, 85%) as clear oil. TLC Rf: 0.53
cyclohexane/AcOEt/NEt3 (5:4:1 v/v/v). 1H NMR (400 MHz, CDCl3) δ 7.41
(m, 2H, Ar), 7.27 (m, 6H, Ar), 7.20 (m, 1H, Ar), 6.81 (m, 4H, Ar), 4.06 (m,
2H, COOCH2), 3.79 (s, 6H, CH3O), 3.71 (m, 2H, POCH2CH2), 3.52 (m, 4H,
CHiPr and DMTrOCH2), 3.01 (m, 2H, POCH2C), 2.68 (m, 2H, COCH2),
2.55 (m, 2H, CH2CN), 2.49 (m, 2H, CH2COO), 2.16 (s, 3H, CH3CO), 1.16
(d, J = 6.8 Hz, 6H, CH3CH), 1.16 (d, J = 6.8 Hz, 6H, CH3CH), 0.98 (d, J=
8.4 Hz, 3H, CH3C). 13C NMR (100 MHz, CDCl3) δ 206.6, 172.6, 158.5,
145.1, 136.2, 130.3, 128.4, 127.8, 126.8, 117.8, 113.1, 85.8, 66.8, 66.1,
64.8, 58.3, 55.3, 43.1, 40.6, 38.0, 30.0, 28.0, 24.7, 20.5, 17.1. 31P NMR
(121 MHz, CDCl3) δ 148.0, 147.90. HR-ESI-QToF MS (positive mode): m/z
calcd. for C40H56N2O9P [M+H3O]+ 739.3723, found 739.3726.Solid phase
synthesis of Cy3-galactocluster siderophore derivatives
237.0991.N-(3-azidopropyl)-2,3-diacetoxybenzamide
3:
N-(3-
azidopropyl)-2,3-dihydroxybenzamide (385 mg, 1.63 mmol) was
2
solubilized in anhydrous pyridine (10 mL) and acetic anhydride (3.0 mL,
32.6 mmol) was added. The solution was stirred for 2h at room
temperature. AcOEt was added (100 mL) and organic layer was washed
twice with a 1M HCl aqueous solution (2 x 60 mL), twice with a saturated
solution of NaHCO3 (2 x 60 mL) and once with brine (60 mL). Organic layer
was dried over Na2SO4 and solvent was evaporated under vacuum. The
crude was purified by silica gel flash chromatography using cyclohexane
and AcOEt (from 20% to 50%) to afford compound 3 (284 mg, 54%) as a
white solid. TLC Rf: 0.22 Cyclohexane/EtOAc (4:6 v/v). 1H NMR (400 MHz,
CDCl3) δ 7.53 (dd, J = 6.1, 3.3 Hz, 1H, Ar), 7.30 – 7.27 (m, 2H, Ar), 6.39
(t, J = 6.6 Hz, 1H, NH), 3.46 (q, J = 6.6 Hz, 2H, NCH2CH2), 3.40 (t, J = 6.6
Hz, 2H, CH2CH2N3), 2.31 (s, 3H, OCCH3), 2.29 (s, 3H, OCCH3), 1.84 (p, J
= 6.6 Hz, 2H, NCH2CH2CH2N3). 13C NMR (101 MHz, CDCl3) δ 168.2, 168.1,
165.4, 143.1, 140.2, 130.5, 126.7, 126.5, 125.9, 49.4, 37.7, 28.8, 20.7,
20.6. HR-ESI-QToF MS (positive mode): m/z calcd. for C14H17N4O5 [M+H]+
321.1199, found 321.1200.
N-acetoxyacetamide 5:[18] Acetohydroxamic acid 4 (1.0 g, 13.3 mmol)
was dispersed in heterogeneous mixture of CH2Cl2/NaOH 2M (1:1 v/v 14
mL). Acetic anhydride (1.9 mL, 19.9 mmol) was added and the mixture
was stirred at room temperature for 2h. Aqueous layer was extracted four
times with CH2Cl2. Organic layer was dried over Na2SO4 and solvent
evaporated (yellow oil). Product 5 was obtained as a colorless oil (1.17 g,
75%) after purification by chromatography on silica gel using CH2Cl2 with
0 to 5% of MeOH. TLC Rf: 0.35 CH2Cl2/MeOH (95:5 v/v). 1H NMR (400
MHz, CDCl3) δ 9.13 (s, 1H, NH), 2.22 (s, 3H, CH3), 2.05 (s, 3H, CH3). 13
C
NMR (101 MHz, CDCl3) δ 169.0, 19.8, 18.4. HR-ESI-QToF MS (negative
mode): m/z calcd for C4H6NO3 [M-H]- 116.0348, found 116.0346.
N-(4-bromobutyl)-N-acetoxyacetamide 6: To N-acetoxyacetamide 5
(400 mg, 3.4 mmol) solubilized in anhydrous DMF (16 mL) Cs2CO3 (2.2 g,
6.8 mmol) was added. The mixture was sonicated for
5 min.
Dibromobutane (4 mL, 34 mmol) was added and the mixture was stirred
at 100 °C for 2h under microwaves assistance. DMF was evaporated and
CH2Cl2 (10 mL) was added, organic layer was washed twice with water.
Organic layer was dried over Na2SO4 and evaporated. The crude was
purified by flash chromatography on silica gel with cyclohexane/AcOEt (1/0
to 0/1) to afford 6 as colorless oil (300 mg, 35%). TLC Rf: 0.35
cyclohexane/AcOEt (2:8, v/v). 1H NMR (500 MHz, CDCl3) δ 4.16 (t, J = 6.3
Hz, 2H, NCH2CH2), 3.43 (t, J = 6.7 Hz, 2H, CH2CH2Br), 2.14 (s, 3H,
OCCH3), 2.03 (s, 3H, NOCCH3), 2.00 – 1.94 (m, 2H, CH2CH2Br), 1.89 –
1.82 (m, 2H, NCH2CH2). 13C NMR (101 MHz, CDCl3) δ 168.8, 168.5, 66.8,
33.2, 29.4, 27.4, 19.7, 15.0. HR-ESI-QToF MS (positive mode): m/z calcd
for C8H14NO3BrNa [M+Na]+ 274.0055, found 274.0055.
General procedure for immobilization on azide solid support 11 of
propargyl α-D-mannoside 12 by CuAAC: An aqueous solution of
propargyl mannoside (0.2M, 300 μL, 60 µmol), freshly prepared aqueous
solutions of CuSO4 (0.04M, 250 μL, 10 µmol) and sodium ascorbate (0.1M,
500 μL, 50 µmol), THPTA (0.1M, 300 μL, 30 µmol) and TEAAc buffer (2M,
500 μL) were added to 10 μmol of azide solid support 11. The mixture was
stirred at room temperature for 2h. The solution was removed, and CPG
beads were washed with H2O (10 mL), MeOH (10 mL), and CH2Cl2 (10
mL) and dried.
N-(4-azidobutyl)-N-acetoxyacetamide
7:
N-(4-bromobutyl)-N-
acetoxyacetamide 6 (88 mg, 0.35 mmol) was coevaporated twice with
anhydrous CH3CN. The residue was solubilized in anhydrous CH3CN (2
mL) and tetramethylguanidinium azide (TMG N3) (111 mg, 0.7 mmol) was
added. The mixture was stirred at 80 °C for 1h30 under microwaves.
CH3CN was evaporated and the residue was purified by flash
chromatography on silica gel with cyclohexane/AcOEt (1/1 to 0/1) to afford
7 as a colorless oil (44 mg, 59%). TLC Rf: 0.19 cyclohexane/AcOEt (3:7
v/v), 1H NMR (400 MHz, CDCl3) δ 4.14 (t, J = 6.2 Hz, 2H, NCH2CH2), 3.31
(t, J = 6.6 Hz, 2H, CH2CH2N3), 2.13 (s, 3H, OCCH3), 2.02 (s, 3H, NOCCH3),
1.80 – 1.72 (m, 1H, NCH2CH2), 1.72 – 1.63 (m, 1H, CH2CH2N3). 13C NMR
(101 MHz, CDCl3) δ 168.8, 168.4, 67.0, 51.1, 26.0, 25.6, 19.6, 14.9. HR-
ESI-QToF MS (positive mode): m/z calcd for C8H14N4O3Na [M+Na]+
37.0964, found 237.0963.
General
procedure
for
incorporation
of
EG2propargyl
phosphoramidite 14: The solid-supported mannoside 13 (1 μmol) was
treated by phosphoramidite chemistry, on a DNA synthesizer (ABI 394),
with EG2propargyl phosphoramidite 14. Only coupling and oxidation steps
were performed. For the coupling step, benzylmercaptotetrazole was used
as an activator (0.3M in anhydrous CH3CN) and EG2Propargyl
phosphoramidite (0.2M in anhydrous CH3CN) was introduced three times
with a 180s coupling time. Oxidation was performed with 0.1M commercial
solution of iodide (0.1M in THF/pyridine/H2O, 78:20:2, v/v/v) for 15s.
