Development and optimization of a competitive binding assay for the galactophilic low affinity lectin LecA from: Pseudomonas aeruginosa

Article abstract of DOI:10.1039/c6ob01313a

Infections with the Gram-negative bacterium Pseudomonas aeruginosa result in a high mortality among immunocompromised patients and those with cystic fibrosis. The pathogen can switch from planktonic life to biofilms, and thereby shields itself against antibiotic treatment and host immune defense to establish chronic infections. The bacterial protein LecA, a C-type lectin, is a virulence factor and an integral component for biofilm formation. Inhibition of LecA with its carbohydrate ligands results in reduced biofilm mass, a potential Achilles heel for treatment. Here, we report the development and optimization of a fluorescence polarization-based competitive binding assay with LecA for application in screening of potential inhibitors. As a consequence of the low affinity of d-galactose for LecA, the fluorescent ligand was optimized to reduce protein consumption in the assay. The assay was validated using a set of known inhibitors of LecA and IC₅₀ values in good agreement with the known K_d values were obtained. Finally, we employed the optimized assay to screen sets of synthetic thio-galactosides and natural blood group antigens and report their structure-activity relationship. In addition, we evaluated a multivalent fluorescent assay probe for LecA and report its applicability in an inhibition assay.

Full text of DOI:10.1039/c6ob01313a

Organic &
Biomolecular Chemistry
View Article Online
View Journal
PAPER
Development and optimization of a competitive
binding assay for the galactophilic low affinity
lectin LecA from Pseudomonas aeruginosa†
Cite this: DOI: 10.1039/c6ob01313a
Ines Joachim,a,b,c Sebastian Rikker,c Dirk Hauck,a,b Daniela Ponader,d
Sophia Boden,e Roman Sommer,a,b Laura Hartmanne and Alexander Titz*a,b,c
Infections with the Gram-negative bacterium Pseudomonas aeruginosa result in a high mortality among
immunocompromised patients and those with cystic fibrosis. The pathogen can switch from planktonic
life to biofilms, and thereby shields itself against antibiotic treatment and host immune defense to estab-
lish chronic infections. The bacterial protein LecA, a C-type lectin, is a virulence factor and an integral
component for biofilm formation. Inhibition of LecA with its carbohydrate ligands results in reduced
biofilm mass, a potential Achilles heel for treatment. Here, we report the development and optimization
of a fluorescence polarization-based competitive binding assay with LecA for application in screening of
potential inhibitors. As a consequence of the low affinity of D-galactose for LecA, the fluorescent ligand
was optimized to reduce protein consumption in the assay. The assay was validated using a set of known
inhibitors of LecA and IC50 values in good agreement with the known Kd values were obtained. Finally, we
employed the optimized assay to screen sets of synthetic thio-galactosides and natural blood group anti-
gens and report their structure–activity relationship. In addition, we evaluated a multivalent fluorescent
assay probe for LecA and report its applicability in an inhibition assay.
Received 17th June 2016,
Accepted 23rd July 2016
DOI: 10.1039/c6ob01313a
www.rsc.org/obc
suffering from cystic fibrosis. With 8% of all hospital-acquired
infections, it is one of the most common bacteria in the
Introduction
Lectins are carbohydrate-binding proteins and often play pro- health care setting.6–8 The bacterium is highly resistant
minent roles in bacterial, viral or parasitic infections.1–5 towards antibiotic treatment which is partially caused by its
Microbial lectins enable infection of the host by the pathogen, ability to form biofilms.9 Bacteria embedded in such biofilms
whereas numerous human lectins of the innate immune are surrounded by a self-produced matrix that shields the
system are defense molecules against pathogens. The Gram- pathogen from host immune defense and also from antibiotic
negative opportunistic pathogen Pseudomonas aeruginosa is a treatment, rendering bacteria 10- to 1000-fold more resistant
major threat for patients with an impaired immune system or towards antibiotics compared to the free floating planktonic
state.10–12 New therapeutic approaches therefore aim to dis-
mantle the biofilm and thereby restore susceptibility towards
antibiotics.13,14
aChemical Biology of Carbohydrates, Helmholtz Institute for Pharmaceutical
LecA and LecB (formerly called PA-IL and PA-IIL) are two
soluble lectins produced by P. aeruginosa which are important
for biofilm formation.15–17 The sequence of LecA is highly con-
served among strains, whereas LecB varies in clinical isolates
and two major types prevail, i.e., the PAO1-type and the PA14-
type.18 In environmental isolates, a third variant is widely
observed (PA7-type).19 In addition, both lectins are virulence
factors that elicit toxic effects to cells and animals and inter-
fere with human airway ciliary beating.20–24 Recently, LecA was
shown to induce glycolipid-dependent membrane invagina-
tions in artificial vesicles, which was proposed as an uptake
mechanism for P. aeruginosa into human cells.25 LecA and
Research Saarland (HIPS), D-66123 Saarbrücken, Germany.
E-mail: alexander.titz@helmholtz-hzi.de
bDeutsches Zentrum für Infektionsforschung (DZIF), Standort Hannover,
Braunschweig, Germany
cDepartment of Chemistry and Graduate School Chemical Biology, University of
Konstanz, D-78457 Konstanz, Germany
dMax Planck Institute of Colloids and Interfaces, Department of Biomolecular
Systems, Research Campus Golm, 14424 Potsdam, Germany
eHeinrich-Heine-University Duesseldorf, Institute of Organic Chemistry and
Macromolecular Chemistry, D-40225 Düsseldorf, Germany
†Electronic supplementary information (ESI) available: 1H and 13C-NMR spectra
of all synthesized compounds; for glycopeptides the 1H-NMR spectrum,
MALDI-MS spectra and HPLC traces. HPLC chromatograms for all fluorescent
ligands. See DOI: 10.1039/c6ob01313a
This journal is © The Royal Society of Chemistry 2016
Org. Biomol. Chem.

View Article Online
Paper
Organic & Biomolecular Chemistry
LecB have a binding specificity for D-galactose- and L-fucose- or pounds presenting galactose on various multivalent scaffolds
D-mannose-containing carbohydrates or their glycoconjugates, have been reported (for a comprehensive overview, the reader
respectively.16 The crystal structures of both lectins show is referred to two recent reviews32,34). Very high avidities due to
homotetrameric assemblies and calcium ions mediating the their multivalency have been achieved. An optimal positioning
recognition of their carbohydrate ligands.18,26–29
of two galactose residues to simultaneously bind to the
In 2001 and 2008, patients with pulmonary P. aeruginosa two adjacent carbohydrate binding sites in the LecA tetramer
infections were treated with an aerosol containing D-galactose was achieved by Pieters and co-workers.48,49 The proposed
and L-fucose and an efficient reduction of bacterial counts in binding mode of such highly potent divalent ligands
patient sputum was reported.30,31 Consequently, the develop- (Kd = 28 nM) was recently experimentally confirmed by X-ray
ment of inhibitors of both lectins with the aim to provide crystallography.50
efficient anti-infectives against this pathogen is a highly active
The potency of published LecA inhibitors is usually quanti-
field which is summarized in recent reviews.13,32–34 Carbo- fied in hemagglutination inhibition assays (HIA), in enzyme-
hydrate-based ligands of LecA and LecB were shown to inhibit linked lectin assays (ELLA), by surface plasmon resonance
biofilm formation or dissolve established biofilms.17,35–38 We (SPR) or by isothermal microcalorimetry (ITC) and IC50s or Kds
have previously reported small drug-like glycomimetic inhibi- are obtained, respectively (cf. all references of the previous
tors of the fucophilic LecB with favorable binding kinetics and paragraph). SPR and ITC are low throughput methods for
thermodynamics as well as activity in a bacterial adhesion obtaining additional data on the kinetics and thermodynamics
model.18,39–43
of binding. Although HIA and ELLA are performed in micro-
The only moderate binding affinity of unmodified D-galac- titre plates and can be rapidly performed in parallel, both
tose (1, Kd = 87.5 µM, Fig. 1) for LecA can be enhanced by the assays have certain drawbacks. In HI assays, IC50 values are
introduction of hydrophobic moieties at the anomeric usually much higher than the thermodynamic Kd values and
center.47 Introducing a methyl-group as aglycon results in ELLAs require numerous incubation and washing steps which
modestly enhanced binding affinities of the derivatives 2 and results in poor reproducibility.
3.44,47 Adding an aromatic ring in β-position to the anomeric
Fluorescence polarization has emerged as a technique
center (4, Kd = 8.8 µM) results in a tenfold increased affinity which allows miniaturization for high throughput and in situ
compared to the unmodified monosaccharide 1.45,47 Recently, detection of the potency of a given competitive inhibitor.
Vidal and co-workers published the monovalent triazol deriva- Despite being established in many fields, its application in the
tive 5 as a potent monovalent inhibitor of LecA with a binding evaluation of carbohydrate–lectin interactions has lagged
affinity of 5.8 µM.46 In addition, Roy and co-workers reported behind. This is probably due to the generally low affinity
naphtyl and umbilliferyl thiogalactosides44 and Reymond and (µM–mM) of such interactions, that require high protein con-
co-workers described tripeptidylphenyl galactosides36 with centrations close to the Kd of the fluorescent tracer. Initially,
affinities in the same range. In general, the increased affinity the method was developed for plant lectins by Khan et al. in
of aromatic β-D-galactosides to LecA was assigned to a CH–π 1981 51 requiring approx. 200 µM protein. In 1995, a fluo-
interaction of a histidine residue in LecA (His50) with the rescence polarization-based assay by Jacob et al.52 required
aromatic glycoside ligand.36,44,45 In addition, numerous com- 166 µM of the human E-selectin and two decades later it was
Fig. 1 Structures of known LecA ligands 1–5 based on D-galactose, dissociation constants are published in the literature.44–46 As fluorescent
tracers for the competitive binding assay, we designed structures based on aliphatic galactosides 6 and 7, and phenyl glycoside 8 derived from
potent ligand 5, as well as its meta-analog 9.
Org. Biomol. Chem.
This journal is © The Royal Society of Chemistry 2016

View Article Online
Organic & Biomolecular Chemistry
Paper
established for galectins by Leffler et al. in 2003 (10 µM protein
For the synthesis of ethyl-linked ligand α-6, bromoethanol
concentration).53,54 The method was then successfully develo- was first glycosylated with D-galactose (1) in a Fischer-type reac-
ped for high affinity interactions of the E. coli lectin FimH at tion (Scheme 1). The resulting inseparable mixture of anomers
200 nM concentration by Hultgren and co-workers55 and by our was treated with β-galactosidase to remove the β-anomer and
group for the P. aeruginosa lectin LecB18,39 using 150 nM LecB pure α-10 was obtained. Nucleophilic displacement of the
in the assay. This technology is now used as a routine screening bromide with sodium azide (10 → 11), hydrogenolysis of the
method for these two high-affinity lectins only.40–43,56
azide (11 → 12) and conjugation with fluoresceine isothio-
Here, we report the development and optimization of a cyanate (FITC) gave α-6 in good overall yield in this protecting
fluorescence polarization-based assay for the bacterial lectin group-free reaction sequence (Scheme 1). The β-anomer 7 was
LecA and its application in the evaluation of potential synthesized starting from protected bromoethyl β-D-galactoside
inhibitors.
13,57 treatment with sodium azide and subsequent Zemplén
deprotection to the unprotected azide (13 → 14, 2 steps). After
hydrogenolysis of the azide 14, the β-ethyl linked reporter
ligand 7 was obtained by reaction of amine 15 with fluores-
ceine isothiocyanate.
Results and discussion
Synthesis of fluorescent reporter ligands
The fluorescent reporter ligands with phenyl spacers 8
and 9 were synthesized starting from pentaacetyl galactose 16
(Scheme 2). For the para-substituted ligand 8, para-nitro-
phenol was galactosylated to give the glycoside 17 in 52%
yield. Then, a three-step one-pot reaction sequence was per-
formed with acetylated 17 according to Cecioni et al.:46 (1)
reduction of the nitro group to an aniline, (2) acetylation of
the aniline with bromoacetyl bromide and (3) nucleophilic
substitution of the bromide with sodium azide. The resulting
acetyl protected azide 18 was obtained in 72% yield over three
steps after one purification. Deacetylation (18 → 19), hydro-
genolysis of the azide (19 → 20), and coupling of 20 with
fluoresceine isothiocyanate yielded para-8 in 73% yield over
3 steps.
Following the same protocol, meta-9 was obtained after
glycosylation of meta-nitrophenol with 16 and subsequent
manipulation of the arene moiety (Scheme 2). In an analogous
three-step one-pot sequence, azide 22 was obtained from nitro-
phenyl glycoside 21 in 87% yield over three steps. The azide
was deacetylated (22 → 23), reduced to the amine 24 and
To establish a fluorescence polarization assay with LecA, suit-
able fluorescently labelled ligands were designed and syn-
thesized. D-Galactose-based probes specifically binding to the
carbohydrate-binding site of LecA were conjugated to fluo-
rescein as fluorophore. To examine the binding properties of
potential reporter ligands, four different linkers were selected
based on literature data of different galactosides, i.e., 1–5
(Fig. 1). In the α-galactoside 6, the fluorescein moiety is
oriented towards the solvent based on the crystal structure of
LecA with 1,26 whereas in its β-anomer 7, an attractive or repul-
sive interaction of the aglycone with the protein surface was
anticipated. Since phenyl β-D-galactosides are superior inhibi-
tors of LecA, reporter ligand 8 based on the potent LecA inhibi-
tor 5 was designed and its meta-analog 9 was included in the
study. The latter isomer was deduced from the crystal structure
of LecA in complex with para-nitrophenyl galactoside (pdb
code 3ZYF36), indicating a solvent exposure for para-substitu-
ents and the possibility of contacts with the protein of substi-
tuents in meta-position.
Scheme 1 Synthesis of α-ethyl linked reporter ligand 6 and β-ethyl linked ligand 7. Reagents and conditions: (a) La(OTf)3, 2-bromoethanol, 70 °C,
24 h, 35%; (b) Aspergillus oryzae β-galactosidase, aq. phosphate buffer, 37 °C, 22 h, 33%; (c) NaN3, DMF, 75 °C, 5–16 h; (d) H2, Pd/C, MeOH, r.t., o.n.;
(e) fluoresceine isothiocyanate, NaHCO3, DMF, r.t., o.n.; (f) NaOMe, MeOH, r.t., 3 h.
This journal is © The Royal Society of Chemistry 2016
Org. Biomol. Chem.

