Selectivity of original C-hexopyranosyl calix[4]arene conjugates towards lectins of different origin

Article abstract of DOI:10.1016/j.carres.2018.08.012

As a part of ongoing activities towards the design of ligands against pathogenic lectins, a synthesis of original α-C-galacto/α-C-manno/α-C-fucopyranosyl glycomimetics based on a calix[4]arene scaffold and their binding evaluation is described. The interactions of the glycomimetics with seven lectins of various origins were carried out using agglutination inhibition assays. The 1,3-alternate tetra-C-fucosylated ligand and its derivative having a tertBu group at the upper rim of the calix[4]arene scaffold were the most potent towards the AAL lectin family (RSL, AFL, AAL, AOL) and BC2L-C. As AFL and RSL originate from important human (Aspergillus fumigatus) and plant (Ralstonia solanacearum) pathogens, the inhibition potency of both leading structures was assessed by surface plasmon resonance. With AFL, both structures exhibited an approximately three orders of magnitude increase in affinity compared to the reference L-fucose. The role of tertBu groups as “aglycon-assisted” events was illustrated by NMR. Furthermore, both compounds showed significantly increased ability to inhibit BC2L-C (from human pathogen Burkholderia cenocepacia) cell agglutination and were able to cross-link whole B. cenocepacia cells. Although the ligands failed to significantly inhibit the agglutination activity of LecA and LecB from Pseudomonas aeruginosa, tetra-C-galactosylated calix[4]arene with tertBu groups at the upper rim of the 1,3-alternate conformation inhibited P. aeruginosa biofilm formation efficiently. This systematic and comprehensive study highlights the fact that hydrolytically stable polyvalent C-glycomimetics should be regarded as potent and selective ligands capable of acting as antiadhesive agents.

Full text of DOI:10.1016/j.carres.2018.08.012

Accepted Manuscript
Selectivity of original C-Hexopyranosyl Calix[4]arene conjugates towards lectins of
different origin
Martina Kašáková, Lenka Malinovská, Tomáš Klejch, Martina Hlaváčková, Hana
Dvořáková, Eva Fujdiarová, Zdeňka Rottnerová, Olga Maťátková, Pavel Lhoták,
Michaela Wimmerová, Jitka Moravcová
PII:
S0008-6215(18)30358-6
10.1016/j.carres.2018.08.012
CAR 7598
DOI:
Reference:
To appear in: Carbohydrate Research
Received Date: 19 June 2018
Revised Date: 1 August 2018
Accepted Date: 19 August 2018
Please cite this article as: M. Kašáková, L. Malinovská, Tomáš. Klejch, M. Hlaváčková, H. Dvořáková,
E. Fujdiarová, Zdeň. Rottnerová, O. Maťátková, P. Lhoták, M. Wimmerová, J. Moravcová, Selectivity
of original C-Hexopyranosyl Calix[4]arene conjugates towards lectins of different origin, Carbohydrate
Research (2018), doi: 10.1016/j.carres.2018.08.012.
This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to
our customers we are providing this early version of the manuscript. The manuscript will undergo
copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please
note that during the production process errors may be discovered which could affect the content, and all
legal disclaimers that apply to the journal pertain.

ACCEPTED MANUSCRIPT

ACCEPTED MANUSCRIPT
Selectivity of Original C-Hexopyranosyl Calix[4]arene Conjugates
towards Lectins of Different Origin
Martina Kašákováa, Lenka Malinovskáb, Tomáš Klejchc, Martina Hlaváčkováa, Hana
Dvořákovád, Eva Fujdiarováb,e, Zdeňka Rottnerováf, Olga Maťátkovág, Pavel Lhotákc,
Michaela Wimmerováb,e,h,*, Jitka Moravcováa,*
aDepartment of Chemistry of Natural Compounds, University of Chemistry and Technology,
Technická 5, 166 28 Prague, Czech Republic
bCentral European Institute of Technology, Masaryk University, Kamenice 5, 625 00 Brno,
Czech Republic
cDepartment of Organic Chemistry, University of Chemistry and Technology, Technická 5,
166 28 Prague, Czech Republic
dLaboratory of NMR Spectroscopy, University of Chemistry and Technology, Technická 5,
166 28 Prague, Czech Republic
eNational Centre for Biomolecular Research, Faculty of Science, Masaryk University,
Kotlarska 2, 611 37 Brno, Czech Republic
fLaboratory of Mass Spectrometry, University of Chemistry and Technology, Technická 5,
166 28 Prague, Czech Republic
gDepartment of Biotechnology, University of Chemistry and Technology, Technická 5, 166 28
Prague, Czech Republic
hDepartment of Biochemistry, Faculty of Science, Masaryk University,Kotlarska 2. 611 37
Brno, Czech Republic
*Corresponding authors. Tel.: + 420 220 444 283. E-mail: Jitka.Moravcova@vscht.cz; +420 54949 8166. E-mail:
michaw@chemi.muni.cz
1

ACCEPTED MANUSCRIPT
Abstract
As a part of ongoing activities towards the design of ligands against pathogenic lectins, a
synthesis of original α-C-galacto/α-C-manno/α-C-fucopyranosyl glycomimetics based on a
calix[4]arene scaffold and their binding evaluation is described. The interactions of the
glycomimetics with seven lectins of various origins were carried out using agglutination
inhibition assays. The 1,3-alternate tetra-C-fucosylated ligand and its derivative having a
tertBu group at the upper rim of the calix[4]arene scaffold were the most potent towards the
AAL lectin family (RSL, AFL, AAL, AOL) and BC2L-C. As AFL and RSL originate from
important human (Aspergillus fumigatus) and plant (Ralstonia solanacearum) pathogens, the
inhibition potency of both leading structures was assessed by surface plasmon resonance.
With AFL, both structures exhibited an approximately three orders of magnitude increase in
affinity compared to the reference L-fucose. The role of tertBu groups as “aglycon-assisted”
events was illustrated by NMR. Furthermore, both compounds showed significant increased
ability to inhibit BC2L-C (from human pathogen Burkholderia cenocepacia) cell
agglutination and were able to cross-link whole B. cenocepacia cells. Although the ligands
failed to significantly inhibit the agglutination activity of LecA and LecB from Pseudomonas
aeruginosa, tetra-C-galactosylated calix[4]arene with tertBu groups at the upper rim of the
1,3-alternate conformation inhibited P. aeruginosa biofilm formation efficiently. This
systematic and comprehensive study highlights the fact that hydrolytically stable polyvalent
C-glycomimetics should be regarded as potent and selective ligands capable of acting as
antiadhesive agents
.
Keywords: glycomimetics; lectin; C-glycosides; polyvalency; calix[4]arene.
2

