A Water Soluble Pd2L4 Cage for Selective Binding of Neu5Ac

Article abstract of DOI:10.1002/chem.202102176

The sialic acid N-acetylneuraminic acid (Neu5Ac) and its derivatives are involved in many biological processes including cell-cell recognition and infection by influenza. Molecules that can recognize Neu5Ac might thus be exploited to intervene in or monitor such events. A key obstacle in this development is the sparse availability of easily prepared molecules that bind to this carbohydrate in its natural solvent; water. Here, we report that the carbohydrate binding pocket of an organic soluble [Pd₂L₄]⁴⁺ cage could be equipped with guanidinium-terminating dendrons to give the water soluble [Pd₂L₄][NO₃]₁₆ cage 7. It was shown by means of NMR spectroscopy that 7 binds selectively to anionic monosaccharides and strongest to Neu5Ac with K_a=24 M^?1. The cage had low to no affinity for the thirteen neutral saccharides studied. Aided by molecular modeling, the selectivity for anionic carbohydrates such as Neu5Ac could be rationalized by the presence of charge assisted hydrogen bonds and/or the presence of a salt bridge with a guanidinium solubilizing arm of 7. Establishing that a simple coordination cage such as 7 can already selectively bind to Neu5Ac in water paves the way to improve the stability, affinity and/or selectivity properties of M₂L₄ cages for carbohydrates and other small molecules.

Full text of DOI:10.1002/chem.202102176

Communication
doi.org/10.1002/chem.202102176
Chemistry—A European Journal
A Water Soluble Pd L Cage for Selective Binding of
2
4
Neu5Ac
[
a]
[a]
[a]
[a]
Xander Schaapkens, Roy N. van Sluis, Eduard O. Bobylev, Joost N. H. Reek, and
Tiddo J. Mooibroek*
[
a]
Ac), which is utilized by influenza or other viruses to enter
Abstract: The sialic acid N-acetylneuraminic acid (Neu5Ac)
and its derivatives are involved in many biological
processes including cell-cell recognition and infection by
influenza. Molecules that can recognize Neu5Ac might thus
be exploited to intervene in or monitor such events. A key
obstacle in this development is the sparse availability of
easily prepared molecules that bind to this carbohydrate in
its natural solvent; water. Here, we report that the
[2]
mammalian cells. Neu5Ac is also found at the end of
X
x
tetrasaccharide Sialyl Lewis 1 (sLe , Figure 1a) and clinical
x
studies showed the importance of sLe in leukocyte adhesion
[3]
[4]
[5]
deficiency, inflammatory response, (in vitro) fertilization,
[6]
[7]
coronavirus binding, and cancer metastasis. An example is
the study of molecular sensors for sLe , which facilitates
x
extravasation of cancer cells out of the blood stream (meta-
[8]
stasis), displaying leukocyte mimicry. While Neu5Ac is typically
4
+
carbohydrate binding pocket of an organic soluble [Pd L ]
[1]
2
4
O-linked to other molecules as α-anomer (axial carboxylate),
cage could be equipped with guanidinium-terminating
dendrons to give the water soluble [Pd L ][NO ] cage 7. It
2
4
3 16
was shown by means of NMR spectroscopy that 7 binds
selectively to anionic monosaccharides and strongest to
À 1
Neu5Ac with K =24 M . The cage had low to no affinity
a
for the thirteen neutral saccharides studied. Aided by
molecular modeling, the selectivity for anionic carbohy-
drates such as Neu5Ac could be rationalized by the
presence of charge assisted hydrogen bonds and/or the
presence of a salt bridge with a guanidinium solubilizing
arm of 7. Establishing that a simple coordination cage such
as 7 can already selectively bind to Neu5Ac in water paves
the way to improve the stability, affinity and/or selectivity
properties of M L cages for carbohydrates and other small
2
4
molecules.
Introduction
Sialic acids, a class of α-keto acid sugars, have been found on
the distal ends of cell surface glycoconjugates and play a wide
[1]
variety of biological roles, especially in cell-to-cell recognition.
The most common member is N-acetylneuraminic acid (Neu5-
[a] X. Schaapkens, R. N. van Sluis, E. O. Bobylev, Prof. Dr. J. N. H. Reek,
Dr. T. J. Mooibroek
Van ‘t Hoff Institute for Molecular Sciences
University of Amsterdam
Science Park 904, 1098 XH, Amsterdam (The Netherlands)
E-mail: t.j.mooibroek@uva.nl
X
x
Figure 1. a) Sialyl Lewis 1 (sLe ) bound to the E-selectin binding site as
found in PDB entry 1G1T with charge assisted HBs highlighted in magenta.