View Article Online
Paper
Organic & Biomolecular Chemistry
Scheme 2 Synthesis of phenyl-linked reporter ligand para-8 and meta-9. Reagents and conditions: (a) p-nitrophenol, BF3·OEt2, CH2Cl2, 0 °C – r.t.,
24 h; (b) H2, Pd/C, CH2Cl2, 12 h; (c) BrCH2COBr, NEt3, CH2Cl2, 0 °C – r.t. 2 h; (d) NaN3, nBu4NI, DMF, 50 °C, 12 h; (e) NaOMe, MeOH, r.t., 2 h; (f) H2,
Pd/C, MeOH, r.t., 2 h; (g) fluoresceine isothiocyanate, NaHCO3, DMF, r.t., 14 h; (h) m-nitrophenol, BF3·OEt2, CH2Cl2, 0 °C – r.t., 24 h.
coupled with fluoresceine isothiocyanate to give meta-9 in 47% tively. This data was consistent with the literature describing
over three steps.
increased affinities of LecA for galactoside ligands with β-linked
aromatic aglycones (see Fig. 1). Interestingly, the absolute
signal intensities vary significantly between the different
ligands from approx. 85 mP for 8 to 200 mP for 6. It should be
noted that the intensity of fluorescence polarization in the
bound state, i.e. at high protein concentrations depends on the
residual mobility of the fluorophore in the bound state (propel-
ler effect). The higher polarization values of 6 indicated a less
flexible fluorophore as observed for compound 8. However, the
amount of protein required for application in a competitive
binding assay largely depends on the affinity of the fluorescent
compound to the protein. Therefore, compounds 8 and 9 were
superior to the alkyl linked tracers 6 and 7.
Binding of fluorescent ligands to LecA observed by
fluorescence polarization
To evaluate the binding of the synthesized galactose-derived
ligands 6–9 to LecA, a direct titration with increasing amounts
of LecA was performed (Fig. 2). All four ligands showed a dose-
dependent increase in fluorescence polarization with increasing
concentrations of LecA. All systems showed sigmoidal curves
with varying maximal fluorescence polarization. The phenyl-
β-galactoside derived ligands had Kd values in the low µM-range
(Kd 8.1 µM for 8, Kd 7.4 µM for 9). The two alkyl-linked ligands
6 and 7 possessed Kd values of 19.2 µM and 27.4 µM, respec-
Fig. 2 Titration of ethyl-linked reporter ligands 6, 7 and phenyl-linked ligands 8, 9 with LecA. Dissociation constants were obtained from a four-
parameter fitting procedure to the dose-dependent increase in fluorescence polarization. Ethyl-linked ligands 6 (Kd 19.2 4.8 µM) and 7 (Kd 27.4
20.3 µM), showed weaker binding affinity than the phenyl-linked ligands 8 (Kd 8.1 2.1 µM) and 9 (Kd 7.4 2.8 µM). One representative titration
experiment of triplicates on one plate is shown. Dissociation constants and standard deviations given were from at least three independent replicates
of triplicates on three plates each.
Org. Biomol. Chem.
This journal is © The Royal Society of Chemistry 2016

View Article Online
Organic & Biomolecular Chemistry
Paper
Validation of the competitive binding assay: determination
of IC50 values of para- and meta-substituted phenyl
β-D-galactosides
Structure–activity relationship of thiogalactoside LecA
inhibitors
After validation of the competitive binding assay using fluo-
In order to determine IC50 values of inhibitors in competition rescent probe 8, a panel of thiogalactosides was screened in a
experiments, the protein concentration should lie within a 384-well microtitre plate format (Table 1). In addition, all thio-
range of 30–90% of the maximal signal intensity in fluo- galactosides were also tested with ethyl-linked 7 as a probe
rescence polarization assays.58,59 Due to the high binding that reached increased fluorescence polarization values when
affinity for the rigid-linked reporter ligands 8 and 9 (Fig. 2), bound to LecA in the direct titration experiment (Fig. 2) to
ligand 8 was chosen as assay probe with the aim to reduce assess its suitability for this assay format. The O-glycosides
the protein consumption needed to reach this signal range. methyl α-D-galactoside (2) and phenyl β-D-galactoside (4) were
In practice, both 8 and 9 are suitable tracers with comparable included as reference compounds. In general, all tested
properties. To validate the competitive binding assay, we galactosides showed inhibition of LecA with IC50s in the range
tested known LecA ligands methyl α-D-galactoside (2),44 of 10 to 75 µM using probe 8. When tested with assay probe 7,
p-nitrophenyl β-D-galactoside (25),47 compound 5 46-derived a similar trend was observed with generally slightly higher IC50
para-substituted phenyl β-D-galactoside 19 and their meta- values (13–146 µM). However, relative potencies of inhibitors
isomers 26 and 23, respectively (Fig. 3). All tested compounds are comparable between both systems. A common trend
inhibited the binding of LecA to the fluorescent tracer 8 in a observed was the increased potencies of phenyl thio-β-D-galac-
dose-dependent manner with IC50s in the one- to two-digit toside derivatives (IC50s 9.3–23.7 µM using fluorescent ligand
µM-range. Methyl galactoside 2 inhibited LecA with an IC50 of 8) over alkyl- or benzyl-derivatives (26.2–74.9 µM). Positioning
38.6 µM which is in good agreement to the corresponding lit- of the aromatic moiety one methylene group further away from
erature known dissociation constant (Kd 50 µM, see Fig. 1). the anomeric center as in phenethyl thio-β-D-galactoside (38),
All phenyl β-D-galactoside derivatives were more potent inhibi- resulted in gain of potency (13.8 µM) compared to the shorter
tors of LecA (IC50s 8.5–19.0 µM), with comparable potencies benzyl derivatives (IC50s 31–64 µM). The latter fact has been
of the meta- and their para-regioisomers: highest inhibition observed for 38 (Kd = 15.6 µM) and was explained by a ben-
was observed for the amido-phenyl derivatives 19 (IC50 eficial coordination of the arene group by molecular model-
9.8 µM) and 23 (IC50 8.5 µM) and nitrophenyl galactosides 25 ing.45 In general, para-tolyl 31 was the most potent LecA
and 26 were two-fold weaker (IC50s 19.0 µM and 13.4 µM, inhibitor in this series with fourfold higher potency in both
respectively).
assay variants (using probe 7 or 8) than reference compound 2
(Fig. 4).
Structure–activity relationship of human blood group antigens
Gilboa-Garber and co-workers tested 90 patient blood samples
for agglutination with LecA and a preference for the blood
group B-type was observed.60 The two LecB variants from
strains PAO1 and PA14 were previously shown to bind strongly
to the Lewisa blood group antigen.18,61 We therefore tested a
set of human blood group antigens from the A-, B-, H- and
Lewis-type blood groups in the established assay (Table 2). In
accordance to the agglutination experiments, a weak competi-
tive inhibition of LecA was observed for blood group B-anti-
gens 43 and 48 with IC50s in the low mM-range and even
weaker for blood group A antigens 42 and 47. All other anti-
gens tested did not show inhibition of LecA binding to the
fluorescent probe 8 at concentrations up to 1 mM of the oligo-
saccharides. Recently, a significant preference for P. aeruginosa
induced sepsis with blood group B infants compared to A-, AB-
and O-type was reported indicating a further potential link to
the role of LecA in the infection process.62
The observed selectivity of LecA for the B-type antigens can
be explained with the presence of one terminal unsubstituted
galactose residue. Despite the fact that in Lewisa and Lewisb
structures also one terminal galactoside is present, no inhi-
bition was observed by these Lewis-type antigens. This could
result from two reasons: (i) that the galactose moiety is
conformationally stacked via its β-face on the α-face of the
Fig. 3 Validation of the competitive binding assay yielded IC50 values
for known LecA inhibitors. Averages and standard deviations were
obtained from three independent experiments.
This journal is © The Royal Society of Chemistry 2016
Org. Biomol. Chem.

View Article Online
Paper
Organic & Biomolecular Chemistry
Table 1 Evaluation of synthetic galactoside inhibitors of LecA using two different fluorescent probes with flexible linker compound 7 or phenyl-
linked 8. Averages and standard deviations were obtained from three independent experiments
Compound
8 IC50 [μM]
Rel. potencya
1
7 IC50 [μM]
Rel. potencya
1
2
4
38.6 16.4
54.4 32.4
13.2 7.7
2.2
24.5 6.4
2.2
27
28
29
30
74.9 46.1
32.4 13.1
26.2 8.2
9.9 4.7
0.5
1.2
1.5
3.8
146.0 69.3
50.1 24.5
38.5 21.9
24.0 14.1
0.4
1.1
1.4
2.3
31
32
9.3 4.2
4.2
3.6
12.5 4.9
4.4
2.1
10.8 2.7
26.5 14.8
33
34
12.1 3.8
23.7 7.2
3.2
1.6
12.5 0.7
25.0 9.9
4.4
2.2
35
36
37
38
63.6 16.4
31.7 22.7
30.9 10.9
13.8 8.2
0.6
1.2
1.2
2.8
49.0 2.8
39.0 0.0
71.0 21.2
28.5 10.6
1.1
1.4
0.8
1.9
a Relative potencies compared to 2.
adjacent fucose residue in these two Lewis-type structures63,64 sides in such LacNAc containing structures as previously
and thus coordination of this moiety to LecA as observed in detected by glycan array experiments65 and inhibition studies
the crystal structures for simple galactosides26,44,45,50,65 is of LecA with structurally related lactose.47 The estimated IC50
probably hindered; or (ii) from a reduced affinity for β-galacto- value for the B-type antigens was approximately thirty-fold
Org. Biomol. Chem.
This journal is © The Royal Society of Chemistry 2016