ACCEPTED MANUSCRIPT
1. Introduction
To date, considerable evidence has been accumulated showing that the recognition at the cell
level is accomplished by the interaction of carbohydrate-binding proteins, e.g. lectins, on the
surface of one type of cell with complementary sugars incorporated into glycoconjugates
present on the surface of another cell. Lectins from the pathogens could be important
virulence factors and therefore they are also a suitable therapeutic target [1]. Lectins are
multivalent proteins frequently displaying an avidity effect, resulting in significantly
increased affinity towards their ligands. Consequently, the synthesis of multivalent
glycosylated constructs and their affinity to lectins have been extensively studied over the past
decade [2].
In the synthesis of complex glycomolecules with controlled topology and valency, one of the
major considerations is the choice of appropriate scaffolds. Recently, there has been a
noticeable growth in interest in using calix[n]arene scaffolds [3]. It appears clear that the easy
availability in particular of calix[4]arene, the possibility of controlling their conformational
properties as well as topological presentation of the epitope, and the possibility of
modifications at the lower and/or upper rim make them attractive candidates for polyvalent
neoglycoconjugate synthesis. Moreover, calix[4]arenes are non-toxic and exhibit good
stability. The only limitation is the low solubility of some glycosylated calix[n]arenes in water
that can be dependent on their conformation [3e]. The NMR study of tetragalactosyl
calix[4]arene in a cone conformation indicated a tendency for the aggregation of these
amphiphilic substances in water [4]. Another important consideration in the synthesis of
polyvalent structures is an appropriate selection of a linker between the sugar moiety and
calixarene scaffold, as the simultaneous interactions of a multivalent ligand to several binding
sites of a multivalent receptor has been proven to have a strong impact. Even when the
3

ACCEPTED MANUSCRIPT
topology and the distance between the binding sites are known, the design of a matching
multivalent ligand does not always meet expectations [5]. The flexibility of a polyvalent
structure can be also influenced by the synthetic approach used for the ligation of the sugar
units. In this respect, a Huisgen cycloaddition reaction, later introduced as Cu(I) catalyzed
azide−alkyne cycloaddition (CuAAC) by Sharpless and Meldal [6], seems to occupy a central
position [7].
The affinity of neoglycoconjugates based on a calix[4]arene scaffold can be illustrated by
several examples from the literature. Isomeric galactosyl calix[4]arenes were designed as
potential antiadhesive drugs of the bacterium Pseudomonas aeruginosa [8]. The 1,3-alternate
tetravalent glycocluster was identified as the best inhibitor of the LecA (PA-IL) lectin with a
800-fold increase in affinity compared to a monovalent galactoside as assessed by isothermal
titration calorimetry (ITC). Subsequently, an aggregative chelate binding mode was evidenced
using atomic force microscopy [9]. The use of these neoglycoconjugates for pharmaceutical
compositions was patented [10]. Calix[4]arene-based glycoclusters with four terminal β-D-
galactopyranosyl units were further examined as ligands for LecA using surface plasmon
resonance (SPR). The best affinity in terms of IC50 was found in the range 0.5 - 1.2 µM for
the 1,3-alternate conformation depending on a linker [11].
The overwhelming majority of polyvalent ligands described so far are based on the
connection of sugar units with a linker through a O-glycosidic bond [2,3]. Generally, the O-
glycosidic bonds are unstable to mild acid and to glycosidase enzymes in vivo, thus this fact
can seriously limit their biomedical applications. C-glycosides are attractive mimetics due to
their potential stability towards acids and hydrolytic enzymes. Although the first C-glycosyl
clustering on a calix[4]arene scaffold was reported in 2002 [12] and a 1,2,3-triazole linker
was introduced in 2006 [13], the synthesis and/or biological assessment of any other
polyvalent C-glycopyranosyl constructs has rarely been published so far. A series of
4

ACCEPTED MANUSCRIPT
multivalent β-C-galactopyranosyl mimetics was synthesized and tested towards LecA [14].
The most potent nonavalent construct exhibited dissociation constants below 500 nM,
corresponding to a 400-fold increase in affinity compared to the methyl β-D-
galactopyranoside using ITC.
Based on the results from the literature [14] and our own findings [15], we introduce here
original polyvalent calix[4]arene-based C-glycomimetics with different three-dimensional
shapes. The readily prepared alkyne functionalized calix[4]arenes, being tuneable platforms,
were coupled with 2-(α-hexopyranosyl)ethylazide with
D-manno, D-galacto, and L-fuco
configurations via CuAAC cycloaddition. To explore whether these constructs are adequate
candidates for the interaction with lectins, we decided to perform a broad initial screening
using an agglutination inhibition test. The collection of lectins consisted of BC2L-C
(Burkholderia cenocepacia), LecA (PA-IL) and LecB (PA-IIL) (P. aeruginosa), RSL
(Ralstonia solanacearum), AFL (Aspergillus fumigatus), AOL (Aspergillus oryzae), and AAL
(Aleuria aurantia). The most promising constructs were subjected to surface plasmon
resonance evaluation, in vivo cross-linking tests with whole cells, and their ability to inhibit
biofilm formation was assessed. This approach allowed us to obtain a deep insight into the
selectivity between lectins and to simultaneously identify promising structures for follow-up
optimization.
2. Results and Discussion
2.1.
Synthesis of calix[4]arene polyvalent glycomimetics
The title glycomimetics were synthesized by a highly efficient copper-catalyzed azide-alkyne
cycloaddition (CuAAC) protocol developed previously [15]. Our strategy required the
preparation of alkynyl calix[4]arene I, II, III, and IV (Scheme 1) and 2-(α-
D
-
-
galactopyranosyl)ethylazide (A), 2-(α- -mannopyranosyl)ethylazide (B), and 2-(α-
D
L
5