General procedure for phosphoramidite chemistry: Solid-supported
(propagylEG2)4-mannoside (1 μmol) was treated by phosphoramidite
chemistry, on a DNA synthesizer. Detritylation step was performed with
3% TCA in CH2Cl2 for 65s. For the coupling step, benzylmercaptotetrazole
(0.3M in anhydrous CH3CN) was used with THME monolevulinyl 10,
2-[(4,4'-Dimethoxytrityl)oxymethyl]-2-methylpropan-3-ol-yl
Levulinate 9: 2-[(4,4'-Dimethoxytrityl)oxymethyl]-2-methylpropane-1,3-
10
This article is protected by copyright. All rights reserved.

10.1002/cbic.202000490
ChemBioChem
FULL PAPER
THME monopropargyl 22a, pentaerythritol dipropargyl 22b, pentaerythritol
tripropargyl 19 phosphoramidite (0.1M in anhydrous CH3CN) with a 60s
coupling time and Cy3 amidite (0.1M in anhydrous CH3CN) with a 180s
coupling time. The capping step was performed with commercial solution
acetic anhydride (Cap A, Ac2O/pyridine/THF, 5:10:85, v/v/v; Cap B, 10%
N-methylimidazole in THF) for 10s. Oxidation was performed with 0.1M
commercial solution of iodide for 15s. Synthesis was performed with Trityl
ON mode.
7.52 (m, 1H, Ar), 7.52 – 7.48 (m, 2H, Ar), 7.45 – 7.36 (m, 3H, Ar), 7.33 (t,
J = 7.9 Hz, 2H, Ar), 6.49 (t, J = 4.8 Hz, 1H, NH), 3.40 (dd, J = 12.7, 6.7 Hz,
2H, NCH2CH2), 3.21 (t, J = 6.7 Hz, 2H, CH2CH2N3), 1.66 (p, J = 6.7 Hz,
2H, NCH2CH2CH2N3). 13C NMR (101 MHz, CDCl3) δ 165.2, 164.4, 164.2,
143.3, 140.4, 134.4, 133.9, 130.9, 130.3, 130.2, 128.9, 128.6, 128.5, 128.1,
127.2, 127.0, 126.1, 49.2, 37.5, 28.8. HR-ESI-QToF MS (positive mode):
m/z calcd for C24H21N4O5 [M + H]+ 445.1512, found 445.1519
Triazolyl-catecholamide mannopyranoside 27: The propargyl-
diethyleneglycol mannopyranoside 26 (520 mg, 1.7 mmol) was solubilized
in dioxane/H2O (3:1 v/v, 17 mL) and N-(3-azidopropyl)-2,3-
dibenzoxybenzamide 25 (978 mg, 2.2 mmol), copper nanopowder (~ 4 mg)
and TEAAc (2M, 500 µL) were added. The mixture was stirred at 55 °C
overnight. Solvent was evaporated under vacuum and the crude was
purified by silica gel flash chromatography (CH2Cl2/MeOH, 75:15 v/v).
Product was treated with Quadrapure® IDA for 6h. The mixture was filtered
and filtrate evaporated under vacuum. The residue was solubilized in a
minimum of dioxane for lyophilization to obtain 27 as a white solid (1.11 g,
87%). TLC Rf: 0.15 CH2Cl2/MeOH (9:1, v/v). 1H NMR (400 MHz, CDCl3) δ
8.05 – 7.94 (m, 4H, Ar), 7.64 (s, 1H, Tz), 1H, 7.59 (d, J = 8.1 Hz, 1H, Ar),
7.55 – 7.47 (m, 2H, Ar), 7.43 (dd, J = 8.1 , 1.3 Hz, 1H, Ar), 7.37 – 7.28 (m,
5H, Ar), 7.24 (t, J = 5.6 Hz, 1H, NH), 4.89 (s, 1H, OH), 4.84 (s, 1H, H1’),
4.75 (s, 1H, OH), 4.57 (s, 2H, OCH2Tz), 4.25 (t, J = 6.5 Hz, 2H, TzCH2CH2),
3.93 – 3.76 (m, 4H, H2’,H4’,H6a’,H3’), 3.76 – 3.66 (m, 2H, H6b’, OCHHCH2),
3.66 – 3.49 (m, 8H, OCHHCH2, OCH2, H5’), 3.27 (dd, J = 11.5, 5.6 Hz, 2H,
NCH2CH2), 2.34 (s, 1H, OH), 1.98 – 1.88 (m, 2H, NCH2CH2CH2Tz). 13C
NMR (126 MHz, CDCl3) δ 166.0, 164.4, 164.4, 144.8, 143.4, 140.5, 134.3,
134.1, 131.0, 130.3, 130.3, 128.9, 128.7, 128.4, 128.2, 126.9, 126.6, 126.0,
123.8, 100.2, 72.4, 71.6, 70.9, 70.7, 70.3, 69.8, 67.0, 66.7, 64.5, 61.4, 47.6,
36.8, 30.1. HR-ESI-QToF MS (positive mode): m/z calcd for C37H43N4O13
[M + H]+ 751.2821, found 751.2848.
General procedure for delevulinylation: The CPG beads were treated
with a solution of 0.5M hydrazinium acetate (H2NNH2-H2O/pyridine/AcOH,
0.124:4:1, v/v/v) for 30min, washed with pyridine, acetonitrile and CH2Cl2,
and dried.
General procedure for introduction of galactoside azide derivative by
CuAAC on solid support: In the column synthesis (1 μmol) were added
the acetylated galactoside azide derivative 17 (0.1M in dioxane, 80 μL, 8
µmol), THPTA (0.1M in H2O, 30 μL, 30 µmol), dioxane (20 μL) and a freshly
prepared aqueous solutions of CuSO4 (0.04M, 25 μL, 0.4 µmol) and
sodium ascorbate (0.1M, 50 μL, 5 µmol), DNA column was vortexed at
60 °C in oven for 3h. The CPG beads were washed with dioxane, H2O,
MeOH, CH2Cl2 and dried under vacuum.
General procedure for deprotection and release from solid support:
The CPG beads were transferred to a 4 mL screw top vial and treated with
2 mL of concentrated aqueous ammonia at room temperature overnight.
The supernatant was withdrawn and evaporated. Crude was purified by
C18 reversed phase HPLC. Pure product was co-evaporated several times
with H2O and then lyophilized.
General procedure for introduction of catechol or hydroxamate azide
by CuAAC in solution: To a solution of Cy3 alkyne-galactocluster (1 mM
in H2O, 200 μL, 200 nmol) were added siderophore azide 3 or 7 (0.1M in
dioxane, 2 eq/alkyne), THPTA (0.1M in H2O, 6 μL, 600 nmol), dioxane (140
μL) and copper nanopowder ( ~ 1 mg). The mixture was stirred at room
temperature for 2h, then after centrifugation, the supernatant was treated
with EDTA solution (complete to 1 mL) and the mixture was purified two
times by steric exclusion column (NAP 10). The conjugate was lyophilized
in H2O and acetyl groups were hydrolyzed by NEt3/MeOH/H2O (700 μL,
1:5:1, v/v/v) under stirring for 2h at room temperature. The siderophore-
glycocluster conjugate was obtained after several co-evaporations with
water and lyophilization from water.
