Fluorinated Carbohydrates as Lectin Ligands: Simultaneous Screening of a Monosaccharide Library and Chemical Mapping by 19F NMR Spectroscopy

Article abstract of DOI:10.1021/acs.joc.0c01830

Molecular recognition of carbohydrates is a key step in essential biological processes. Carbohydrate receptors can distinguish monosaccharides even if they only differ in a single aspect of the orientation of the hydroxyl groups or harbor subtle chemical

Full text of DOI:10.1021/acs.joc.0c01830

pubs.acs.org/joc
Article
Fluorinated Carbohydrates as Lectin Ligands: Simultaneous
Screening of a Monosaccharide Library and Chemical Mapping by
19F NMR Spectroscopy
̃
J. Daniel Martínez, Ana I. Manzano, Eva Calvino, Ana de Diego, Borja Rodriguez de Francisco,
̀
́
́
Cecilia Romano, Stefan Oscarson, Oscar Millet, Hans-Joachim Gabius, Jesus Jimenez-Barbero,*
̃
and Francisco J. Canada*
Cite This: https://dx.doi.org/10.1021/acs.joc.0c01830
Read Online
sı
*
Metrics & More
Article Recommendations
Supporting Information
ACCESS
ABSTRACT: Molecular recognition of carbohydrates is a key step in essential
biological processes. Carbohydrate receptors can distinguish monosaccharides even if
they only differ in a single aspect of the orientation of the hydroxyl groups or harbor
subtle chemical modifications. Hydroxyl-by-fluorine substitution has proven its merits
for chemically mapping the importance of hydroxyl groups in carbohydrate−receptor
interactions. 19F NMR spectroscopy could thus be adapted to allow contact mapping
together with screening in compound mixtures. Using a library of fluorinated glucose
(Glc), mannose (Man), and galactose (Gal) derived by systematically exchanging every
hydroxyl group by a fluorine atom, we developed a strategy combining chemical
mapping and 19F NMR T2 filtering-based screening. By testing this strategy on the
proof-of-principle level with a library of 13 fluorinated monosaccharides to a set of
three carbohydrate receptors of diverse origin, i.e. the human macrophage galactose-
type lectin, a plant lectin, Pisum sativum agglutinin, and the bacterial Gal-/Glc-binding
protein from Escherichia coli, it became possible to simultaneously define their monosaccharide selectivity and identify the essential
hydroxyls for interaction.
sugar ring.10−19 This protocol synthetically eliminates or
modifies hydroxyl groups (deoxygenation, methylation, ex-
INTRODUCTION
■
Molecular recognition events are at the heart of health and
change by halogens).20 In particular, hydroxyl-by-fluorine
disease. From the chemical perspective, understanding the
substitution has been used to trace key hydroxyl groups for
details of interactions for the underlying functional pairings
contact with either lectins, antibodies, transporters, or
may provide key information for innovative drug discovery and
enzymes.21
design. In this context, carbohydrate oligomers (saccharides,
Fluorine can be considered as an isosteric mimic of the
glycans) are ubiquitous in nature, commonly presented on cell
hydroxyl group, although without the capacity to act as a
surfaces by protein and lipid scaffolds.1−4 Structurally, an
hydrogen-bond donor and with a diminished hydrogen-bond
exceptionally large diversity can be generated by simply
acceptor competence.22−24 Additionally, its particular phys-
exploiting permutations of linkage points and anomeric
icochemical properties25 introduce electronic and polar-
position at each glycosidic linkage.5 As a consequence, glycans
hydrophobic effects.26 Indeed, fluorine modulates the pop-
are “ideal for generating compact units with explicit informa-
ulation of the conformational space27,28 and lipophilicity of
tional properties”,6 and this information is being disclosed to
fluorine-containing carbohydrates.29 Smart use of these
be “read” and “translated” into (patho)physiological processes
features has already allowed development of new molecules
by lectins.4,5 Thus, the analysis of glycan-lectin recognition has
that efficiently act as substrates30 and inhibitors of
become a topic with biomedically promising perspective7,8 and
glycosidases.31,32 Advances for fluorine introduction into
a fructiferous foundation to enhance the symbiosis of
Chemistry and Biology as Lemieux asked for.9
From the molecular recognition perspective, different
approaches have been tested to examine the relevance of
Special Issue: A New Era of Discovery in Carbohy-
drate Chemistry
hydroxyl groups from saccharide units in binding to receptors.
Received: July 30, 2020
One extensively applied approach rests on screening a given set
of available closely related saccharides that display different
stereochemistry and/or substitutions at a certain site within the
https://dx.doi.org/10.1021/acs.joc.0c01830
© XXXX American Chemical Society
J. Org. Chem. XXXX, XXX, XXX−XXX
A

The Journal of Organic Chemistry
pubs.acs.org/joc
Article
organic molecules have made available a large variety of
mono-33 and polyfluorinated saccharides,33−40 which are
highly attractive as chemical probes from different point of
views.21 It is also well-known that fluorine-containing
molecules are extensively used in bioorganic and medicinal
chemistry.23,41 Many of these studies have driven the
development of 19F NMR-spectroscopy methodologies as
valuable tools to study molecular recognition events or to
screen compound libraries.42,43
In this context, we and others have applied 19F-observed
NMR strategies to study glycan−protein interactions by means
of saturation transfer difference (STD NMR-spectroscopy)
measurements using 1D44 and 2D34 experimental designs, by
monitoring chemical shifts perturbations and exchange
kinetics,45,46 by observing line broadening of the 19F NMR
signals,47−49 or by employing relaxation filtering protocols.50,51
Herein, we propose a robust and general method to
efficiently pick up and study the interactions of a library of
fluorinated sugars with a given receptor. By taking advantage of
the large chemical shift range of the 19F nucleus and its
sensitivity, monitoring sugar−protein interactions by a panel of
13 different monofluorinated sugars (with up to 26 well
resolved 19F NMR signals considering the presence of the α
and β-anomers for each sugar, Figure 1, Table S1) provides
information on the selectivity of the binding event in a single
setup. This methodology extends the applicability of the
1
reported T2-filtering strategy and overcomes the limits of H
NMR resolution (see spectrum in Figure 1b).
As proof-of-concept, two lectins and a sugar transporter of
diverse origins and selectivities have been chosen: the human
Macrophage Galactose-type Lectin (MGL, CLEC10A,
CD301), a C-type lectin binding N-acetylgalactosamine in O-
glycans (Tn antigen, CD175) and in N-glycans (Lacdi-
NAc);52−54 Pisum sativum agglutinin (PSA), a plant lectin
selective for α-mannopyranosides and -glucopyranosides;15,55
Figure 1. (a) 19F NMR (1H-decoupled) spectra recorded for each
individual monofluorinated monosaccharide as anomeric mixture.
The α/β anomeric ratios are given between brackets. Lower panel, 19F
NMR spectrum of the full library. Each peak is numbered from lower
to higher field. (b) 1H NMR spectrum of the mixture of the 13
monosaccharides. (c) Representation of the structures of the different
monodeoxy-monofluorinated monosaccharides present in the library.
The corresponding peak number for each anomer in the 19F NMR
spectrum is indicated.
and the glucose/galactose-binding protein (GGBP),56−58
a
bacterial sensor for free monosaccharides. From the analysis of
data from simple 1D 19F NMR experiments by applying
transversal relaxation filters, screening and chemical mapping
are simultaneously achieved. In essence, information on the
monosaccharide selectivity for a particular sugar receptor is
obtained (screening) together with the direct identification of
hydroxyls that are essential for binding and those that can be
chemically substituted or modified without critically com-
promising the binding event (chemical mapping).
the binding site makes contact with its Gal/GalNAc ligands
through the equatorial/axial OH-3 and OH-4 groups.53 In
order to study the importance of each hydroxyl group of the
Gal moiety, the binding of the four possible monodeoxy-
monofluorinated Gal analogues (at positions 2, 3, 4, and 6),
keeping the anomeric position free, was tested to detect those
hydroxyl-to-fluorine substitutions that impair binding. A
similar strategy, using an extended mixture of mono- and
polyfluorinated galactopyranosides and applying a diversity of
techniques, has allowed identification of OH-3 and -4 as the
coordinating groups in a calcium-dependent bacterial
galactophilic lectin.33
Since every monosaccharide exists as a mixture of its α and β
anomers in equilibrium, eight different molecules are present in
solution. The 19F NMR spectrum of the mixture is pleasingly
simple, just showing eight individual 19F NMR signals (Figure
2a), one for each monofluorinated Gal anomer in the mixture.
Their intensities are governed by the anomer ratio at
equilibrium.60
RESULTS
■
Three different types of sugar receptors are deliberately
selected herein to illustrate broad applicability, i.e. a human
lectin involving Ca2+ for direct ligand contact, a plant
agglutinin, and a bacterial sugar transporter.