ACCEPTED MANUSCRIPT
fucopyranosyl)ethylazide (C). The synthesis of calix[4]arenes I, II, and IV was carried out
using a modified version of reported procedures. Thus, the slow addition of propargyl
bromide (3.5 eq.) to a stirred solution of starting tert-butylcalix[4]arene 1 (1 eq.) and K2CO3
(2.4 eq.) in acetone at room temperature, and subsequent reflux of the reaction mixture
overnight gave disubstituted compound I in an 80% yield [16]. Tetrapropargyl derivative II
was obtained by the alkylation of I with propargyl bromide (3.5 eq.) in acetone under reflux,
this time using Cs2CO3 (3 eq.) as a base. Under these conditions, the corresponding 1,3-
alternate conformer II was obtained in a 60% yield [17]. Analogous conditions (excess of
propargyl bromide (16 eq.) and Cs2CO3 (20 eq.) in refluxing acetone) were used for the
preparation of derivative IV starting from calix[4]arene 2. The mixture of the partial cone and
1,3-alternate conformations was separated using column chromatography on silica gel to give
compound IV in a 28% yield, and the corresponding partial cone analogue V in a 41% yield
[18]. The synthesis of the cone conformer III bearing four acetylene units on the upper rim
started with the alkylation of basic calix[4]arene 2 with an excess of n-propyl iodide in DMF
in the presence of sodium hydride (Scheme 1) [19]. Due to the template effect of the sodium
cation, the cone product VI was obtained in an excellent yield (95%) after precipitation with
methanol. Refluxing VI with hexamethylenetetramine in trifluoroacetic acid (Duff reaction)
gave tetraformyl derivative VII in a good yield of 59% after precipitation from methanol [20].
This compound could be transformed into derivative VIII using the so-called Corey-Fuchs
procedure. The reaction with 3 equiv. of dibromomethylene triphenylphosphorane, generated
in situ from Zn, CBr4 and PPh3, led to compound VIII in a 71% yield. The reaction could be
scaled up to 2.0 g of starting compound, as no chromatographic separation was necessary for
the isolation of product VIII, which was simply obtained by recrystallization from a
DCM/MeOH mixture. The bromoalkene VIII was further treated with n-BuLi (8 equiv.) to
6

ACCEPTED MANUSCRIPT
give the ethynyl-substituted calix[4]arene III [21] in the cone conformation in a 53% yield
after precipitation from methanol (Scheme 1).
Scheme 1. Synthesis of propargyl calix[4]arene scaffolds
The conformation of calix[4]arenes was established using the 1H NMR spectra of compounds.
Starting compounds II and IV immobilised in the 1,3-alternate conformation possessed a
typical singlet of methylene bridges at around 3.50
-
3.70 ppm. The
7

ACCEPTED MANUSCRIPT
corresponding cone conformer III exhibited typical doublets of the Ar-CH2-Ar moiety at 4.37
ppm (axial) and 3.10 ppm (equatorial) with a geminal interaction constant 13.5 Hz.
The preparation of pyranosyl azides A, B and C has already been described. Briefly,
galactosyl azide A [22] and mannosyl azide B [15] were prepared from the respective α-C-
allyl glycoside [23] by one-pot ozonolysis and reduction [24] followed by azidation with
phosphoryl azide [22] in an overall yield of 45%. Fucopyranosyl azide C was synthesized
from L-fucose by a sequence of acetylation, C-allylation, ozonolysis and reduction in a one-
pot arrangement, azidation and deacetylation in an overall yield of 23% [15].
The synthesis of all title glycomimetics followed an optimized procedure using CuI as the
catalyst, N,N-diisopropylethylamine (DIPEA) as the base and N,N-dimethylformamide (DMF)
as the solvent. Microwave irradiation was applied to accelerate the reaction rate (Scheme 2).
8

ACCEPTED MANUSCRIPT
Scheme 2. Synthesis of glycomimetics Ia-c, IIa-c, IIIa-c and IVc
All final compounds were purified by preparative HPLC and characterized by mass
spectrometry, optical rotation, 1H and 13C NMR spectrometry. The isolated yields varied from
1
46 to 77%. Interestingly, both H and 13C NMR spectra revealed a doubling of certain
resonances as the result of the diastereotopicity, due to the presence of chiral sugar moieties at
the linker. The fact that the diastereotopicity has been observed not only on methylene protons
9

ACCEPTED MANUSCRIPT
but also on other protons and carbons of the molecule (e.g. cavity aromatic protons H3 and
carbons C2 and C3 and bridge carbons) is unique.
Besides the glycomimetics depicted (Scheme 2), the synthesis of tetravalent glycomimetics
with tetrapropargyloxy calix[4]arene in the 1,2-alternate and cone conformation derivatized at
the lower rim was tried, but all attempts failed. Only inseparable mixtures of partially C-
glycosylated calix[4]arene were obtained. Such failure is known from the literature and is
attributed to the high density of alkyne which promotes the side reaction, the homocoupling of
scaffolded alkynes [25]. Because the preliminary agglutination test indicated that L-fucosyl
glycomimetic IIc could be a promising ligand, a pool of structures was completed with the
additional ligand IVc.
2.2.
We chose a pool of economically and/or medically important lectins from bacteria and fungi
that are selective for -fucose or -mannose or -galactose.
Lectins
L
D
D
The Gram-negative bacterium Pseudomonas aeruginosa is an opportunistic pathogen causing
lethal airway infections in cystic fibrosis and immunosuppressed patients [26]. P. aeruginosa
produces two soluble lectins, LecA (PA-IL) and LecB (PA-IIL), specific for
-fucose, respectively [27]. The crystallographic structures of both lectins revealed a
tetrameric arrangement with a requirement for calcium in the carbohydrate binding site [28].
It has been found that LecB exhibits a clear preference for the α-anomer; methyl α-
fucopyranoside was 770 times more potent a ligand than the β-anomer [29]. Furthermore,
D-galactose and
L
L-
LecB is considered to be
[28c] .
L-fucose-specific but it can also bind
D-mannose with high affinity
10

ACCEPTED MANUSCRIPT
Burkholderia cenocepacia is a Gram-negative bacterium and belongs to the B. cepacia
complex containing at least 18 species [30]. B. cenocepacia causes lung infections in
immunosuppressed patients, especially those suffering from cystic fibrosis [31]. Both of the
above-mentioned pathogens often infect patients simultaneously [32]. Looking at B.
cenocepacia, three soluble lectins were identified and designed as BC2L-A, -B, and -C. Each
of them contains at least one domain with a strong sequence similarity to LecB from P.
aeruginosa [27b]. Lectin BC2L-C has two distinct domains with unique selectivity [33]. The
N-terminal domain is a novel TNF-α-like
L-fucose binding domain, while the C-terminal part
exhibits selectivity for
manner.
D-mannose and
L
-glycero- -manno-heptose in a calcium-dependent
D
Another opportunistic human pathogen associated with cystic fibrosis is Aspergillus
fumigatus, which is also responsible for invasive pulmonary aspergillosis in immunodeficient
patients [34]. It produces a soluble L-fucose-specific lectin AFL, which crystallizes as a
homodimer and contains six non-equivalent binding sites per monomer [35].
Aspergillus oryzae is a fungus widely used in the fermentation industry and its L-fucose-
specific lectin AOL might be involved in A. oryzae-induced allergic responses [36]. The
crystal structure of AOL was recently solved in a complex with seleno α- and β-fucosides
[37]. There are six L-fucose-binding sites in the monomer of AOL, and each of them has a
different carbohydrate binding affinity.
Ralstonia solanacearum is one of the most destructive phytopathogenic bacteria, causing
great agricultural and economic losses in potato and tomato production [38]. Its L-fucose
binding lectin RSL is believed to be responsible for the infection, although its function was
not yet elucidated. However, it could be involved in the adhesion, possibly via binding to the
terminal fucosides of xyloglucans [39]. RSL is a trimer with two almost identical sugar-
11