b) Two artificial carbohydrate receptors that bind to Neu5Ac in part due to
[11]
Supporting information for this article is available on the WWW under
https://doi.org/10.1002/chem.202102176
[12]
1
charge assisted HBs. R =guanindinium terminating dendron. c) Left:
macrocycle 4 selective for GlcNAc-β-OMe. R =carboxylate terminating
2
©
2021 The Authors. Chemistry - A European Journal published by Wiley-
[
13]
VCH GmbH. This is an open access article under the terms of the Creative
Commons Attribution Non-Commercial License, which permits use, dis-
tribution and reproduction in any medium, provided the original work is
properly cited and is not used for commercial purposes.
dendrimer. Right: Pt
2 4
L cage 5 selective for disaccharides where
3 [14]
L=anthracene functionalized dipyridyl ligand and R =O(CH
2
)
2
OMe. d)
) where L=isophthalamide linked
dipyridyl ligand and R a solubilizing group.
À
[15]
À
3
Pd
2
L
4
cages 6 (X=BF
4
)
and 7 (X=NO
4
Chem. Eur. J. 2021, 27, 1–7
1
© 2021 The Authors. Chemistry - A European Journal published by Wiley-VCH GmbH
��
These are not the final page numbers!

Communication
doi.org/10.1002/chem.202102176
Chemistry—A European Journal
the chemical antecedent to such linkages is cytidine-5’-mono-
We thus wondered what the binding properties of a cage
such as 6 would be in aqueous solution, in particular for anionic
carbohydrates like Neu5Ac. To this end, the solubility handles of
the ligands in 6 were replaced by guanidinium-terminating
[1,9]
phospho-β-Neu5Ac.
Unlinked Neu5Ac is predominantly
present as the β-anomer in solution and has a rich invivo
[1,10]
chemistry.
Molecules that can bind selectively to Neu5Ac
1
6+
and its derivatives might thus be exploited to understand,
monitor or intervene in a range of biological processes.
Inspiration for the development of Neu5Ac binders can be
drawn from selectins, a subclass of lectins (carbohydrate bind-
ing proteins). As is illustrated in Figure 1a for crystal structure
dendrons to make the [Pd L ]
cage 7 (Figure 1d). Herein, we
2
4
report that 7 has selective affinity for anionic sugars, particularly
for Neu5Ac, and that 7 has very low to no affinity for common
neutral mono- and disaccharides.
1
G1T (human E-selectin), there is a high degree of interaction
x [11]
complementarity in the binding mode with sLe . Notably, the Results and Discussion
anionic Neu5Ac fragment of sLe forms a salt-bridge with an
x
arginine residue and the galactose fragment has strong charge
assisted hydrogen bonds (HBs) with the same arginine residue
The synthesis of the ligand precursor to cage 7 (penta nitric
acid salt 13) is shown in Scheme 1. The starting trimesic
pentafluorophenyl (PFP) ester 8 and amine 9 were synthesized
(highlighted in magenta).
[19]
The beneficial effect of employing charge assisted HBs to
according to literature procedures and then coupled to each
bind sialic acid derivatives has been mimicked by the artificial
other to form bis-PFP ester 10 in 62% yield by using a
[12a]
[12b]
[15,19d]
receptors 2
and 3
shown in Figure 1b. The benzoborox-
previously reported protocol.
Subsequently, the remaining
ole-based receptor 2 can bind covalently to Neu5Ac with its
PFP esters of 10 were substituted by 3-aminopyridine to afford
[15]
borane part and binding is further enhanced by the presence of
11.
Deprotection of the Boc groups of 11 followed by
[12a]
the nearby guanidinium group.
Pyrenyl platform 3 was
basification and treatment with bis-boc-pyrazolocarboxamidine
[20]
intended to bind carbohydrates in water using CH···π inter-
afforded hexa-boc guanidine 12 in 69% yield. The desired
guanidinium ligand 13 could be obtained in 74% yield after
treatment of 12 with 1 M nitric acid in a water/1,4-dioxane
solvent mixture. The pyridyl rings in 13 could be selectively
deprotonated by the addition of two equivalents of sodium
hydroxide. The subsequent addition of a Pd(NO3)2 solution
actions and showed enhanced affinity for Neu5Ac when
1
[12b]
equipped with guanidinium-terminating dendrons (R ).
An-
other binding strategy is the use of covalent macrocyclic
compounds that can encapsulate a carbohydrate in aqueous
[13,16]
media.
[
13]
1
For example, macrocycle 4 (Figure 1c, left)
is highly
(0.55 eq.) gave cage 7 in a quantitative yield based on H NMR.
À 1
selective for GlcNAc-β-OMe (K �18,000 M
in water) by
[Note: As is detailed in Section S2 of the Supporting
a
encapsulating the carbohydrate by regular HBs and CH···π
However, such covalent macrocycles are not
selective for Neu5Ac or related anionic carbohydrates.
can be rationalized by the presence of anionic dendrimers (R )
used to solubilize the hydrophobic binding pockets.
