Amide, amino, hydroxy and aminopyridine groups as building blocks for carbohydrate receptors

Article abstract of DOI:10.1002/ejoc.200701097

Receptors 1-3, incorporating primary amide, hydroxy or amino groups as recognition units used in nature, as well as 2-aminopyridine units as heterocyclic analogues of the asparagine/glutamine primary amide side chains, were prepared and their binding prop

Full text of DOI:10.1002/ejoc.200701097

FULL PAPER
DOI: 10.1002/ejoc.200701097
Amide, Amino, Hydroxy and Aminopyridine Groups as Building Blocks for
Carbohydrate Receptors
Monika Mazik*[a] and Matthias Kuschel[a]
Keywords: Molecular recognition / Receptors / Carbohydrates / Supramolecular chemistry / Hydrogen bonds
Receptors 1–3, incorporating primary amide, hydroxy or
amino groups as recognition units used in nature, as well as
2-aminopyridine units as heterocyclic analogues of the as-
stoichiometries, with an overall β21 binding constant of 107–
108 –2. Both hydrogen bonding and interactions of the sugar
CH groups with the central phenyl rings of the receptors con-
M
paragine/glutamine primary amide side chains, were pre- tribute to the stabilisation of the receptor–sugar complexes.
pared and their binding properties towards neutral sugar
molecules were studied. The design of these receptors was
inspired by the binding motifs observed in the crystal struc-
tures of protein–carbohydrate complexes. The binding stud-
The syntheses, molecular modeling studies and binding
properties of the receptors 1–3 are described, as well as com-
parative binding studies with receptor 4, lacking the third
recognition site.
ies with β-glucopyranoside 5 indicated the formation of com- (© Wiley-VCH Verlag GmbH & Co. KGaA, 69451 Weinheim,
plexes with 1:1 and 2:1 receptor–monosaccharide binding
Germany, 2008)
the receptor–sugar complexes[6,7] (packing of an aromatic
ring from the protein against sugars is observed in most
carbohydrate-binding proteins[3]).
Introduction
The design of artificial carbohydrate receptors operating
through noncovalent interactions remains a subject of in-
tensive current research.[1,2] Consideration of the crystal
structures of protein–carbohydrate complexes[3] provides
much of the inspiration for the development of such recep-
tors. Our previous studies have shown that mimicking of
the binding motifs observed in the crystal structures of pro-
tein–carbohydrate complexes, through the use of natural re-
cognition groups or their analogues,[2e,2i,2q] represents a
powerful strategy for the design of effective and selective
carbohydrate receptors.
The aim of this work was to explore the potential of re-
ceptors 1–3 (Scheme 1) – each containing two different
types of neutral hydrogen-bonding sites – in carbohydrate
recognition; the design of these receptors was inspired by
the binding motifs shown in Figure 1. Receptors 1–3 incor-
porate primary amide, hydroxy or amino groups as recogni-
tion units used in nature, as well as 2-aminopyridine units[4]
as heterocyclic analogues of the asparagine/glutamine pri-
mary amide side chains.[5] As in natural complexes, the par-
ticipation of different types of hydrogen-bonding groups in
the recognition process was expected to be favourable for
achieving high binding affinities and selectivities of the arti-
ficial receptors. Furthermore, the interactions between the
central phenyl rings of the receptors and the sugar CH
groups were expected to provide additional stabilization of
To compare the binding properties of receptors 1–3 with
the properties of the previously published receptors, octyl
β--glucopyranoside (5) and octyl α--glucopyranoside (6)
were selected as substrates for binding studies in organic
media. In addition, the binding properties of receptors 1–3
were compared with the properties of compound 4 (Fig-
ure 2), lacking the third hydrogen-bonding site. X-ray crys-
tallographic data revealed that the hydrogen bonds between
sugar-binding proteins and essential recognition determi-
nants on sugars are shielded from bulk solvent, meaning
that they exist in an environment with a lower dielectric
constant[3] (see also ref.[8]). Thus, investigations with syn-
thetic receptors in a medium with a lower dielectric con-
stant, such as chloroform, make an important contribution
to our understanding of the complex carbohydrate binding
processes in nature.[9]
Results and Discussion
Synthesis of the Receptors
The basis for the synthesis of compounds 1–3 was com-
pound 7 (see Scheme 1), prepared by treatment of 1,3,5-
tris(bromomethyl)-2,4,6-triethylbenzene[10] with 2 equiv. of
2-amino-4,6-dimethylpyridine (see Exp. Sect.). The reaction
of 7 with 1 equiv. of diethyl iminodiacetate provided the
diester 8, which was further transformed into the diamide
1 by treatment with methanolic ammonia. Treatment of 7
with aqueous ammonia gave the amino derivative 2, while
[a] Institut für Organische Chemie der Technischen Universität
Braunschweig,
Hagenring 30, 38106 Braunschweig
Supporting information for this article is available on the
WWW under http://www.eurjoc.org or from the author.