ACCEPTED MANUSCRIPT
binding sites on each monomer [39]. RSL was found to interact with both the α- and the β-
anomer of
The orange peel fungus Aleuria aurantia possesses an
widely used as a specific probe for -fucose [41]. AAL has been described as a dimer in
solution [42] and also in its crystal structure [43]. The crystal structure of the complex
between AAL and -fucose revealed five -fucose residues bound per monomer [43a], while a
L-fucose as well as
D-arabinose and
L-galactose in the NMR study [40].
L-fucose-specific lectin AAL that is
L
L
L
crystal with a different space group contained only three
[43b].
L-fucose molecules per monomer
2.3.
Agglutination tests
The cone calix[4]arene ligands I displaying only two sugar units at the lower rim were not
soluble enough in water. Even their solution in 50% dimethyl sulfoxide (DMSO) in water had
a tendency to precipitate. Glycomimetics III were soluble in 50% DMSO, while the 1,3-
alternate-based glycomimetics II exhibited a higher water solubility and no aggregation
activity. Glycomimetic IVc was fully soluble in water. The low solubility of calix[4]arene-
based glycoclusters, in particular those in a cone conformation, was already cited in the
literature [8,11,44].
Our initial screening was based purely on a simple and robust agglutination inhibition test,
using red blood cells or yeast if the solubility of ligands was insufficient and DMSO had to be
added. Microscopy was utilised as the detection method as described previously [45]. This
method is quick and the consumption of lectins and inhibitors is significantly lower compared
to classical hemagglutination in a microtiter plate. The lowest concentration of inhibitor able
to completely inhibit hemagglutination/yeast agglutination was determined, and the potency
of the inhibitors was calculated via comparison with a standard monosaccharide.
12

ACCEPTED MANUSCRIPT
The inhibition potency of tetragalactosyl-C-calix[4]arene IIa towards lectin LecA was 16-
times higher than that of
potency of tetrafucosyl-C-calix[4]arene IIc towards LecB was observed. However, even the
monovalent compound C showed twice higher MIC compared to -fucose (Table 1)
suggesting additional interactions with the lectin in comparison to free fucose. For -mannose
D-galactose (Table 1), only a two times increase in inhibition
L
D
and LecB, inhibitory effect of B also increased two times. Unfortunately, the tetramannosyl
ligand IIb was not completely soluble in the buffer at the concentration necessary for the
inhibition of LecB and precipitated. Compounds IIIb and IIIc were not directly soluble in the
working buffer. To test the inhibitory potential of our ligands against the LecB lectin, they
were dissolved in DMSO and MIC was determined via yeast agglutination in the buffer and in
a 50% DMSO environment (Table 2). The corresponding control values were routinely
determined under identical conditions so that the test system was able to answer the question
of whether the sugar moieties in the calix[n]arenes are capable of acting as lectin inhibitors.
As the concentration of inhibitors necessary for the yeast agglutination was lower than in the
hemagglutination experiment, the affinity of ligand IIb was also evaluated in the buffer. The
results revealed a low inhibition potency of all the tested compounds for LecB.
13

ACCEPTED MANUSCRIPT
Table 1. Minimal inhibitory concentration (MIC) for inhibition of hemagglutination caused
by LecA and LecB lectins (P. aeruginosa), and BC2L-C (B. cenocepacia) and potency of
tested ligands (MIC of standard/MIC of ligand)
Lectin
Inhibitor
MIC
(mM)
6.25
Potency Lectin
Inhibitor
MIC
(mM)
50
Potency
LecA
D
-galactosea
1
2
LecB
D
-mannosea
1
2
1
2
2
2
3.125
0.391
50
25
A
B
16
1
L
-fucosea
0.781
0.391
0.391
0.391
IIa
BC2L-C
L
-fucosea
C
25
2
C
IIc
IVc
0.195
1.563
256
32
IIc
IVc
aStandard
Table 2. MIC (minimal inhibitory concentration) values for inhibition of yeast agglutination
caused by LecB and potency of tested ligands (MIC of standard/MIC of ligand)a
Buffer
50% DMSO
Inhibitor
-mannoseb
MIC (mM)
6.25
Potency
1
MIC (mM)
0.391
Potency
1
D
(0.195)c
0.391
3.125
1.5625
n.d.d
2
1
1
B
4
0.391
IIb
IIIb
n.d.
6.25
0.06
(1.5625)e
0.09766
0.09766
0.195
(0.25)e
1
L
-fucoseb
0.195
0.09766
0.195
n.d.
1
2
1
C
1
0.5
0.25
IIc
IIIc
n.d.
0.391
aExperiments were performed in the buffer (20 mM Tris/HCl, 150 mM NaCl, 5 mM CaCl2,
b
c
pH 7.5) and in the mixture buffer : DMSO (1 : 1). Standard. MIC was not clearly
d
e
distinguishable. MIC not detected due to solubility problems. Observed agglutination was
weak until concentration of IIIb dropped to 0.781 mM. The inhibitory effect at the higher
concentration was therefore significant, but MIC was not clearly determined.
14

ACCEPTED MANUSCRIPT
Based on the results, we focused exclusively on the affinity of fucopyranosyl calix[4]arene
constructs IIc and IVc. Four homologous representatives of the fucose-specific AAL lectin
family (AFL, RSL, AAL and AOL) were subjected to hemagglutination tests (Table 3). The
construct IIc exhibited a sizeable affinity to all these fucose-binding lectins and was the best
ligand for RSL compared to the standard. Moreover, the relative potency of IIc decreased in
the order RSL>AFL~AAL>AOL. The affinity of IVc was very similar to that of IIc, giving a
relative potency in the order RSL~AAL>AOL>AFL.
Table 3. MIC (minimal inhibitory concentration) values for inhibition of hemagglutination
caused by AFL, RSL, AAL, and AOL lectins and potency (MIC of standard/MIC of ligand)
of tested ligands IIc and IVc
Inhibitor
MIC (mM)
Potency
MIC (mM)
Potency
Lectin AFL
Lectin RSL
L
-fucosea
12.5
6.25
25
1
2
1.5625
1.5625
3.125
1
1
C
L
-fucosea,b
1
1
0.781
12.5
0.781
32
1
0.04883
0.781
64
1
IIc
L
-fucosea,b
16
0.0122
64
IVc
Lectin AAL
Lectin AOL
L
-fucosea
25
1
4
12.5
6.25
1
2
6.25
C
0.781
0.391
32
64
0.781
0.391
16
32
IIc
IVc
aStandard. bNew batch of lectins was used.
15