[17]
interactions.
[
16c–e]
This
2
Contrarily, coordination cages based on a dipyridyl ligand
8
2+
2+
(
L) an a square planar d metal (M, for example Pd or Pt ) are
positively charged and are known to have affinity for anionic
[18]
guests. Another advantage of such coordination cages is the
reversibility of the pyridine-metal bond. This allows for non-
productive oligomerization products to become intermediates
towards the desired macrocycle, thus evading low-yielding
macrocyclization reactions needed in the synthesis of covalent
cages. Recently, two examples of coordination cages with the
4
+
structure [M L ]
have been reported with affinity for
As is exemplified in Figure 1c (right),
2
4
[
14–15]
carbohydrates.
4
+
[
Pt L ] cage 5 is based on dipyridyl ligands separated by
2 4
anthracene moieties. This cage bound selectively to D-sucrose
À 1
(
K �1,000 M ) by virtue of shape-complementarity and multi-
a
ple CH···π interactions between the carbohydrate and the
polyaromatic cavity of 5. A similar [M L ] cage was reported
[14]
4+
2
4
Scheme 1. Synthesis of cage 7 from ligand 13, prepared from previously
[15,19–20]
reported building blocks following (adjusted) literature protocols.
(
6 in Figure 1d) where the dipyridyl ligands are separated by
[15]
PFP=pentafluorophenyl, Boc=tert-Butyloxycarbonyl, Sol=guanidinium
solubilizing group. Conditions: i) N,N-diisopropylethylamine, 42 h at room
temperature (RT) in tetrahydrofuran; ii) 6 eq. 3-aminopyridine, 41 h at 100°C
isophtalamides, similar to macrocycle 4.
organic solubility handle and could be studied in CD Cl
containing 10% DMSO-d , where selectivity towards n-octyl-β-
Cage 6 had an
2
2
in pyridine; iii) 4 h at RT in 4 M HCl in dioxane/water; iv) neutralization with
6
À 1
À 1
NaOH and basification with NEt
3
; v) 6 eq. bis-Boc-pyrazolocarboxamide, 20 h
glucoside (K �51 M ) versus n-octyl-β-galactoside (K �29 M )
a
a
°
at RT (with dichloromethane); vi) 1 M HNO
3
, 22 h at 50 C in dioxane/water;
was observed.
vii) 2 eq. NaOH in D O;viii) 0.55 eq. Pd(NO ) in D O (see also Figure 2). See
2
3
2
2
Section S2 for experimental details and full characterizations.
Chem. Eur. J. 2021, 27, 1–7
www.chemeurj.org
2
© 2021 The Authors. Chemistry - A European Journal published by Wiley-VCH GmbH
��
These are not the final page numbers!

Communication
doi.org/10.1002/chem.202102176
Chemistry—A European Journal
À
Information, an alternative route to prepare ligand 13 was
unproductive due to reaction compatibility issues between PFP-
esters and the Boc-linked NH of diBoc-protected guanidines.]
includes some deuterium and Cl because the solution was
measured from a D O sample in undeuterated solvent contain-
ing trace amounts of salts (see Supporting Information and
2
The synthesis of 7 from 13 could also be followed in D O by
H NMR, as shown by the stacked spectra in Figure 2a (top).
Figures S74–S90 for details).
2
1
16+
With the water soluble [Pd L ]
cage 7 in hand, the
2
4
Upon addition of 2 equivalents of NaOH the resonances
belonging to the pyridyl ring in 13 were found significantly
upfield, which is in line with deprotonation of the pyridyl
nitrogens.
binding affinity for the carbohydrates listed in Table 1 was
1
investigated by H NMR titration experiments in D O. Titrations
2
with charge neutral carbohydrates 14–26 to about 140 mM
only resulted in minor near-linear peaks shifts of some
max
Subsequent stepwise addition of Pd(NO ) resulted in the
resonances of 7 with Δδ �0.02 p.p.m. on average (see
3
2
disappearance of dipyridyl ligand signals with the proportional
appearance of a new set of resonances.
Figures S92–S104). These shifts could not result from the
dilution of 7, as a dilution study in the concentration range
used during titrations revealed that all resonances remained
stationary (see Figure S91). Attempts to fit these shifts to a
binding model was not feasible and could only be roughly
modelled (not fitted) to binding with an affinity around or
The emerging Pd-complex and its parent ligand thus appear
to be in slow exchange relative to the NMR time scale. All
proton resonances of the resulting well-defined spectrum could
be identified and are consistent with that of cage 7. In particular
the large downfield shifts of proton resonances such as a
À 1
below the detection limit of ~3 M . We thus interpret these
À 1
(
8.41!8.86) and d (8.77!9.74) are indicative of pyridyl-
shifts as resulting from very weak binding ofꢀ3 M (entry 1,
[21]
palladium coordination.