MGL. MGL belongs to the C-type lectin family charac-
terized by containing a calcium cation at the binding site,
directly involved in carbohydrate recognition by coordination
bonding.52,53 MGL, like the hepatic asialoglycoprotein
receptor, is a transmembrane protein with the carbohydrate
recognition domain (CRD) on top of its extracellular stalk that
oligomerizes to trimers.59 GalNAc in α/β linkage are the
preferred ligands (KD = 12 μM for methyl α-N-acetyl-
galactosaminide, Me α-GalNAc), galactose being a weaker
binder (KD = 0.9 mM for Me α-Gal).53 To perform the
recognition studies, the soluble extracellular ectodomain
containing the CRD was used. It is known that the Ca2+ in
The transverse relaxation time (T2) for each compound was
measured in the free state in the absence of lectin, ranging
between 1.2 to 1.8 s for Gal derivatives (Table S1). The
monofluorinated Gal mixture was added to a solution of MGL,
B
https://dx.doi.org/10.1021/acs.joc.0c01830
J. Org. Chem. XXXX, XXX, XXX−XXX

The Journal of Organic Chemistry
pubs.acs.org/joc
Article
(β = 57%, α = 73%) and 4F-Gal (β = 98%, α = 80%) are
significantly less altered.
To confirm that this selective signal reduction is due to the
binding of the 19F-containing Gal entities to the CRD, a
competition experiment was performed by adding Me α-
GalNAc to the mixture. The recovery of the 2F-Gal and 6F-Gal
signals was evident, indicating that they are displaced from the
binding site by the strong competitor (Figure 2b, upper panel).
Drawing a conclusion from chemical mapping30,49 is
straightforward: the modification of either hydroxyl at 3 or 4
eliminates a coordination bond in the interaction between
sugar and Ca2+. Therefore, the signals corresponding to 3F-Gal
and 4F-Gal are not affected by the lectin and do not show
significant signal reduction. On the contrary, the hydroxyls at
positions 2 and 6 can be substituted by fluorine. Their 19F
NMR signals are clearly reduced in the presence of the MGL
due to binding.
Since the broad dispersion of 19F NMR chemical shifts of
the four anomeric pairs of the monofluorinated Gal analogues
is more than 30 ppm (between −199 and −230 ppm), the
feasibility to test a broad panel of monofluorinated
monosaccharides was envisioned. Thus, the four monodeoxy-
monofluorinated D-glucoses (2F-Glc, 3F-Glc, 4F-Glc, and 6F-
Glc) and D-mannoses (2F-Man, 3F-Man, 4F-Man, and 6F-
Man) together with 2-deoxy-2-fluoro-L-fucose (2F-Fuc) were
added to provide a library with 13 different anomeric pairs of
monofluorinated monosaccharide (Figure 1). All compounds
were available from commercial sources except 6F-Man that
was chemically synthesized (see Experimental Section).
This mixture with the 13 fluorinated monosaccharide
anomeric pairs gives a very crowded 1H NMR spectrum
(Figure 1b). In contrast, its proton-decoupled 19F NMR
spectrum presents well-resolved individual signals for each of
the 26 different molecules in the sample, which are spread over
40 ppm (Figure 1a). Thus, the extended compound library,
including the monofluorinated Gal, Glc, and Man analogues,
was now tested with MGL, applying again the T2-filtering
strategy (Figure 3). For qualitative visualization of the NMR
Figure 2. (a) 19F NMR (1H-decoupled) spectrum of the F-Galactose
mixture (2F-Gal (0.5 mM); 3F-Gal (0.68 mM), 4F-Gal (0.37 mM),
and 6F-Gal (0.45 mM) in absence of the protein, without T2
relaxation filter (lower panel), and after applying a 800 ms T2
relaxation filter (upper panel). (b) 19F NMR (1H-decoupled)
spectrum of the same mixture in the presence of 0.015 mM MGL
before (lower panel) and after (mid panel) the application of a 800
ms T2 relaxation filter. The upper panel shows the spectrum after the
addition of 0.5 mM Me α-GalNAc with the same relaxation filter.
and the T2 filtering strategy50,61,62 was applied to identify the
binders. Briefly, those molecules that bind to the protein
drastically change their hydrodynamic behavior in the bound
state, and thus their rotational motion correlation time
increases toward that of the large protein, with a concomitant
reduction in T2. Additionally, the effective transverse relaxation
is also affected by the kinetics of the chemical exchange process
between the free and bound states, further reducing the
observed T2, especially if the system no longer follows the fast
chemical exchange regime. This reduction in T2, which is in
the first instance manifested in standard 1D NMR spectra as
signal broadening, can be easily transformed into a signal-
intensity reduction by the application of a standard Carr−
Purcell−Meiboom−Gill (CPMG) spin echo pulse train
sequence before acquisition. The filtered NMR spectrum
displays the NMR signals of the binders significantly reduced
or even suppressed, compared to those of the unbound
compounds.
Figure 3. 19F NMR (1H-decoupled) T2-filtered spectra recorded for
the fluorinated monosaccharide library (Man, Glc, and Gal analogues)
in the presence of MGL (30 μM). (a) Spectrum acquired with a short
8 ms T2 filter. (b) Spectrum acquired with long 160 ms T2 filter
factored 1.1 times. (c) Difference spectrum.
Figure 2b shows the comparison of the 19F NMR spectra
recorded for the mixture of monofluorinated galactose
derivatives in the presence of MGL (lower panel) with that
obtained by applying a spin−echo filter of 800 ms (central
panel). The drastic reduction of the intensity of signals in the
presence of protein (Figure 2b central panel) relative to the
experiment in its absence (Figure 2a upper panel) correspond-
ing to 2F-Gal (β = 2%, α = 4%) and 6F-Gal (β = 1%, α = 7%)
is clearly observed, while the NMR signals obtained for 3F-Gal
signals affected by the lectin, the obtained filtered 19F NMR
spectrum was subtracted from the nonfiltered one following
the protocol described in the Experimental Section (a
correction factor f was applied to the filtered spectrum to
account for the signal reduction due to transversal relaxation
unrelated with the presence of the protein). Only the 19F NMR
C
https://dx.doi.org/10.1021/acs.joc.0c01830
J. Org. Chem. XXXX, XXX, XXX−XXX

The Journal of Organic Chemistry
pubs.acs.org/joc
Article
signals corresponding to binders should appear in the
difference spectrum. Indeed, the peaks corresponding to 2F-
Gal (peaks 13 and 14) and 6F-Gal (peaks 21 and 22) are
clearly displayed in the difference spectrum (Figure 3) in
accordance with the results of the experiment described above
for the smaller sized Gal library.
Given the encouraging results for the first system, the
suitability of the monofluorinated monosaccharide library for
simultaneous ligand screening and chemical mapping was
further tested with two other types of carbohydrate receptors
with different sugar selectivities.
Figure 5. Close-up view of selected 19F NMR peaks recorded in the
19F NMR T2-filtered spectra (720 ms) of the fluorinated
monosaccharide library in the absence of lectin (green), in the
presence of PSA (blue), and when adding different concentrations of
Me α-Man.
PSA. Pisum sativum agglutinin (PSA), a leguminous lectin
with a “jelly roll” fold,63 was also tested. PSA is selective for
Man/Glc-containing oligosaccharides without involvement of
Ca2+ in contact with the sugar, but displays weak affinity for
single monosaccharides: 0.53 and 1.15 mM for methyl α-
mannoside (Me α-Man) and methyl α-glucoside (Me α-Glc),
respectively.55
the competitor (Figure 5), indicating that this molecule is not
a binder of the lectin. Very likely, the signal observed in the
difference spectrum described above is due to the intrinsic fast
relaxation of 4F-α-Man (19F-T2,free = 0.70 s, Table S1).
Analogous results were observed in the absence of the lectin;
i.e., T2 of 4F-Man is not affected by the presence of PSA. A
mixed situation took place for the 6F-Man and 6F-Glc
derivatives (Figure 5). In these cases, the observed signals in
the difference experiment are due to ligand binding and to fast
Several 19F NMR peaks diminished (Figure 4a,b) when the
T2 filter was applied. Those present in the difference spectrum
relaxation. In fact, the intrinsic T2,free for the corresponding 19
F
signals of the 6F-Man (19F-T2,free = 1.00 s and 19F-T2,free = 0.82
s for the β and α anomers, respectively) and 6F-Glc (19F-T2,free
= 0.92 s and 19F-T2,free = 0.90 s for the β and α anomers,
respectively) derivatives is also rather short. On the other side,
indeed, the initial decrease in signal intensity induced by the
presence of the protein was subtle, but recovery was almost
complete after addition of a small concentration of competitor,
thus also confirming affinity, although likely weaker.