ACCEPTED MANUSCRIPT
Finally, the affinity of IIc and IVc towards the lectin BC2L-C (B. cenocepacia) was also
established in the hemagglutination test. An MIC value of 50 mM was determined for L-
fucose, whilst 0.195 mM was the value found for IIc, resulting in a 256-fold affinity increase
in inhibitory potency. An MIC value of 1.563 mM was found for IVc, which corresponds to a
32-fold increase in inhibitory potency.
2.4.
Surface plasmon resonance
As AFL and RSL originate from important human (A. fumigatus) and plant (R. solanacearum)
pathogens, their interactions with calix[4]arene tetravalent glycomimetics IIc and IVc were
further examined in detail by SPR inhibition assay using immobilized α-L-fucopyranoside
(Figure 1, Table 4). The SPR inhibition assay can be considered to be a better model for lectin
binding than agglutination tests, due to the fact that the binding to a complex glycan surface in
the flow-through arrangement mimics interactions with host tissues. The binding potency of
IIc and IVc for AFL was approximately three orders of magnitude higher than that of L-
fucose. With RSL, the increase in the potency was much lower. This observation could reflect
possible differences in the affinity of native ligands and particular functions of lectins in host-
pathogen interactions. These data correspond with the affinities of both lectins, when AFL has
a lower affinity to monosaccharides than RSL, and also suggest that AFL exhibits a higher
avidity (or multivalency) effect than RSL [39,46]. Therefore, compounds IIc and IVc are
suitable candidate inhibitors of AFL, which may be able to compete with natural complex
ligands on host cells.
Table 4. IC50 and potency (IC50 standard/IC50 inhibitor) for fucosyl azide C and construct IIc
and IVc for inhibition of binding of AFL and RSL to immobilized α-L-fucopyranoside
16

ACCEPTED MANUSCRIPT
Inhibitor
Lectin AFL
Lectin RSL
IC50 (µM)
IC50 (µM)
Potency
Potency
1
L
-fucosea
3,110
716
3.0
1
4
12.3
16.1
3.6
0.8
C
1,037
915
3.4
IIc
3.4
2.5
4.9
IVc
aStandard.
Figure 1. Inhibitory effects of monosaccharides and compounds IIc and IVc on RSL (A) and
AFL (B) binding to immobilized α- -fucopyranoside assessed by SPR. The monosaccharide
-galactose, which is not specifically recognized by lectins, was used as a negative control.
L
D
17

ACCEPTED MANUSCRIPT
2.5.
STD NMR
To obtain a deeper insight into the interactions of both IIc and IVc with lectins AFL and RSL,
an STD NMR experiment was performed. STD experiments revealed the interaction of the
nonpolar fucosyl C5-methyl group of both calix[4]arene glycomimetics IIc and IVc with both
types of lectins AFL and RSL, which corresponds with published crystal structures [39,46]. In
addition, the tert-Bu group in IIc behaves as an “aglycon-assisted” moiety, and this
interaction with both lectins is even stronger than the interaction of the fucopyranosyl methyl
group. This was documented by the binding experiment of IIc with RSL, where the ratio of
integral intensities of the C5-methyl group and tert-Bu groups in the STD experiment was
1
3:18, while in the H NMR spectrum it was 3:9 (Figure 2). A slightly weaker STD effect
(3:12) was observed for IIc interacting with AFL. Furthermore, it was found that the methyl
group of
L-fucopyranose of both IIc and IVc interact similarly with both lectins. On the other
hand, no interaction with the aromatic protons of the calix[4]arene scaffold was detected.
18

ACCEPTED MANUSCRIPT
Figure 2. STD NMR screening of ligands IIc and IVc (500 µM) and 5µM RSL; a)
Reference 1H NMR spectrum of IIc; b) STD spectrum of IIc with RSL selectively saturated
1
at 0 Hz; c) Reference H NMR spectrum of IVc; d) STD spectrum of IVc with RSL
selectively saturated at 0 Hz.
2.6.
Cross-linking tests
As the lectin BC2L-C was confirmed to be immunogenic and located on the bacterial surface
[33], inhibitors IIc and IVc, which showed increased inhibitory potency, were further
investigated directly via cross-linking of whole B. cenocepacia cells to determine whether
they could interact with BC2L-C at the cell level. Compound IIc at a concentration of 10 mM
was able to aggregate 5% (w/w) B. cenocepacia cells grown in LB medium. Significantly
19

ACCEPTED MANUSCRIPT
better results were obtained with cells cultivated in minimal M9 medium. Both compounds
IIc and IVc were able to cross-link B. cenocepacia cells in a concentration-dependent manner
with IIc (Figure 3), resulting in an apparently higher aggregation rate than IVc. Cross-linking
and aggregation was not observed using simple monovalent
of 1M. These results are in agreement with the hemagglutination inhibitory assay, where
compound IIc is a 256-fold better inhibitor than -fucose, whereas IVc is only 32-fold better.
L-fucose even at a concentration
L
Figure 3. Cross-linking of Burkholderia cenocepacia cells with compound IIc. B.
cenocepacia cells alone and in presence of -fucose were used as a control. Cells were
L
cultivated in minimal M9 medium. Magnification 100x, bright field, background subtraction
in GIMP, triplicates.
2.7.
Biofilm inhibition
20

ACCEPTED MANUSCRIPT
The ability of the ligands IIa, IIc, and IVc to inhibit the biofilm formation of the reference
strain P. aeruginosa ATCC 10145 was investigated by Microtiter Dish Biofilm Formation
Assay [47].
Figure 4. P. aeruginosa biofilm formation in the presence of compounds IIa and IIc (control
1, cells in medium without inhibitor; control 2, cells in medium with 5% DMSO without
inhibitor)
The data showed that both substances IIa and IIc (Figure 4) were able to inhibit the biofilm
formation of P. aeruginosa in a dose-dependent manner. Over the concentration range 0.05–
0.78 mM, the percentage of biofilm inhibition increased from 9% to 88% for the
galactopyranosyl ligand IIa. Fucopyranosyl ligand IIc also inhibited the biofilm formation,
but to a lesser degree – the inhibition was from 26 % to 70 % over the same concentration
range. In contrast, fucopyranosyl ligand IVc was only a weak inhibitor of biofilm formation.
50% inhibition was achieved with a concentration of 6.24 mM. Therefore, although only a
21