Table 1), spanning only the very start of possible binding
curves.
The 2D DOSY NMR of this sample reveals that the diffusion
constant (D) of 7 (log(D)=À 9.83) is substantially larger than
that of the neutralized ligand (log(D)=À 9.57) which is also
consistent with cage formation (see bottom of Figure 2a).
Furthermore, the isotope distribution of a species with largest
monoisotopic mass m/z=476.1798 (Figure 2b) measured with
CSI HRMS is in agreement with a simulated distribution of
Addition of neutralized solutions of 27–29, did result in
significant non-linear shifting of the resonances of 7, with clear
sings of saturation (Figures S105–S107). These shifts could be
fitted accurately to a 1:1 binding model resulting in the binding
constants listed in entries 2–4 of Table 1.
Selected spectra of the titration with 29 are shown in
Figure 3a. With increasing concentration of Neu5Ac 29, aro-
7
+
[
4
7(NO ) Cl ]
with largest monoisotopic mass of m/z=
3
3
6
7
+
76.1823. The modelled molecular formula of [7(NO ) Cl ]
3 3 6
Table 1. Overview of binding studies performed with 7 and the structures
of titrants 14–29.
À 1 [a]
Entry
Guest
K
a
(M
)
Correlation
of fit (r )
2
[b]
1
2
3
4
14–26
27
28
ꢀ3
–
6.6ꢁ0.2
8.3ꢁ0.3
24.0ꢁ0.2
0.9974
0.9609
0.9981
29
1
[22]
[a] Obtained by curve fitting H NMR resonance shifts using HypNMR;
b] Correlation of fit computed over all fitted data points.
[
Figure 2. a) Top: stepwise formation of 7 by deprotonation of the pyridyls in
3 (top two spectra) followed by stepwise addition of Pd(NO as a 10 mM
1
3 2
)
solution. Bottom: overlay of the 2D DOSY NMR spectra of 7 and 13 with
deprotonated pyridyl rings. b) representation of 7 with labelled protons and
CSI HRMS isotope distribution with indicated highest isotopic mass as
2
measured from a D O solution (top, blue) and simulated (bottom, green).
Chem. Eur. J. 2021, 27, 1–7
www.chemeurj.org
3
© 2021 The Authors. Chemistry - A European Journal published by Wiley-VCH GmbH
��
These are not the final page numbers!

Communication
doi.org/10.1002/chem.202102176
Chemistry—A European Journal
1
Figure 3. a) Partial H NMR spectra of 7 titrated with Neu5Ac 29 in D
2
O, a spectrum at the end of the titration after heating (‘A. H.’) at 60 °C, and a spectrum of
1
(
3+29 at the same concentration as at the end of the titration. Assignment of the proton signals: [7·29] major symmetric species (a-g), asymmetric species
*), [13·29] minor symmetrical species (a
using a 1:1 binding model to the shifting of indicated resonances (grey symbols) of the symmetrical signals of 7 to give K
goodness of for (r ). The speciation is also shown as unbound 7 (host, green line) and the [7·29] adduct (blue line).
L
–g
L
), and a small impurity present in 29 (I and II) that does not bind (see Figure S108); b) HypNMR fit (black lines)
À 1
a
=24.0ꢁ0.2 M with indicated
2
matic signals a, b, d, f and g shifted downfield and broadened
slightly. In the presence of a large excess of 29, a minor species
with lower symmetry arose, marked with blue asterisks in
Figure 3a. This species disappeared after heating at 60°C
(‘A.H.’), while the major species of [7·29] persisted and another
minor symmetrical species appeared (marked with diamonds).
This new set of resonances was nearly identical to a solution of
deprotonated ligand 13 and neutralized 29 at the same
concentration as present in the titration (shown at the top of
Figure 3a and assigned with subscript ‘L’). The minor sym-
metrical species (diamonds) present after heating is thus
probably ligand bound to 29, with the ligand originating from
cage decomposition (likely driven by Pd-plating). To quantify
binding of the major symmetrical species 7, the peak shifting
[22]
could be fitted to a 1:1 model using HypNMR as shown in
Figure 3b. This fitting gave the association constant (K ) of
a
À 1
2
4.0ꢁ0.2 M listed in entry 4 of Table 1 with an excellent
1
Figure 4. a) structures of 7 and β-Neu5Ac 29 with reference H NMR
spectrum. Assignment of 29 in main text is with a ‘c’ for ‘carbohydrate’ in
front of the number to distinguish it from other assignments. b) 1D selective
2
goodness of fit of r =0.9981.