Interestingly, it has been described that hydroxyls at
positions 2 and 3 of glucose can be substituted by fluorine
while retaining binding by PSA; however, when F is at the 6
position the reported binding was minimal and modifications
at OH-4 abolished the binding,15,64 supporting the results
presented here regarding PSA selectivity.
Figure 4. 19F NMR (1H-decoupled) T2-filtered spectra recorded for
the fluorinated monosaccharide library in the presence of PSA (25
μM, ligand/protein ratio around 36:1). (a) Spectrum acquired with a
short 8 ms T2 filter. (b) Spectrum acquired with 720 ms of T2 filter
factored 1.6 times. (c) Difference spectrum. The peak corresponding
to 4F-α-Man, in red, is a “false positive” (see text). **2-Fluoroethanol
added as internal reference.
GGBP. The third receptor is the bacterial GGBP. It is
involved in chemotaxis and sugar transport in bacteria and has
a very high affinity for Glc (0.04 μM) and Gal (0.13 μM)65,66
typical for bacterial binders of free monosaccharides. Its
structure consists of two globular Rossman fold domains, and
differently from the tested lectins, GGBP presents a deep
binding pocket at the hinge connecting and closing both
domains around the monosaccharide ligand.56,57 When GGBP
was added to the monofluorinated monosaccharide library, the
signals belonging to Glc and Gal molecules with F atoms at
positions 4 or 6 showed reduced peak intensities in the T2-
filtered spectrum (Figure 6). On the contrary, those signals
corresponding to Glc and Gal moieties substituted at either
position 2 or 3 were not affected by the presence of GGBP.
This evidence indicates that the OH groups at those 2 and 3
positions are required for the binding to take place and cannot
be substituted by a fluorine atom. In the difference 19F NMR
spectrum, signals for 2F-α-Man, 4F-α-Man, and 6F-α-Man also
appear (Figure 6c). However, when Glc was added to the
library/receptor mixture as a competitor, the corresponding
signals of those fluoromannoses were not recovered (Figure 7).
In the cases of 4F-α-Man and 6F-α-Man, as for PSA, this
(Figure 4c) correspond to 2F-Glc (peaks 5 and 6) and 2F-Man
(11 and 20), 3F-Glc (1 and 7), 3F-Man (3 and 10), 6F-Glc
(25 and 26), and 6F-Man (23 and 24). Neither 4F-Glc nor any
Gal derivatives were observed in the difference 19F NMR
spectrum. However, a limitation of the method was detected.
The difference 19F NMR spectrum also displays “false
positives” corresponding to fast relaxing signals (see below),
especially when the applied T2 filter is long enough (720 ms in
this experiment). This was the case for 4F-α-Man (peak 12,
19F-T2,free = 0.7 s), whose signal relaxation was significantly
faster than that of the other signals of the molecules present in
the mixture (Table S1).
To confirm specific binding, the difference 19F NMR
spectrum was again complemented with competition experi-
ments (Figure 5) in the presence of a known ligand (Me α-
Man). The signals corresponding to the 2F- and 3F- Man/Glc
derivatives were now clearly observed, indicating that 2F-/3F-
Man/Glc are indeed displaced from the binding site by Me α-
Man. On the contrary, no difference in the intensities of the
4F-α-Man signal was observed in the absence and presence of
D
https://dx.doi.org/10.1021/acs.joc.0c01830
J. Org. Chem. XXXX, XXX, XXX−XXX

The Journal of Organic Chemistry
pubs.acs.org/joc
Article
Both rotational motion and exchange effects add together in
the T2 filtering strategy and allow the efficient detection of
medium- to low-affinity binders (from low micromolar to
millimolar KD), even using high ligand/protein ratios.62 The
application of the CPMG-based T2 filtering scheme is fairly
straightforward, and usually a reasonable number of spin−echo
loops before acquisition is sufficient to obtain highly sensitive
NMR spectra with the required information discriminating
binders from nonbinders. From the practical perspective, the
current library renders very well resolved 19F NMR spectra
with separated signals for all different monosaccharide moieties
in the mixture. Obviously, other fluorinated saccharides could
well be added to the mixture increasing the screening power of
the concept. As an added value for the 19F observation, the
experiments do not require any deuterated buffer, thus
simplifying the experimental setup.
Regarding the screening process, in the first instance, and
assuming that all 19F nuclei in the library have similar T2
relaxation times when free in solution, it should be possible to
qualitatively visualize those signals that are affected by the
protein. To do so, a difference NMR spectrum is obtained by
subtracting the spectrum recorded using a short spin−echo
delay from a second one measured employing a longer delay.
However, the 19F NMR signals of some molecules, such as 4F-
α-Man, 6F-α/β-Man, and 6F-α/β-Glc (Table 1), relax
significantly faster (T2 < 1 s) than the others (T2 > 1.2 s)
and their peaks consistently appear in the difference spectrum
when long spin−echo relaxation delays are used. Therefore, to
unambiguously assess the existence of specific binders at the
carbohydrate-binding site, the difference 19F spectrum should
be complemented with the information provided by additional
competition experiments carried out by adding a known ligand
of the lectin. The comparison of the recovered 19F NMR
signals in the presence of an excess of the competitor in the
lectin/library sample can be expressed as the ratio of signal
intensities, It(+C/−C), measured in spectra acquired with a
relaxation filter t in presence (+C) and absence (-C) of
competitor (C), thus highlighting the specific binders (Figure
8).
At least qualitatively, these signal recovery data allow the
specificities of the three sugar receptors to be distinguished,
correlating them with their known monosaccharide selectivity:
MGL only recognizes Gal moieties and PSA binds Glc and Gal
analogues, while GGBP interacts with Glc and Gal
monosaccharides. Additionally, information on the selectivity
for the anomeric configuration can be gleaned from the signal
recovery data in the T2-filtered competition experiments. For
instance, for PSA, the α-anomers show a higher recovery ratio
than their corresponding β-anomers in accordance with
previous reports. On the contrary, based on X-ray and NMR
structural data, GGBP has been described to display specificity
for β-anomers.57 The recovery ratio data here presented not
only are in agreement with that selectivity but also show that
the α-anomers are binders, as previously suggested by means of
ligand-binding kinetic experiments.65,66
As mentioned above, the OH by F substitution has been
extensively used in carbohydrate chemistry to map the key
hydroxyl groups of a given sugar that are involved in their
recognition by lectins, antibodies, transporters, or en-
zymes.10−20 The methodology presented herein, which
employs a rationally assembled collection of monofluorinated
monossaccharides for which their hydroxyl groups have been
systematically substituted by fluorine atoms, allows dissecting
Figure 6. 19F NMR (1H-decoupled) T2-filtered spectra recorded for
the fluorinated monosaccharide library in the presence of GGBP (25
μM, ligand/protein ratio ca. 36:1). (a) Spectrum acquired with a
short 8 ms T2 filter, (b) spectrum acquired with long 400 ms of T2
filter factored 1.3 times, (c) difference spectrum. For peaks labeled in
green and red, see text. **2-Fluoroethanol added as internal
reference.
Figure 7. Close-up view of selected 19F NMR peaks recorded in the
19F NMR T2-filtered spectra of the fluorinated monosaccharide library
in the absence of receptor (green), in the presence of GGBP (blue),
and when adding different concentrations of the Glc competitor (0.23
mM orange and 0.9 mM black). In all cases, the NMR experiments
were acquired using 720 ms of T2 filter.
behavior is again due to the intrinsic fast relaxation of the 4F-
α-Man and 6F-α-Man 19F signals. Interestingly, 2F-α-Man is a
special case; its signal reduction only takes place in the
presence of the protein, and it is not affected by glucose
(Figure 7). This result suggests 2F-α-Man interacts with
GGBP but at a location different from the canonical sugar-
binding site.
DISCUSSION
■
The tested screening method is based on the dramatic
differences in transverse relaxation observed for binders within
a library of fluorinated monosaccharides, when acquiring NMR
spectra in the absence or presence of a carbohydrate-binding
protein. The transversal relaxation time is related to the
rotational motion correlation time of the molecule, and it
sharply decreases as the correlation time increases. When
monosaccharides interact with a large receptor, they adopt the
correlation time of the macromolecule during the time the
complex is associated and, thus, undergo a critical decrease of
their T2. This change in T2 may be followed in a
straightforward manner under fast exchange conditions
between bound and free states. Thus, only a single 19F NMR
signal appears in the spectrum at the averaged chemical shift of
the exchanging states weighted by their corresponding molar
fractions. In fact, the observed effective T2 also depends on the
kinetics of the exchange between the free and bound forms.