View Article Online
Organic & Biomolecular Chemistry
Paper
Table 2 Evaluation of various human blood group antigens as ligands
of LecA in the competitive binding assay with phenyl-linked 8 as fluo-
rescent probe. n.i. = no inhibition. Averages and standard deviations
were obtained from triplicates on one plate
Compound
Inhibition
n.i.
39 Lewisa
40 Lewisb
n.i.
Fig. 4 Competitive inhibition of LecA and fluorescent probe 8 with
selected O- and S-galactosides. One representative titration of tripli-
cates on one plate is shown. IC50s given in Table 1 are calculated from
three independent experiments on three plates.
41 H type I
n.i.
higher than the one for methyl galactoside 2, which may also
result from steric hindrance induced by the neighboring glyco-
side residues in these epitopes.
42 A type I
26% at 1 mM
Can multivalent fluorescent probes improve the assay and
reduce protein consumption?
In high affinity lectin–carbohydrate interactions, protein con-
sumption in fluorescence polarization-based competitive
binding assays can be drastically reduced into the nanomolar
range.18,39,43,55 Such high affinity systems are suitable assay
conditions for large screenings where protein consumption is
an important factor. The binding of LecA to phenyl galacto-
sides reaches only low 5–10 µM affinities (Fig. 1) and thus,
10–15 µM protein was needed in the assay described above.
The 12.9 kDa polypeptide chain of LecA forms a homotetra-
meric assembly with four carbohydrate binding sites.26 Poly-
valent display of galactosides was shown to significantly
increase binding affinity of ligands towards this protein.32,34,49
The use of oligo- or multivalent fluorescently-labelled assay
probes could thus further decrease the amount of LecA
needed in the competitive binding assay. For this purpose, we
synthesized the divalent LecA inhibitor 49 in a solid phase
assembly according to a previously published procedure66 and
attached fluoresceine to the free amino group to give the bi-
valent assay probe 50 (Scheme 3). 49 was identified from a
series of sequence-defined glycopolymer segments presenting
galactose in analogy to Ponader et al. for mannose and the
multivalent lectin ConA66 and using a short ethyl spacer build-
ing block (SDS) as potential divalent ligand of LecA.
43 B type I
44% at 1 mM
44 Lewisx
n.i.
n.i.
45 Lewisy
46 H type II
47 A type II
n.i.
32% at 1 mM
Binding to the target protein was then examined by titration
of 50 with increasing amounts of LecA (Fig. 5A). A Kd of
1.54 µM was determined for the divalent ligand and the calcu-
lated Z′ factor for this ligand 50 was 0.93. The applicability of
50 in a competitive binding assay at optimal protein con-
centration (2–5 µM) showed a good agreement of the deter-
mined IC50 values of selected LecA inhibitors (Fig. 5B) with
those obtained from the assay using the monovalent ligands 7
and 8.
48 B type II
45% at 1 mM
This journal is © The Royal Society of Chemistry 2016
Org. Biomol. Chem.

View Article Online
Paper
Organic & Biomolecular Chemistry
Scheme 3 Synthesis of bivalent fluorescent assay probe 50 from its glycopeptide precursor 49. Reagents and conditions (a) 6-FITC, PBS, pH 9,
DMF, o.n., 36%.
Fig. 5 A. Direct titration of the bivalent ligand 50 with increasing amounts of LecA; one site binding fit gave a determined Kd of 1.54 0.13 µM. One
representative titration of triplicates on one plate is shown. The Kd and standard deviation were calculated from three independent experiments on
three plates. B. Competitive binding assay to LecA (2 µM) of five different galactosides with the bivalent ligand 50. All potencies were calculated with
respect to 2. One representative titration of triplicates on one plate is shown. IC50s and standard deviations given were calculated from at least four
independent experiments on four plates for compound 2 and 30; or from two independent experiments for 4, 19 and 25.
LecA inhibitors and good agreement of the obtained IC50
Conclusion
values with published Kd values was achieved. We then
Competitive fluorescence polarization-based assays are valu- screened a set of thiogalactosides with various aglycones for
able tools for rapid screening of potential inhibitors. Due to inhibition and established a structure–activity relationship
the generally low affinity nature of lectin–carbohydrate inter- that compares to previously published O-linked analogs. Aryl
actions, the technique requires high amounts of protein in the thioglycosides were potent inhibitors whereas alkyl or benzyl
range of the Kd to achieve sufficient fluorescence polarization derivatives showed a reduced inhibition. We furthermore
of the probe. Here, we developed a competitive binding assay tested human blood group antigen epitopes for inhibition of
for the bacterial lectin LecA. Four different fluorescein-linked LecA but only a weak inhibition by blood group B was
monovalent galactosides were evaluated and the dissociation observed. To further reduce the required amount of LecA in
constants determined were in the 10–25 µM range. Due to the assay system, the fluorescently-labelled divalent LecA
their increased potency, phenyl-linked 8 and 9 required less inhibitor 50 was designed and synthesized. The successful use
protein in the competitive binding assay compared to ethyl- of this bivalent assay probe in inhibitor screening is a promis-
linked 6 or 7. Therefore, the competitive binding assay was ing possibility for the low affinity system galactose-LecA to
established for assay probe 8 (Z′ = 0.99) with a set of known further reduce the amount of protein (2–5 µM) needed in high
Org. Biomol. Chem.
This journal is © The Royal Society of Chemistry 2016

View Article Online
Organic & Biomolecular Chemistry
Paper
throughput assays. Here, comparable IC50 values were 37 were purchased from Carbosynth (Compton, UK) and
obtained with two monovalent and the bivalent assay probe. thioglycosides 34 and 38 from Sigma-Aldrich (Steinheim,
Due to the presence of an additional linking moiety in 50 Germany). Blood group oligosaccharides 39–48 were purchased
between its two galactose residues, inhibitors binding to LecA from Elicityl Oligotech (Crolles, France).
at positions occupied by this linker could be detected as hits
2-Bromoethyl α-D-galactopyranoside (10). D-Galactose (1,
despite a binding pose outside the carbohydrate binding site. 360 mg, 2 mmol) and La(OTf)3 (38 mg, 65 µmol) were sus-
This could theoretically yield inhibitors of the association of pended in 2-bromoethanol (2.1 mL, 30 mmol), heated to 70 °C
the linker with LecA rather than functional inhibitors of the under nitrogen atmosphere and subsequently stirred for 24 h
carbohydrate-binding functionality of LecA.
in analogy to Dasgupta et al.68 The resulting solution was
In summary, a robust competitive binding assay for LecA cooled to r.t., directly loaded onto a silica gel column and puri-
(Z′ up to 0.99) was developed and optimized with a set of assay fied by MPLC (SiO2, CH2Cl2/EtOH, 5–20%) to give an anomeric
probes. All synthesized galactose-based fluorescent tracers mixture of the glycoside as inseparable syrup (202 mg, 35%,
showed good performance in binding to LecA and three were α/β ratio = 7/3). The mixture (150 mg) was dissolved in 10 mL
tested in the competitive binding assay. Aryl-linked ligands, aqueous phosphate buffer (20 mM NaH2PO4, pH = 6.0) and
such as 8 and 9 (Kds of 8.1 and 7.4 µM, respectively), showed Aspergillus oryzae β-galactosidase (110 mg lactrase, 18%
highest potency in the monovalent series, further reduction of enzyme, Pro Natura, Germany) was added. The reaction was
protein consumption from 18 µM to 2–5 µM could be achieved incubated at 37 °C and shaking at 180 rpm for 22 h. The
by employing the potent divalent assay probe 50 (Kd = 1.5 µM). mixture was filtered over celite, the solvents were removed
The assay can be used for these potent monovalent and bi- in vacuo, the residue was purified by MPLC (SiO2,
valent probes in a 384-well format for miniaturization and CH2Cl2/EtOH, 5–15%) and pure α-galactoside 10 (48.5 mg,
protein-economic use in inhibitor screening.
33%) was obtained. 1H NMR (500 MHz, MeOH-d4) δ 4.93 (d,
J = 3.4 Hz, 1H, H-1), 4.02 (dt, J = 11.3, 6.5 Hz, 1H, 1H of
OCH2
̲
CH2Br), 3.98–3.95 (m, 1H, H-5), 3.94 (dd, J = 3.0, 1.3 Hz,
CH2Br),
1H, H-4), 3.89 (dt, J = 11.5, 5.9 Hz, 1H, 1H of OCH2
̲
Experimental details
Chemical synthesis
3.82 (dd, J = 10.1, 3.4 Hz, 1H, H-2), 3.79 (dd, J = 10.3, 3.0 Hz,
1H, H-3), 3.75 (dd, J = 11.5, 6.7 Hz, 1H, H-6a), 3.73 (dd, J =
General experimental details. Nuclear magnetic resonance 11.4, 5.4 Hz, 1H, H-6b), 3.64 (t, J = 6.2 Hz, 2H, OCH2CH2
(NMR) spectroscopy was performed on a Bruker Avance III 400, 13C NMR (126 MHz, MeOH-d4) δ 100.73 (C-1), 72.71 (C-5),
500 or 600 UltraShield spectrometer at 400/500/600 MHz (1H) 71.40 (C-3), 71.04 (C-4), 70.12 (C-2), 69.62 (OC
H2CH2Br), 62.70
H2Br). ESI-MS [C8H15BrO6 + Na]+ calcd
̲ Br);
̲
or 101/126/151 MHz (13C). Chemical shifts are given in ppm (C6), 31.33 (OCH2C
̲
and were calibrated on residual solvent peaks.67 Multiplicities 308.99, found 309.0. The synthesis of 10 was first reported by
were specified as s (singlet), d (doublet), t (triplet), q (quartet), Grandjean et al. without disclosure of analytical data.69
m (multiplet) or interpreted according to 1st order and higher
2-Azidoethyl α-D-galactopyranoside (11). Bromide 10 (48.5 mg,
order where possible. The signals were assigned with the 0.17 mmol) was dissolved in DMF (1.7 mL) and NaN3 (54 mg,
help of 1H,1H-COSY, DEPT-135-edited 1H,13C-HSQC and 0.84 mmol) was added. The reaction was stirred at 75 °C for
1H,13C-HMBC experiments. High resolution mass spectra were 16 h, cooled to r.t., filtered over celite and the volatiles were
obtained on a ESI Bruker micrOTOF II spectrometer. Analytical removed in vacuo. The product was purified by MPLC (SiO2,
HPLC-MS was performed on a Thermo Dionex Ultimate 3000 CH2Cl2/EtOH, 5–15%) and the pure product was obtained as
1
HPLC coupled to a Bruker Amazon SL ESI-MS system. Data solid (30.9 mg, 74%). H NMR (500 MHz, MeOH-d4) δ 4.86 (d,
were analyzed using DataAnalysis from Bruker. Thin layer
chromatography (TLC) was performed using silica gel 60 OCH2
J
=
2.9 Hz, 1H, H-1), 3.94–3.89 (m, 2H, H-4, 1H of
CH2N3), 3.87 (ddd, J = 6.8, 5.3, 1.3 Hz, 1H, H-5), 3.80 (dd,
̲
coated aluminum sheets containing fluorescence indicator J = 9.8, 2.9 Hz, 1H, H-2), 3.77 (dd, J = 9.8, 2.5 Hz, 1H, H-3), 3.74
(Merck KGaA, Darmstadt, Germany) using UV light (254 nm) (dd, J = 11.5, 6.8 Hz, 1H, H-6a), 3.70 (dd, J = 11.5, 5.6 Hz, 1H,
and by charring either in aqueous KMnO4 solution or in a H-6b), 3.64 (ddd, J = 10.6, 5.6, 3.5 Hz, 1H, 1H of OCH2
molybdate solution (a 0.02 M solution of ammonium cerium 3.57 (ddd, J = 13.2, 7.6, 3.5 Hz, 1H, 1H of OCH2CH2N3), 3.39
sulfate dihydrate and ammonium molybdate tetrahydrate in (ddd, J = 13.2, 5.7, 3.4 Hz, 1H, 1H of OCH2CH2
N3); 13C NMR
aqueous 10% H2SO4) with heating. Medium pressure liquid (126 MHz, MeOH-d4) δ 100.70 (C-1), 72.64 (C-5), 71.31 (C-3),
chromatography (MPLC) was performed on a Teledyne Isco 71.09 (C-4), 70.06 (C-2), 68.11 (OCH2CH2N3), 62.76 (C-6), 51.82
H2N3); ESI-MS [C8H15N3O6 + Na]+ calcd 272.09, found:
̲ CH2N3),
̲
̲
̲
Combiflash Rf200 system using pre-packed silica gel 60 (OCH2C
̲
columns from Teledyne Isco, SiliCycle or Macherey-Nagel. 272.1; 11 was first reported by Wang,70 the NMR data corres-
Commercial chemicals and solvents were used without further pond to those reported by Park et al. in D2O.71
purification. D-Galactose was purchased from Dextra Labora-
2-Aminoethyl α-D-galactopyranoside (12). Azide 11 (30.9 mg,
tories (Reading, UK) and fluorescein isothiocyanate isomer I 124 µmol) was dissolved in MeOH (1.0 mL) and Pd/C (3.1 mg,
(FITC) from Serva Biochemicals (Heidelberg, Germany). 10 wt%) was added. The reaction vessel was flushed several
Deuterated solvents were from Eurisotop (Saarbrücken, times with hydrogen and subsequently stirred under hydrogen
Germany). Methyl galactoside 2 and thioglycosides 27–33, 36, atmosphere (1 bar) over night. The reaction was filtered over
This journal is © The Royal Society of Chemistry 2016
Org. Biomol. Chem.