To further understand binding of 7 with 29, selective 1D
nuclear Overhauser effect (nOe) NMR spectra were recorded
after heating at 60°C, as shown in Figure 4b. For reference
1
5+
NOESY spectrum of [7·29]
with t
m
=500 ms after heating at 60°C. See
section S3 for full structural elucidation of β-Neu5Ac 29.
1
purposes, the H NMR spectrum of 29 is shown in Figure 4a
together with an assignment that is based on a full structure
elucidation of 29 (see Section S3 for full details). As is shown in
Figure 4b, clear nOe signals were observed between irradiated
cage protons d, f/a and c/g and carbohydrate protons c3, c9
and c11. Proton d of 7 also shifted the most when binding to
molecular models were generated of cage 7 bound to the β-
anomers of anionic carbohydrates 27, 28 and 29 using density
functional theory (DFT, see Section S4 for details).
Shown in the top of Figure 5a is the space filling
representation of a model of 7 that fully encapsulates
glucoronate 27, with OH-3 protruding from one of the portals.
Interestingly, as is shown in the bottom of Figure 5a, OH-3 is
not involved in a HB to an amide, while all the other hydroxyl
groups are. Rather, OH-3 is involved in the only (and weak)
charge assisted HB involving the pyridyl CÀ H (blue arrow). The
carboxylate is furthermore held in place by four amidic HBs
with H···O distances in the range of 2.0–2.3 Å. This four-
pronged charge assisted HB interaction might rationalize the
selectivity of 7 for the anionic glucoronate 27 over the neutral
glucose 14. The model of 7 with galacturonate 28 is very similar
2
9 (see Figure 3). The much weaker nOe with proton b suggest
that these are further away from the CÀ H protons of Neu5Ac
9. It is worth pointing out that protons c3, c9, and c11 are
2
located on different sides of 29 and that c3 and c9 are even
opposite to each other (see Figure 4a). The fact that nOe’s were
observed to these signals thus implies that 29 is bound inside
7
.
The titration data (Table 1) clearly show that 7 is selective
for anionic monosaccharides, in particular for Neu5Ac 29, and
that 7 binds about equally well to glucoronate 27 and
galacturonate 28. To rationalize these observations, some
Chem. Eur. J. 2021, 27, 1–7
www.chemeurj.org
4
© 2021 The Authors. Chemistry - A European Journal published by Wiley-VCH GmbH
��
These are not the final page numbers!

Communication
doi.org/10.1002/chem.202102176
Chemistry—A European Journal
1
6+
form the [Pd L ]
cage 7. Titrations of 7 with common
2
4
carbohydrates revealed selectivity for anionic carbohydrates.
While the binding affinities towards neutral mono- and
disaccharides was near or under the detection limit of Ka
À 1
�
3 M , the binding affinities for anionic carbohydrates 27, 28
À 1
and 29 were found to be 6.6, 8.3 and 24 M , respectively. Cage
7
is thus at least 8 times more selective for 29 than for neutral
saccharides. This selectivity could be rationalized based on DFT
modelling and likely originates from complementary ion-pair
formation between the anionic sugars and cationic 7 by various
(charge assisted) HBs, much like those found in nature.
The selectivity for anionic carbohydrates in aqueous
[12a]
solution is rare,
and has not yet been reported with similar
Figure 5. Models of 7 bound to anionic carbohydrates calculated by DFT/
ωB97XÀ D/6–31G*. a) top: space filling representation of 7 bound to 27;
bottom: hydrogen bonding pattern of the [7·27] model overlaid with the
isophthalamide macrocyclic structures held together by either
[13,16a–k]
[15]
covalent
or coordination
bonds. While the affinities
[
7·28] model. The C-atoms of 27 and 28 are labelled in red and the
found are on the lower end, this study establishes the principle
that a coordination cage such as 7 can selectively bind to
anionic carbohydrates in water. This finding thus paves the way
for further improvements of coordination cages to enhance
their stability, affinity and/or selectivity properties. Ultimately,
such developments could lead to selective synthetic lectins for
Neu5Ac and its α/β-derivatives that can be used in biological
studies such as Western blotting, targeted drug delivery.
distances (in Å) are averages for both models. See Figure S116 for more
details; b) top: space filling model of Neu5Ac 29 bound to a model of 7 with
a guanidinium arm incorporated; bottom: hydrogen bonding pattern of
[
7·29] with distances in Å and C-atoms of 29 labelled in red. The structure is
similar to a structure without such a guanidiunium arm, see Figure S117.
to the [7·27] model, which is evident from the overlaid HB
patterning at the bottom of Figure 5a (see also Figure S116).