E
https://dx.doi.org/10.1021/acs.joc.0c01830
J. Org. Chem. XXXX, XXX, XXX−XXX

The Journal of Organic Chemistry
pubs.acs.org/joc
Article
sugar selectivity of the tested receptor, (ii) detecting its
anomer preference, and (iii) identifying the key hydroxyls for
binding, distinguishing them from those that can be chemically
modified in the quest to find new binders. Extending this
approach to other saccharides (aminosugars and sialosides)
and to synthetic libraries of disaccharides will especially be
attractive to screen a variety of carbohydrate−receptor families,
on the way “from biology to drug target”.68 In this sense,
Siglecs, sialoside receptors proposed to act as “immune cell
checkpoints in disease”,69 or the multifunctional galectins,70,71
look like exciting targets to start with. The versatility of the
described strategy is evident: it shows applicability to lectins
and sensor/transport proteins and proved to be suitable to
cover diverse selectivities and wide-ranging affinities, from sub-
micromolar (40 nM for the Glc-GGBP complex) to over
millimolar (1.15 mM for the Me β-Glc/PSA complex)
dissociation constants. Thus, the method is robust and
envisioned to find wide application.
Figure 8. Competition experiments and 19F NMR T2-filtered spectra
for the analysis of the interaction of the fluorinated monosaccharides
library with the different lectins. The x axis corresponds to the signal
intensity recovery, It(+C/−C), expressed as the ratio between the relative
decay at time t in the presence (+) and absence (−) of competitor
(C). From left to right: (a) MGL, T2 filter t = 720 ms. MGL (30 μM)
and competitor Me α-GalNAc (0.9 mM); (b) PSA, T2 filter t = 720
ms, PSA (25 μM) and competitor Me α-Man (18 mM); and (c)
GGBP, T2 filter t = 400 ms, GGBP (25 μM) and competitor Glc (0.9
EXPERIMENTAL SECTION
■
Materials. PSA was from a commercial source (Sigma-Aldrich-
Merck) and dissolved in phosphate-buffered saline at pH 7.5. MGL
ectodomain was recombinantly produced in E. coli and routinely
checked for purity and activity as previously described, including
ascertaining GalNAc-inhibitable histochemical staining.53,72 The
samples for NMR were prepared in deuterated Tris buffer (10
mM), containing CaCl2 (1 mM) and NaCl (75 mM) at pH 7.5 by
means of five ultrafiltration−dilution buffer exchange steps with a 10
kDa cutoff membrane. GGBP was expressed in E. coli and purified as
previously described,56 and the samples were prepared in 20 mM Tris,
containing 150 mM NaCl and 10 mM CaCl2 at pH 7.0. Protein
concentrations were measured by UV spectrometry.
mM). In all cases the mixture with 0.9
monosaccharide was used.
0.3 mM of each
chemical mapping information regarding the importance of
each individual hydroxyl group in the interaction with its
receptor. For MGL, the same experiment allows identifying its
selectivity for Gal moieties and simultaneously shows that
hydroxyls at positions 3 and 4 are essential to keep the
interaction ability of the Gal analogue, while hydroxyls 2 and 6
can be modified while still maintaining binding to MGL. PSA
can recognize Man and Glc, epimers at position 2. Thus, the
orientation of OH-2, axial in Man, equatorial in Glc, is not
essential for binding, and consequently, both fluorinated
epimers 2F-Man and 2F-Glc are recognized. Moreover, it
can be inferred that modifications at OH-3 are tolerated, as F
to OH substitution at this position does not block binding to
the lectin. On the contrary, OH-4 is essential for binding while
the modification at position 6 still sustains a weak interaction.
Finally, for GGBP, which also recognizes two monosaccharides
that share the equatorial configuration at C2, i.e. Glc and Gal,
OH-2 and OH-3 are necessary for binding, while OH-4 (either
axial in Gal or equatorial in Glc) and OH-6 can be modified.
Thus, the binding pattern is completely opposite to that
observed for PSA. Additionally, for GGBP, the possibility of a
secondary binding site has been deduced, given the existence
of binding to 2F-α-Man (see Figure 6; 2F-α-Man is marked in
green) and the fact that this interaction is not abolished by Glc
(see Figure 7, the signal intensity of 2F-α-Man is not recovered
after addition of Glc), the canonical ligand of GGBP. The
implications of this result remain to be explored.
The monofluorinated monosaccharide mixtures were prepared
from concentrated stock solutions of each individual monosaccharide
depending on their availability, either commercial or from synthesis.
Final concentrations in the mixtures were centered around 0.5 mM or
0.9 mM, depending on the experiment, with variations in 35% range.
Given the intrinsic different equilibrium populations of the different
anomers for a given monosaccharide, it is impossible to use the same
concentration for each individual species.
Fluorinated Monosaccharides. 2-Deoxy-2-fluoro-glucose, 3-
deoxy-3-fluoro-glucose, 4-deoxy-4-fluoro-glucose, 6-deoxy-6-fluoro-
glucose, 2-deoxy-2-fluoro-galactose, 3-deoxy-3-fluoro-galactose, 4-
deoxy-4-fluoro-galactose, 6-deoxy-6-fluoro-galactose, 2-deoxy-2-fluo-
ro-mannose, 3-deoxy-3-fluoro-mannose, 4-deoxy-4-fluoro-mannose, 2-
deoxy-2-fluoro-fucose, 2-fluoroethanol, Me α-N-acetylgalactosami-
nide, and Me α-mannopyranoside were from commercial sources
(Sigma-Aldrich Merck, Spain; Carbosynth, UK). 6-Deoxy-6-fluoro-
mannose was synthesized as described in the Supporting Information,
and its analytical data were consistent with literature values.73
Characterizations of the intermediates in reaction steps in the
synthesis are described below. NMR peak assignments were made
using correlation spectroscopy (COSY) and heteronuclear single-
quantum coherence (HSQC).
Methyl 2,3,4-Tri-O-benzoyl-6-deoxy-6-fluoro-α-D-manno-
pyranoside (2). Compound 174 (50 mg, 0.098 mmol) was dissolved
in dry CH2Cl2(1.2 mL), then the solution was cooled to −78 °C, and
DAST (98 μL, 0.74 mmol) was slowly added dropwise. The reaction
mixture was kept at −78 °C for 30 min, then warmed to rt, and left
stirring overnight. The solution was then cooled to −20 °C, and the
reaction was quenched with MeOH. The solvents were evaporated,
and the residue was purified by silica gel column chromatography
(toluene/EtOAc, 98:2 → 8:2, v/v) to give 2 as a yellowish solid (37
Although T2 filtering has been merely applied herein from a
qualitative perspective, the obtained data clearly pave the way
to perform further quantitative affinity studies. In fact, such
values could be in principle deduced for each isolated
monosaccharide from competition experiments, using a
competitor with a known affinity constant.67
1
mg, 0.07 mmol, 74%). Rf = 0.8, toluene/EtOAc 8:2; H NMR (400
MHz, CDCl3): δ 8.14−8.05 (m, 2H, HBz), 8.02−7.94 (m, 2H, HBz),
7.86−7.78 (m, 2H, HBz), 7.64−7.59 (m, 1H, HBz), 7.56−7.46 (m, 3H,
HBz), 7.46−7.36 (m, 3H, HBz), 7.29−7.23 (m, 2H, HBz), 5.94−5.85
(m, 2H, H-3, H-4), 5.68 (dd, J = 3.0, 1.8 Hz, 1H, H-2), 5.02 (d, J =
In summary, using this 19F NMR-based T2-filtering strategy
using a library of fluorinated monosaccharides generated
through systematic OH-to-F substitutions allows (i) defining
F
https://dx.doi.org/10.1021/acs.joc.0c01830
J. Org. Chem. XXXX, XXX, XXX−XXX

The Journal of Organic Chemistry
pubs.acs.org/joc
Article
1.8 Hz, 1H, H-1), 4.64 (dt, J = 46.9, 3.7 Hz, 2H, H-6ab), 4.34−4.21
(m, 1H, H-5), 3.55 (s, 3H, OCH3). 19F NMR (376 MHz, CDCl3): δ
−231.70 (td, J = 47.2, 22.9 Hz). All analytical data were consistent
with literature values.74
around 50 mM in deuterated water of each fluorinated mono-
saccharide (glucose, galactose, mannose, and L-fucose) were prepared.
These stock solutions were appropriately mixed and diluted to the
final concentration used in each experiment. The concentrations were
estimated by integrating the corresponding signals in the 19F-
spectrum. 2-Fluoroethanol was added to the mixture as the internal
reference. To prepare the samples of the monosaccharide library in
the presence of proteins, 0.2 or 0.5 mL (for using 2 mm and 5 mm
NMR tubes, respectively) aliquots of the mixture with 0.9 mM of each
fluorinated monosaccharide were dried in a speed-vac, and the
resulting powder reconstituted with the same volume of the
corresponding buffer with and without protein. T2 values were
obtained from a series of CPMG experiments recorded with
increasing number n (spin echo loops). Experiments with up to 16
different spin echo total relaxation times ranging from 8 to 8000 ms
were determined.