View Article Online
Paper
Organic & Biomolecular Chemistry
celite, the volatiles were removed in vacuo and the pure azide 14 was obtained in 55% over 2 steps. 1H NMR data
product was obtained as an oil (24.4 mg, 88%) which was used corresponds to the assignment reported by Susaki et al.
without further purification in the next step. 1H NMR 13C NMR (101 MHz, MeOH) δ 105.10 (C-1), 76.75 (C-5), 74.95
(500 MHz, MeOH-d4) δ 4.85 (1H, H-1, overlapped by HDO (C-3), 72.45 (C-2), 70.29 (C-4), 69.20 (OC
peak, assigned by HSQC and COSY), 3.89 (dd, J = 2.8, 1.2 Hz, 52.11 (OCH2CH2N3).
̲H2CH2N3), 62.49 (C-6),
̲
1H, H-4), 3.87–3.81 (m, 2H, H-5, 1H of OCH2
̲
CH2NH2),
2-Aminoethyl β-D-galactopyranoside (15), first reported by
3.81–3.78 (m, 1H, H-2), 3.76 (dd, J = 9.8, 2.7 Hz, 1H, H-3), 3.73 Chiang et al.,75 was synthesized from azide 14 (200 mg,
(dd, J = 11.0, 6.5 Hz, 1H, H-6a), 3.69 (dd, J = 11.4, 5.2 Hz, 1H, 0.8 mmol) by hydrogenation (1 bar) over Pd/C (10 wt%, 24 mg)
H-6b), 3.50 (ddd,
OCH2CH2NH2), 2.96 (ddd, J = 13.2, 5.5, 3.6 Hz, 1H, 1H of through a pad of celite and the volatiles were removed in vacuo
OCH2CH2
OCH2CH2
J = 10.5, 7.4, 3.7 Hz, 1H, 1H of in MeOH (4.8 mL) at r.t. over night. The solution was filtered
̲
̲
̲
NH2), 2.91 (ddd, J = 13.5, 7.3, 3.7 Hz, 1H, 1H of to give the title compound as white solid (190 mg, 99%).
NH2); 13C NMR (126 MHz, MeOH-d4) δ 100.54 (C-1), 1H NMR (400 MHz, D2O) δ 4.44 (d, J = 7.9 Hz, 1H, H-1),
72.62 (C-5), 71.47 (C-3), 71.09 (C-4), 70.30 (C-2), 69.03 4.06–3.97 (m, 1H, 1H of OCH2
(OCH2CH2NH2), 62.84 (C-6), 41.74 (OCH2CH2NH2). 0.7 Hz, 1H, H-4), 3.88–3.75 (m, 3H, H-6a, -6b, 1H of
N-(Fluorescein-5-yl)-N′-(α-D-galactopyranosyl-O-ethyl)-thiocarb- OCH2CH2NH2), 3.73 (ddd, J = 7.8, 4.3, 0.8 Hz, 1H, H-5), 3.68
amide (6). Amine 12 (23 mg, 103 µmol) was dissolved in DMF (dd, J = 9.9, 3.5 Hz, 1H, H-3), 3.57 (dd, J = 9.9, 7.8 Hz, 1H, H-2),
(4.0 mL) and NaHCO3 (0.093 mmol, 7.8 mg) was added. After 2.93 (ddd, J = 6.7, 4.6, 2.0 Hz, 2H, OCH2CH2NH2). NMR data
addition of FITC (103 µmol, 40 mg), the reaction was protected were in agreement with Susaki et al.74
from light and stirred at r.t. over night. The solvent was N-(Fluorescein-5-yl)-N′-(β-D-galactopyranosyl-O-ethyl)-thiocarb-
̲ CH2NH2), 3.95 (dd, J = 3.4,
̲
̲
̲
̲
removed under reduced pressure and the residue was purified amide (7). Amine 15 (0.235 mmol, 52 mg) was dissolved in
by MPLC (SiO2, CH2Cl2/EtOH/5% AcOH, gradient of 5%–40% DMF (1.5 mL), FITC (0.235 mmol, 86 mg) and NaHCO3
EtOH) to give 6 as an orange solid (16.2 mg, 26%). 1H NMR (2.15 mmol, 180 mg) were added. The reaction was stirred at
(500 MHz, MeOH-d4) δ 8.12 (d, J = 2.1 Hz, 1H, ArH), 7.77 (d, J = r.t. over night, filtered and the remaining solids were washed
8.9 Hz, 1H, ArH), 7.17 (d, J = 8.2 Hz, 1H, ArH), 6.81 (d, J = with DMF and filtered. The solvent of the combined filtrates
8.8 Hz, 2H, ArH), 6.68 (d, J = 2.3 Hz, 2H, ArH), 6.58 (dd, J = 8.8, was removed under reduced pressure. The residue was purified
2.4 Hz, 2H, ArH), 4.89 (d, J = 3.5 Hz, 1H, H-1), 3.99–3.92 (m, by MPLC (SiO2, CH2Cl2/EtOH/2% AcOH, gradient of 5%–40%
2H, 1H of OCH2
2H, H-4, -5), 3.84–3.78 (m, 3H, 1H of OCH2CH2
3.78–3.73 (m, 2H, 1H of OCH2
̲
CH2NHR, 1H of OCH2CH2
̲
NHR), 3.92–3.85 (m, EtOH) to give the title compound as orange foam after lyophili-
̲
NHR, H-2, -3), zation from water (76 mg, 55%). 1H NMR (400 MHz, MeOH-d4)
̲
CH2NHR, H-6a), 3.70 (dd, J = δ 8.23 (s, 1H, ArH), 7.84 (dd, J = 8.3, 1.9 Hz, 1H, ArH), 7.15 (d,
11.4, 4.9 Hz, 1H, H-6b). 13C NMR (126 MHz, MeOH-d4) J = 8.3 Hz, 1H, ArH), 6.69–6.64 (m, 4H, ArH), 6.54 (dd, J = 8.7,
δ 182.89 (CvS), 130.92 (2C, ArCH), 130.0 (ArCH, from HSQC), 2.2 Hz, 2H, ArH), 4.30 (d, J = 7.5 Hz, 1H, H-1), 4.05–3.88 (m,
126.8 (ArCH, from HSQC), 120.8 (ArCH, from HSQC), 112.61 3H, 2H of OCH2
(2C, ArCH), 103.61 (2C, ArCH), 100.79 (C-1), 72.69 (C-5), 71.47 3H, H-4, -6a,b), 3.66–3.53 (m, 2, -2, -5H), 3.50 (dd, J = 9.7,
(C-3), 71.12 (C-4), 70.21 (C-2), 67.98 (OC
H2CH2NHR), 62.93 3.3 Hz, 1H, H-3). 13C NMR (101 MHz, MeOH-d4) δ 182.72
(C-6), 45.59 (OCH2C
H2NHR). ESI-MS [C29H28N2O11S + Na]+ calcd (CvS), 175.26 (CvO), 171.30 (CvO), 154.14 (ArC), 130.35
635.13, found 613.1; Rf 0,45 (CH2Cl2/EtOH 2 : 1 + 2% AcOH). (ArC), 113.56 (ArCH), 111.45 (2C, ArCH), 105.53 (ArCH), 103.48
2-Bromoethyl 2,3,4,6-tetra-O-acetyl-β-D-galactopyranoside (C-1), 101.39 (ArCH), 76.90 (C-5), 74.88 (C-3), 72.55 (C-2), 70.37
̲ CH2NHR, 1H of OCH2CH̲2NHR), 3.86–3.74 (m,
̲
̲
(13). first reported by Coles et al.,72 was synthesized according (C-4), 70.15 (OC
̲H2CH2NH) 62.81 (C-6), 46.11 (OCH2C̲H2NH);
to Dahmen et al.57 1H NMR (400 MHz, CDCl3) δ 5.38 (d, J = Rf = 0.25 (CH2Cl2/EtOH 3 : 1 + 2% AcOH); HRMS calcd
3.3 Hz, 1H, H-4), 5.22 (dd, J = 10.4, 8.0 Hz, 1H, H-2), 5.02 (dd, C29H27N2O11S−: 611.1341 [M − H]−; found: 611.1345.
J = 10.5, 3.4 Hz, 1H, H-3), 4.53 (d, J = 7.9 Hz, 1H, H-1),
4.20–4.08 (m, 3H, H-6a, -6b, 1H of OCH2
CH2Br), 3.88–3.92 (m, thesized according to Cohen et al.76
1H, H-5), 3.85–3.76 (m, 1H, 1H of OCH2CH2Br), 3.50–3.43 (m, p-Nitrophenyl 2,3,4,6-tetra-O-acetyl-β-D-galactopyranoside (17).
2H, OCH2CH2Br), 2.14 (s, 3H), 2.07 (s, 3H), 2.04 (s, 3H), 1.98 Molecular sieves (5 g, 3 Å) were dried under vacuum at 350 °C
(s, 3H). 13C NMR (101 MHz, CDCl3) δ 170.49, 170.32, 170.24, in a two-necked flask for 30 min. After cooling to r.t., the flask
169.64 (4C, CH3CO), 101.66 (C-1), 70.96 (C-5), 70.87 (C-3), was flushed with argon and dry CH2Cl2 (25 mL) was added.
69.87 (OCH2CH2Br), 68.68 (C-2), 67.09 (C-4), 61.38 (C-6), 30.04 Galactosyl donor 16 (5.0 g, 12.8 mmol) and p-nitrophenol
(OCH2CH2Br), 20.97, 20.79, 20.77, 20.69 (4C, CH3CO). (2.2 g, 15.9 mmol) were added, the reaction mixture was
1,2,3,4,6-Penta-O-acetyl-β-D-galactopyranoside (16) was syn-
̲
̲
̲
̲
̲
̲
̲
2-Azidoethyl β-D-galactopyranoside (14). 14 was synthesized cooled (0 °C) and BF3·OEt2 (8.2 mL, 65.3 mmol) was added
from 13 by nucleophilic displacement with sodium azide dropwise. Afterwards, the mixture was allowed to warm to r.t.
to give 2-azidoethyl-2,3,4,6-tetra-O-acetyl-β-D-galactopyranoside and stirred for 1 d. The reaction was poured onto ice water
according to D’Agata et al.73 The azido galactoside was de- (200 mL) and the aqueous phase was extracted with aqueous
protected under Zemplén conditions as previously reported by saturated NaHCO3 (5 × 50 mL). The combined organic layers
Susaki et al.74 with the following modifications: the reaction were washed with brine (25 mL) and dried over anhydrous
was stopped by neutralization with Amberlite IR-120H+, fil- Na2SO4. After filtration, the solvent was removed in vacuo. The
tered and the volatiles were removed in vacuo. Deprotected crude product was dissolved in warm EtOH and left at 4 °C
Org. Biomol. Chem.
This journal is © The Royal Society of Chemistry 2016