Eur. J. Org. Chem. 2008, 1517–1526
© 2008 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
1517

M. Mazik, M. Kuschel
FULL PAPER
Figure 1. Examples of hydrogen bonds in the complexes of: a) Galanthus nivalis lectin with mannose,[3f,3a] b) -galactose-binding protein
with -glucose,[3b] and c) -arabinose-binding protein with -arabinose.[3b]
Figure 2. Structures of receptors 1–4.
compound 3 was synthesized by treatment of 7 with sodium Binding Studies
hydroxide (see Scheme 1 and Experimental Section). The
synthesis of 4 involves the reaction of 1,3-bis(bromometh-
The interactions of the receptors 1–4 and carbohydrates
yl)-2,4,6-triethylbenzene (9)[10a] with 2 equiv. of 2-amino- 5 and 6 were investigated by 1H NMR spectroscopy in
1
4,6-dimethylpyridine.
[D1]chloroform or in water-containing [D1]chloroform. H
1518
www.eurjoc.org
© 2008 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Org. Chem. 2008, 1517–1526

Building Blocks for Carbohydrate Receptors
Scheme 1. a) HN(CH2COOCH2CH3)2, THF/CH3CN, K2CO3, 24 h, yield 60%; b) NH3/CH3OH, 48 h; yield 39%; c) NH3/H2O (25%
solution), THF/CH3OH, 12 h; yield 72%; d) NaOH aq, THF, 5 h; yield 75%.
NMR titration experiments[11] were carried out by addition found to be 144520 –1 (K11) and 4330 –1 (K21) (β21
=
of increasing amounts of the sugar to a CDCl3 solution 6.25ϫ108 –2; Table 1).[11c] Similar complexation behaviour
of the receptor. In addition, inverse titrations in which the of β-glucopyranoside 5 could also be observed in water-
concentration of sugar was held constant and that of the containing CDCl3 ([receptor]/[water] = 1:5).
receptor was varied were performed. The 1H NMR binding
In addition, the interactions between β-glucopyranoside
titration data were analysed by use of the Hostest 5.6 pro-
gram[12] (stoichiometries of the receptor–sugar complexes
were determined by mole ratio plots and by the curve-fit-
ting analysis of the titration data).
5 and the receptor 1 were investigated on the basis of in-
verse titrations in which the concentration of sugar 5 was
held constant and that of receptor 1 was varied. During the
titration of 5 with 1 the signals due to the OH protons of
5 were shifted downfield with strong broadening and were
unobservable after the addition of about 0.1 equiv. of 1 (in
the case of 6-OH, after the addition of about 0.5 equiv. of
1), indicating important contribution of the OH groups of
5 to the complex formation. The complexation between 5
and the receptor 1 was further evidenced by chemical shift
changes of the CH units of 5 (see Figure 3, c). The signals
due to the 1-, 2-, 3- and 4-CH protons of 5 were shifted
upfield by 0.06, 0.15, 0.20 and 0.25, respectively. The best
fit of the titration data was obtained with the “mixed” 1:1
and 1:2 sugar–receptor binding model (see Figure 4, c);
thus, the inverse titrations supported the existence of 1:1
and 2:1 receptor–sugar complexes in chloroform solution,
with the stronger association constant for 1:1 binding and
a weaker association constant for the 2:1 receptor–sugar
complex. The association constants obtained on the base of
these titrations are identical within the limits of uncertainty
to those determined from titrations in which the roles of
receptor and substrate were reversed.