ACCEPTED MANUSCRIPT
weak inhibitory activity of calix[4]arene glycomimetics was observed during agglutination
inhibition assay for LecA and LecB, these compounds could inhibit biofilm formation. The
higher efficiency of galactosylated inhibitors compared to fucosylated suggests that LecA is a
more suitable target for inhibition of biofilm formation than LecB.
3. Conclusions
This article addresses one limitation that can be found in the literature published so far: the
selectivity of ligand-lectin interactions. To address this issue, we synthesized ten new
polyvalent glycomimetics built on calix[4]arene scaffolds using a CuAAC procedure.
D-
Galactose, -mannose and -fucose, being frequently occurring sugar epitopes, were attached
D
L
through potentially stable C-glycosidic bonds. Their affinity towards a pool of seven different
lectins was assessed by agglutination inhibition tests, and the most promising candidates were
subjected to SPR. The biological evaluation was completed with tests at the cell level. We
discovered three promising tetravalent glycomimetics based on calix[4]arene (1,3-alternate):
IIa against LecA (P. aeruginosa), IIc against BC2L-C (B. cenocepacia) and both IIc and IVc
against AFL (A. fumigatus) as well as RSL (R. solanacearum) lectins. Work is currently in
progress on modulating the structural parameters of ligands IIa, IIc and IVc, which could
lead to even more potent candidates for further research, especially at the cell level and in
vivo.
4. Experimental
4.1.
General Methods
Optical rotations were measured with a Autopol VI (Rudolph Research Analytical, USA)
digital polarimeter in appropriate solvents, at temperature 25 ºC and 589 nm sodium line, in 1
1
dm cuvette and are given at 10-1.deg.cm².g-1. Concentrations (c) are given in g/100 ml. H,
22

ACCEPTED MANUSCRIPT
13C, COSY, HMQC and HMBC spectra were measured on a Bruker DPX-300, DRX-400,
DRX-500, or Bruker Advance III 600 (Bruker Corporation, Germany) spectrometer. All
spectra were acquired at 298 K. Chemical shifts are given in δ-units (ppm) and are referenced
to TMS. Coupling constants (J) are reported in Hz. The ESI high resolution mass spectra were
measured with a LTQ Velos Orbitrap XL (Thermo Fisher Scientific, UK) instrument
equipped with LockSpray in ES+ and ES- modes with methanol as a mobile phase at flow rate
150 µL.min-1. Nominal and exact m/z values are reported in Daltons and were locked at
palmitic acid (255.2329) and diisooctyl phthalate mass (413.2662) for ES- and ES+ modes,
respectively. For preparative HPLC, a chromatograph Waters 600 equipped with UV/VIS
detector was used. The instrument was connected to a column (250 x 25 mm) packed with
LiChrosorb RP-18 (5 µm; Merck, Germany).
4.2.
General procedure for the CuAAC reaction
In the optimized procedure, the free azide derivative (1.1 eq. per alkyne), alkynyl
calix[4]arene (1 eq.), CuI (0.5 eq.) were dissolved in DMF. Subsequently N,N-
diisopropylethylamine (DIPEA) (5 eq.) was added. Reaction mixture was heated at 110 °C for
30 min under microwave irradiation. Then the mixture was concentrated and a residue
purified on column chromatography (SiO2, CHCl3 : CH3OH = 4:1) followed by a preparative
HPLC (C18, CH3OH : H2O = 7:3) unless otherwise stated. Numbering of nuclei for NMR
assignment of pseudoglycoconjugates is depicted in Figure 5.
23

ACCEPTED MANUSCRIPT
Figure 5. Numbering of nuclei for NMR shift assignment in synthesised
pseudoglycoconjugates.
4.2.1.
25,27-Bis-{1N-[2-(α-
dihydroxy-5,11,17,23-tetra-tertbutyl-calix[4]arene (cone) (Ia)
Starting from 2-azido-(α- -galactopyranosyl)ethane (A) (110 mg, 0.47 mmol)
D-galactopyranosyl)ethyl]-triazol-4-yl}-methoxy-26,28-
D
pseudoglycoconjugate Ia (131 mg, 53 %) was isolated as a glassy product: [α]D25 21.7 (c 1.9,
24

ACCEPTED MANUSCRIPT
1
MeOH); H NMR (CD3OD): δ 8.30 (s, 2H, triazolyl-H), 7.15 (s, 4H; HA-3), 7.00 (s, 4H; HB-
3), 5.17 (s, 4H; O-CH2), 4.70-4.60 (m, 4H; H-2´), 4.28 (2 x d, J 13.0 Hz, 4H; H bridge ax),
4.10 (ddd, J1,2 5.2 Hz, J1,1´ 10.9 Hz, J1,1´ 4.3 Hz, 2H; H-1), 3.96 (dd, 4H, J3,4 = J4,5 2.8 Hz, 2H;
H-4), 3.92 (dd, J1,2 5.2 Hz, J4,3 8.5 Hz, 2H; H-2), 3.88 (dd, J6,6´ 11.5 Hz, J5,6 7.6 Hz, 2H; H-6),
3.83-3.81 (m, 2H; H-5), 3.73-3.69 (m, 4H; H-3, H-6), 3.36-3.31 (m, 4H; H bridge eq), 2.40-
2.33 (m, 2H; H-1´), 2.31-2.25 (m, 2H; H-1´), 1.32 (s, 18H; tertBu-A), 1.07 (s, 18H; tertBu-
B);13C NMR (CD3OD), δ (ppm): 149.68 (C1-A) 149.44 (C1-B), 147.70 (C4-B), 143.10
(Ctriazol) and 142.04 (C4-A), 133.19 (C2-B), 128.15 (C2-A), 125.56 (C3-B), 125.23 (CHtriazol),
124.82 (C3-A), 73.47 (C-5), 70.98 (C-1), 70.62 (C-3), 68.80 (C-2), 68.72 (CH2-O) 68.66 (C-
4), 60.82 (C-6), 47.19 (C-2´), 33.61, 33.34 (C(CH3)3), 31.14, 31.12 (Cbridge), 30.72, 30.22
(C(CH3)3), 26.51 (C-1´); HRMS/ESI (positive mode): m/z Calcd. for C66H91N6O14 (M+H+):
1191.6588; Found: 1191.6604; Calcd. for C66H90N6NaO14 (M+Na+): 1213.6407; Found:
1213.6418; Calcd. for C66H90KN6O14 (M+K+): 1229.6147; Found: 1229.6082.
4.2.2.
25,27-Bis-{1N-[2-(α-D-mannopyranosyl)ethyl]-triazol-4-yl}-methoxy-26,28-
dihydroxy-5,11,17,23-tetra-tertbutyl-calix[4]arene (cone) (Ib)
Starting from 2-azido-(α-
D
-mannopyranosyl)ethane (B) (76 mg, 0.32 mmol)
25
pseudoglycoconjugate Ib (57 mg, 56 %) was isolated as a glassy product: [α]D 13.4 (c 1.2,
1
MeOH); H NMR (CD3OD): δ 8.35 (s, 2H, triazolyl-H), 7.13 (s, 4H, HA-3), 7.05 (s, 4H, HB-
3), 5.21 (s, 4H, O-CH2), 4.71 (dd, 4H, J 6.0 Hz, J 7.8 Hz, H-2´), 4.27 and 4.25 (2 x d, J 12.9
Hz, 4H, H bridge ax), 3.88-3.86 (m, 2H, H-1), 3.85 (dd, J 11.9 Hz, J5,6 6.8 Hz, 2H, H-6),
6,6
3.79 (dd, J6,6 11.9 Hz, J5,6 2.8 Hz, 2H, H-6) 3.77-3.74 (m, 4H, H-2, H-3), 3.68 (dd, J3,4 7.3 Hz,
2H, H-4), 3.66-3.60 (m, 2H, H-5), 3.35-3.30 (m, 4H, H bridge eq), 2.42-2.36 (m, 2H, H-1´),
2.28-2.22 (m, 2H, H-1´), 1.25 (s, 18H, tertBu-A), 1.08 (s, 18H, tertBu-B); 13C NMR
(CD3OD): δ 149.63 (C1-A) 149.41 (C1-B), 147.79 (C4-B), 143.15 (Ctriazol), 142.09 (C4-A),
133.25 (C2-B), 128.15 and 128.13 (C2-A), 125.61 and 125.58 (C3-B), 125.28 (CHtriazol),
25