Here too the carboxylate is held in place by four amidic HBs
and OH-3 is the only hydroxyl that has no HB to an amide and Acknowledgements
only weakly to a pyridyl CÀ H (blue arrow). This absence of an
amidic HB with OH-3 in both 27 and 28 might explain the lack
in selectivity observed between these two carbohydrates. As
can be seen in the top of Figure 5b, modelling the binding of 7
to Neu5Ac 29 resulted in a complex where the anomeric center
This research was financially supported by the Netherlands
Organization for Scientific Research (NWO) with VIDI grant
number 723.015.006.
(
(
C2) protrudes through one of the four portals of 7. The glyceryl
C7–C9) and acetyl (C10, C11) fragments point out of two other Conflict of Interest
portals (see Figure S117 for details). Presumably, the carbox-
ylate and hydroxyl group on C2 of 29 are too large to be
accommodated in the interior of 7 in the same manner as
modelled for the carboxylates in 27 and 28 (Figure 5a). More-
over, the guanidinium arm that was incorporated in this model
of 7 formed a strong salt-bridge with the carboxylate of 29. This
is particularly evident from the HB pattern in the [7·29] model
shown in the bottom of Figure 5b. Highlighted in magenta are
the three strong charge assisted HBs with H···O distances in the
range of 1.8–2.1 Å, which resemble the charge assisted HBs
found in PDB entry 1G1T (Figure 1a). Additionally, there are
three amidic HBs and four relatively strong HBs involving
pyridyl CÀ H’s with H···O�2.2 Å (blue arrows). The salt-bridge
formation, together with the encapsulation of 29 via three of
the four portals of 7, the three amidic HBs and the four HBs
with pyridyl CÀ H’s observed in the model can offer a rationale
for the ~3–4 fold selectivity observed for 29 over 27/28 (see
Table 1).
There are no conflicts to declare.
Keywords: carbohydrate recognition · host-guest systems ·
molecular recognition · sialic acids · self-assembly
[
1] T. Angata, A. Varki, Chem. Rev. 2002, 102, 439–469.
[2] S. X. Tong, X. Y. Zhu, Y. Li, M. Shi, J. Zhang, M. Bourgeois, H. Yang, X. F.
Chen, S. Recuenco, J. Gomez, PLoS Pathog. 2013, 9, Art. Nr.: e1003657.
[
3] S. Hanna, A. Etzioni, N. Y. A. S. Annals, Ann. N. Y. Acad. Sci. 2012, 1250,
0–55.
[11]
5
[4] a) M. Fukuda, N. Hiraoka, J. C. Yeh, J. Cell Biol. 1999, 147, 467–470;
b) S. R. Barthel, J. D. Gavino, L. Descheny, C. J. Dimitroff, Expert Opin.
Ther. Targets 2007, 11, 1473–1491.
[
5] P. C. Pang, P. C. N. Chiu, C. L. Lee, L. Y. Chang, M. Panico, H. R. Morris,
S. M. Haslam, K. H. Khoo, G. F. Clark, W. S. B. Yeung, Science 2011, 333,
1761–1764.
[
6] a) M. A. Tortorici, A. C. Walls, Y. F. Lang, C. Y. Wang, Z. S. Li, D. Koerhuis,
G. J. Boons, B. J. Bosch, F. A. Rey, R. J. de Groot, Nat. Struct. Mol. Biol.
2019, 26, 481–489; b) W. T. Li, R. J. G. Hulswit, I. Widjaja, V. S. Raj, R.
McBride, W. J. Peng, W. Widagdo, M. A. Tortorici, B. van Dieren, Y. Lang,
Proc. Natl. Acad. Sci. USA 2017, 114, E8508–E8517.
[
7] a) F. Q. Jin, F. S. Wang, Glycoconjugate J. 2020, 37, 277–291; b) S. S. Li,
C. K. M. Ip, M. Y. H. Tang, M. K. S. Tang, Y. Tong, J. W. Zhang, A. A.
Hassan, A. S. C. Mak, S. Yung, T. M. Chan, Nat. Commun. 2019, 10, Art.
Nr.: 2406; c) S. Julien, A. Ivetic, A. Grigoriadis, D. Qize, B. Burford, D.
Sproviero, G. Picco, C. Gillett, S. L. Papp, L. Schaffer, Cancer Res. 2011,
71, 7683–7693; d) H. Läubli, L. Borsig, Semin. Cancer Biol. 2010, 20, 169–
Conclusions
4
+
A previous reported [Pd L ] cage was rendered water soluble
by introducing guanidinium solubility groups on the ligand to
2
4
Chem. Eur. J. 2021, 27, 1–7
www.chemeurj.org
5
© 2021 The Authors. Chemistry - A European Journal published by Wiley-VCH GmbH
��
These are not the final page numbers!

Communication
doi.org/10.1002/chem.202102176
Chemistry—A European Journal
1
77; e) C. Ohyama, S. Tsuboi, M. Fukuda, EMBO J. 1999, 18, 1516–1525;
f) R. Kannagi, M. Izawa, T. Koike, K. Miyazaki, N. Kimura, Cancer Sci. 2004,
5, 377–384.