Binding Properties of Receptor 1
During the titration of 1 with β-glucopyranoside 5 the
signal due to the amide NH of 1 moved downfield by about
1.1 ppm (see Figure 3, a), with saturation occurring after
the addition of about 1 equiv. of 5 (see Figure 4, a). The
amine NHA signal showed very strong broadening and was
unobservable after the addition of only 0.1 equiv. of 5.
1
Furthermore, the H NMR spectra showed changes in the
chemical shifts of the CH3 (protons E, F; for labelling, see
Figure 2), CH2 (protons B, C, D) and pyridine CH reso-
nances of 1. The signal for the protons B were moved up-
field by 0.23 ppm with broadening and splitting (see Fig-
ure 3, b). The splitting of the CH2C signal of 1 was observed
after the addition of about 0.1 equiv. of 5, as shown in part
a of Figure S1 (see Supporting Information), while the sig-
D
nal due to CH2 was shifted downfield by 0.10 ppm with
E,F
splitting (see Figure S1, b). The signals due to the CH3
(see Figure S1, c) and pyridine CH protons were shifted up-
Similarly to the binding studies between 1 and β-gluco-
and downfield in ranges of 0.02–0.06 ppm. The amide NH, pyranoside 5, complexation between 1 and the α anomer 6
B
F
CH2 and CH3 signals were monitored for the determi- was evidenced by several changes in the NMR spectra, as
nation of the binding constants; a typical titration curve is shown in Figure 3 (d) and Figure S2 (see Supporting Infor-
shown in Figure 4 (a). The best fit of the titration data was mation). The signal due to the amide NH of 1 was moved
obtained with the “mixed” 1:1 and 2:1 receptor–sugar bind- downfield by about 1.1 ppm; the amine NHA signal showed
ing model; this model was further supported by mole ratio very strong broadening and was unobservable after the ad-
B
D
plots (see Figure S3a). The binding constants for 1·5 were dition of about 1 equiv. of 5. The CH2 and CH2 (for
Eur. J. Org. Chem. 2008, 1517–1526
© 2008 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjoc.org
1519

M. Mazik, M. Kuschel
FULL PAPER
Figure 3. a, b) Partial 1H NMR spectra (400 MHz, CDCl3) of 1 after addition of (from bottom to top) 0.00–3.03 equiv. of β-glucopyrano-
B
1
side 5 ([1] = 1.00 m). Shown are chemical shifts of the amide NH and CH2 resonances of 1 (for labelling, see formula 1). c) Partial H
NMR spectra (500 MHz. CDCl3) of β-glucopyranoside 5 after addition of 0.00–4.39 equiv. of 1 ([5] = 0.68 m). Shown are chemical
1
shifts of the CH-2, -3, -4 and CH-5 resonances of 5 (as well as shifts of the OCH proton of 5). d) Partial H NMR spectra (400 MHz.
CDCl3) of receptor 1 after addition of 0.00–4.95 equiv. of α-glucopyranoside 6 ([1] = 0.85 m). Shown are chemical shifts of the CH3
E,F
resonances of 1.
Figure 4. a, b) Plot of the observed (ϫ) and calculated (–) chemical shifts of the amide NH signal of 1 (1.05 and 0.85 m) as a function
of added β-glucopyranoside 5 (a) or α-glucopyranoside 6 (b); the [receptor]:[sugar] ratio is marked. c) Chemical shift changes observed
for 1-CH of 5 during the titration of 5 (0.68 m) with 1 (inverse titration); the [sugar]:[receptor] ratio is marked.