View Article Online
Organic & Biomolecular Chemistry
Paper
overnight. Precipitated pure 17 (3.1 g, 6.6 mmol, 52%) was compound was obtained as a colorless oil (22 mg, 96%).
1
obtained as light yellow amorphous solid. H-NMR (400 MHz, 1H NMR (500 MHz, D2O) δ 7.37–7.32 (m, 2H, Ar-H), 7.13–7.08
CDCl3): δ 8.25–8.17 (m, 2H, ArH), 7.10–7.05 (m, 2H, ArH), 5.51 (m, 2H, Ar-H), 5.00 (d, J = 7.5 Hz, 1H, H-1), 3.96 (d, J = 3.2 Hz,
(dd, J = 10.4, 7.9 Hz, 1H, H-2), 5.47 (d, J = 3.3 Hz, 1H, H-4), 1H, H-4), 3.82 (t, J = 6.20 Hz, 1H, H-5), 3.79–3.69 (m, 4H, H-2,
5.17 (d, J = 7.9 Hz, 1H, H-1), 5.13 (dd, J = 10.4 3.4 Hz, 1H, H-3), -3, -6a,b), 3.46 (s, 2H, CH
4.25–4.10 (m, 3H, H-5, -6a,b), 2.18 (s, 3H, CH3) 2.06 (s, 6H, (CO), 154.25, (ArC), 131.50 (ArC), 123.84 (2C, ArCH), 117.02
CH3) 2.01 (s, 3H, CH3
). 13C-NMR (101 MHz, CDCl3): δ 170.41 (2C, ArCH), 100.92 (C1), 75.42 (C5), 72.53 (C3), 70.51 (C2),
̲
2); 13C NMR (126 MHz, D2O) δ 173.56
̲
̲
̲
̲
(CvO), 170.23 (CvO), 170.15 (CvO), 169.38 (CvO), 161.30 68.45 (C4), 60.72 (C6), 43.84 (CH̲
2); MS: [C14H20N2O7 + H]+
(ArC), 143.31 (ArC), 125.91 (2C, ArCH), 116.68 (2C, ArCH), calcd 329.1, found 329.1.
98.69 (C-1), 71.57 (C-5), 70.68 (C-3), 68.37 (C-2), 66.79 (C-4),
para-Substituted phenyl-linked reporter ligand 8. Amine 20
̲H3); (22 mg, 0.07 mmol) was dissolved in dry DMF (4 mL). Fluores-
61.46 (C-6), 20.82 (CH3), 20.79 (CH3), 20.76 (CH3), 20.68 (C
̲
̲
̲
Rf = 0.27 (CH2Cl2/MeOH 12 : 1); ESI-MS: [C20H23NO12 + Na]+ ceine isothiocyanate (26.1 mg, 0.07 mmol, 1 eq.) and NaHCO3
calcd 492.1 found 492.1. 17 was first described by Goebel and (5 mg, 0.06 mmol, 0.9 eq.) were dissolved in dry DMF (1 mL)
Avery,77 NMR data obtained are in agreement with the litera- and this solution was added to 20 and stirring was continued
ture 1H-NMR data by Apparu et al.78
at r.t. for 14 h. The solvent was removed and the crude product
p-(α-Azidoacetamido)-phenyl 2,3,4,6-tetra-O-acetyl-β-D- was purified by MPLC, (SiO2, CH2Cl2/EtOH containing 1%
galactopyranoside (18) was synthesized from protected 17 in 3 HOAc, gradient of 5 to 60% EtOH). To remove the residual
chemical steps in analogy to the reaction sequence of Vidal AcOH the residue was co-evaporated with hexane (3 × 25 mL)
and co-workers.46 1H-NMR (400 MHz, CDCl3) δ 8.01 (s, 1H, and subsequently lyophilized. 8 was obtained as an orange
1
NH
̲
(dd, J = 10.3, 7.9 Hz, 1H, H-2), 5.44 (dd, J = 3.4, 1.2 Hz, 1H, δ 10.44 (br s, 1H, NH/OH), 10.08 (s, 1H, NH/OH), 8.40–8.22 (m,
H-4), 5.09 (dd, J = 10.5, 3.4 Hz, 1H, H-3), 4.99 (d, J = 8.0 Hz, 2H, ArH, NH), 7.79 (d, J = 7.5 Hz, 1H, ArH), 7.51 (d, J = 8.9 Hz,
1H, H-1), 4.21 (dd, J = 11.3, 7.0 Hz, 1H, H-6a), 4.13 (s, 2H, 2H, ArH), 7.15 (d, J = 8.2 Hz, 1H, Ar-H), 7.00 (d, J = 8.9 Hz, 2H,
CH
(s, 3H, CH
CH
3); 13C-NMR (101 MHz, CDCl3) δ 170.50 (CH3C
(CH3CO), 170.25 (CH3CO), 169.53 (CH3CO), 164.63 (C
̲
2N3), 4.18–4.10 (m, 1H, H-6b), 4.06–4.01 (m, 1H, H-5), 2.17 ArH), 6.73–6.62 (m, 3H, ArH), 6.54–6.33 (m, 4H, ArH), 4.75 (d,
3), 2.07 (s, 3H, CH3), 2.05 (s, 3H, CH3), 2.00 (s, 3H, J = 7.7 Hz, 1H, H-1), 4.37 (s, 2H, CH2NHR), 3.69 (d, J = 3.1 Hz,
O), 170.37 1H, H4), 3.58–3.46 (m, 5H, H-2, -3, -5, -6a,b6); 13C NMR
ONH), (126 MHz, DMSO-d6) δ 180.70 (CS), 166.76 (CO), 153.47 (ArC),
̲
̲
̲
̲
̲
̲
̲
̲
̲
̲
154.12 (ArC), 132.33 (ArC), 121.81 (2C, ArCH), 117.75 (2C, 133.00 (ArC), 129.68 (2C, ArCH), 120.38 (2C, ArCH), 116.55
ArCH), 100.05 (C-1), 71.17 (C-5), 70.90 (C-3), 68.71 (C-2), 66.97 (2C, ArCH), 110.35 (ArC), 102.49 (ArCH), 101.41 (C-1), 75.47
(C-4), 61.48 (C-6), 53.01 (C
20.79 (CH3), 20.72 (CH3); Rf = 0.23 (PE/EtOAc 1 : 1); ESI-MS (C-6), 47.47 (C
[C22H26N4O11 + H]+ calcd 523.17, found 523.1. Our data match found 718.1661; Rf = 0.14 (CH2Cl2/EtOH/AcOH 3 : 1 : 0.08).
the previously published NMR spectra by Cecioni et al.,46
m-Nitrophenyl 2,3,4,6-tetra-O-acetyl-β-D-galactopyranoside
though the peak assignment by Cecioni et al. differs from the (21). Molecular sieves (5 g, 3 Å) were dried under vacuum at
one reported here. 350 °C in a two-necked flask for 30 min. After cooling to r.t.,
̲H2N3), 20.88 (C̲H3), 20.81 (C̲
̲
̲
̲
p-(2-Azidoacetamido)-phenyl β-D-galactopyranoside (19). A the flask was flushed with argon and dry CH2Cl2 (25 mL) was
small portion of sodium was dissolved in dry MeOH (5 mL) added. Galactosyl donor 16 (5.0 g, 12.8 mmol) and m-nitro-
and tetraacetate 18 (164 mg, 0.31 mmol) was added. The solu- phenol (2.2 g, 15.9 mmol) were added, the reaction mixture
tion was stirred for 2 h at r.t., neutralized with Amberlite was cooled (0 °C) and BF3·OEt2 (8.2 mL, 65.3 mmol) was
IR-120/H+ (Merck), the resin was filtered and the solvent was added dropwise. Afterwards, the mixture was allowed to warm
removed in vacuo to give 19 (102 mg, 0.31 mmol, 99%) as a to r.t. and stirred for 1 d. The reaction was poured onto ice
1
white solid. H NMR (400 MHz, MeOH-d4) δ 7.49–7.43 (m, 2H, water (250 mL) and the aqueous phase was extracted with
ArH), 7.14–7.04 (m, 2H, ArH), 4.82 (d, J = 7.7 Hz, 1H, H-1), 3.99 aqueous saturated NaHCO3 (5 × 50 mL). The combined
(s, 2H, CH2), 3.90 (d, J = 3.5 Hz, 1H, H-4), 3.82–3.72 (m, 3H, organic layers were washed with brine (25 mL) and dried over
H-2, -6a,b), 3.70–3.64 (m, 1H, H-5), 3.57 (dd, J = 9.7, 3.4 Hz, anhydrous Na2SO4. After filtration, the solvent was removed in
1H, H-3); 13C NMR (101 MHz, MeOH-d4) δ 168.25 (CvO), vacuo. The crude product was dissolved in warm EtOH and left
156.15 (ArC), 133.68 (ArC), 122.82 (2C, ArCH), 118.16 (2C, at 4 °C overnight. Precipitated pure 21 (3.3 g, 7.1 mmol, 55%)
ArCH), 103.28 (C-1), 76.99 (C-5), 74.85 (C-3), 72.27 (C-2), 70.23 was obtained as light yellow amorphous solid. 1H-NMR
(C-4), 62.43 (C-6), 53.23 (CH2N3); HRMS: [C14H18N4O7 + Na]+ (400 MHz, CDCl3): δ 7.94 (ddd, J = 8.2, 2.1, 0.9 Hz, 1H, ArH),
calcd: 377.1068; found: 377.1056; Rf = 0.68 (EtOAc/EtOH 5 : 3).
7.88 (t, J = 2.3 Hz, 1H, ArH), 7.46 (t, J = 8.2 Hz, 1H, ArH), 7.31
p-(α-Aminoacetamido)-phenyl β-D-galactopyranoside (20) (ddd, J = 8.3, 3.5, 0.9 Hz, 1H, ArH), 5.52 (dd, J = 10.5, 7.9 Hz,
was synthesized from the azide 19(25 mg, 0.07 mmol) by 1H, H-2), 5.48 (d, J = 3.4 Hz, 1H, H-4), 5.14 (d, J = 7.7 Hz, 1H,
hydrogenation (1 bar) over Pd–C (8 wt%, 2 mg) in MeOH for H-1), 5.12 (dd, J = 10.5, 3.4 Hz, 1H, H-3), 4.24–4.13 (m, 3H,
2 h. The solution was filtered through a pad of celite and H-5, -6a,b), 2.18 (s, 3H, CH
the volatiles were removed in vacuo. After MPLC, (SiO2, CH3), 2.01 (s, 3H, CH3
); 13C-NMR (400 MHz, CDCl3): δ 170.73
EtOAc/EtOH/5% NH4OH, gradient 2–100% EtOH) the title (CvO), 170.27 (CvO), 170.16 (CvO), 169.44 (CvO), 157.15
3̲ ), 2.10 (s, 3H, CH̲3), 2.08 (s, 3H,
̲
̲
This journal is © The Royal Society of Chemistry 2016
Org. Biomol. Chem.