ACCEPTED MANUSCRIPT
124.79 (C3-A), 75.85 (C-5), 72.53 (C-1), 71.20 (C-2), 70.77 (C-3), 68.67 (CH2-O), 68.34 (C-
4), 61.18 (C-6), 46.87 (C-2´), 33.63 (C(CH3)3-B), 33.33 (C(CH3)3-A), 31.15 and 31.14
(Cbridge), 30.66 (C(CH3)3-A), 30.20 (C(CH3)3-B), 29.68 (C-1´); HRMS/ESI (positive mode):
m/z Calcd. for C66H90N6NaO14 (M+Na+): 1213.6407; Found: 1213.6409.
4.2.3.
25,27-Bis-{1N-[2-(α-
dihydroxy-5,11,17,23-tetra-tertbutyl-calix[4]arene (cone) (Ic)
Starting from 2-azido-(α- -fucopyranosyl)ethane (C) (66 mg, 0.30 mmol)
L-fucopyranosyl)ethyl]-triazol-4-yl}-methoxy-26,28-
L
pseudoglycoconjugate Ic (107 mg, 67 %) was isolated as a glassy product: [α]D25 -21.6 (c 1.2,
MeOH); 1H NMR (CD3OD): δ 8.24 (s, 2H, triazolyl-H), 7.13 (d, J 2.3 Hz, 2H, HA-3), 7.11 (d,
J 2.3 Hz, 2H, HA-3), 7.03 (d, J 2.2 Hz, 2H, HB-3), 7.02 (d, J 2.2 Hz, 2H, HB-3), 5.19 (d, JHa,Hb
11.8 Hz, 2H, O-CHaHb), 5.15 (d, JHa,Hb 11.8 Hz, 2H, O-CHaHb), 4.52 – 4.65 (m, 4H, H-2´),
4.29 and 4.27 (2 x d, JHax,Heq 12.9 Hz, 4H, H bridge ax), 3.99-3.92 (m, 2H, H-1), 3.95-3.90 (m,
2H, H-2), 3.88-3.83 (m, 2H, H-5) 3.71-3.60 (m, 4H, H-3, H-4), 3.35-3.30 (2x overlapped by
MeOH, 4H, H bridge eq), 2.38-2.35 (m, 2H, H-1´), 2.31-2.29 (m, 2H, H-1´), 1.27 (s, 18H,
tertBu-A), 1.16 (d, J6,5 6.3 Hz, 6H, H-6), 1.06 (s, 18H, tertBu-B); 13C NMR (CD3OD): δ
149.70 (C1-A), 149.44 (C1-B), 147.70 (C4-B), 143.05 (Ctriazol), 142.01 (C4-A), 133.18 (C2-
B), 128.10 and 128.09 (C2-A), 125.59 and 125.53 (C3-B), 124.99 (CHtriazol), 124.83 and
124.81 (C3-A), 72.23 (C-1), 71.14 (C-3), 70.84 (C-4), 68.79 (CH2-O), 68.26 (C-2), 67.81 (C-
5), 47.18 (C-2´) 33.61 and 33.33 (C(CH3)3), 31.13 and 31.10 (Cbridge), 30.71 (C(CH3)3-A),
30.20 (C(CH3)3-B), 25.96 (C-1´), 15.32 (C-6); HRMS/ESI (positive mode): m/z Calcd. for
C66H91N6O12 (M+H+): 1159.6690; Found: 1159.6710 Calcd. for C66H90N6NaO12 (M+Na+):
1181.6509; Found: 1181.6525.
4.2.4.
25,26,27,28-Tetrakis-{1N-[2-(α-D-galactopyranosyl)ethyl]-triazol-4-yl}-
methoxy-5,11,17,23-tetra-tertbutyl-calix[4]arene (1,3-alternate) (IIa)
26