Mannopyranoside mono-catechol 28: To triazolyl-catecholamide
mannopyranoside 27 (232 mg, 0.3 mmol) in anhydrous CH3CN (1.5 mL),
propargyldiethyleneglycol phosphoramidite 14 (707 mg, 2 mmol) was
added and mixture was dried over molecular sieve 3Å for 1h. A solution of
tetrazole (0.4M, 7.5 mL, 3 mmol) was added and the mixture was stirred
at room temperature for 4h. Water (1 mL) was added and after 5min
Amberlyst ® A26 IO4- resin (2.49 mmol/g, 1.2 g, 3 mmol) was added, the
mixture was stirred for 2h. After filtration and dilution in CH2Cl2, the organic
layer was washed with saturated solution of NaHCO3 and brine. Organic
layer was dried over Na2SO4 and solvent was evaporated under vacuum
to afford crude 28 (yellow oil, 614 mg, quantitative) which was used for the
next step without further purification. TLC Rf: 0.22 CH2Cl2/MeOH (95:5,
2,3-Dibenzoxybenzoic acid 24: 2,3-Dihydroxybenzoic acid (100 mg, 0.6
mmol) was solubilized in pyridine (4 mL). Benzoic anhydride (440 mg, 1.9
mmol) was added and the reaction mixture was stirred at room
temperature overnight. MeOH (6 mL) was added and solvent evaporated
under vacuum. The residue was purified on flash silica gel chromatography
with cyclohexane and AcOEt (1:0 to 0:1 v/v) to afford 24 as a colorless oil
(201 mg, 93%). TLC Rf: 0.23 cyclohexane/AcOEt (3:7, v/v). 1H NMR (400
MHz, CDCl3) δ 8.08 – 8.00 (m, 5H, Ar), 7.62 (dd, J = 8.0, 1.5 Hz, 1H, Ar),
7.57 – 7.48 (m, 2H, Ar), 7.43 (t, J = 8.0 Hz, 1H, Ar), 7.39 – 7.31 (m, 4H,
Ar). 13C NMR (101 MHz, CDCl3) δ 168.3, 164.4, 164.3, 144.2, 143.5, 134.0,
133.7, 130.5, 130.4, 130.4, 129.7, 128.7, 128.7, 128.6, 126.4. HR-ESI-
QToF MS (negative mode): m/z calcd for C21H13O6 [M - H]- 361.07176,
found 361.07025
1
v/v). H NMR (400 MHz, CD3CN) δ 8.03 – 7.95 (m, 4H, Ar), 7.69 (s, 1H,
Tz), 7.65 – 7.58 (m, 3H, Ar), 7.55 (dd, J = 8.1, 1.8 Hz, 1H, Ar), 7.52 – 7.38
(m, 5H, Ar), 7.15 (t, J = 6.1 Hz, 1H, NH), 5.16 – 5.07 (m, 1H, H1’), 4.86 –
4.79 (m, 1H, H3’), 4.75 – 4.51 (m, 5H, H4’, H2’, OCH2Tz), 4.42 – 3.75 (m,
40H, OCH2, CH2Tz, CH2CH2CN, POCH2, CH2CCH, H5’), 3.75 – 3.44 (m,
44H, H6’, OCH2), 3.26 (m, 2H, NCH2CH2), 2.91 – 2.76 (m, 8H, CH2CH2CN),
2.71 (t, J = 3.6 Hz, 4H, CH2CCH), 2.01 (p, J = 6.8 Hz, 2H, NCH2CH2CH2Tz).
13C NMR (101 MHz, CD3CN) δ 166.0, 165.2, 164.9, 145.3, 144.4, 141.4,
135.1, 135.1, 132.5, 130.9, 130.8, 129.8, 129.8, 129.5, 129.3, 127.8, 127.3,
126.7, 124.5, 118.7, 98.2, 80.9, 75.8, 75.6, 75.0, 71.8, 71.1, 70.9, 70.5,
69.9, 68.73, 68.4, 68.3, 66.6, 64.8, 63.9, 63.4, 58.7, 48.3, 37.4, 30.8, 20.2.
31P NMR (162 MHz, CD3CN) δ -1.69, -1.74, -2.18, -2.35 (PO), -2.42, -2.66,
-2.73 (POO-). HR-ESI-QToF MS (positive mode): m/z calcd for C77H99
N8O33P4 [M + H]+ 1787.5265, found 1787.5258.
N-(3-Azidopropyl)-2,3-dibenzoxybenzamide 25: To a cold solution (~ -
10 °C) of 24 (2.2 g, 6 mmol) in dry CH2Cl2 (60 mL) and DIEA (2.1 mL, 12
mmol) was added dropwise ethyl chloroformate (914 µL, 9.6 mmol). After
15min stirring at -10 °C, azidopropylamine (901 mg, 9 mmol) and DIEA
(1.05 mL, 6 mmol) were added. 15min after addition the solution was
allowed to warm up to room temperature and stirred for 1h then the
solution was applied on a silica gel column and chromatographed using an
increasing amount of AcOEt (0 to 80%) in cyclohexane to obtain
compound 25 (white solid, 1.8 g, 67%). TLC Rf: 0.54 cyclohexane/AcOEt
(4:6, v/v). 1H NMR (400 MHz, CDCl3) δ 8.06 (dd, J = 8.3, 1.2 Hz, 2H, Ar),
7.99 (dd, J = 8.3, 1.2 Hz, 2H, Ar), 7.75 (dd, J = 7.7, 1.7 Hz, 1H, Ar), 7.60 –
Galactocluster mono-catechol G1C: Tetra-propargyl mannopyranoside
mono-catechol 28 (536 mg, 0.3 mmol) was solubilized in dioxane /H2O (1:1
v/v, C = 0.05M, 6 mL) and acetylated galacto-azide derivative 17a (784
mg, 1.5 mmol) was added followed by copper nanopowder (~ 2 mg) and
TEAAc (2M, 1 mL). Mixture was stirred at 55 °C for 24h (the reaction was
monitored by MALDI-TOF spectrometry). The crude was purified by silica
gel flash chromatography (CH2Cl2/MeOH, 85:15 v/v) to obtain protected
galactocluster mono-catechol (yellow solid, 863 mg, 74%) after treatment
with Quadrapure® IDA (500 mg) overnight. Solution was filtered and
11
This article is protected by copyright. All rights reserved.

10.1002/cbic.202000490
ChemBioChem
FULL PAPER
filtrate lyophilized. 1H NMR (400 MHz, CD3CN) δ 9.02 (s, 4H, NH), 8.00 –
7.94 (m, 4H, Ar), 7.92 – 7.80 (m, 4H, Tz), 7.69 (s, 1H, Tz), 7.64 – 7.32 (m,
17H, Ar), 7.25 (s, 1H, NH), 6 .97 (d, J = 8.9 Hz, 8H, Ar), 5.44 – 5.37 (m,
4H, H4’gal), 5.31 – 5.06 (m, 25H, H2’gal, H3’gal, OCCH2Tz, H1’gal, H1’man),
4.87 – 4.80 (m, 1H, H2’man), 4.73 – 4.63 (m, 1H, H3’man), 4.63 – 4.44 (m,
11H, OCH2Tz, H4’man), 4.32 – 4.00 (m, 38H, TzCH2CH2, CH2CH2CN,
OPCH2CH2, H6’gal, H5’gal, H6’man, H-5’man), 3.68 – 3.46 (m, 49H, OCH2),
3.24 (q, J = 6.1 Hz, 2H, NCH2CH2), 2.85 – 2.70 (m, 8H, CH2CH2CN), 2.19
– 2.08 (m, 14H, NCH2CH2CH2Tz, OCCH3), 2.02 (s, 12H, OCCH3), 1.97 (s,
12H, OCCH3), 1.94 (s, 12H, OCCH3). 13C NMR (101 MHz, CD3CN) δ 171.2,
171.1, 170.8, 170.5, 165.1, 154.5, 145.6, 135.1, 134.5, 130.8, 129.8, 127.8,
127.3, 126.2, 122.3, 118.7, 100.3, 72.0, 71.5, 71.1, 70.3, 69.6, 68.3, 67.7,
64.8, 64.0, 62.3, 53.4, 48.4, 37.5, 30.8, 20.9, 20.2. 31P NMR (202 MHz,
CDCl3) δ -2.70 (PO). MALDI-TOF MS (negative mode, THAP): m/z calcd
for C165H201N24O77P4Na [M – 2H + Na]- 3899.40, found 3900.25, for
C162H198N23O77P4 [(M – Cne) - H]- 3823.35, found 3822.20. HR-ESI-QToF
MS (positive mode): m/z calcd for C165H204N24O77P4 [(M + 2H)/2]+
1938.5868, found 1938.5840.
phosphoramidite chloride (3.7 mmol, 875 µL) was added dropwise and the
mixture was stirred for 1h at room temperature. H2O (1 mL) was added,
after 5min organic layer was washed twice with a saturated solution of
NaHCO3. Organic layer was dried over Na2SO4 and solvent evaporated
under vacuum. The residue was purified on silica gel flash chromatography
(CH2Cl2/AcOEt/10% NEt3, 9:1 v/v 10% NEt3) to obtain phosphoramidite 30
(white solid, 2.28 g, 78%). TLC Rf: 0.2 (CH2Cl2/AcOEt/10% NEt3, 7:2, v/v
10% NEt3). 1H NMR (400 MHz, CDCl3) δ 8.45 (s, 1H, NH), 7.80 (s, 1H, Tz),
7.41 (d, J = 9.0 Hz, 2H, Ar), 6.94 (d, J = 9.0 Hz, 2H, Tz), 5.51 – 5.40 (m,
2H, H2’, H4’), 5.17 (s, 2H, OCCH2Tz), 5.09 (dd, J = 10.5, 3.4 Hz, 1H, H3’),
4.98 (d, J = 8.0 Hz, 1H, H1’), 4.70 (s, 2H, OCH2Tz), 4.27 – 4.10 (m, 2H,
H6’), 4.04 (t, J = 6.9 Hz, 1H, H5’), 3.89 – 3.75 (m, 2H, POCH2CH2CN), 3.75
– 3.62 (m, 8H, OCH2), 3.62 – 3.50 (m, 2H, NCH), 2.62 (t, J = 6.0 Hz, 2H,
POCH2CH2CN), 2.16 (s, 3H, OCCH3), 2.05 (s, 3H, OCCH3), 2.04 (s, 3H,
OCCH3), 2.00 (s, 3H, OCCH3), 1.16 (d, J = 6.7 Hz, 6H, NCHCH3), 1.14 (d,
J = 6.7 Hz, 6H, NCHCH3). 13C NMR (101 MHz, CDCl3) δ 170.5, 170.5,
170.2, 169.5, 163.3, 154.1, 145.8, 132.6, 124.8, 121.9, 118.1, 117.7, 100.1,
71.4, 71.2, 70.9, 70.7, 70.1, 68.8, 67.0, 64.7, 62.8, 62.6, 61.5, 58.7, 58.5,
53.5, 43.2, 43.1, 24.8, 24.7, 24.7, 20.9, 20.8, 20.8, 20.7, 20.5, 20.4. 31P
NMR (162 MHz, CDCl3) δ 148.51. HR-ESI-QToF MS (positive mode): m/z
calcd for C38H56N6O15P [M + H]+ 867.3536, found = 867.3525.