1520
www.eurjoc.org
© 2008 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Org. Chem. 2008, 1517–1526

Building Blocks for Carbohydrate Receptors
Table 1. Association constants[a,b] for receptors 1–4 and carbo- described three-armed pyridine-based analogue {1,3,5-tris-
hydrates 5 and 6.
[(4,6-dimethylpyridin-2-yl)aminomethyl]-2,4,6-triethyl-
benzene[4a]}.
[e]
Receptor– K11
K21[c]or β21 = K11K21 or ∆δobs
[d]
substrate
complex
K12
β12 = K11K12
[–1]
[–1]
[–2]
[ppm]
Binding Properties of Receptors 2 and 3
1·5
144520
4330[c]
6.25ϫ108
amide-
NH: 1.12;
The addition of β-glucopyranoside 5 to a solution of the
receptor 2 caused a significant downfield shift of the amine
NHA signal of 2 (see Figure 5, b), as well as changes in the
CH2B: –0.23;
CH3F: 0.06;
1-CH: –0.06;
2-CH: –0.20;
4-CH: –0.25
amide-
E,F
B,C
chemical shifts of the CH3 (see Figure 5, a), CH2
(see
Figure S4) and pyridine CH resonances. The NHA signal
was moved downfield by about 1.9 ppm with broadening,
whereas the NH2 signal of 2 broadened during the titration
and was unobservable after the addition of only 0.1 equiv.
1·6
2·5
24880
39800
1750[d]
1610[c]
4.35ϫ107
6.41ϫ107
NH: 1.06;
CH2B: –0.14;
CH3F: 0.04
NHA: 1.87;
CH2B: –0.18;
CH3E: –0.07;
1-CH: –0.23;
2-CH: –1.75
NHA: 1.21;
CH2B: –0.13;
CH3F: –0.05
NHA: 1.47;
CH2B: –0.17;
CH3E: –0.06;
1-CH: –0.19;
2-CH: –0.96
NHA: 1.22;
CH2B: –0.14;
CH3F: –0.05
NHA: 0.99;
CH2B: –0.13;
CH3E: –0.04
B
of 5. The CH2 signal was shifted upfield by 0.18 ppm,
C
while splitting of the CH2 signal was observed after the
addition of 0.1 equiv. of 5, as shown in Figure S4. Both the
curve fitting of the titration data and the mole ratio plots
suggested the existence of 1:1 and 2:1 receptor–sugar com-
plexes in the chloroform solution, with a stronger associa-
tion constant for 1:1 binding and a weaker association con-
stant for the 2:1 receptor–sugar complex. The association
constants for 2·5 were determined to be 39800 –1 (K11)
and 1610 –1 (K21) (β21 = 6.41ϫ107 –2; Table 1). Similar
complexation behaviour of β-glucopyranoside 5 could be
also observed in water-containing CDCl3 ([receptor]/[water]
= 1:6).
2·6
3·5
2280
18900
2850[c]
5.38ϫ107
3·6
4·5
1840
1330
During the titrations of 3 with β-glucopyranoside 5 the
signal due to the amine NHA of 3 (see Figure 5, d) was
shifted downfield by about 1.5 ppm with strong broaden-
ing, whereas the signals for the CH3E,F, CH2
(see Fig-
B,C
[a] Average Ka values from multiple titrations in CDCl3. [b] Errors
in Ka are less than 10%. [c] K21 corresponds to the 2:1 receptor–
sugar association constant. [d] K12 corresponds to the 1:2 receptor–
sugar association constant. [e] Largest change in chemical shift ob-
served during the titration for NH, CH2 and CH3 signals of the
receptor (down- and upfield shifts; the concentration of the recep-
tor was kept constant and that of the sugar varied), as well as for
the CH groups of the sugar in the case of inverse titrations (upfield
shifts; the concentration of the sugar was kept constant and that
of the receptor varied).
ure 5, c) and pyridine CH protons were moved up- and
downfield in ranges of 0.02–0.17 ppm (Table 1). The analy-
sis of the titration data again indicated the formation of
complexes with 1:1 and 2:1 receptor–sugar stoichiometries;
the binding constants for 3·5 were found to be 18900 –1
(K11) and 2850 –1 (K21) (β21 = 5.38ϫ107 –2).