View Article Online
Paper
Organic & Biomolecular Chemistry
(ArC), 149.20 (ArC), 130.34 (ArCH), 123.88 (ArCH), 118.33 109.97 (ArCH), 102.88 (C-1), 76.97 (C-4), 74.88 (C-2), 72.24
(ArCH), 111.38 (ArCH), 99.32 (C-1), 71.81 (C-5), 70.76 (C-3), (C-3), 70.21 (C-5), 62.41 (C-6), 53.32 (CH2N3); HRMS:
68.44 (C-2), 67.07 (C-4), 61.94 (C-6), 20.87 (C
20.75 (CH3), 20.70 (C
H3); ESI-MS: [C20H23NO12 + Na]+ calcd (EtOAc/EtOH 4 : 1).
492.1, found 492.1; Rf = 0.33 (PE/EtOAc 3 : 2). The title com- m-(α-Aminoacetamido)-phenyl β-D-galactopyranoside (24).
pound was first synthesized by Iversen and Johansson starting Azide 23 (30 mg, 0.08 mmol) was hydrogenolyzed under H2
from tetraacetylgalactosyl bromide.79
(1 atm) over Pd–C (10 wt%, 3 mg) in MeOH for 2 h. The solu-
̲H3), 20.79 (C̲
H3), [C14H18N4O7 + Na]+ calcd 377.1068, found 377.1053; Rf = 0.47
̲
̲
m-(α-Azidoacetamido)-phenyl 2,3,4,6-tetra-O-acetyl-β-D-galac- tion was filtered through a pad of celite and the volatiles were
topyranoside (22). Nitrophenyl galactoside 21 (650 mg, removed in vacuo. After MPLC, (SiO2, EtOAc/EtOH/5% NH4OH,
1.39 mmol) was dissolved in dry CH2Cl2 (30 mL) and Pd/C gradient 2–100% EtOH) compound 24 was obtained as a color-
(60 mg) was added. After three vacuum/H2 cycles the reaction less oil (23 mg, 88%). 1H NMR (500 MHz, D2O) δ 7.40 (t, J =
was stirred under H2 atmosphere (1 atm) for 16 h at r.t. After 8.2 Hz, 1H, ArH), 7.33 (t, J = 2.2 Hz, 1H, ArH), 7.17 (dd, J = 8.1,
cooling to 0 °C, NEt3 (230 µL, 1.67 mmol) followed by bromo- 1.1 Hz, 1H, ArH), 7.00 (dd, J = 8.4, 2.5 Hz, 1H, ArH), 5.08 (d, J =
acetylbromide (145 µL, 1.67 mmol) were added dropwise. After 7.5 Hz, 1H, H-1), 4.03 (dd, J = 3.2, 1.0 Hz, 1H, H-4), 3.92–3.87
stirring for 1 h at 0 °C, it was warmed to r.t. and stirred for an (m, 1H, H-5), 3.85–3.76 (m, 4H, H-2, -3, -6a,b), 3.56 (s, 2H,
additional hour. The reaction was filtered and the organic CH̲
2); 13C NMR (126 MHz, D2O) δ 157.73 (ArC), 138.590
phase was washed 1 N HCl (2 × 50 mL), saturated NaHCO3 (ArC), 130.94 (ArCH), 116.67 (ArCH), 113.88 (ArCH), 110.49
(3 × 50 mL) and water (2 × 50 mL). After drying of the organic (ArCH), 101.33 (C-1), 76.10 (C-5), 73.18 (C-3), 71.14 (C-2),
layer over Na2SO4 the solvent was removed in vacuo. Crude 69.10 (C-4), 61.39 (C-6), 44.44 (CH̲2N3); HRMS: [C14H20N2O7 +
bromoacetylated galactoside was obtained as yellow foam and H]+ calcd 329.1343, found 329.1327; Rf = 0.2 (EtOH/NH4OH
dissolved in dry DMF (20 mL). NaN3 (452 mg, 6.95 mmol) and 6 : 1).
the catalyst Bu4NI (52 mg, 0.14 mmol) were added and the
meta-Substituted phenyl-linked reporter ligand 9. Amine 24
reaction was stirred for 16 h at 50 °C. After filtration, the (23 mg, 0.07 mmol) was dissolved in dry DMF (4 mL). Fluores-
organic phase was washed with saturated aqueous NaHCO3 ceine isothiocyanate (26.1 mg, 0.07 mmol) and NaHCO3 (5 mg,
(2 × 50 mL), water (2 × 50 mL), and brine (2 × 50 mL). The 0.06 mmol) were separately dissolved in dry DMF (1 mL) and
organic layer was dried over Na2SO4, filtered and the volatiles added to 24 under stirring at r.t. After 14 h, the solvent was
removed in vacuo. The crude yellow oil was purified by MPLC removed and the crude product was purified by MPLC (SiO2,
(SiO2, PE/EtOAc 0–100%) to give pure azide 22 as a solid CH2Cl2/EtOH/1% AcOH, gradient of 5 to 60% EtOH). For
(615 mg, 1.19 mmol, 86% over three steps). 1H-NMR removal of the remaining AcOH the residue was co-evaporated
(400 MHz, CDCl3) δ 8.22 (s, 1H, NH), 7.40 (t, 2.2 Hz, 1H, ArH), with hexane (3 × 25 mL) and lyophilized. The title compound
1
7.24 (t, J = 8.1 Hz, ArH), 7.16 (ddd, J = 8.2, 2.0, 1.0 Hz, 1H, 9 (38 mg, 0.05 mmol, 71%) was obtained as orange solid. H
ArH), 6.78 (ddd, J = 8.2, 2.5, 1.0 Hz, 1H, ArH), 5.46 (dd, J = NMR (500 MHz, DMSO-d6) δ 10.40 (s, 1H, NH/OH), 10.20 (s,
10.6, 8.0 Hz, 1H, H-2), 5.44 (d, J = 3.2 Hz, 1H, H-4), 5.09 (dd, 1H, NH/OH), 10.10 (s, 1H, NH/OH), 8.36 (s, 1H, ArH), 8.25 (br
J = 10.5, 3.5 Hz, 1H, H-3), 5.06 (d, J = 7.9 Hz, 1H, H-1), s, 1H, CH2NH
4.24–4.06 (m, 5H, H-5, -6a,b, CH2N3), 2.16 (s, 3H, CH3), 2.06 (s, ArH), 7.35–7.18 (m, 4H, NH, ArH), 6.78–6.74 (m, 1H, ArH),
3H, CH3), 2.04 (s, 3H, CH3), 2.00 (s, 3H, CH3
); 13C-NMR 6.70–6.53 (m, 7H, ArH), 5.18 (d, J = 5.25 Hz, 1H, OH-2), 4.86 (d,
(101 MHz, CDCl3) δ 170.85 (CH3CO), 170.68 (CH3CO), 170.54 J = 5.35 Hz, 1H, OH-3), 4.78 (d, J = 7.70 Hz, 1H, H-1), 4.68–4.61
(CH3CO), 169.87 (CH3CO), 163.00 (CONH), 157.77 (ArC), (m, 1H, OH-6), 4.51 (d, J = 4.65 Hz, 1H, OH-4), 4.43–4.37 (m,
138.60 (ArC), 130.35 (ArC), 115.07 (ArCH), 113.74 (ArCH), 2H, CH2NH), 3.73–3.69 (m, 1H, H4), 3.60–3.51 (m, 3H, H-2, -5,
109.23 (ArCH), 99.93 (C-1), 71.50 (C-5), 71.23 (C-3), 68.99 (C-2), -6a), 3.51–3.45 (m, 1H, H-6b), 3.43–3.39 (m, 1H, H-3); 13C NMR
67.30 (C-4), 61.75 (C-6), 53.32 (CH2N3), 21.19 (CH3), 21.10 (126 MHz, DMSO-d6) δ 180.69 (CS), 168.50 (CO), 167.12 (CO),
(CH3), 21.07 (CH3), 21.02 (C
H3); ESI-MS [C22H26N4O11 + Na]+ 159.55 (ArC), 157.87 (ArC), 151.92 (ArC), 141.22 (ArC), 139.77
calcd 545.15, found 545.1; Rf = 0.21 (PE/EtOAc 1 : 1). (ArC), 129.51 (ArCH), 129.08 (ArCH), 126.70 (ArC), 124.23
̲), 7.79 (dd, J = 8.3, 1.9 Hz, 1H, ArH), 7.39 (s, 1H,
̲
̲
̲
̲
̲
̲
̲
̲
̲
̲
̲
̲
̲
̲
̲
m-(α-Azidoacetamido)-phenyl β-D-galactopyranoside (23). A (ArCH), 116.57 (ArCH), 112.64, 110.97, 109.73, 107.61, 102.25
small portion of sodium was dissolved in dry MeOH (5 mL) (ArCH), 101.15 (C-1), 75.38 (C-2/C-5), 73.29 (C-3), 70.29
and tetraacetate 22 (100 mg, 0.28 mmol) was added. The (C-2/C-5), 67.95 (C-4), 60.11 (C-6), 47.65 (CH2NH); ESI-MS
solution was stirred for 2 h at r.t., then neutralized with [C35H31N3O12S + H]+ calcd 718.17, found 718.2; Rf = 0.12
Amberlite IR-120/H+ (Merck), and the resin was filtered. The (CH2Cl2/EtOH/AcOH 3 : 1 : 0.08).
solvent was removed in vacuo to give 23 (91 mg, 0.31 mmol,
p-Nitrophenyl β-D-galactopyranoside (25) was obtained after
1
98%) as a white solid. H-NMR (400 MHz, MeOH-d4) δ 7.44 (t, deprotection of acetate 17 following Zemplén conditions. 25
J = 2.0 Hz, 2H, ArH), 7.26–7.19 (m, 2H, ArH, NH), 6.91–6.86 was first described by Goebel and Avery.77
(m, 1H, ArH), 4.87 (d, J = 7.8 Hz, 1H, H-1), 4.00 (s, 2H, CH
̲
2),
m-Nitrophenyl β-D-galactopyranoside (26) was obtained
3.91 (dd, J = 3.4, 1.0 Hz, 1H, H-4), 3.82–3.73 (m, 3H, H-2, -6a,b), after deprotection of acetate 21 following Zemplén conditions.
3.72–3.68 (m, 1H, H-5), 3.58 (dd, J = 9.7, 3.4 Hz, 1H, H-3); 26 was first described by Csuros et al.80
13C-NMR (101 MHz, MeOH-d4) δ 168.40 (C
̲
Phenyl β-D-galactopyranoside (4) was prepared from galac-
Org. Biomol. Chem.
This journal is © The Royal Society of Chemistry 2016

View Article Online
Organic & Biomolecular Chemistry
Paper
Benzyl β-D-thiogalactopyranoside (35). Galactose pentaace-
Cleavage from solid support. 30% TFA in DCM was added
tate 16 was treated with thioacetic acid and BF3·OEt2 according to the resin and was shaken for 1 h. The filtrate was added to
to Caraballo et al.82 After deprotection of 1-thio-galactose cold Et2O (40 mL). The resulting precipitate was centrifuged
pentaacetate under Zemplén conditions, the resulting 1-thio- and the ether was decanted. The product was redissolved in
galactose was treated with sodium hydride and benzyl bromide MeOH and was again precipitated in cold Et2O, centrifuged
to give thiogalactoside 35 after chromatographic purification.83 and decanted.
1H NMR (400 MHz, MeOH-d4) δ 7.39–7.34 (m, 2H, ArH),
β-D-Gal(1,4)-4-SDS (49) was synthesized following the above
7.32–7.25 (m, 2H, ArH), 7.24–7.18 (m, 1H, ArH), 4.15 (d, J = described general coupling and deprotection protocol with the
9.7 Hz, 1H, H-1), 4.03 (d, J = 12.9 Hz, 1H, 1H of PhCH2), building block sequence TDS, SDS, SDS, TDS. The primary
3.89–3.83 (m, 2H, H-4, 1H of PhCH2), 3.79 (dd, J = 11.5, 6.9 Hz, amine end group was capped by the reaction with acetic anhy-
1H, H-6a), 3.71 (dd, J = 11.5, 5.1 Hz, 1H, H-6b), 3.64–3.53 (m, dride. Two β-D-galactosetetraacetate units were subsequently
1H, H-2), 3.45 (t, J = 6.0 Hz, 1H, H-5), 3.38 (dd, J = 9.2, 3.4 Hz, conjugated to the polymeric scaffold by applying general
1H, H-3); 13C NMR (101 MHz, MeOH-d4) δ 139.53 (ArC), 130.28 CuAAC protocol followed by sugar deprotection. Final cleavage
(2C, ArCH), 129.41 (2C, ArCH), 127.93 (ArC), 85.79 (C-1), 80.66 from solid support gave the desired glycooligomer 49 with a
(C-5), 76.28 (C-3), 71.45 (C-2), 70.59 (C-4), 62.73 (C-6), 34.38 yield of 58 mg (0.041 mmol, 82%) and 92% purity.
(PhC̲
H2). The title compound was first described by Helferich MALDI-TOF-MS: m/z [C58H98N18O23 + H]+ calcd 1415.71; found
and Türk.84
1415.7. RP-HPLC: (5%/95% MeCN/H2O to 95%/5% MeCN/H2O
in 30 min): tr = 4.91 min.
β-D-Gal(1,4)-4-SDS-FITC (50). 19.27 mg (0.014 mmol) of β-D-
Experimental procedure for β-D-Gal(1,4)-4-SDS-FITC (50)
The synthesis is following previously established protocols for Gal(1,4)-4-SDS 49 were dissolved in 1.36 mL PBS buffer (pH
the solid phase synthesis of glycomacromolecules applying 9.0) to give a 10 mM solution. 1.36 mL of a 10 mM solution of
alkyne-functionalized TDS building block and short ethyl FITC (5.29 mg, 0.014 mmol) in DMF were added. The reaction
spacer building block SDS.66 Commercially available Tentagel mixture was stirred overnight. The solvent was removed under
chlorotrityl resin modified with ethylene diamine (EDA) was reduced pressure and the obtained crude product was purified
used as resin for solid phase synthesis. After swelling by preparative HPLC (5%/95% MeCN/H2O to 95%/5%
0.05 mmol of resin in DCM for 30 min, the initial coupling MeCN/H2O in 10 min). The final product 50 was obtained as
was performed by dissolving 0.25 mmol (5 eq.) of building yellow powder after lyophilization with 96% purity. 1H-NMR
block and 0.25 mmol (5 eq.) of PyBOP in DMF (2 mL), followed (600 MHz, D2O): δ (ppm) 8.03 (s, 1H, FITC-Ar-H), 7.86 (s, 1H,
by the addition of 1.0 mmol (20 eq.) DIPEA. This mixture was Ar-H), 7.80 (s, 1H, Ar-H), 7.60 (s, 1H, FITC-ArH), 6.94 (s, 1H,
shaken for 30 s and purged with nitrogen for 1 min. Then the FITC-Ar-H), 6.73 (s, 2H, FITC-Ar-H), 6.67–6.47 (m, 4H, FITC-Ar-
mixture was added to the resin and was shaken for 1.5 h. After H), 4.63–4.50 (m, 4H, N
CH2O), 4.36–4.29 (m, 2H, Gal-H),
that, the resin was washed from unreacted reagent 10 times 4.28–3.96 (m, 4H, N(CH2)2O), 3.92–3.88 (m, 2H, Gal-H),
with DMF. The Fmoc protecting group was then cleaved by 3.76–3.16 (m, 36H, Gal-H, NH(CH2)2NH), 2.99–2.81 (m, 4H, NC
adding a solution of 0.4 M LiCl in DBU/piperidine/DMF (O)CH
(1 : 1 : 48, v : v) for 20 min and 25% piperidine in DMF for 20 16H, C(O)CH
and then 10 min. After the last deprotection step the resin was 1.5 H, NHC(O)CH
washed 10 times with DMF.
Na]+ calcd 1826.73; found 1826.7. RP-HPLC: (5%/95%
̲
2CH2), 2.78–2.62 (m, 4H, NC(O)CH2CH2
2CH2C(O)), 1.91 (s, 1.5 H, NHC(O)CH3
3). MALDI-TOF-MS: m/z [C79H109N19O28S +
̲
), 2.52–2.35 (m,
̲
̲
̲
), 1.89 (s,
̲
Capping of N-terminal SITE. After successful assembly of MeCN/H2O to 95%/5% MeCN/H2O in 30 min): tr = 12.96 min.
the desired number of building blocks on solid phase, the
N-terminal site was capped with an acetyl group. For that, a
Recombinant expression and purification of LecA
0.5 M solution of DIPEA and acetic anhydride in DMF was The protein LecA was expressed and purified as described pre-
shaken with the resin for 1 h. Subsequently, it was washed viously.65 Briefly, E. coli BL21(DE3) carrying the plasmid
10 times with DMF.
pET25pa1l were grown in 1 L LB supplemented with ampicillin
General CuAAC protocol. To 0.05 mmol of resin loaded with (100 µg mL−1) to an OD600 = 0.5–0.6 at 37 °C and 180 rpm.
the desired number of building blocks, 0.4 mmol (8 eq.) of Expression was induced with IPTG (0.25 mM final concen-
2-azidoethyl 2,3,4,6-tetra-O-acetyl-β-D-galactopyranoside85 dis- tration) and bacteria were cultured for 4 h at 30 °C and 180
solved in 1.5 mL DMF was added. 20 mol% sodium ascorbate rpm. The cells were then harvested by centrifugation (3000g,
and 20 mol% CuSO4 were dissolved in 0.5 mL water and were 10 min) and the pellet was washed with PBS. The cells were
also added to the resin. The mixture was shaken overnight and resuspended in 25 mL TBS/Ca (20 mM Tris, 137 mM NaCl,
was subsequently washed in cycles with a 23 mM solution of 2.6 mM KCl at pH 7.4 supplemented with 100 μM CaCl2) with
sodium diethyldithiocarbamate in DMF, water, DMF and PMSF (1 mM) and lysozyme (0.4 mg mL−1) and subsequently
DCM.
disrupted using a homogenizer (5 cycles, microfluidics). Cell
The deprotection of the sugar moieties was performed debris was removed by centrifugation (60 min, 10 000g) and
using 5 mL of 0.2 M suspension of NaOMe in MeOH. The the supernatant was loaded on galactosylated sepharose
resin was shaken with this mixture for 1 h. Subsequently it was CL-6B.86 The column was washed with TBS/Ca and LecA was
washed with water and DCM.
eluted by addition of 100 mM D-galactose to the buffer. The
This journal is © The Royal Society of Chemistry 2016
Org. Biomol. Chem.