8] a) G. Kaur, H. Fang, X. M. Gao, H. B. Li, B. H. Wang, Tetrahedron 2006, 62,
583–2589; b) W. Q. Yang, H. Y. Fan, X. M. Gao, S. H. Gao, V. V. R. Karnati,
W. J. Ni, W. B. Hooks, J. Carson, B. Weston, B. H. Wang, Chem. Biol. 2004,
1, 439–448; c) W. Q. Yang, S. H. Gao, X. M. Gao, V. V. R. Karnati, W. J. Ni,
B. H. Wang, W. B. Hooks, J. Carson, B. Weston, Bioorg. Med. Chem. Lett.
002, 12, 2175–2177; d) G. S. Jacob, C. Kirmaier, S. Z. Abbas, S. C.
Howard, C. N. Steininger, J. K. Welply, P. Scudder, Biochemistry 1995, 34,
210–1217.
9] a) F. Amann, C. Schaub, B. Muller, R. R. Schmidt, Chem. Eur. J. 1998, 4,
106–1115; b) M. Gilbert, R. Bayer, A. M. Cunningham, S. Defrees, Y. H.
Gao, D. C. Watson, N. M. Young, W. W. Wakarchuk, Nat. Biotechnol. 1998,
6, 769–772; c) D. N. Kwon, K. H. Lee, M. J. Kang, Y. J. Choi, C. Park, J. J.
2019, 11, 52–56; g) N. Chandramouli, Y. Ferrand, G. Lautrette, B.
Kauffmann, C. D. Mackereth, M. Laguerre, D. Dubreuil, I. Huc, Nat. Chem.
2015, 7, 334–341; h) P. Mateus, B. Wicher, Y. Ferrand, I. Huc, Chem.
Commun. (Camb.) 2018, 54, 5078–5081; i) O. Francesconi, F. Cicero, C.
Nativi, S. Roelens, ChemPhysChem 2020, 21, 257–262; j) O. Francesconi,
M. Martinucci, L. Badii, C. Nativi, S. Roelens, Chem. Eur. J. 2018, 24,
6828–6836; k) T. J. Mooibroek, M. P. Crump, A. P. Davis, Org. Biomol.
Chem. 2016, 14, 1930–1933; l) P. Rios, T. J. Mooibroek, T. S. Carter, C.
Williams, M. R. Wilson, M. P. Crump, A. P. Davis, Chem. Sci. 2017, 8,
4056–4061.
9
[
[
2
1
2
1
[17] K. L. Hudson, G. J. Bartlett, R. C. Diehl, J. Agirre, T. Gallagher, L. L.
Kiessling, D. N. Woolfson, J. Am. Chem. Soc. 2015, 137, 15152–15160.
[18] a) H. Dasary, R. Jagan, D. K. Chand, Inorg. Chem. 2018, 57, 12222–12231;
b) S. Yi, V. Brega, B. Captain, A. E. Kaifer, Chem. Commun. (Camb.) 2012,
48, 10295–10297; c) D. W. Zhang, T. K. Ronson, J. Mosquera, A. Martinez,
J. R. Nitschke, Angew. Chem. Int. Ed. 2018, 57, 3717–3721; Angew. Chem.
2018, 130, 3779–3783; d) R. Custelcean, P. V. Bonnesen, N. C. Duncan,
X. H. Zhang, L. A. Watson, G. Van Berkel, W. B. Parson, B. P. Hay, J. Am.
Chem. Soc. 2012, 134, 8525–8534; e) Q. Q. Wang, S. Gonell, S. Leenders,
M. Durr, I. Ivanovic-Burmazovic, J. N. H. Reek, Nat. Chem. 2016, 8, 225–
230.
1
1
Whyte, A. N. Brown, J. H. Kim, M. Samuel, J. D. Mao, Sci. Rep. 2013, 3, Art.
Nr.: 01981.
[
10] a) I. C. Blomfield, Microbiol. Spectr. 2015, 3, Art. Nr.: MBP-0015–2014;
b) A. Bell, E. Severi, M. Lee, S. Monaco, D. Latousakis, J. Angulo, G. H.
Thomas, J. H. Naismith, N. Juge, J. Biol. Chem. 2020, 295, 13724–13736.
11] W. S. Somers, J. Tang, G. D. Shaw, R. T. Camphausen, Cell 2000, 103,
[
[
4
67–479.
[19] a) H. A. Bruson, T. W. Riener, J. Am. Chem. Soc. 1943, 65, 23–27; b) G. R.