The protected galactocluster mono-catechol (0.15 mmol, 595 mg) was
stirred at room temperature overnight in NEt3/MeOH/H2O (1:5:1 v/v/v, 21
mL). Solvents were evaporated under vacuum. The crude was purified on
reverse phase flash chromatography (H2O/1% CH3CN 25 mM TEAAc -
H2O/20% CH3CN 25 mM TEAAc). Triethylammonium ion were exchanged
to Na+ by treatment with DOWEX® 50W X8 Na+ form resin. The product
was lyophilized in H2O to give galactocluster mono-catechol G1C (beige
solid, 321 mg, 74%). 1H NMR (400 MHz, D2O) δ 8.18 – 7.75 (m, 5H, Tz),
7.42 – 7.24 (m, 8H, Ar), 7.11 – 6.88 (m, 11H, Ar), 6.86 – 6.60 (m, 1H, NH),
5.48 – 5.16 (m, 8H, OCCH2Tz), 5.16 – 5.05 (m, 1H, H1’man), 5.02 – 4.88
(m, 5H, sugar), 4.73 – 4.25 (m, 21H, OPCH2, OCH2Tz, sugar), 4.20 – 3.40
(m, 88H, OCH2, sugar), 3.40 – 3.30 (m, 2H, NCH2CH2), 2.23 – 2.08 (m,
2H, NCH2CH2CH2Tz). 13C NMR (101 MHz, D2O) δ 181.5, 165.8, 154.2,
144.1, 131.4, 126.5, 123.1, 117.0, 101.0, 75.3, 72.5, 70.5, 70.3, 69.5, 68.9,
68.5, 64.9, 63.1, 60.7, 52.4, 23.3. 31P NMR (162 MHz, D2O) δ 0.69, -0.60,
-0.85 (PO). C18 HPLC (1% to 24% CH3CN 50 mM TEAAc over 20min):
Rt= 13.2 min. MALDI-TOF MS (negative mode, THAP): m/z calcd for
C107H149N20O59P4 [M-H]- 2783.34, found 2783.22, and calcd for
Galactocluster diethyleneglycol 32: The propargyl-diethyleneglycol
mannopyranoside 26 (92 mg, 0.3 mmol) and galactopyranoside
phosphoramidite 30 (1.81 g, 2.1 mmol) were solubilized in dry CH3CN (12
mL) and mixture was dried over molecular sieve 3Å for 1h. Then BMT (519
mg, 2.7 mmol) as activator was added and the mixture was stirred at room
-
temperature for 4h. H2O was added and after 5min, Amberlyst ® A26 IO4
resin (2.49 mmol/g, 1 g) was added, the mixture was stirred for 2h. After
filtration, solvent was evaporated under vacuum (MALDI-TOF MS (positive
mode, THAP): m/z calcd for C141H183N20O72P4 [M+H]+ 3433.98, found
3434.05). The crude product was treated with 30% ammonia for 2h at room
temperature. After evaporation under vacuum, the residue was purified on
reverse phase flash chromatography (H2O/1% CH3CN 25 mM TEAAc -
80% CH3CN 25 mM TEAAc) and lyophilized in H2O to afford product 32
1
(yellow solid, 409 mg, 54%). H NMR (500 MHz, D2O) δ 8.08 – 8.03 (m,
4H, Tz), 7.33 (d, J = 8.9 Hz, 8H, Ar), 7.04 (d, J = 7.7 Hz, 8H, Ar), 5.36 –
5.28 (m, 8H, OCCH2Tz), 5.01 (s, 1H, H1’man), 4.92 (d, J = 7.6 Hz, 4H, H-
1’gal), 4.70 – 4.54 (m, 9H, OCH2Tz, H’man), 4.40 – 4.28 (m, 1H, H’man),
4.28 – 4.20 (m, 1H, H’man), 4.12 (d, J = 2.2 Hz, 2H, CH2CCH), 4.08 – 3.98
(m, 6H, POCHH, H’man), 4.00 – 3.90 (m, 7H, POCHH, H’gal), 3.91 – 3.83
(m, 2H, H’man), 3.83 – 3.51 (m, 54H, H’gal, OCH2), 2.81 (t, J = 2.2 Hz, 1H,
CH2CCH). 13C NMR (126 MHz, D2O) δ 180.7, 165.1, 153.4, 143.4, 137.0,
130.5, 125.7, 122.4, 116.1, 100.1, 97.0, 78.6, 71.7, 69.7, 69.3, 68.7, 68.6,
68.1, 67.7, 67.6, 64.0, 63.9, 62.2, 59.8, 57.0, 51.6, 36.7. 31P NMR (202
MHz, D2O) δ 0.52, -0.74, -0.80, -0.97. HPLC C18 (1% to 24% CH3CN 50
mM TEAAc over 20min): 11.8 min. MALDI-TOF MS (negative mode,
THAP): m/z calcd for C97H137N16O56P4 [M-H]- 2547.11, found 2547.38. HR-
ESI-QToF MS (positive mode): m/z calcd for C97H140N16O56P4 [(M + 2H)/2]+
1274.3769, found 1274.3739.
C
107H149N20O59P4Na [M - 2H + Na]- 2805.33, found 2805.18. HR-ESI-QToF
MS (positive mode): m/z C107H152N20O59P4 calcd for [(M+2H)/2]+
1392.4230, found 1392.4210 and calcd for C107H153N20O59P4 [(M+3H)/3]+
928.6179, found 928.6185.
Triazolyl-diethyleneglycol
galactopyranoside
29:
Propargyl
diethyleneglycol (1 mmol, 144.2 mg) was solubilized in dioxane/H2O (5:1
v/v, 12 mL) and galactoside azide derivative 17a (1 mmol, 522 mg) was
added followed by copper nanopowder (0.016 mmol, 1 mg) and TEAAc
(2M, 200 µL). The mixture was stirred at 55 °C overnight. Solvent was
evaporated under vacuum and the crude was purified on silica gel flash
chromatography (CH2Cl2/MeOH, 95:5 v/v) to obtain 29 (white solid, 545
mg, 82%). TLC Rf: 0.2 (CH2Cl2/MeOH, 9:1 v/v). 1H NMR (400 MHz, CDCl3)
δ 8.61 (s, 1H, NH), 7.81 (s, 1H, Tz), 7.41 (d, J = 8.9 Hz, 2H, Ar), 6.94 (d, J
= 8.9 Hz, 2H, Ar), 5.50 – 5.39 (m, 2H, H2’, H4’), 5.18 (s, 2H, OCCH2Tz),
5.10 (dd, J = 10.5, 3.4 Hz, 1H, H3’), 4.99 (d, J = 7.9 Hz, 1H, H1’), 4.70 (s,
2H, OCH2Tz), 4.21 (dd, J = 11.3, 7.0 Hz, 1H, H6a’), 4.14 (dd, J = 11.3, 6.2
Hz, 1H, H6b’), 4.04 (t, J = 6.7 Hz, 1H, H5’), 3.79 – 3.65 (m, 6H, OCH2), 3.65
– 3.52 (m, 2H, OCH2), 3.04 (s, 1H, OH), 2.17 (s, 3H, OCCH3), 2.06 (s, 3H,
OCCH3), 2.04 (s, 3H, OCCH3), 2.00 (s, 3H, OCCH3). 13C NMR (101 MHz,
CDCl3) δ 170.5, 170.4, 170.3, 169.6, 163.4, 154.1, 145.5, 132.7, 125.1,
122.0, 117.7, 100.1, 72.6, 71.2, 71.0, 70.5, 70.1, 68.8, 67.0, 64.5, 61.7,
61.5, 53.6, 20.9, 20.8, 20.8, 20.7. HR-ESI-QToF MS (positive mode): m/z
calcd for C29H39N4O14 [M + H]+ 667.2457, found 667.2461.