The interactions between β-glucopyranoside 5 and the
receptors 2 and 3 were also investigated on the basis of
inverse titrations (see Figure 6). During the titrations of 5
labelling, see Figure 2) signals were moved up- and down- with 2 or 3 the signals due to the OH protons of 5 were
field by 0.21 and 0.09 ppm, respectively (see parts a and c shifted downfield with strong broadening and were unob-
in Figure S2, Supporting Information). The splitting of the servable after the addition of about 0.1 equiv. of the recep-
B
C
CH2 and CH2 signals of 1 was observed after the ad- tor (see Figure 6, b and d), indicating important contri-
dition of only 0.1 equiv. of 5, as shown in parts a and b butions of the OH groups of 5 to the complex formation
of Figure S2 (see Supporting Information). In addition, the (similar to the titrations of 5 with 1). The complexation
E
F
CH3 and CH3 signals were shifted up- and downfield by between 5 and the receptor 2 or 3 was further evidenced by
about 0.04 ppm (see Figure 3, d). Curve fitting of the ti- significant chemical shift changes of the CH units of 5 (see
tration data suggested the existence of 1:1 and 1:2 receptor– Figure 6). For example, during the titrations of 5 with 2 the
sugar complexes in the chloroform solution (different bind- signals due to the 1- and the 2-CH protons of 5 were shifted
ing model from that determined for 1·5); a typical titration upfield by 0.23 and 1.75 ppm, respectively. The correspond-
curve is shown in Figure 4 (b). The binding constants for ing titrations of 5 with 3 resulted in upfield shifts of the 1-
1·6 were found to be 24880 –1 (K11) and 1750 –1 (K21) and 2-CH signals of 5 by 0.19 and 0.95 ppm, respectively.
(β21 = 4.35ϫ107 –2). The binding studies with α-gluco- Among the CH signals, the signal due to the 2-CH proton
pyranoside 6 thus showed the interactions of receptor 1 of 5 shows the largest shift, suggesting a particularly impor-
with this monosaccharide to be less favourable than those tant contribution of the CH units to the complex stabilisa-
with β-glucopyranoside 5. However, the affinity of 1 toward tion (through CH···π interactions with the phenyl ring of
the α anomer 6 is much higher than that of the previously the receptor). The participation of the CH units of 5 in
Eur. J. Org. Chem. 2008, 1517–1526
© 2008 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjoc.org
1521

M. Mazik, M. Kuschel
FULL PAPER
Figure 5. a, b) Partial 1H NMR spectra (500 MHz, CDCl3) of 2 after addition of (from bottom to top) 0.00–3.55 equiv. of β-glucopyrano-
E,F
1
side 5 ([2] = 0.99 m). Shown are chemical shifts of the CH3
and NHA resonances of 2 (for labelling, see Figure 2). c, d) Partial H
B,C
NMR spectra of 3 after addition of (from bottom to top) 0.00–3.52 equiv. of 5 ([3] = 0.99 m). Shown are chemical shifts of the CH2
and NHA resonances of 3.