Detection of Ligand Binding by T2-Filtered Experiments. A
general protocol was followed using a protein-containing solution
with a concentration between 10 μM and 30 μM. The mixtures of
monofluorinated monosaccharides were prepapred by mixing aliquots
of each monosaccharide from highly concentrated. The final
concentration of each monosaccharide in the mixture was around
0.9 mM ([α]+[β]) ranging between 0.6 and 1.2 mM depending their
availability. The ligand to protein ratio (L/P) was maintained
between a 20- to 150-fold excess, optimized in each case to yield
comparable T2 decay responses between the three systems (PSA,
GGBP, and MGL). CPMG experiments were carried out as previously
described, but recording an initial reference experiment with 2 CPMG
loops with τ = 2 ms (8 ms total relaxation time) and one to five
additional experiments with CPMG filters between 16 and 400
CPMG echo loops (64 to 1600 ms, respectively; the exact values of
the spectra selected are indicated in each experiment).
1-O-Acetyl-2,3,4-tri-O-benzoyl-6-deoxy-6-fluoro-α-D-man-
nopyranoside (3). Compound 2 (180 mg, 0.35 mmol) was
dissolved in Ac2O/AcOH (2:1, 3.5 mL). H2SO4 (4 μL, 0.07 mmol)
was slowly added dropwise at 0 °C, and the mixture was stirred for 5
h. The reaction was then diluted with AcOEt and washed with sat.
NaHCO3(aqueous). The organic layer was dried over MgSO4,
filtered, and concentrated. The residue was purified by silica gel flash
column chromatography (Toluene/EtOAc, 98:2 → 8:2, v/v) to give 3
as a white powder (150 mg, 0.28 mmol, 80%). Rf = 0.62, Tol/AcOEt
9:1; [α]2D0 = −64.1 (c = 0.6, CHCl3); 1H NMR (500 MHz, CDCl3): δ
8.13−8.06 (m, 2H, HBz), 8.02−7.94 (m, 2H, HBz), 7.86−7.77 (m, 2H,
HBz), 7.66−7.60 (m, 1H, HBz), 7.56−7.52 (m, 1H, HBz), 7.50 (t, J =
7.8 Hz, 2H, HBz), 7.47−7.43 (m, 1H, HBz), 7.40 (t, J = 7.8 Hz, 2H,
HBz), 7.28 (t, J = 7.9 Hz, 2H, HBz), 6.39 (d, J = 2.0 Hz, 1H, H-1), 5.99
(t, J = 10.1 Hz, 1H, H-4), 5.92 (dd, J = 10.1, 3.3 Hz, 1H, H-3), 5.72
(dd, J = 3.4, 2.0 Hz, 1H, H-2), 4.69−4.56 (m, 2H, H-6ab), 4.34 (ddt,
J = 23.3, 10.1, 3.2 Hz, 1H, H-5), 2.28 (s, 3H, OCOCH3). 13C{1H}
NMR (126 MHz, CDCl3): δ 168.25 (OCOCH3), 165.70 (COBz),
165.40 (COBz), 165.35 (COBz), 133.86 (CBz), 133.78 (CBz), 133.53
(CBz), 130.13 (2CBz), 129.94 (2CBz), 129.89 (2CBz), 129.02 (CBz),
128.89 (CBz), 128.85 (CBz), 128.81 (2CBz), 128.67 (2CBz), 128.51
(2CBz), 90.81 (C-1), 81.34 (d, J = 176.4 Hz, C-6), 71.92 (d, J = 19.3
Hz, C-5), 69.72 (C-3), 69.30 (C-2), 65.78 (d, J = 7.0 Hz, C-4), 21.11
(OCOCH3). 19F NMR (376 MHz, CDCl3): δ −232.65 (td, J = 47.0,
23.3 Hz). HRMS (ESI-TOF) m/z [M + Na]+ calcd for C29H25FO9Na
559.1380; found 559.1396.
6-Deoxy-6-fluoro-D-mannose (4). Compound 3 (150 mg, 0.28
mmol) was dissolved in dry MeOH (2 mL), and then solid sodium
methoxide was added until pH = 10−11. The reaction was stirred for
3 h, then quenched with Dowex 50WX8 H+ form, filtered, and
concentrated. The residue was purified by silica gel flash column
chromatography (CH2Cl2/MeOH, 98:5, v/v) to give 4 as a white
solid (40 mg, 0.22 mmol, 78%, α:β 9:1). Rf = 0.2, CH2Cl2/MeOH
9:1; 1H NMR (500 MHz, CD3OD): δ 5.11 (d, J = 1.7 Hz, 1H, H-1α),
4.73−4.54 (m, 2H, H-6a, H-6b), 3.91 (dddd, J = 26.0, 10.0, 4.7, 2.0
Hz, 1H, H-5), 3.83 (dd, J = 3.4, 1.7 Hz, 1H, H-2), 3.80 (dd, J = 9.2,
3.4 Hz, 1H, H-3), 3.69 (t, J = 9.6 Hz, 1H, H-4). 19F NMR (470 MHz,
CD3OD) δ −233.99 (td, J = 47.9, 23.8 Hz), −234.81 (td, J = 47.8,
25.8 Hz). All analytical data were consistent with literature values.73
NMR Experiments. All NMR spectra were recorded on a 500
MHz Bruker spectrometer (470.56 MHz for fluorine) equipped with a
19F probe (19F, 1H SEF from Bruker) at 298 K in D2O unless
In order to obtain the difference spectrum, the T2 filtered spectra
were multiplied by a factor f to correct the signal decay in the absence
of protein. f is defined as the mean value of the ratio of 19F signal
intensities after the first (I1) and the last (It) CPMG experiments for
all the fluorinated monosaccharides in the mixture: f = I /It. The first
1
CPMG spectrum is acquired with t1 = 8 ms, and the last one at t =
160, 720, and 400 ms in each case, yielding a factor f of 1.1, 1.6, and
1.2 as shown in Figures 3, 4, and 6, respectively.
Detection of Ligand Binding by Competition Experiments.
Competition (displacement) experiments were performed by adding
an excess of a cognate sugar to the lectin/monofluorinated
monosaccharide mixtures. In particular, Me α-N-acetylgalactosami-
nide (12 μM Kd) up to 1 mM for MGL,53 Me α-mannopyranoside
(530 μM)55 up to 18 mM for PSA, and glucose (0.04 μM)65 up to 1
mM for GGBP. Equivalent experiments, with the same CPMG
relaxation filter parameters to those used for detection of ligand
binding, were carried out to observe recovery in signals that had
previously diminished as a consequence of binding. Each experiment
was repeated upon sequential addition of the competing ligand.
The signal recovery ratio represented in Figure 8 for each
fluorinated monosaccharide in the presence of the lectin after
relaxation time ti, Iti(+C/−C), with (+C) or without (−C)
competitor was calculated from the ratio of relative signal decays in
the presence (Iti/It1)+C and in the absence (Iti/It1)−C of competitor for
t1 = 8 ms and ti = 720, 720, and 400 ms for MGL, PSA, and GGBP,
respectively.
1
otherwise is indicated. Standard pulse sequences 1D H with and
without decoupling 19F and 1D-19F with and without decoupling 1H
included in Topspin acquisition software were used. For measuring
transversal relaxation times, T2, the CPMG pulse sequence was
used.75,76 It was as follows: [D-90x-(τ-180y-τ)n-acquire], with a
prescan delay of 4 s and a pre and post 180° pulse echo delay τ of 2
ms. The number n of echo loops varies from 2 to 2000, depending on
the experiment. The 90x and 180y pulse durations were calculated for
each sample. The total time used for the relaxation filter corresponds
to n times the spin echo pulse was applied: n(2τ+180y) (typically
between 8 ms to 8 s).
ASSOCIATED CONTENT
* Supporting Information
The Supporting Information is available free of charge at
https://pubs.acs.org/doi/10.1021/acs.joc.0c01830.
■
sı
Includes the scheme of synthesis for 6F-Man and
characterization spectra of intermediates and final
product, the individual 19F spectra for each monofluori-
nated monosaccharide, and a table with their 19F
transversal relaxation times. (PDF)
19F was set as the observed nucleus, and proton decoupling was
carried out during acquisition using the WALTZ-16 scheme.