Tosyl-triethyleneglycol 2-cyanotethyl diisopropyl phosphoramidite
34: To a solution of tosyl-triethyleneglycol[33] (460 mg 1.5 mmol) and DIEA
(392 µL, 2.25 mmol) in anhydrous CH2Cl2 (25 mL) was added 2-
cyanoethyl-N,N-diisopropylchloro phosphoramidite (334 µL, 1.5 mmol).
The resulting mixture was stirred for 1h at room temperature. Water (1 mL)
was added and the solution was diluted with CH2Cl2 (40 mL), and washed
with a saturated solution of NaHCO3 (40 mL) then with brine (42 mL). The
organic layer was dried (Na2SO4), filtered, and concentrated. The crude
product was purified by silica gel column chromatography using
cyclohexane and ethyl acetate 90:10 to 70:30 v/v with 4% of triethylamine
affording tosyl-triethyleneglycol phosphoramidite 34 (486 mg, 64%) as a
colorless oil. 1H-NMR (CDCl3, 300 MHz) 7.80 (d, 2H, J = 8.3Hz, Ar), 7.34
(d, 2H, J = 8.2Hz, Ar), 4.2-4.18 – 4.09 (m, 2H, TosOCH2), 3.91 - 3.44 (m,
14H, NCH, POCH2CH2CN, OCH2), 2.64 (t, 2H, J = 6.5Hz, POCH2CH2CN),
2.45 (s, 3H, PhCH3), 1.18 (d, 6H, J = 6.5Hz, CHCH3), 1.17 (d, 6H, J =
6.5Hz, CHCH3). 13C-NMR (CDCl3, 101 MHz):  145.0, 133.1, 130.0,
128.1, 118.0, 71.4, 71.4, 71.0, 70.7, 69.4, 68.8, 62.8, 62.6, 58.7, 58.5, 53.6,
43.2, 43.1, 24.8, 24.7, 24.8, 21.8, 20.5, 20.4. 31P-NMR (CDCl3, 121 MHz):
Triazolyl-diethyleneglycol galactopyranoside phosphoramidite 30:
The galactoside derivative 29 (3.36 mmol, 2.24 g) was co-evaporated
twice with anhydrous CH3CN, and product was solubilized in anhydrous
CH2Cl2 (45 mL). Anhydrous DIEA (4.7 mmol, 820 µL) was added and
mixture was dried over molecular sieve (4 Å) for 2h under argon
atmosphere and CaCl2 guard. At
0 °C, cyanoethyl-N,N-diisopropyl
12
This article is protected by copyright. All rights reserved.

10.1002/cbic.202000490
ChemBioChem
FULL PAPER
δ 148.7 ppm. HR-ESI-QToF MS (positive mode): m/z calcd for
give 37 (white solid, 447 mg, 29%). 1H NMR (CDCl3, 400 MHz) δ 8.03 –
7.93 (m, 12 H, Ar), 7.84 (s, 3H, Tz), 7.67 – 7.60 (m, 3H, Ar), 7.55 – 7.46
(m, 6H, Ar), 7.43 – 7.37 (m, 3H, Ar), 7.37 – 7.28 (m, 15H, Ar), 4.58 – 4.50
(m, 6H, OCH2Tz), 4.33 – 4.21 (m, 6H, TzCH2CH2), 3.98 – 3.86 (m, 2H,
POCH2), 3.85 – 3.79 (m, 2H, CH2CH2N3), 3.65 – 3.54 (m, 8H, POCH2C,
OCH2), 3.45 (s, 6H, CCH2O), 3.36 – 3.20 (m, 10H, CH2CH2N, OCH2), 2.78
(q, J = 7.2 Hz, 12H, CH2CH3 triethyl ammonium), 2.06 – 1.89 (m, 6H,
TzCH2CH2CH2N), 1.12 (t, J = 7.3 Hz, 18H, CH2CH3 triethyl ammonium).
13C NMR (CDCl3, 101 MHz) δ 176.4, 166.3, 166.2, 164.4, 164.2, 145.0,
143.2, 140.4, 134.1, 134.0, 131.1, 130.3, 130.1, 128.8, 128.6, 128.2, 128.1,
126.6, 126.3, 125.6, 123.5, 70.4, 69.8, 68.6, 66.1, 64.4, 49.2, 49.0, 48.9,
48.7, 47.5, 45.2, 36.6, 36.5, 29.8, 22.2, 8.3, 8.2. 31P NMR (202 MHz,
CDCl3) δ -0.26. C18 HPLC (32% to 80% CH3CN in 50 mM TEAAc over
20min): 15min. MALDI-TOF MS (negative mode, THAP): m/z calcd for
C92H89N15O24P [M - H]- 1819.78, found 1819.76. HR-ESI-QToF MS
(positive mode): m/z Calcd for C92H91N15O24P [M + H]+ 1820.6099, found
1820.6093, m/z calcd for C92H92N15O24P [(M + 2H)/2]+ 910.8088, found
910.8104.
C22H40N2O8PS [M+H3O]+ 523.2243, found 523.2223.
Tripropargyl-pentaerythrityl
tosyl-triethyleneglycol
cyanoethyl
phosphate 35: To tripropargyl pentaerythritol 33 (112 mg, 0.45 mmol) in
anhydrous CH3CN (2 mL), tosyl-triethyleneglycol phosphoramidite 34 (297
mg, 0.6 mmol) was added and mixture was dried over molecular sieve 3Å
for 1h. A solution of tetrazole (0.4M, 3 mL, 1.2 mmol) was added and the
mixture was stirred at room temperature for 2h. Water was added (1 mL)
-
and after 5 min Amberlyst ® A26 IO4 resin (2.49 mmol/g, 482 mg, 1.2
mmol) was added, the mixture was stirred for 2h. After filtration and dilution
in CH2Cl2, the organic layer was washed with saturated solution of
NaHCO3 and brine. Organic layer was dried over Na2SO4 and solvent was
evaporated under vacuum. Crude was purified on silica gel flash
chromatography (CH2Cl2/MeOH, 95:5 v/v) to obtain 35 (yellow oil, 176 mg,
59%). TLC Rf: 0.43 (CH2Cl2/MeOH, 95:5, v/v). 1H NMR (600 MHz, CDCl3)
δ 7.80 (d, J = 8.3 Hz, 2H, Ar), 7.35 (d, J = 8.0 Hz, 2H, Ar), 4.32 – 4.18 (m,
4H, POCH2CH2CN, POCH2), 4.18 – 4.06 (m, 10H, TosOCH2, POCH2C,
OCH2CCH), 3.72 – 3.67 (m, 4H, OCH2), 3.63 – 3.58 (m, 4H, OCH2), 3.53
(s, 6H, CCH2O), 2.79 (t, J = 6.4 Hz, 2H, POCH2CH2CN), 2.45 (s, 3H,
PhCH3), 2.43 (t, J = 2.4 Hz, 3H, CH2CCH). 13C NMR (151 MHz, CDCl3) δ
145.0, 133.1, 130.0, 128.1, 116.8, 79.8, 74.6, 70.9, 70.7, 69.4, 69.0, 68.3,
67.3, 67.2, 62.0, 58.9, 45.0, 44.9, 21.8, 19.7. 31P NMR (162 MHz, CDCl3)
δ -1.78. HR-ESI-QToF MS (positive mode): m/z calcd for C30H41NO12PS
[M + H]+ 670.2082, found 670.2077.
Benzoylated galactocluster tri-catechol BzG3C: To a solution of
galactocluster mannoside alkyne 32 (400 mg, 0.156 mmol) and tri-catechol
platform 37 (312 mg, 0.172 mmol) in dioxane/H2O (5:1 v/v, 12 mL) and
TEAAc (2M, 60 µL), THPTA (68 mg, 0.156 mmol) and copper nanopowder
(4 mg, 0.063 mmol) were added and stirred for 6 days. After filtration and
evaporation, the crude dissolved in CH3CN/H2O, (1:1 v/v, 5mL) and was
treated with Quadrapure® (800 mg) for 6h. The crude was purified on
reverse phase by flash chromatography (H2O/1% CH3CN 25 mM TEAAc -
80% CH3CN 25 mM TEAAc). The product was freeze-dried to afford
compound BzG3C together with partially deprotected products. A pure
fraction was isolated for characterization. HPLC C18 (1% to 24% CH3CN
50 mM TEAAc over 12 min and 24% to 80% CH3CN 50 mM TEAAc over
7 min): 17.5 min. 1H NMR (600 MHz, D2O) δ 8.12 – 7.96 (m, 8H, Tz, NH),
7.92 – 7.86 (m, 3H, Tz), 7.76 – 7.55 (m, 11H, Ar), 7.54 – 7.40 (m, 19H, Ar),
7.40 – 7.27 (m, 18H, Ar), 7.27 – 7.12 (m, 12H, Ar), 7.14 – 6.96 (m, 25H,
Ar), 6.96 – 6.68 (m, 18H, Ar), 5.39 – 5.14 (m, 17H, OCCH2Tz), 5.10 (s, 3H,
H1’man), 4.98 – 4.89 (m, 11H, H1’gal, H’man), 4.71 – 4.22 (m, 46H,
CCH2Tz, OCH2CH2Tz), 4.19 – 3.92 (m, 39H, CH2CH2Tz, H’gal, POCH2),
3.92 – 3.03 (m, 156H, OCH2, H’gal, POCH2C, CCH2O, NCH2CH2), 1.88 –
1-60 (m, 6H, NCH2CH2CH2Tz). 13C NMR (151 MHz, D2O) δ 181.5, 175.8,
165.3, 154.2, 144.2, 136.3, 131.6, 131.2, 129.7, 128.8, 128.3, 127.4, 126.5,
123.0, 117.0, 101.0, 75.3, 72.6, 70.5, 69.5, 69.0, 68.4, 64.8, 64.7, 64.5,
63.7, 63.1, 60.7, 52.4, 49.8, 47.5, 46.6, 36.4, 29.1, 23.3. 31P NMR (202
MHz, D2O) δ 0.57, 0.34, -0.82, -1.02 (PO). MALDI-TOF MS (positive mode,
THAP): m/z calcd for C189H229N31O80P5 [M+H]+ 4369.92, found 4370.49.