CH···π interactions with the central phenyl ring of the re- 6 chemical shift changes continue to higher [sugar]:[recep-
ceptor was also indicated by molecular modelling (Fig- tor] ratios, indicating lower affinities of 2 and 3 toward 6.
ure 7). In both cases (5·2 and 5·3), the best fits of the ti- During the titrations of 2 and 3 with α-glucopyranoside 6
tration data were obtained with the “mixed” 1:1 and 1:2 the signals due to the amine NHA of the receptors were
sugar–receptor binding model; thus, the inverse titrations moved downfield by about 1.2 ppm (after the addition of
fully confirmed the binding model determined through the about 4.5 equiv. of 6; see Figure S5, a). Furthermore, the
titrations of the receptor 2 or 3 with sugar 5. The associa- 1H NMR titrations of 2 or 3 with 6 produced chemical shift
B,C
tion constants obtained on the base of these titrations are changes in the CH2
(see, for example, b in Figure S5),
E,F
again identical within the limits of uncertainty to those de- CH3
termined from titrations in which the roles of receptor and 0.14 ppm). The curve fitting of the titration data indicated
substrate was reversed. the formation of complexes with 1:1 receptor–sugar stoichi-
and pyridine CH groups (in ranges of 0.03–
Binding studies with α-glucopyranoside 6 showed the in- ometry. The binding model is different from that deter-
teractions of the receptors 2 and 3 with this monosaccha- mined for the receptors 2/3 and β-glucopyranoside 5. The
ride to be less favourable than those with β-glucopyranoside binding constants were found to be 2180 –1and
5. Similarly to the binding studies with sugar 5, the com- 1840 –1 for 2·6 and 3·6, respectively (see Table 1). Thus,
plexation between receptors 2 and 3 and glucopyranoside 6 the complexes formed between the receptors 2/3 and
was evidenced by several changes in the NMR spectra (see, α-glucopyranoside 6 are much less stable than those formed
for example, Figure S5, Supporting Information). However, with the β-glucopyranoside 5. Like receptor 1, compounds
whereas after the addition of about 2 equiv. of β-glucopyr- 2 and 3 show high β vs. α binding selectivity in the recogni-
anoside 5 almost no more change in the chemical shifts of tion of glucopyranosides; however, they show significantly
the receptor signals was observed, with the monosaccharide lower affinity than the receptor 1 for 5 and 6 (see Table 1).
1522
www.eurjoc.org
© 2008 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Org. Chem. 2008, 1517–1526

Building Blocks for Carbohydrate Receptors
Figure 6. a, b) Partial 1H NMR spectra (500 MHz. CDCl3) of β-glucopyranoside 5 after addition of (from bottom to top) 0.00–4.94 equiv.
of 2 ([5] = 0.69 m). c and d) Partial 1H NMR spectra (500 MHz, CDCl3) of β-glucopyranoside 5 after addition of (from bottom to top)
0.00–4.91 equiv. of 3 ([5] = 0.70 m).
It should be also noted that the receptors 2 and 3 exhibit a
level of affinity towards α-glucopyranoside 6 similar to that
shown by the previously described three-armed pyridine-
based analogue[4a]
.
Comparative Binding Studies with Compound 4
In contrast with the strong binding of receptors 1–3 with
β-glucopyranoside 5, the interactions between compound 4,
lacking the third hydrogen-bonding site, and the sugar 5
are, as might be expected, relatively weak. 1H NMR ti-
trations of 4 with 5 produced several spectral changes (see
Figure 8); however, saturation did not occur before the ad-
1
dition of more than 5 equiv. of 5. The H NMR spectra
showed changes in the chemical shifts of the NHA (down-
field shift by 0.99 ppm, after the addition of 7.5 equiv. of
5), CH2B (upfield shift by 0.13 ppm), pyridine CH and CH3
(up- and downfield shifts in ranges of 0.02–0.05 ppm) reso-
nances of 4. The best fit of the titration data was obtained
with the 1:1 binding model; the binding constant for 4·5
was found to be 1330 –1. Thus, the removal of the third
hydrogen-bonding site results in a significant fall in the
binding constant.
Figure 7. Energy-minimized structure of a) the 1:1 complex formed
between receptor 2 and β-glucopyranoside 5, b) the 2:1 receptor–
sugar complex between 2 and 5, c) the 1:1 complex between recep-
tor 3 and β-glucopyranoside 5, and d) the 2:1 receptor–sugar com-
plex formed between 3 and 5 (MacroModel V.8.5, OPLS-AA
forcefield, MCMM, 50000 steps). Color code: receptor C, blue; re-
ceptor N, green; sugar O, red; sugar C,H, grey.