Transverse Relaxation Time of F-Monosaccharides. To carry
out the relaxation filtered experiments, individual stock solutions
G
https://dx.doi.org/10.1021/acs.joc.0c01830
J. Org. Chem. XXXX, XXX, XXX−XXX

The Journal of Organic Chemistry
pubs.acs.org/joc
Article
facility at the CIB Margarita Salas. The referees are
acknowledged for helping in improving the manuscript
AUTHOR INFORMATION
■
Corresponding Authors
́
́
Jesus Jimenez-Barbero − CIC bioGUNE, Basque Research
Technology Alliance, BRTA, 48160 Derio, Spain; Ikerbasque,
Basque Foundation for Science, 48009 Bilbao, Spain;
Department of Organic Chemistry II, Faculty of Science and
Technology, UPV-EHU, 48940 Leioa, Spain; orcid.org/
0000-0001-5421-8513; Email: jjbarbero@cicbiogune.es
ABBREVIATIONS
■
NMR, nuclear magnetic resonance; CPMG, Carr−Purcell−
Meiboom−Gill; STD, saturation transfer difference; 1D, one-
dimensional; MGL, Macrophage Galactose-type Lectin; PSA,
Pisum sativum agglutinin; GGBP, glucose/galactose-binding
protein; CRD, carbohydrate recognition domain; CLEC, C-
Type lectin; CD, cluster of differentiation
Francisco J. Canada − Centro de Investigaciones Biológicas
̃
Margarita Salas, CSIC, 28040 Madrid, Spain; Centro de
Investigación Biomédica en Red-Enfermedades Respiratorias
(CIBERES), 28029 Madrid, Spain; orcid.org/0000-
0003-4462-1469; Email: jcanada@cib.csic.es
REFERENCES
■
(1) Ginsburg, V.; Neufeld, E. F. Complex Heterosaccharides of
Animals. Annu. Rev. Biochem. 1969, 38, 371.
(2) Varki, A. Biological roles of glycans. Glycobiology 2017, 27, 3.
(3) Gagneux, P.; Aebi, M.; Varki, A. In Essentials of Glycobiology, 3rd
ed.; Varki A, C. R., Esko, J. D., et al., Eds.; Cold Spring Harbor
Laboratory Press: Cold Spring Harbor (NY), 2017; p 253.
(4) Kaltner, H.; Abad-Rodriguez, J.; Corfield, A. P.; Kopitz, J.;
Gabius, H.-J. The sugar code: letters and vocabulary, writers, editors
and readers and biosignificance of functional glycan-lectin pairing.
Biochem. J. 2019, 476, 2623.
Authors
J. Daniel Martínez − CIC bioGUNE, Basque Research
Technology Alliance, BRTA, 48160 Derio, Spain
Ana I. Manzano − Centro de Investigaciones Biológicas
Margarita Salas, CSIC, 28040 Madrid, Spain
Eva Calvino − Centro de Investigaciones Biológicas Margarita
̃
Salas, CSIC, 28040 Madrid, Spain
̈
(5) Bohm, M.; Bohne-Lang, A.; Frank, M.; Loss, A.; Rojas-Macias,
M. A.; Lutteke, T. Glycosciences.DB: an annotated data collection
Ana de Diego − Centro de Investigaciones Biológicas
Margarita Salas, CSIC, 28040 Madrid, Spain
Borja Rodriguez de Francisco − Centro de Investigaciones
Biológicas Margarita Salas, CSIC, 28040 Madrid, Spain;
orcid.org/0000-0003-4778-313X
̈
linking glycomics and proteomics data (2018 update). Nucleic Acids
Res. 2019, 47, D1195.
(6) Winterburn, P. J.; Phelps, C. F. The Significance of Glycosylated
Proteins. Nature 1972, 236, 147.
̀
Cecilia Romano − Centre for Synthesis and Chemical Biology,
́
(7) Solís, D.; Bovin, N. V.; Davis, A. P.; Jimenez-Barbero, J.;
University College Dublin, Belfield, Dublin 4, Ireland
Stefan Oscarson − Centre for Synthesis and Chemical Biology,
University College Dublin, Belfield, Dublin 4, Ireland;
orcid.org/0000-0002-8273-4918
Romero, A.; Roy, R.; Smetana, K., Jr.; Gabius, H.-J. A guide into
glycosciences: How chemistry, biochemistry and biology cooperate to
crack the sugar code. Biochim. Biophys. Acta, Gen. Subj. 2015, 1850,
186.
(8) Agre, P.; Bertozzi, C.; Bissell, M.; Campbell, K. P.; Cummings, R.
D.; Desai, U. R.; Estes, M.; Flotte, T.; Fogleman, G.; Gage, F.;
Ginsburg, D.; Gordon, J. I.; Hart, G.; Hascall, V.; Kiessling, L.;
Kornfeld, S.; Lowe, J.; Magnani, J.; Mahal, L. K.; Medzhitov, R.;
Roberts, R. J.; Sackstein, R.; Sarkar, R.; Schnaar, R.; Schwartz, N.;
Varki, A.; Walt, D.; Weissman, I. Training the next generation of
biomedical investigators in glycosciences. J. Clin. Invest. 2016, 126,
405.
Oscar Millet − CIC bioGUNE, Basque Research Technology
Alliance, BRTA, 48160 Derio, Spain; orcid.org/0000-
0001-8748-4105
Hans-Joachim Gabius − Institute of Physiological Chemistry,
Faculty of Veterinary Medicine, Ludwig-Maximilians-
University Munich, 80539 Munich, Germany; orcid.org/
0000-0003-3467-3900
(9) Lemieux, R. U. Toward Enhanced symbiosis of Chemistry and
Biology. FASEB J. 1994, 8, 273.
Complete contact information is available at:
https://pubs.acs.org/10.1021/acs.joc.0c01830
(10) Bessell, E. M.; Foster, A. B.; Westwood, J. H. The use of
deoxyfluoro-D-glucopyranoses and related compounds in a study of
yeast hexokinase specificity. Biochem. J. 1972, 128, 199.
(11) Glaudemans, C. P. J. Mapping of subsites of monoclonal anti-
carbohydrate antibodies using deoxy and deoxyfluorosugars. Chem.
Rev. 1991, 91, 25.
Funding
́
Agencia Estatal de Investigacion (Spain) Grants CTQ2015-
64597-C2-1-P and 2-P, RTI2018-094751-B-C21 and C22,
Severo Ochoa Excellence Accreditation (SEV-2016-0644)
European Research Council (RECGLYCANMR, Advanced
Grant No. 788143), and CIBERES, an initiative from the
Spanish Institute of Health Carlos III. Science Foundation
Ireland, SFI Award 13/IA/1959.
́
́
(12) Fernandez, P.; Jimenez-Barbero, J.; Martín-Lomas, M. Synthesis
of all the possible monomethyl ethers and several deoxyhaloanalogs of
methyl n-lactoside as ligands for the Ricinus communis lectins.
Carbohydr. Res. 1994, 254, 61.
(13) Hevey, R. Bioisosteres of carbohydrate functional groups in
glycomimetic design. Biomimetics 2019, 4, 53.
Notes
(14) Barnett, J. E. G.; Holman, G. D.; Munday, K. A. Structural
requirements for binding to sugar-transport system of human
erythrocyte. Biochem. J. 1973, 131, 211.
(15) Vanwauwe, J. P.; Loontiens, F. G.; Debruyne, C. K.
Carbohydrate binding specificity of lectin from pea (Pisum sativum).
Biochim. Biophys. Acta, Protein Struct. 1975, 379, 456.
(16) Bhattacharyya, L.; Brewer, C. F. Lectin-carbohydrate
interactions. Studies of the nature of hydrogen-bonding between D-
galactose and certain D-galactose-specific lectins, and between D-
mannose and concanavalin-A. Eur. J. Biochem. 1988, 176, 207.
(17) Withers, S. G.; Street, I. P.; Percival, M. D. Fluorinated
carbohydrates as probes of enzyme specificity and mechanism. ACS
Symp. Ser. 1988, 374, 59.
The authors declare no competing financial interest.
ACKNOWLEDGMENTS
■
́
We thank Agencia Estatal de Investigacion (Spain) for grants
CTQ2015-64597-C2-1-P and 2-P, RTI2018-094751-B-C21
and C22, and for FPI fellowship to J.D.M., and for the Severo
Ochoa Excellence Accreditation (SEV-2016-0644). J.J.-B. also
thanks the European Research Council (RECGLYCANMR,
Advanced Grant No. 788143) and CIBERES, an initiative from
the Spanish Institute of Health Carlos III. S. O. thanks the SFI
Award 13/IA/1959. We acknowledge the access to NMR
H
https://dx.doi.org/10.1021/acs.joc.0c01830
J. Org. Chem. XXXX, XXX, XXX−XXX

The Journal of Organic Chemistry
pubs.acs.org/joc
Article
(18) Nycholat, C. M.; Bundle, D. R. Synthesis of monodeoxy and
mono-O-methyl congeners of methyl beta-D-mannopyranosyl-(1 ->
2)-beta-D-mannopyranoside for epitope mapping of anti-Candida
albicans antibodies. Carbohydr. Res. 2009, 344, 555.