MALDI-TOF MS (negative mode, THAP): m/z calcd for C189H227N31O80P5
[M-H]- 4367.92, found 4367.0. HR-ESI-QToF MS (positive mode): m/z
calcd for C189H231N31O80P5 [(M + 3H)/3]+ 1456.4544, found 1456.4528.
[Tri-(2,3-dibenzoxybenzamide propyl triazol)-pentaerythrityl] tosyl-
triethyleneglycol cyanoethyl phosphate 36: Compound 35 (1.17 g, 1.7
mmol) was solubilized in dioxane/H2O (3:1 v/v, 17 mL) and N-(3-
azidopropyl)-2,3-dibenzoxybenzamide 25 (3.02 g, 6.8 mmol), copper (4
mg, 0.068 mmol) and TEAAc (2M, 500 µL) were added. The mixture was
stirred at 55 °C overnight. After filtration and evaporation, the residue was
solubilized in a minimum of CH3CN and treated with Quadrapure® IDA
(500 mg) for 3h. The mixture was filtered and filtrate evaporated. The crude
was purified on silica gel flash chromatography (CH2Cl2/MeOH, 95:5 v/v).
Residue was solubilized in a minimum of dioxane for lyophilization to
obtain 36 (white solid, 2.21 g, 65%). TLC Rf: 0.21 (CH2Cl2/MeOH, 95:5
1
v/v). H NMR (400 MHz, CDCl3) δ 8.05 – 7.94 (m, 12H, Ar), 7.75 (d, J =
8.3 Hz, 2H, Ar), 7.67 (s, 3H, Tz), 7.58 (dd, J = 7.8, 1.6 Hz, 3H, Ar), 7.55 –
7.48 (m, 6H, Ar), 7.44 – 7.39 (m, 3H, Ar), 7.38 – 7.27 (m, 17H, Ar), 7.21 (t,
J = 5.9 Hz, 3H, NH), 4.53 (s, 6H, OCH2Tz), 4.26 (t, J = 6.7 Hz, 6H,
TzCH2CH2), 4.15 – 3.97 (m, 8H, POCH2, TosOCH2, POCH2CH2, POCH2C),
3.66 – 3.58 (m, 4H, OCH2), 3.56 – 3.52 (m, 4H, OCH2), 3.44 (s, 6H,
CCH2O), 3.28 (dd, J = 12.6, 6.2 Hz, 6H, CH2CH2N), 2.72 – 2.63 (m, 2H,
POCH2CH2CN), 2.41 (s, 3H, PhCH3), 2.04
–
1.92 (m, 6H,
TzCH2CH2CH2N). 13C NMR (101 MHz, CDCl3) δ 165.9, 164.4, 164.3,
145.0, 143.5, 140.6, 134.2, 134.0, 133.0, 131.3, 130.3, 130.3, 130.1, 128.8,
128.7, 128.5, 128.3, 128.1, 126.7, 126.6, 125.9, 123.4, 117.3, 70.8, 70.5,
69.9, 69.5, 68.8, 68.3, 67.3, 65.0, 62.2, 47.5, 45.2, 36.8, 30.1, 21.8, 19.6.
31P NMR (162 MHz, CDCl3) δ -2.17, -2.21, -2.25. HR-ESI-QToF MS
(positive mode): m/z calcd for C102H101N13O27PS [M + H]+ 2002.6388,
found 2002.6431, m/z calcd for C102H102N13O27PS [(M + 2H)/2]+ 1001.8233,
found 1001.8260, m/z calcd for C102H103N13O27PS [(M + 3H)/3]+ 668.2181,
found 668.2202.
Galactocluster tri-catechol G3C:
A solution of the benzoylated
galactocluster tricatechol BzG3C (~ 300 mg) in NEt3/MeOH/H2O (1:5:1
v/v/v, 30 mL) was kept at room temperature overnight, then extracted three
times with AcOEt and triethyl ammonium was exchanged with Dowex 50
W X8 Na+ to give after lyophilization G3C (brown solid, 220 mg, 37%, for
two steps). 1H NMR (600 MHz, D2O) δ 8.22 – 7.75 (m, 8H, Tz), 7.51 – 6.73
(m, 19H, Ar), 5.39 (s, 8H, OCCH2Tz), 5.11 (s, 1H, H1’man), 5.05 – 4.85 (m,
4H, H1’gal), 4.75 – 4.24 (m, 29H, H’man, TzCH2O, CH2CH2Tz), 4.23 – 2.96
(m, 104H, POCH2, POCH2C, OCH2, H’gal, H’man, NCH2CH2) 2.31 – 1.94
(m, 6H, NCH2CH2CH2Tz). 13C NMR (151 MHz, D2O) δ 181.5, 169.9, 165.9,
154.2, 144.2, 131.4, 128.8, 128.3, 126.5, 124.6, 123.1, 117.0, 101.0, 97.7,
75.3, 72.6, 70.5, 70.3, 69.5, 69.0, 68.5, 64.9, 64.7, 63.7, 63.1, 60.7, 52.4,
49.8, 36.6, 28.7, 23.3. 31P NMR (202 MHz, D2O) δ 0.59, 0.31, -0.75, -1.00
(PO). MALDI-TOF MS (negative mode, THAP): m/z calcd for
C147H203N31O74P5 [M-H]- 3743.25, found 3743.27, MALDI-TOF MS
(positive mode, THAP): m/z calcd for C147H205N31O74P5 [M+H]+ 3745.27,
found 3745.75. HR-ESI-QToF MS (positive mode): m/z calcd for
C147H207N31O74P5 [(M + 3H)/3]+ 1248.4020, found 1248.4016.
[Tri-(2,3-dibenzoxybenzamide propyl triazol) pentaerythrityl] azido-
triethyleneglycol phosphate 37: Compound 36 (1.93 g, 0.96 mmol) was
solubilized in NEt3/MeOH/H2O mixture (1:5:1 v/v/v, 100 mL) and mixture
was stirred at room temperature overnight. Reaction was monitored by
MALDI-TOF spectrometry. Solvents were evaporated under vacuum and
residue was solubilized in DMF (20 mL). TMGN3 (184 mg, 1.16 mmol) was
added and the solution was heated to 55 °C overnight. Solvent was
evaporated and residue was solubilized in anhydrous pyridine (20 mL) and
cooled to -5 °C then benzoyl chloride was added dropwise (744 µL, 6.4
mmol) and crude was stirred for 1h at 0 °C. Solvent was evaporated and
residue was purified on reverse phase flash chromatography (H2O/32%
CH3CN 50 mM TEAAc - 80% CH3CN 25 mM TEAAc). Tampon was co-
evaporated with H2O and CH3CN. Residue was lyophilized in dioxane to
13
This article is protected by copyright. All rights reserved.

10.1002/cbic.202000490
ChemBioChem
FULL PAPER
Fluorescence quantification of bacterial labelling by cy3-
galacto/fucoclusters. Stationary phase growing bacteria were adjusted
to an OD620nm of 3 in PBS 1X and labelled with 10 μg/mL of 4′,6′-diamidino-
[1] C. T. Walsh and T. A. Wencewicz, Nat. Prod. Rep. 2013, 30, 175-200.
[2] A. Bernardi, J. Jimenez-Barbero, A. Casnati, C. De Castro, T. Darbre, F.