Eur. J. Org. Chem. 2008, 1517–1526
© 2008 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjoc.org
1523

M. Mazik, M. Kuschel
FULL PAPER
Figure 8. a) Partial 1H NMR spectra (500 MHz, CDCl3) of 4 after addition of (from bottom to top) 0.00–7.42 equiv. of β-glucopyranoside
B
5 ([4] = 0.80 m). Shown are chemical shifts of the CH2 and NHA resonances of 4 (for labelling, see Figure 4). b) Plot of the observed
(ϫ) and calculated (–) chemical shifts of the NHA signal of 4 as a function of added 5.
complexes (as also indicated by molecular modelling calcu-
lations; see Figure 7).
Conclusions
Nature’s use of Asn, Gln, Tyr, Ser and Lys to bind carbo-
hydrates, as shown in Figure 1, suggests that combining of
primary amide, amino or hydroxy groups (or analogues of
the natural recognition groups) should lead to effective
carbohydrate receptors. Binding studies with receptors 1–
3, containing such hydrogen-bonding groups, as well as 2-
aminopyridine units as heterocyclic analogues of the aspar-
agine/glutamine primary amide side chains, have confirmed
this hypothesis.
The comparison of the binding properties of the amide-/
aminopyridine-based receptor 1 with those of the pre-
viously described three-armed pyridine-based analogue
{1,3,5-tris[(4,6-dimethylpyridin-2-yl)aminomethyl]-2,4,6-tri-
ethylbenzene}[4a] shows that the incorporation of the two
amide units into the acyclic receptor structure significantly
affects the binding affinity of the new receptor. Receptor 1
has the tendency to form strong 1:1 and 2:1 receptor–sugar
complexes with β-glucopyranoside 5 with an overall bind-
ing constant β21 ≈ 108 –2 (see Table 1). Comparison of the
binding isotherms calculated on the base of the stepwise
binding constants and the analysis of the mole ratio plots
suggested that the 1:1 receptor–monosaccharide complexes
predominate in chloroform solution. The interactions of re-
ceptor 1 with α-glucopyranoside 6 are less favourable than
those with β-glucopyranoside 5; however, the affinity of 1
toward the α anomer 6 (see Table 1) is much higher than the
affinities of the previously described three-armed pyridine-
based receptors (see ref.[4a,4c]).
The removal of the third hydrogen-bonding site results
in a significant fall in the binding constant, as shown by
comparative binding studies with compound 4, which in-
corporates only two aminopyridine groups as hydrogen-
bonding units (see Table 1). The incorporation of the pri-
mary amide, amino or hydroxy group into the receptor
structure thus significantly affects the binding affinities of
the receptors 1–3.
The selective and effective recognition of neutral sugars
by artificial receptors still represents a significant challenge,
so synthetic receptors using noncovalent interactions pro-
vide valuable model systems to study the basic molecular
features of carbohydrate recognition. Both our previous
studies and the present investigations show that the acyclic
receptors represent particularly interesting objects for such
studies. The acyclic scaffold provides simplicity in the syn-
thetic plan for many modifications of the receptor structure,
supplying a base for systematic studies toward recognition
motifs for carbohydrates.
Experimental Section
General: Analytical TLC was carried out on silica gel 60 F254 plates
with ethyl acetate/toluene (3:1, v/v) or chloroform/methanol
(7:1, v/v) as the mobile phase. Melting points are uncorrected.