(19) McCarter, J. D.; Adam, M. J.; Withers, S. G. Binding energy and
catalysis. Fluorinated and deoxygenated glycosides as mechanistics
probes of Escherichia-coli (LacZ) β-galactosidase. Biochem. J. 1992,
286, 721.
(38) Card, P. J. Synthesis of fluorinated carbohydrates. J. Carbohydr.
Chem. 1985, 4, 451.
(39) Dax, K.; Albert, M.; Ortner, J.; Paul, B. J. Synthesis of
deoxyfluoro sugars from carbohydrate precursors. Carbohydr. Res.
2000, 327, 47.
(40) Denavit, V.; Laine, D.; St-Gelais, J.; Johnson, P. A.; Giguere, D.
A Chiron approach towards the stereoselective synthesis of
polyfluorinated carbohydrates. Nat. Commun. 2018, 9, 11.
(41) Wang, J.; Sanchez-Rosello, M.; Acena, J. L.; del Pozo, C.;
Sorochinsky, A. E.; Fustero, S.; Soloshonok, V. A.; Liu, H. Fluorine in
pharmaceutical industry: fluorine-containing drugs introduced to the
market in the last decade (2001−2011). Chem. Rev. 2014, 114, 2432.
(42) Dalvit, C. Ligand- and substrate-based 19F NMR screening:
Principles and applications to drug discovery. Prog. Nucl. Magn. Reson.
Spectrosc. 2007, 51, 243.
(20) Lemieux, R. U. The origin of the specificity in the recognition
of oligosaccharides by proteins. Chem. Soc. Rev. 1989, 18, 347.
́
(21) Linclau, B.; Arda, A.; Reichardt, N.-C.; Sollogoub, M.; Unione,
́
L.; Vincent, S. P.; Jimenez-Barbero, J. Fluorinated carbohydrates as
chemical probes for molecular recognition studies. Current status and
perspectives. Chem. Soc. Rev. 2020, 49, 3863.
́
́
(22) Begue, J.-P.; Bonnet-Delpon, D. l. Bioorganic and medicinal
chemistry of fluorine; John Wiley & Sons: Hoboken, NJ, 2008.
(23) Gillis, E. P.; Eastman, K. J.; Hill, M. D.; Donnelly, D. J.;
Meanwell, N. A. Applications of Fluorine in Medicinal Chemistry. J.
Med. Chem. 2015, 58, 8315.
(43) Dalvit, C.; Vulpetti, A. Ligand-Based Fluorine NMR Screening:
Principles and Applications in Drug Discovery Projects. J. Med. Chem.
2019, 62, 2218.
́ ́
(44) Diercks, T.; Ribeiro, J. P.; Canada, F. J.; Andre, S.; Jimenez-
̃
(24) Dalvit, C.; Invernizzi, C.; Vulpetti, A. Fluorine as a hydrogen-
bond acceptor: experimental evidence and computational calculations.
Chem. Eur. J. 2014, 20, 11058.
(25) Dunitz, J. D. Organic fluorine: odd man out. ChemBioChem
2004, 5, 614.
(26) Biffinger, J. C.; Kim, H. W.; DiMagno, S. G. The polar
hydrophobicity of fluorinated compounds. ChemBioChem 2004, 5,
622.
Barbero, J.; Gabius, H.-J. Fluorinated Carbohydrates as Lectin
Ligands: Versatile Sensors in 19F-Detected Saturation Transfer
Difference NMR Spectroscopy. Chem. Eur. J. 2009, 15, 5666.
(45) Matei, E.; Andre, S.; Glinschert, A.; Infantino, A. S.; Oscarson,
S.; Gabius, H.-J.; Gronenborn, A. M. Fluorinated Carbohydrates as
Lectin Ligands: Dissecting Glycan-Cyanovirin Interactions by Using
F-19 NMR Spectroscopy. Chem. Eur. J. 2013, 19, 5364.
(46) Shishmarev, D.; Fontenelle, C. Q.; Kuprov, I.; Linclau, B.;
Kuchel, P. W. Transmembrane Exchange of Fluorosugars: Character-
ization of Red Cell GLUT1 Kinetics Using(19)F NMR. Biophys. J.
2018, 115, 1906.
(27) Unione, L.; Xu, B.; Díaz, D.; Martín-Santamaría, S.; Poveda, A.;
Sardinha, J.; Rauter, A. P.; Bleriot, Y.; Zhang, Y.; Canada, F. J.;
̃
́
Sollogoub, M.; Jimenez-Barbero, J. Conformational Plasticity in
Glycomimetics: Fluorocarbamethyl-L-idopyranosides Mimic the
Intrinsic Dynamic Behaviour of Natural Idose Rings. Chem. - Eur. J.
2015, 21, 10513.
(47) Unione, L.; Alcala, M.; Echeverría, B.; Serna, S.; Arda, A.;
́
Franconetti, A.; Canada, F. J.; Diercks, T.; Reichardt, N.; Jimenez-
̃
Barbero, J. Fluoroacetamide Moieties as NMR Spectroscopy Probes
for the Molecular Recognition of GlcNAc-Containing Sugars:
Modulation of the CH-p Stacking Interactions by Different
Fluorination Patterns. Chem. Eur. J. 2017, 23, 3957.
́
̂
́
́
̀
(28) St-Gelais, J.; Cote, E.; Laine, D.; Johnson, P. A.; Giguere, D.
Addressing the Structural Complexity of Fluorinated Glucose
Analogues: Insight into Lipophilicities and Solvation Effects. Chem.
- Eur. J. 2020, DOI: 10.1002/chem.202002825.
(48) Katiyar, S.; Singh, A.; Surolia, A. The interaction of N-
trifluoroacetylgalactosamine and its derivatives with winged bean
(Psophocarpus tetragonolobus) basic agglutinin reveals differential
mechanism of their recognition: a fluorine-19 nuclear magnetic
resonance study. Glycoconjugate J. 2014, 31, 537.
(49) Krishna Sastry, M. V.; Swamy, M. J.; Surolia, A. Analysis of
dynamics and mechanism of ligand binding to Artocarpus integrifolia
agglutinin. A 13C and 19F NMR study. J. Biol. Chem. 1988, 263, 14826.
(50) Martínez, J. D.; Valverde, P.; Delgado, S.; Romano, C.; Linclau,
(29) Linclau, B.; Wang, Z.; Compain, G.; Paumelle, V.; Fontenelle,
C. Q.; Wells, N.; Weymouth-Wilson, A. Investigating the Influence of
(Deoxy) fluorination on the Lipophilicity of Non-UV-Active
Fluorinated Alkanols and Carbohydrates by a New log P
Determination Method. Angew. Chem., Int. Ed. 2016, 55, 674.
(30) Williams, S. J.; Withers, S. G. Glycosyl fluorides in enzymatic
reactions. Carbohydr. Res. 2000, 327, 27.
(31) Withers, S. G.; Street, I. P.; Bird, P.; Dolphin, D. H. 2-Deoxy-2-
Fluoroglycosides. A Novel Class of Mechanism-Based Glucosidase
Inhibitors. J. Am. Chem. Soc. 1987, 109, 7530.
(32) Tysoe, C.; Withers, S. G. Fluorinated Mechanism-Based
Inhibitors: Common Themes and Recent Developments. Curr. Top.
Med. Chem. 2014, 14, 865.
(33) Denavit, V.; Laine, D.; Bouzriba, C.; Shanina, E.; Gillon, E.;
Fortin, S.; Rademacher, C.; Imberty, A.; Giguere, D. Stereoselective
Synthesis of Fluorinated Galactopyranosides as Potential Molecular
Probes for Galactophilic Proteins: Assessment of Monofluorogalacto-
side-LecA Interactions. Chem. - Eur. J. 2019, 25, 4478.
́
B.; Reichardt, N. C.; Oscarson, S.; Arda, A.; Jimenez-Barbero, J.;
Canada, F. J. Unraveling Sugar Binding Modes to DC-SIGN by
̃
Employing Fluorinated Carbohydrates. Molecules 2019, 24, 2337.
(51) Wamhoff, E. C.; Hanske, J.; Schnirch, L.; Aretz, J.; Grube, M.;
Silva, D. V.; Rademacher, C. F-19 NMR-Guided Design of
Glycomimetic Langerin Ligands. ACS Chem. Biol. 2016, 11, 2407.
(52) van Vliet, S. J.; Saeland, E.; van Kooyk, Y. Sweet preferences of
MGL: carbohydrate specificity and function. Trends Immunol. 2008,
29, 83.
(53) Marcelo, F.; García-Martín, F.; Matsushita, T.; Sardinha, J.;
Coelho, H.; Oude-Vrielink, A.; Koller, C.; Andre, S.; Cabrita, E. J.;
́
(34) Diercks, T.; Infantino, A. S.; Unione, L.; Jimenez-Barbero, J.;
Oscarson, S.; Gabius, H.-J. Fluorinated Carbohydrates as Lectin
Ligands: Synthesis of OH/F-Substituted N-Glycan Core Trimanno-
side and Epitope Mapping by 2D STD-TOCSYreF NMR spectros-
copy. Chem. - Eur. J. 2018, 24, 15761.