Fieschi, J. Finne, H. Funken, K.-E. Jaeger, M. Lahmann, T. K. Lindhorst,
M. Marradi, P. Messner, A. Molinaro, P. V. Murphy, C. Nativi, S. Oscarson,
S. Penades, F. Peri, R. J. Pieters, O. Renaudet, J.-L. Reymond, B. Richichi,
J. Rojo, F. Sansone, C. Schaeffer, W. B. Turnbull, T. Velasco-Torrijos, S.
Vidal, S. Vincent, T. Wennekes, H. Zuilhof and A. Imberty, Chem. Soc. Rev.
2013, 42, 4709-4727.
2-phenylindole (DAPI, Sigma).
1
μM of cy3-galacto/fucoclusters
harbouring 0 to 3 catechols/hydroxamates was added to the bacterial
suspension and allow to interact for 1h. Initial fluorescence (Cy3 and DAPI)
associated with each sample was measured using a Clariostar microplate
reader (BMG Labtech). Samples were centrifuged (5000 g, 3 min) and
washed in PBS 1X until Cy3 fluorescence of the supernatant was
undetectable. Cy3 (ex. 550 nm/em. 570 nm) and DAPI (ex. 350 nm/em.
460 nm) fluorescence of each bacterial suspension was finally measured
using the Clariostar fluorescence plate reader. Ratio of Cy3/DAPI
fluorescence was considered as specific bacterial labelling by the cy3-
galacto/fucoclusters with or w/o catechols or hydroxamate and adjusted to
100% for the control labelling (G0 or F0). Correction of the labelling ratio
was done if the initial Cy3 fluorescence of samples were different.
[3] J. J. Lundquist and E. J. Toone, Chem. Rev. 2002, 102, 555-578.
[4] a) F. D. Sottile, T. J. Marrie, D. S. Prough, C. D. Hobgood, D. J. Gower, L. X.
Webb, J. W. Costerton and A. G. Gristina, Crit. Care Med. 1986, 14, 265-
270; b) J. L. Vincent, D. J. Bihari, P. M. Suter, H. A. Bruining, J. White, R.,
M. H. Nicolas-Chanoin, M. Wolff, R. C. Spencer and M. Hemmer, J. Am.
Med. Assoc. 1995, 274, 639-644.
[5] a) N. Garber, U. Guempel, A. Belz, N. Gilboagarber and R. J. Doyle, Biochim.
Biophys. Acta 1992, 1116, 331-333; b) E. Mitchell, C. Houles, D.
Sudakevitz, M. Wimmerova, C. Gautier, S. Perez, A. M. Wu, N. Gilboa-
Garber and A. Imberty, Nat. Struct. Biol. 2002, 9, 918-921; c) G. Cioci, E.
P. Mitchell, C. Gautier, M. Wimmerova, D. Sudakevitz, S. Perez, N. Gilboa-
Garber and A. Imberty, FEBS Lett. 2003, 555, 297-301; d) R. Loris, D.
Tielker, K. E. Jaeger and L. Wyns, J. Mol. Biol. 2003, 331, 861-870; e) A.
Imberty, M. Wimmerova, E. P. Mitchell and N. Gilboa-Garber, Microbes
Infect. 2004, 6, 221-228; f) D. Tielker, S. Hacker, R. Loris, M. Strathmann,
J. Wingender, S. Wilhelm, F. Rosenau and K. E. Jaeger, Microbiology 2005,
151, 1313-1323.
Bacterial and cell culture. The mucoepidermoid pulmonary carcinoma
cell line NCI-H292 (A.T.C.C. cell line CRL-1848) was kindly provided by Dr
Jean-Marc Lo Guidice (EA4483, Lille, France). Cells were grown in DMEM
(Dulbecco's Modified Eagle Medium, Biowest, Denmark) supplemented
with 10% (v/v) fetal calf serum (FCS, Gibco BRL, USA),
2 mM
Ultraglutamine (Lonza, Switzerland), 100 units/mL penicillin, and 0.1
mg/mL streptomycin in a humidified atmosphere of 5% CO2 at 37 °C. For
maintenance, cells were grown to confluence and subcultured every 2–3
days at a split ratio of 1:4.
[6] a) C. Chemani, A. Imberty, S. de Bentzmann, M. Pierre, M. Wimmerova, B.
P. Guery and K. Faure, Infect. Immun. 2009, 77, 2065-2075; b) T. Eierhoff,
B. Bastian, R. Thuenauer, J. Madl, A. Audfray, S. Aigal, S. Juillot, G. E.
Rydell, S. Mueller, S. de Bentzmann, A. Imberty, C. Fleck and W. Roemer,
Proc. Natl. Acad. Sci. U. S. A. 2014, 111, 12895-12900.
PA strains, wild type PAO1 (kindly provided by Pr. Reuben Ramphal,
University of Florida Gainsville, FL, USA) and lecA, lecB, fpvA or exbB1
mutants (obtained from the Pseudomonas Transposon Mutant Collection,
UW Genome Sciences, Washington, USA) were cultured for 16 h at 37 °C
in LB medium under 150 rev/min agitation. For adhesion assays, bacteria
were washed in DMEM medium without FCS and diluted in the same
medium at working concentrations (1x106 UFC/mL)
[7] J. Glick and N. Garber, J. Gen. Microbiol. 1983, 129, 3085-3090.
[8] a) S. Cecioni, A. Imberty and S. Vidal, Chem. Rev. 2015, 115, 525-561; b)
C. Ligeour, O. Vidal, L. Dupin, F. Casoni, E. Gillon, A. Meyer, S. Vidal, G.
Vergoten, J. M. Lacroix, E. Souteyrand, A. Imberty, J. J. Vasseur, Y.
Chevolot and F. Morvan, Org. Biomol. Chem. 2015, 13, 8433-8444; c) R.
Visini, X. Jin, M. Bergmann, G. Michaud, F. Pertici, O. Fu, A. Pukin, T. R.
Branson, D. M. E. Thies-Weesie, J. Kemmink, E. Gillon, A. Imberty, A.
Stocker, T. Darbre, R. J. Pieters and J.-L. Reymond, ACS Chem. Biol. 2015,
10, 2455-2462; d) G. Michaud, R. Visini, M. Bergmann, G. Salerno, R.
Bosco, E. Gillon, B. Richichi, C. Nativi, A. Imberty, A. Stocker, T. Darbre
and J.-L. Reymond, Chem. Sci. 2016, 7, 166-182; e) A. Angeli, L. Dupin,
M. Madaoui, M. Li, G. Vergoten, S. Wang, A. Meyer, T. Gehin, S. Vidal, J.-
J. Vasseur, Y. Chevolot and F. Morvan, Chemistryselect 2017, 2, 10420-
10427; f) K. Buffet, I. Nierengarten, N. Galanos, E. Gillon, M. Holler, A.
Imberty, S. E. Matthews, S. Vidal, S. P. Vincent and J.-F. Nierengarten,
Chem. Eur. J. 2016, 22, 2955-2963; g) A. Angeli, M. Li, L. Dupin, G.
Vergoten, M. Noel, M. Madaoui, S. Wang, A. Meyer, T. Gehin, S. Vidal, J.-
J. Vasseur, Y. Chevolot and F. Morvan, Chembiochem 2017, 18, 1036-
1047; h) M. Donnier-Marechal, S. Abdullayev, M. Bauduin, Y. Pascal, M.-
Q. Fu, X.-P. He, E. Gillon, A. Imberty, E. Kipnis, R. Dessein and S. Vidal,
Org. Biomol. Chem. 2018, 16, 8804-8809; i) K. S. Buecher, N. Babic, T.
Freichel, F. Kovacic and L. Hartmann, Macromol. Biosci. 2018, 18; j) G. Yu,
D. M. E. Thies-Weesie and R. J. Pieters, Helv. Chim. Acta 2019, 102; k) G.
Yu, A. C. Vicini and R. J. Pieters, J. Org. Chem. 2019, 84, 2470-2488; l) D.
Goyard, B. Thomas, E. Gillon, A. Imberty and O. Renaudet, Front. Chem.
2019, 7.
Infection assays/Gentamicin protection assay. NCI-H292 cells were
seeded into 12-well plates and grown to 80% confluence for 48-72 h in
complete DMEM. Medium was renewed every 24h. After two washes with
fresh DMEM without FCS, cells were incubated with bacterial suspension
in the same medium (1×106 UFC/mL and MOI of 5) for 2 h at 37 °C.
Unbound bacteria were removed by two washes with 1 ml of DMEM
without FCS. Then, cells were incubated 1h with fresh DMEM, without FCS,
complemented with gentamcin (200 μg/mL) in order to kill bound bacteria
not internalized in the cell. Cells were washed four more times with DMEM
without FCS, and lysed using deionized water containing 0.02% Triton X-
100. Serial dilution of cell lysates in DPBS were then prepared and plated
on to LB agar to quantify the rate of infection by comparison with the control,
untreated PAO1. When indicated, galactoclusters with or w/o catechols,
were added to the medium during the 2h of infection to a final
concentration ranging from 100 to 500 μM.