Newkome, C. N. Moorefield, K. J. Theriot, J. Org. Chem. 1988, 53, 5552–
5554; c) C. Ji, P. A. Miller, M. J. Miller, J. Am. Chem. Soc. 2012, 134, 9898–
9901; d) H. Destecroix, C. M. Renney, T. J. Mooibroek, T. S. Carter, P. F. N.
Stewart, M. P. Crump, A. P. Davis, Angew. Chem. Int. Ed. 2015, 54, 2057–
2061; Angew. Chem. 2015, 127, 2085–2089.
12] a) A. Di Pasquale, S. Tommasone, L. L. Xu, J. Ma, P. M. Mendes, J. Org.
Chem. 2020, 85, 8330–8338; b) T. S. Carter, T. J. Mooibroek, P. F. N.
Stewart, M. P. Crump, M. C. Galan, A. P. Davis, Angew. Chem. Int. Ed.
2
016, 55, 9311–9315; Angew. Chem. 2016, 128, 9457–9461.
[
13] P. Rios, T. S. Carter, T. J. Mooibroek, M. P. Crump, M. Lisbjerg, M.
Pittelkow, N. T. Supekar, G. J. Boons, A. P. Davis, Angew. Chem. Int. Ed.
[20] a) A. Unciti-Broceta, E. Holder, L. J. Jones, B. Stevenson, A. R. Turner, D. J.
Porteous, A. C. Boyd, M. Bradley, J. Med. Chem. 2008, 51, 4076–4084;
b) K. Robinson, C. J. Easton, A. F. Dulhunty, M. G. Casarotto, ChemMed-
Chem 2018, 13, 1957–1971.
2
016, 55, 3387–3392; Angew. Chem. 2016, 128, 3448–3453.
14] M. Yamashina, M. Akita, T. Hasegawa, S. Hayashi, M. Yoshizawa, Sci. Adv.
017, 3, Art. Nr.: e1701126.
[
[
[
2
15] X. Schaapkens, E. O. Bobylev, J. N. H. Reek, T. J. Mooibroek, Org. Biomol.
Chem. 2020, 18, 4734–4738.
16] a) A. P. Davis, Chem. Soc. Rev. 2020, 49, 2531–2545; b) O. Francesconi, S.
Roelens, ChemBioChem 2019, 20, 1329–1346; c) Y. Ferrand, E. Klein, N. P.
Barwell, M. P. Crump, J. Jimenez-Barbero, C. Vicent, G. J. Boons, S. Ingale,
A. P. Davis, Angew. Chem. Int. Ed. 2009, 48, 1775–1779; Angew. Chem.
[21] a) N. L. S. Yue, D. J. Eisler, M. C. Jennings, R. J. Puddephatt, Inorg. Chem.
2004, 43, 7671–7681; b) N. Yue, Z. Q. Qin, M. C. Jennings, D. J. Eisler, R. J.
Puddephatt, Inorg. Chem. Commun. 2003, 6, 1269–1271; c) N. L. S. Yue,
M. C. Jennings, R. J. Puddephatt, Inorg. Chim. Acta 2016, 445, 37–45.
[22] C. Frassineti, S. Ghelli, P. Gans, A. Sabatini, M. S. Moruzzi, A. Vacca, Anal.
Biochem. 1995, 231, 374–382.
2
009, 121, 1807–1811; d) J. D. Howgego, C. P. Butts, M. P. Crump, A. P.
Davis, Chem. Commun. (Camb.) 2013, 49, 3110–3112; e) P. Stewart, C. M.
Renney, T. J. Mooibroek, S. Ferheen, A. P. Davis, Chem. Commun. (Camb.)
2
018, 54, 8649–8652; f) R. A. Tromans, T. S. Carter, L. Chabanne, M. P.
Manuscript received: June 18, 2021
Version of record online: ■■■, ■■■■
Crump, H. Y. Li, J. V. Matlock, M. G. Orchard, A. P. Davis, Nat. Chem.
Chem. Eur. J. 2021, 27, 1–7
www.chemeurj.org
6
© 2021 The Authors. Chemistry - A European Journal published by Wiley-VCH GmbH
��
These are not the final page numbers!

COMMUNICATION
X. Schaapkens, R. N. van Sluis, E. O.
Bobylev, Prof. Dr. J. N. H. Reek, Dr. T. J.
Mooibroek*
1
– 7
A Water Soluble Pd L Cage for
2
4
Selective Binding of Neu5Ac
The carbohydrate binding pocket of
scopy that the cage binds selectively
to anionic monosaccharides and
4
+
an organic soluble [Pd L ] cage
2
4
À 1
could be equipped with guanidinium-
terminating dendrons to give the
water soluble [Pd L ][NO ] cage. It
strongest to Neu5Ac with K =24 M .
a
The cage had low to no affinity for
the thirteen neutral saccharides
studied.
2
4
3 16
was shown by means of NMR spectro-