́
Gabius, H.-J.; Nishimura, S.-I.; Jimenez-Barbero, J.; Canada, F. J.
̃
Delineating Binding Modes of Gal/GalNAc and Structural Elements
of the Molecular Recognition of Tumor-Associated Mucin Glyco-
peptides by the Human Macrophage Galactose-Type Lectin. Chem. -
Eur. J. 2014, 20, 16147.
́
(35) N’Go, I.; Golten, S.; Arda, A.; Canada, F. J.; Jimenez-Barbero,
̃
J.; Linclau, B.; Vincent, S. P. Tetrafluorination of Sugars as Strategy
for Enhancing Protein-Carbohydrate Affinity: Application to UDP-
Galp Mutase Inhibition. Chem. Eur. J. 2014, 20, 106.
(54) Pirro, M.; Rombouts, Y.; Stella, A.; Neyrolles, O.; Burlet-
Schiltz, O.; van Vliet, S. J.; de Ru, A. H.; Mohammed, Y.; Wuhrer, M.;
van Veelen, P. A.; Hensbergen, P. J. Characterization of Macrophage
Galactose-type Lectin (MGL) ligands in colorectal cancer cell lines.
Biochim. Biophys. Acta, Gen. Subj. 2020, 1864, 129513.
(36) Cheng, Y.; Guo, A.-L.; Guo, D.-S. Recent Progress in Synthesis
and Applications of Fluorinated Carbohydrates. Curr. Org. Chem.
2010, 14, 977.
(55) Schwarz, F. P.; Puri, K. D.; Bhat, R. G.; Surolia, A.
Thermodynamics of monosaccharide binding to concanavalin A,
(37) Uhrig, M. L.; Lantano, B.; Postigo, A. Synthetic strategies for
fluorination of carbohydrates. Org. Biomol. Chem. 2019, 17, 5173.
I
https://dx.doi.org/10.1021/acs.joc.0c01830
J. Org. Chem. XXXX, XXX, XXX−XXX

The Journal of Organic Chemistry
pubs.acs.org/joc
Article
pea (Pisum-sativum) lectin, and lentil (Lens-culinaris) lectin. J. Biol.
Chem. 1993, 268, 7668.
̃
(56) Ortega, G.; Castano, D.; Diercks, T.; Millet, O. Carbohydrate
glycoclusters by using tissue sections as an assay platform. RSC Adv.
2018, 8, 28716.
(73) Khedri, Z.; Muthana, M. M.; Li, Y.; Muthana, S. M.; Yu, H.;
Cao, H.; Chen, X. Probe sialidase substrate specificity using
chemoenzymatically synthesized sialosides containing C9-modified
sialic acid. Chem. Commun. 2012, 48, 3357.
(74) Esmurziev, A. M.; Simic, N.; Hoff, B. H.; Sundby, E. Synthesis
and Structure Elucidation of Benzoylated Deoxyfluoropyranosides. J.
Carbohydr. Chem. 2010, 29, 348.
(75) Carr, H. Y.; Purcell, E. M. Effects of Diffusion on Free
Precession in Nuclear Magnetic Resonance Experiments. Phys. Rev.
1954, 94, 630.
Affinity for the Glucose-Galactose Binding Protein Is Regulated by
Allosteric Domain Motions. J. Am. Chem. Soc. 2012, 134, 19869.
(57) Borrok, M. J.; Kiessling, L. L.; Forest, K. T. Conformational
changes of glucose/galactose-binding protein illuminated by open,
unliganded, and ultra-high-resolution ligand-bound structures. Protein
Sci. 2007, 16, 1032.
(58) Johan; Mowbray, S. L. Sugar Recognition by a Glucose/
Galactose Receptor. Evaluation of Binding Energetics from Molecular
Dynamics Simulations. J. Biol. Chem. 1995, 270, 9978.
(76) Meiboom, S.; Gill, D. Modified Spin-Echo Method for
Measuring Nuclear Relaxation Times. Rev. Sci. Instrum. 1958, 29, 688.
́
(59) Jegouzo, S. A. F.; Quintero-Martínez, A.; Ouyang, X.; dos
́
Santos, A.; Taylor, M. E.; Drickamer, K. Organization of the
extracellular portion of the macrophage galactose receptor: a trimeric
cluster of simple binding sites for N-acetylgalactosamine. Glycobiology
2013, 23, 853.
(60) Kerins, L.; Byrne, S.; Gabba, A.; Murphy, P. V. Anomer
Preferences for Glucuronic and Galacturonic Acid and Derivatives
and Influence of Electron-Withdrawing Substituents. J. Org. Chem.
2018, 83, 7714.
(61) Hajduk, P. J.; Olejniczak, E. T.; Fesik, S. W. One-dimensional
relaxation- and diffusion-edited NMR methods for screening
compounds that bind to macromolecules. J. Am. Chem. Soc. 1997,
119, 12257.
(62) Dalvit, C.; Piotto, M. F-19 NMR transverse and longitudinal
relaxation filter experiments for screening: a theoretical and
experimental analysis. Magn. Reson. Chem. 2017, 55, 106.
(63) Srinivas, V. R.; Reddy, G. B.; Ahmad, N.; Swaminathan, C. P.;
Mitra, N.; Surolia, A. Legume lectin family, the ‘natural mutants of the
quaternary state’, provide insights into the relationship between
protein stability and oligomerization. Biochim. Biophys. Acta, Gen. Subj.
2001, 1527, 102.
(64) Schwarz, F. P.; Misquith, S.; Surolia, A. Effect of substituent on
the thermodynamics of D-glucopyranoside binding to concanavalin A,
pea (Pisum sativum) lectin and lentil (Lens culinaris) lectin. Biochem. J.
1996, 316 (Pt 1), 123.
(65) Miller, D. M.; Olson, J. S.; Quiocho, F. A. The Mechanism of
Sugar binding to the Periplasmic Receptor for Galactose Chemotaxis
and Transport in Escherichia coli. J. Biol. Chem. 1980, 255, 2465.
(66) Vyas, M. N.; Vyas, N. K.; Quiocho, F. A. Crystallographic
analysis of the epimeric and anomeric specificity of the periplasmic
transport/chemosensory protein receptor for D-glucose and D-
galactose. Biochemistry 1994, 33, 4762.
́
(67) Dalvit, C.; Parent, A.; Vallee, F.; Mathieu, M.; Rak, A. Fast
NMR Methods for Measuring in the Direct and/or Competition
Mode the Dissociation Constants of Chemical Fragments Interacting
with a Receptor. ChemMedChem 2019, 14, 1115.
(68) O’Sullivan, J. A.; Chang, A. T.; Youngblood, B. A.; Bochner, B.
S. Eosinophil and mast cell Siglecs: From biology to drug target. J.
Leukocyte Biol. 2020, 108, 73.
(69) Duan, S.; Paulson, J. C. Siglecs as Immune Cell Checkpoints in
Disease. Annu. Rev. Immunol. 2020, 38, 365.
(70) Garcia Caballero, G.; Kaltner, H.; Kutzner, T. J.; Ludwig, A.-K.;
Manning, J. C.; Schmidt, S.; Sinowatz, F.; Gabius, H.-J. How galectins
have become multifunctional proteins. Histol. Histopathol. 2020, 35,
509.
(71) Ludwig, A.-K.; Michalak, M.; Xiao, Q.; Gilles, U.; Medrano, F.
J.; Ma, H.; FitzGerald, F. G.; Hasley, W. D.; Melendez-Davila, A.; Liu,
M.; Rahimi, K.; Kostina, N. Y.; Rodriguez-Emmenegger, C.; Moeller,
M.; Lindner, I.; Kaltner, H.; Cudic, M.; Reusch, D.; Kopitz, J.;
Romero, A.; Oscarson, S.; Klein, M. L.; Gabius, H.-J.; Percec, V.
Design-functionality relationships for adhesion/growth-regulatory
galectins. Proc. Natl. Acad. Sci. U. S. A. 2019, 116, 2837.
(72) Kaltner, H.; Manning, J. C.; Caballero, G. G.; Di Salvo, C.;
Gabba, A.; Romero-Hernandez, L. L.; Knospe, C.; Wu, D.; Daly, H.
C.; O’Shea, D. F.; Gabius, H.-J.; Murphy, P. V. Revealing biomedically
relevant cell and lectin type-dependent structure-activity profiles for
J
https://dx.doi.org/10.1021/acs.joc.0c01830
J. Org. Chem. XXXX, XXX, XXX−XXX