ACCEPTED MANUSCRIPT
Starting from 2-azido-(α-D-galactopyranosyl)ethane (A) (133 mg, 0.57 mmol)
pseudoglycoconjugate IIa (115 mg, 47 %) was isolated as a glassy product. A gradient elution
25
was used for a column chromatography (SiO2, CHCl3 : CH3OH 4:1→2:1): [α]D 64.4 (c 0.8,
MeOH); 1H NMR (CD3OD): δ 7.36 (s, 4H, triazolyl-H), 6.96 (d, J 2.6 Hz, 4H, HA-3), 6.94 (d,
J 2.6 Hz, 4H, HA-3), 4.67 (s, 8H, O-CH2), 4.58-4.50 (m, 8H, H-2´), 3.99-3.97 (m, 8H, H-4, H-
1), 3.92- 3.87(m, 8H, H-6, H-2) 3.83-3.68 (m, 4H, H-5), 3.72-3.67 (m, 8H, H-3, H-6), 3.61
and 3.51 (2 x s, 8H, H bridge), 2.37-2.28 (m, 4H, H-1´), 2.18-2.11 (m, 4H, H-1´), 1.08 (s,
13
36H, tertBu); C NMR (CD3OD), δ (ppm): 153.28 (C-1A), 144.61 (C-4A), 144.36 (Ctriazol),
133.46 and 133.18 (C-2A), 126.41 and 126.31 (C-3A), 124.47 (CHtriazol), 73.86 (C-5), 70.71
(C-3), 70.11 (C-1), 69.06 (C-2), 68.27 (C-4), 63.59 (OCH2), 60.38 (C-6), 46.99 (C-2´), 37.66
and 37.60 (Cbridge), 33.35 (C(CH3)3), 30.73 (C(CH3)3), 27.13 (C-1´); HRMS/ESI (negative
mode): m/z Calcd. for C88H122N12O24 (M-2H+): 865.4353; Found: 865.4345; Calcd. for
C88H123N12O24 (M-H+): 1731.8779; Found: 1731.8774.
4.2.5.
25,26,27,28-Tetrakis-{1N-[2-(α-
methoxy-5,11,17,23-tetra-tertbutyl-calix[4]arene (1,3-alternate) (IIb)
Starting from 2-azido-(α- -mannopyranosyl)ethane (B) (117 mg, 0.50 mmol)
D-mannopyranosyl)ethyl]-triazol-4-yl}-
D
pseudoglycoconjugate IIb (166 mg, 77 %) was isolated as a glassy product. A gradient
elution was used for a column chromatography (SiO2, CHCl3 : CH3OH 4:1→2:1): [α]D25 21.2
1
(c 1.7, MeOH); H NMR (CD3OD), δ (ppm): 7.39 (s, 4H, triazolyl-H), 6.95 (d, J 2.4 Hz, 4H,
HA-3), 6.94 (d, J 2.4 Hz, 4H, HA-3), 4.66 (d, JHa,Hb 12.2 Hz, 4H, O-CHaHb), 4.62 (d, JHa,Hb
12.2 Hz, 4H, O-CHaHb), 4.60-4.50 (m, 8H, H-2´), 3.87-3.82 (m, 8H, H-1, H-6a), 3.78 (dd,
4H, J5,6a 2.8 Hz, J6a,6b 11.9 Hz, H-6b), 3.75-3.72 (m, 8H, H-2, H-3), 3.67 (dd, J3,4 7.3 Hz, J4,5
7.3 Hz, 4H, H-4), 3.61 (s, 4H, H bridge), 3.62-3.55 (m, 4H, H-5), 3.54 (s, 4H, H bridge), 2.40-
2.30 (m, 4H, H-1´), 2.18-2.11 (m, 4H, H-1´), 1.12-1.03 (s, 36H, tertBu); 13C NMR (CD3OD):
δ 153.42 (C-1A), 144.55 and 144.45 (C-4A; Ctriazol), 133.39 and 133.23 (C-2A), 126.37(C-
27

ACCEPTED MANUSCRIPT
3A), 124.39 (CHtriazol), 75.75 (C-5), 72.59 (C-1), 71.25 (C-2), 70.84 (C-3), 68.26 (C-4), 63.72
(OCH2), 61.18 (C-6), 46.62 (C-2´), 37.70 (Cbridge), 33.35 (C(CH3)3), 30.70 (C(CH3)3), 29.79
(C-1´); HRMS/ESI (positive mode): m/z Calcd. for C88H124N12Na2O24 (M+2Na+): 889.4318;
Found: 889.9331; m/z Calcd. for C88H124N12NaO24 (M+Na+): 1755.8744; Found: 1755.8731.
4.2.6.
25,26,27,28-Tetrakis-{1N-[2-(α-L-fucopyranosyl)ethyl]-triazol-4-yl}-
methoxy-5,11,17,23-tetra-tertbutyl-calix[4]arene (1,3-alternate) (IIc)
Starting from 2-azido-(α-l-fucopyranosyl)ethane (C) (110 mg; 0.50 mmol)
pseudoglycoconjugate IIc (136 mg, 68 %) was isolated as a glassy product. A gradient elution
was used for a column chromatography (SiO2, CHCl3 : CH3OH 4:1→2:1): [α]D25 -50.2 (c 1.0,
1
MeOH); H NMR (CD3OD): δ 7.26 (s, 4H, triazolyl-H), 6.95 and 6.97-6.91 (m, 8H, HA-3),
4.67 (s, 8H, OCH2), 4.52-4.43 (m, 8H, H-2´), 4.02-3.98 (m, 4H, H-1), 3.92 (dd, J1,2 7.5 Hz,
J2,3 8.9 Hz, 4H, H-2), 3.88 (dq, J4,5 1.9 Hz, J5,6 6.5 Hz, 4H, H-5), 3.73 (br dd, 4H, H-4), 3.70
(dd, J2,3 8.9 Hz, J3,4 3.4 Hz, 4H, H-3), 3.60 (s, 4H, H bridge), 3.54 (s, 4H, H bridge), 2.38-2.31
(m, 4H, H-1´), 2.28-2.21 (m, 4H, H-1´), 1.25 (d, J5,6 6.5 Hz, 12H, H-6), 1.08 (s, 36H, tertBu);
13C NMR (CD3OD): δ 153.35 (C-1A), 144.65 and 144.30 (C-4A, Ctriazol), 133.42 and 133.27
(C-2A), 126.26 and 126.24 (C-3A), 124.19 (CHtriazol), 71.83 (C-1), 70.93 (C-3, C-4), 68.31
(C-2), 67.93 (C-5), 63.61 (OCH2), 47.18 (C-2´), 37.76 and 37.72 (Cbridge), 33.35 (C(CH3)3)
30.71 (C(CH3)3), 26.26 (C-1´), 15.27 (C-6); HRMS/ESI (positive mode): m/z Calcd. for
C88H124N12Na2O20 (M+2Na+): 857.9420; Found: 857.9431; Calcd. for C88H125N12NaO20
(M+H++Na+): 846.9549; Found: 846.9526; Calcd. for C88H124N12NaO20 (M+Na+): 1691.8947;
Found: 1691.8960.
4.2.7.
5,11,17,23-Tetrakis-{1N-[2-(α,
25,26,27,28-tetrapropoxy-calix[4]arene (cone) (IIIa)
Starting from 2-azido-(α- -galactopyranosyl)ethane (A) (114 mg, 0.49 mmol)
D-galactopyranosyl)ethyl]}-triazol-4-yl-
D
pseudoglycoconjugate IIIa (132 mg, 67 %) was isolated as a glassy product. A gradient
28