Statistical analysis. Values presented for fluorescence quantification and
Infection assays are means ± SD of three independent experiments and
were tested by One-way ANOVA multiple comparison, Tukey-Test, using
GraphPad Prism 8.4.
[9] a) A. M. Boukerb, A. Rousset, N. Galanos, J. B. Mear, M. Thepaut, T.
Grandjean, E. Gillon, S. Cecioni, C. Abderrahmen, K. Faure, D.
Redelberger, E. Kipnis, R. Dessein, S. Havet, B. Darblade, S. E. Matthews,
S. de Bentzmann, B. Guery, B. Cournoyer, A. Imberty and S. Vidal, J. Med.
Chem. 2014, 57, 10275-10289; b) A. Palmioli, P. Sperandeo, A. Polissi and
C. Airoldi, Chembiochem 2019, 20, 2911-2915; c) M. Taouai, K. Chakroun,
R. Sommer, G. Michaud, D. Giacalone, M. A. Ben Maaouia, A. Vallin-
Butruille, D. Mathiron, R. Abidi, T. Darbre, P. J. Cragg, C. Mullie, J.-L.
Reymond, G. A. O'Toole and M. Benazza, J. Med. Chem. 2019, 62, 7722-
7738; d) G. Granata, S. Stracquadanio, G. M. L. Consoli, V. Cafiso, S.
Stefani and C. Geraci, Carbohydr. Res. 2019, 476, 60-64.
Acknowledgements
We thank the Agence Nationale de la Recherche
"investissements d'avenir" LABEX CheMISyst, ANR-10-LABX-
05-01 for financial support and for the award of a research
studentship of M.M. F. M. is a member of Inserm.
Keywords: glycoclusters • siderophore • Pseudomonas
aeruginosa • Lectins • Iron transport
[10] a) 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 and S. Vidal,
14
This article is protected by copyright. All rights reserved.

10.1002/cbic.202000490
ChemBioChem
FULL PAPER
Org. Biomol. Chem. 2017, 15, 10037-10043; b) F. Zuttion, C. Ligeour, O.
Vidal, M. Waelte, F. Morvan, S. Vidal, J.-J. Vasseur, Y. Chevolot, M.
Phaner-Goutorbe and H. Schillers, Nanoscale 2018, 10, 12771-12778; c)
L. Malinovska, L. Son Thai, M. Herczeg, M. Vaakova, J. Houser, E.
Fujdiarova, J. Komarek, P. Hodek, A. Borbas, M. Wimmerova and M.
Csavas, Biomolecules 2019, 9.
[11] a) B. Gerland, A. Goudot, G. Pourceau, A. Meyer, V. Dugas, S. Cecioni, S.
Vidal, E. Souteyrand, J. J. Vasseur, Y. Chevolot and F. Morvan,
Bioconjugate Chem. 2012, 23, 1534-1547; b) B. Gerland, A. Goudot, C.
Ligeour, G. Pourceau, A. Meyer, S. Vidal, T. Gehin, O. Vidal, E. Souteyrand,
J. J. Vasseur, Y. Chevolot and F. Morvan, Bioconjugate Chem. 2014, 25,
379-392.
[12] a) M. Miethke and M. A. Marahiel, Microbiol. Mol. Biol. Rev. 2007, 71, 413-
451; b) H. Drakesmith and A. Prentice, Nat. Rev. Microbiol. 2008, 6, 541-
552.
[13] R. C. Hider and X. L. Kong, Nat. Prod. Rep. 2010, 27, 637-657.
[14] I. J. Schalk and L. Guillon, Amino Acids 2013, 44, 1267-1277.
[15] a) C. Ji, R. E. Juarez-Hernandez and M. J. Miller, Future Med. Chem. 2012,
4, 297-313; b) S. Rebuffat, Biochem. Soc. Trans. 2012, 40, 1456-1462; c)
G. L. A. Mislin and I. J. Schalk, Metallomics 2014, 6, 408-420; d) T. F.
Zheng, J. L. Bullock and E. M. Nolan, J. Am. Chem. Soc. 2012, 134, 18388-
18400; e) T. F. Zheng and E. M. Nolan, J. Am. Chem. Soc. 2014, 136,
9677-9691; f) P. Chairatana, T. F. Zheng and E. M. Nolan, Chem. Sci. 2015,
6, 4458-4471.
[16] P. Cornelis and J. Dingemans, Front.Cell.Infect.Microbiol. 2013, 3.
[17] G. Pourceau, A. Meyer, Y. Chevolot, E. Souteyrand, J. J. Vasseur and F.
Morvan, Bioconjugate Chem. 2010, 21, 1520-1529.
[18] Z. H. Zhang, Y. Yu and L. S. Liebeskind, Org. Lett. 2008, 10, 3005-3008.
[19] J. Lietard, A. Meyer, J. J. Vasseur and F. Morvan, J. Org. Chem. 2008, 73,
191-200.
[20] T. Hasegawa, M. Numata, S. Okumura, T. Kimura, K. Sakurai and S.
Shinkai, Org. Biomol. Chem. 2007, 5 2404-2412.
[21] G. Pourceau, A. Meyer, J. J. Vasseur and F. Morvan, J. Org. Chem. 2009,
74, 6837-6842.
[22] J. Katajisto, P. Virta and H. Lonnberg, Bioconjugate Chem. 2004, 15, 890-
896.
[23] C. Ligeour, A. Meyer, J. J. Vasseur and F. Morvan, Eur. J. Org. Chem. 2012,
1851-1856.
[24] F. Morvan, A. Meyer, A. Jochum, C. Sabin, Y. Chevolot, A. Imberty, J. P.
Praly, J. J. Vasseur, E. Souteyrand and S. Vidal, Bioconjugate Chem. 2007,
18, 1637-1643.
[25] H.-P. Wu, T.-L. Cheng and W.-L. Tseng, Langmuir 2007, 23, 7880-7885.
[26] J. Greenwald, M. Nader, H. Celia, C. Gruffaz, V. Geoffroy, J. M. Meyer, I.
J. Schalk and F. Pattus, Mol. Microbiol. 2009, 72, 1246-1259.
[27] a) A. Ito, T. Nishikawa, S. Matsumoto, H. Yoshizawa, T. Sato, R. Nakamura,
M. Tsuji and Y. Yamano, Antimicrob. Agents Chemother. 2016, 60, 7396-
7401; b) G. G. Zhanel, A. R. Golden, S. Zelenitsky, K. Wiebe, C. K.
Lawrence, H. J. Adam, T. Idowu, R. Domalaon, F. Schweizer, M. A. Zhanel,
P. R. S. Lagace-Wiens, A. J. Walkty, A. Noreddin, J. P. Lynch and J. A.
Karlowsky, Drugs 2019, 79, 271-289.
[28] E. K. Dolence, C. E. Lin, M. J. Miller and S. M. Payne, J. Med. Chem. 1991,
34, 956-968.
[29] N. Noinaj, M. Guillier, T. J. Barnard and S. K. Buchanan in TonB-Dependent
Transporters: Regulation, Structure, and Function, Vol. 64 Eds.: S.
Gottesman and C. S. Harwood), 2010, pp. 43-60.
[30] S. P. Diggle, R. E. Stacey, C. Dodd, M. Camara, P. Williams and K. Winzer,
Environ. Microbiol. 2006, 8, 1095-1104.
[31] V. Percec, P. Leowanawat, H. J. Sun, O. Kulikov, C. D. Nusbaum, T. M.
Tran, A. Bertin, D. A. Wilson, M. Peterca, S. D. Zhang, N. P. Kamat, K.
Vargo, D. Moock, E. D. Johnston, D. A. Hammer, D. J. Pochan, Y. C. Chen,
Y. M. Chabre, T. C. Shiao, M. Bergeron-Brlek, S. Andre, R. Roy, H. J.
Gabius and P. A. Heiney, J. Am. Chem. Soc. 2013, 135, 9055-9077.
[32] O. Bajolet-Laudinat, S. Girod-de Bentzmann, J. M. Tournier, C. Madoulet,
M. C. Plotkowski, C. Chippaux and E. Puchelle, Infect. Immun. 1994, 62,
4481-4487.
[33] W. H. A. Kuijpers and C. A. A. Vanboeckel, Tetrahedron 1993, 49, 10931-
10944.
15
This article is protected by copyright. All rights reserved.

10.1002/cbic.202000490
ChemBioChem
FULL PAPER
Entry for the Table of Contents
Insert graphic for Table of Contents here.
Trojan horse strategy: Glycocluster-siderophores were synthesized to target LecA and LecB lectins contained in the outer
membrane and in the bacterium of Pseudomonas aeruginosa. The galactoclusters with catechol moieties were able to hijack the iron
transport and exhibited a 46 to 63% inhibitory effect at 100 mM. This result shows that LecA localized in the outer membrane of P.
aeruginosa is involved in the infection process.
16
This article is protected by copyright. All rights reserved.