View Article Online
Paper
Organic & Biomolecular Chemistry
eluted fractions were extensively dialyzed against distilled and to Laura Becker (University of Konstanz) for the synthesis
water and then, the protein was lyophilized. Between 20 and of phenyl β-D-galactoside. We wish to further acknowledge the
35 mg LecA per liter bacterial culture were obtained. The protein support of Anke Friemel (University of Konstanz), Dr Josef
was dissolved in TBS/Ca before use and after centrifugation the Zapp (Saarland University) and Dr Holger Bußkamp (Univer-
concentration was determined by UV spectroscopy at 280 nm sity of Konstanz) in measuring NMR and HR-MS spectra. Gen-
using a molar extinction coefficient of 27 385 M−1 cm−1 87
.
erous funding was received from the Helmholtz-Association
(grant no. VH-NG-934), the Zukunftskolleg and the Deutsche
Forschungsgemeinschaft (grant no. Ti 756/2-1 and HA 5950/1-1).
Direct binding of fluorescent ligands to LecA
The fluorescent ligands 6–9 were dissolved in DMSO to a final
concentration of 3 mg mL−1. A 1 μM dilution stock was pre-
pared in TBS/Ca. Approx. 5 mg LecA was dissolved in 1 mL of
TBS/Ca and incubated with shaking for 2 h at r.t. Afterwards,
the solution was centrifuged for 30 min at 25 000g and
23 °C and the concentration of LecA in the supernatant was
determined by UV absorbance at 280 nm (ε = 27 385 M−1
cm−1). A serial dilution of LecA was added in triplicates to one
384-well plate (Greiner Bio-One, Germany, cat no 781900). The
solution of one fluorescent ligand was added to a final concen-
tration of 10 nM, and after incubation for 1 h at r.t. fluo-
rescence polarization was determined using a PheraStar FS
microplate reader (BMG Labtech GmbH, Germany). The data
were analyzed using a four parameter fit calculated with MARS
Data Analysis Software (BMG Labtech GmbH, Germany). A
minimum of three independent measurements on three plates
was performed for each ligand.
Notes and references
1 H. Lis and N. Sharon, Chem. Rev., 1998, 98(2), 637–674.
2 N. Sharon, FEBS Lett., 1987, 217(2), 145–157.
3 A. Varki, Essentials of glycobiology, Cold Spring Harbor
Laboratory Press, Cold Spring Harbor, N.Y., 2nd edn,
2009.
4 W. A. Petri Jr., R. Haque and B. J. Mann, Annu. Rev. Micro-
biol., 2002, 56, 39–64.
5 J. A. Rodrigues, A. Acosta-Serrano, M. Aebi, M. A. J. Ferguson,
F. H. Routier, I. Schiller, S. Soares, D. Spencer, A. Titz,
I. B. H. Wilson and L. Izquierdo, PLoS Pathog., 2015, 11(11),
e1005169.
6 A. I. Hidron, J. R. Edwards, J. Patel, T. C. Horan,
D. M. Sievert, D. A. Pollock, S. K. Fridkin and National
Healthcare Safety Network Team, Participating National
Healthcare Safety Network Facilities, Infect. Control Hosp.
Epidemiol., 2008, 29(11), 996–1011.
Competitive binding assays
Typically, to 20 μL of a concentrated stock solution of LecA
(final assay concentrations of LecA: 20 µM for 7, or 18 µM for
8, 2–5 µM for 50) and fluorescent ligand (final assay concen-
trations of fluorescent ligands 7, 8, or 50: 5–10 nM) in TBS/Ca
were added 10 µL serial dilutions (10 mM to 128 μM) of test
compounds in TBS/Ca in triplicates in black 384-well micro-
titer plates (Greiner Bio-One, Germany, cat no 781900). After
addition of the reagents, the microtiter plates were centrifuged
at 800 rpm for 1 min at 23 °C and subsequently incubated for
4–6 h at r.t. Fluorescence was measured on a PheraStar FS
plate reader (BMG Labtech GmbH, Germany) or an Infinite
F500 (Tecan Deutschland GmbH, Germany) with excitation
filters at 485 nm and emission filters at 535 nm. The measured
intensities were reduced by the values of only LecA in buffer.
The data were analyzed with MARS Data Analysis Software
(BMG Labtech GmbH, Germany) or Graphpad Prism and fitted
according to the four parameter variable slope model. Bottom
and top plateaus were defined by the standard compounds
included as controls in each assay (methyl α-D-galactoside (2)
and phenyl β-D-galactoside (4), respectively) and the data was
reanalyzed with these values fixed. A minimum of three inde-
pendent measurements on three plates was performed for
each inhibitor.
7 K. G. Kerr and A. M. Snelling, J. Hosp. Infect., 2009, 73(4),
338–344.
8 A. Y. Peleg and D. C. Hooper, N. Engl. J. Med., 2010,
362(19), 1804–1813.
9 K. Poole, Front Microbiol., 2011, 2, 65.
10 H.-C. Flemming and J. Wingender, Nat. Rev. Microbiol.,
2010, 8(9), 623–633.
11 D. Davies, Nat. Rev. Drug Discovery, 2003, 2(2), 114–122.
12 T. Bjarnsholt, O. Ciofu, S. Molin, M. Givskov and N. Høiby,
Nat. Rev. Drug Discovery, 2013, 12(10), 791–808.
13 S. Wagner, R. Sommer, S. Hinsberger, C. Lu,
R. W. Hartmann, M. Empting and A. Titz, J. Med. Chem.,
2016, 59(13), 5929–5969.
14 R. Sommer, I. Joachim, S. Wagner and A. Titz, Chimia,
2013, 67(4), 286–290.
15 D. Tielker, S. Hacker, R. Loris, M. Strathmann,
J. Wingender, S. Wilhelm, F. Rosenau and K.-E. Jaeger,
Microbiology, 2005, 151(Pt 5), 1313–1323.
16 N. Gilboa-Garber, Methods Enzymol., 1982, 83, 378–385.
17 S. P. Diggle, R. E. Stacey, C. Dodd, M. Cámara, P. Williams
and K. Winzer, Environ. Microbiol., 2006, 8(6), 1095–1104.
18 R. Sommer, S. Wagner, A. Varrot, C. M. Nycholat,
A. Khaledi, S. Haussler, J. C. Paulson, A. Imberty and
A. Titz, Chem. Sci., 2016, 7, 4990–5001.
19 A. M. Boukerb, A. Decor, S. Ribun, R. Tabaroni, A. Rousset,
L. Commin, S. Buff, A. Doléans-Jordheim, S. Vidal,
A. Varrot, A. Imberty and B. Cournoyer, Front Microbiol.,
2016, 7, 811.
Acknowledgements
The authors are grateful to Dr Anne Imberty (CERMAV Greno-
ble) for providing the expression plasmid of LecA pET25pa1l
Org. Biomol. Chem.
This journal is © The Royal Society of Chemistry 2016

View Article Online
Organic & Biomolecular Chemistry
Paper
20 C. Chemani, A. Imberty, S. de Bentzmann, M. Pierre,
M. Wimmerová, B. P. Guery and K. Faure, Infect. Immun.,
2009, 77(5), 2065–2075.
21 O. Bajolet-Laudinat, S. Girod-de Bentzmann, J. M. Tournier,
C. Madoulet, M. C. Plotkowski, C. Chippaux and E. Puchelle,
Infect. Immun., 1994, 62(10), 4481–4487.
22 E. C. Adam, B. S. Mitchell, D. U. Schumacher, G. Grant and
U. Schumacher, Am. J. Respir. Crit. Care Med., 1997, 155(6),
2102–2104.
23 M. Mewe, D. Tielker, R. Schönberg, M. Schachner,
K.-E. Jaeger and U. Schumacher, J. Laryngol. Otol., 2005,
119(8), 595–599.
and J.-L. Reymond, Angew. Chem., Int. Ed., 2011, 50(45),
10631–10635.
37 A. M. Boukerb, A. Rousset, N. Galanos, J.-B. Méar,
M. Thepaut, T. Grandjean, E. Gillon, S. Cecioni,
C. Abderrahmen, K. Faure, D. Redelberger, E. Kipnis,
R. Dessein, S. Havet, B. Darblade, S. E. Matthews,
S. de Bentzmann, B. Guéry, B. Cournoyer, A. Imberty and
S. Vidal, J. Med. Chem., 2014, 57(24), 10275–10289.
38 C. Ligeour, O. Vidal, L. Dupin, F. Casoni, E. Gillon,
A. Meyer, S. Vidal, G. Vergoten, J.-M. Lacroix,
E. Souteyrand, A. Imberty, J.-J. Vasseur, Y. Chevolot and
F. Morvan, Org. Biomol. Chem., 2015, 13, 8433–8444.
24 R. S. Laughlin, M. W. Musch, C. J. Hollbrook, F. M. Rocha, 39 D. Hauck, I. Joachim, B. Frommeyer, A. Varrot, B. Philipp,
E. B. Chang and J. C. Alverdy, Ann. Surg., 2000, 232(1),
133–142.
H. M. Möller, A. Imberty, T. E. Exner and A. Titz, ACS
Chem. Biol., 2013, 8(8), 1775–1784.
25 T. Eierhoff, B. Bastian, R. Thuenauer, J. Madl, A. Audfray, 40 R. Sommer, T. E. Exner and A. Titz, PLoS One, 2014, 9(11),
S. Aigal, S. Juillot, G. E. Rydell, S. Müller, S. de Bentzmann, e112822.
A. Imberty, C. Fleck and W. Römer, Proc. Natl. Acad. 41 A. Hofmann, R. Sommer, D. Hauck, J. Stifel, I. Göttker-
Sci. U. S. A., 2014, 111(35), 12895–12900. Schnetmann and A. Titz, Carbohydr. Res., 2015, 412, 34–42.
26 G. Cioci, E. P. Mitchell, C. Gautier, M. Wimmerová, 42 R. Sommer, D. Hauck, A. Varrot, S. Wagner, A. Audfray,
D. Sudakevitz, S. Pérez, N. Gilboa-Garber and A. Imberty,
FEBS Lett., 2003, 555(2), 297–301.
A. Prestel, H. M. Möller, A. Imberty and A. Titz, Chemistry-
Open, 2015, 4(6), 756–767.
27 R. Loris, D. Tielker, K.-E. Jaeger and L. Wyns, J. Mol. Biol., 43 G. Beshr, R. Sommer, D. Hauck, D. C. B. Siebert,
2003, 331(4), 861–870.
28 E. Mitchell, C. Houles, D. Sudakevitz, M. Wimmerova,
A. Hofmann, A. Imberty and A. Titz, MedChemComm, 2016,
7, 519–530.
C. Gautier, S. Pérez, A. M. Wu, N. Gilboa-Garber and 44 J. Rodrigue, G. Ganne, B. Blanchard, C. Saucier,
A. Imberty, Nat. Struct. Biol., 2002, 9(12), 918–921.
D. Giguère, T. C. Shiao, A. Varrot, A. Imberty and R. Roy,
29 E. P. Mitchell, C. Sabin, L. Snajdrová, M. Pokorná,
Org. Biomol. Chem., 2013, 11(40), 6906–6918.
S. Perret, C. Gautier, C. Hofr, N. Gilboa-Garber, J. Koca, 45 R. U. Kadam, D. Garg, J. Schwartz, R. Visini, M. Sattler,
M. Wimmerová and A. Imberty, Proteins, 2005, 58(3), 735–
746.
A. Stocker, T. Darbre and J.-L. Reymond, ACS Chem. Biol.,
2013, 8(9), 1925–1930.
30 P. von Bismarck, R. Schneppenheim and U. Schumacher, 46 S. Cecioni, J.-P. Praly, S. E. Matthews, M. Wimmerová,
Klin. Padiatr., 2001, 213(5), 285–287.
A. Imberty and S. Vidal, Chem. – Eur. J., 2012, 18(20), 6250–
6263.
47 N. Garber, U. Guempel, A. Belz, N. Gilboa-Garber and
R. J. Doyle, Biochim. Biophys. Acta, 1992, 1116(3), 331–333.
31 H.-P. Hauber, M. Schulz, A. Pforte, D. Mack, P. Zabel and
U. Schumacher, Int. J. Med. Sci., 2008, 5(6), 371–376.
32 A. Bernardi, J. Jiménez-Barbero, A. Casnati, C. De Castro,
T. Darbre, F. Fieschi, J. Finne, H. Funken, K.-E. Jaeger, 48 F. Pertici, N. J. de Mol, J. Kemmink and R. J. Pieters, Chem.
M. Lahmann, T. K. Lindhorst, M. Marradi, P. Messner, – Eur. J., 2013, 19(50), 16923–16927.
A. Molinaro, P. V. Murphy, C. Nativi, S. Oscarson, 49 F. Pertici and R. J. Pieters, Chem. Commun., 2012, 48(33),
S. Penadés, F. Peri, R. J. Pieters, O. Renaudet, 4008–4010.
J.-L. Reymond, B. Richichi, J. Rojo, F. Sansone, C. Schäffer, 50 R. Visini, X. Jin, M. Bergmann, G. Michaud, F. Pertici,
W. B. Turnbull, T. Velasco-Torrijos, S. Vidal, S. Vincent,
T. Wennekes, H. Zuilhof and A. Imberty, Chem. Soc. Rev.,
2012, 42(11), 4709–4727.
O. Fu, A. Pukin, T. R. Branson, D. M. E. Thies-Weesie,
J. Kemmink, E. Gillon, A. Imberty, A. Stocker, T. Darbre,
R. J. Pieters and J.-L. Reymond, ACS Chem. Biol., 2015,
10(11), 2455–2462.
33 J.-L. Reymond, M. Bergmann and T. Darbre, Chem. Soc.
Rev., 2013, 42(11), 4814–4822.
34 S. Cecioni, A. Imberty and S. Vidal, Chem. Rev., 2015, 115(1),
525–561.
35 E. M. V. Johansson, S. A. Crusz, E. Kolomiets, L. Buts,
R. U. Kadam, M. Cacciarini, K.-M. Bartels, S. P. Diggle,
51 M. I. Khan, N. Surolia, M. K. Mathew, P. Balaram and
A. Surolia, Eur. J. Biochem., 1981, 115(1), 149–152.
52 G. Jacob, C. Kirmaier, S. Abbas, S. Howard, C. Steininger,
J. Welply and P. Scudder, Biochemistry, 1995, 34(4), 1210–
1217.
M. Cámara, P. Williams, R. Loris, C. Nativi, F. Rosenau, 53 P. Sörme, B. Kahl-Knutsson, M. Huflejt, U. J. Nilsson and
K.-E. Jaeger, T. Darbre and J.-L. Reymond, Chem. Biol.,
2008, 15(12), 1249–1257.
36 R. U. Kadam, M. Bergmann, M. Hurley, D. Garg,
H. Leffler, Anal. Biochem., 2004, 334(1), 36–47.
54 P. Sörme, B. Kahl-Knutson, U. Wellmar, U. J. Nilsson and
H. Leffler, Methods Enzymol., 2003, 362, 504–512.
M. Cacciarini, M. A. Swiderska, C. Nativi, M. Sattler, 55 Z. Han, J. S. Pinkner, B. Ford, R. Obermann, W. Nolan,
A. R. Smyth, P. Williams, M. Cámara, A. Stocker, T. Darbre
S. A. Wildman, D. Hobbs, T. Ellenberger, C. K. Cusumano,
This journal is © The Royal Society of Chemistry 2016
Org. Biomol. Chem.

View Article Online
Paper
Organic & Biomolecular Chemistry
S. J. Hultgren and J. W. Janetka, J. Med. Chem., 2010, 68 S. Dasgupta, V. K. Rajput, B. Roy and B. Mukhopadhyay,
53(12), 4779–4792. J. Carbohydr. Chem., 2007, 26(2), 91–106.
56 S. Kleeb, L. Pang, K. Mayer, D. Eris, A. Sigl, R. C. Preston, 69 C. Grandjean, C. Rommens, H. Gras-Masse and O. Melnyk,
P. Zihlmann, T. Sharpe, R. P. Jakob, D. Abgottspon, Tetrahedron Lett., 1999, 40(40), 7235–7238.
A. S. Hutter, M. Scharenberg, X. Jiang, G. Navarra, 70 L.-X. Wang, US 2005/244424A1, 2005.
S. Rabbani, M. Smiesko, N. Lüdin, J. Bezençon, 71 S. Park and I. Shin, Org. Lett., 2007, 9(9), 1675–1678.
O. Schwardt, T. Maier and B. Ernst, J. Med. Chem., 2015, 72 H. W. Coles, M. L. Dodds and F. H. Bergeim, J. Am. Chem.
58(5), 2221–2239.
57 J. Dahmen, T. Frejd, G. Gronberg, T. Lave, G. Magnusson 73 R. D’Agata, G. Grasso, G. Iacono, G. Spoto and G. Vecchio,
and G. Noori, Carbohydr. Res., 1983, 116(2), 303–307. Org. Biomol. Chem., 2006, 4(4), 610–612.
58 X. Huang and A. Aulabaugh, Methods Mol. Biol., 2009, 565, 74 H. Susaki, K. Suzuki, M. Ikeda, H. Yamada and
Soc., 1938, 60(5), 1020–1022.
127–143.
H. K. Watanabe, Chem. Pharm. Bull., 1994, 42(10), 2090–
59 G. S. Sittampalam, N. P. Coussens, H. Nelson, M. Arkin,
2096.
D. Auld, C. Austin, B. Bejcek, M. Glicksman, J. Inglese, 75 C. K. Chiang, M. McAndrew and R. Barker, Carbohydr. Res.,
P. W. Iversen, Z. Li, J. McGee, O. McManus, L. Minor, 1979, 70(1), 93–102.
A. Napper, J. M. Peltier, T. Riss, O. J. Trask Jr. and 76 R. B. Cohen, K.-C. Tsou, S. H. Rutenburg and A. M. Seligman,
J. Weidner, Assay Guidance Manual, 2012. Available at:
http://www.ncbi.nlm.nih.gov/books/NBK53196/.
60 N. Gilboa-Garber, D. Sudakevitz, M. Sheffi, R. Sela and
C. Levene, Glycoconjugate J., 1994, 11(5), 414–417.
J. Biol. Chem., 1952, 195(1), 239–249.
77 W. F. Goebel and O. T. Avery, J. Exp. Med., 1929, 50(4), 521–
531.
78 M. Apparu, M. Blanc-Muesser, J. Defaye and H. Driguez,
Can. J. Chem., 1981, 59(2), 314–320.
61 S. Perret, C. Sabin, C. Dumon, M. Pokorná, C. Gautier,
O. Galanina, S. Ilia, N. Bovin, M. Nicaise, M. Desmadril, 79 T. Iversen and R. Johansson, Synthesis, 1979, 823–824.
N. Gilboa-Garber, M. Wimmerová, E. P. Mitchell and 80 Z. Csuros, G. Deak and M. Haraszthy-Papp, Acta Chim.
A. Imberty, Biochem. J., 2005, 389(Pt 2), 325–332.
62 K.-C. Kuo, H.-C. Kuo, L.-T. Huang, C.-S. Lin and S.-N. Yang, 81 P. Curcio, C. Zandanel, A. Wagner, C. Mioskowski and
J. Microbiol. Immunol. Infect., 2013, 46(2), 109–114. R. Baati, Macromol. Biosci., 2009, 9(6), 596–604.
63 A. Titz, A. Marra, B. Cutting, M. Smieško, G. Papandreou, 82 R. Caraballo, M. Sakulsombat and O. Ramström, Chem-
Acad. Sci. Hung., 1964, 42(3), 263–267.
A. Dondoni and B. Ernst, Eur. J. Org. Chem., 2012, (28),
5534–5539.
BioChem, 2010, 11(11), 1600–1606.
83 M. Cerný, J. Stanek and J. Pacák, Monatsh. Chem., 1962, 94
64 A. Imberty, C. Breton, R. Oriol, R. Mollicone and S. Pérez,
Adv. Macromol. Carbohydr. Res., 2003, 2, 67–130.
65 B. Blanchard, A. Nurisso, E. Hollville, C. Tétaud, J. Wiels,
(1), 290–294.
84 B. Helferich and D. Türk, Chem. Ber., 1956, 89(10), 2215–
2219.
M. Pokorná, M. Wimmerová, A. Varrot and A. Imberty, 85 L. Wu and N. S. Sampson, ACS Chem. Biol., 2014, 9(2), 468–
J. Mol. Biol., 2008, 383(4), 837–853. 475.
66 D. Ponader, F. Wojcik, F. Beceren-Braun, J. Dernedde and 86 N. Fornstedt and J. Porath, FEBS Lett., 1975, 57(2), 187–191.
L. Hartmann, Biomacromolecules, 2012, 13(6), 1845–1852.
67 H. E. Gottlieb, V. Kotlyar and A. Nudelman, J. Org. Chem.,
1997, 62(21), 7512–7515.
87 M. R. Wilkins, E. Gasteiger, A. Bairoch, J. C. Sanchez,
K. L. Williams, R. D. Appel and D. F. Hochstrasser, Methods
Mol. Biol., 1999, 112, 531–552.
Org. Biomol. Chem.
This journal is © The Royal Society of Chemistry 2016