1-(Bromomethyl)-3,5-bis[(4,6-dimethylpyridin-2-yl)aminomethyl]-
2,4,6-triethylbenzene (7): A CH3CN (10 mL) solution of 2-amino-
4,6-dimethylpyridine (3.16 g, 25.69 mmol) was added to a mixture
of 1,3,5-tris(bromomethyl)-2,4,6-triethylbenzene (6.00 g, 13.60
mmol) and K2CO3 (3.76 g, 27.20 mmol) in CH3CN/THF (1:1, v/v;
40 mL).The mixture was stirred at room temperature for 48 h. Af-
ter filtration and removal of solvents, the crude product was puri-
fied by column chromatography (ethyl acetate/toluene, 1:3, v/v).
Like compound 1, the receptors 2 and 3 show high β vs.
α binding selectivity in the recognition of glucopyranosides.
The binding affinities of the two receptors for the β anomer
5 are high, but significantly lower than that of 1 (see
Table 1). Both hydrogen bonding and interactions of the
sugar CH groups with the central phenyl rings of the recep-
1
Yield 2.13 g (30%); m.p. 78–79 °C. H NMR (400 MHz, CDCl3):
tors contribute to the stabilisation of the receptor–sugar δ = 1.22 (t, J = 7.5 Hz, 3 H), 1.29 (t, J = 7.5 Hz, 6 H), 2.24 (s, 6
1524
www.eurjoc.org
© 2008 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Org. Chem. 2008, 1517–1526

Building Blocks for Carbohydrate Receptors
H), 2.36 (s, 6 H), 2.73 (q, J = 7.5 Hz, 2 H), 2.85 (q, J = 7.5 Hz, 4
H), 4.23 (brs, 2 H), 4.37 (d, J = 4.2 Hz, 4 H), 4.62 (s, 2 H), 6.10
(s, 2 H), 6.35 (s, 2 H) ppm. 13C NMR (100 MHz, CDCl3): δ = 16.4,
16.7, 21.1, 22.8, 23.0, 24.1, 29.6, 40.5, 103.6, 113.9, 131.9, 133.4,
143.8, 144.9, 148.9, 156.5, 158.0 ppm. HR-MS calcd. for
C29H39BrN4: 522.2353; found 522.2360. Rf = 0.12 (ethyl acetate/
toluene, 1:3).
tion (7  in methanol, 10 mL) was added to a solution of com-
pound 8 (400 mg, 0.63 mmol) in methanol (10 mL). The reaction
mixture was stirred for 48 h. During this time a solid precipitated.
The solid was filtered off and was purified by column chromatog-
raphy (CHCl3/CH3OH, 7:1). Yield 0.14 g (39%); m.p. 106–107 °C;
1H NMR (400 MHz, [D6]DMSO): δ = 1.10 (t, J = 7.3 Hz, 6 H),1.16
(t, J = 7.4 Hz, 3 H), 2.09 (s, 6 H), 2.26 (s, 6 H), 2.66 (q, J = 7.4 Hz,
2 H), 2.85 (q, J = 7.3 Hz, 4 H), 3.33 (s, 4 H), 3.74 (s, 2 H), 4.38 (d,
J = 4.0 Hz, 4 H), 6.03 (brs, 2 H), 6.19 (s, 2 H), 6.22 (s, 2 H), 7.09
(s, 2 H), 7.46 (s, 2 H) ppm. 13C NMR (100 MHz, [D6]DMSO): δ =
16.37, 16.43, 20.53, 21.97, 22.56, 23.97, 38.87, 51.36, 56.69, 105.37,
112.18, 130.92, 132.95, 143.11, 143.90, 146.90, 154.99, 158.14,
172.69 ppm. HR-MS calcd. for C33H47N7O2: 573.3785; found
573.3777. Rf = 0.16 (CHCl3/CH3OH, 7:1).
1-(Aminomethyl)-3,5-bis[(4,6-dimethylpyridin-2-yl)aminomethyl]-
2,4,6-triethylbenzene (2): Aqueous ammonia solution (25%, 25 mL)
was added to a solution of 7 (776 mg, 1.48 mmol) in THF/MeOH
(1:1, v/v, 20 mL). This mixture was stirred at room temperature for
12 h. After evaporation of solvents, water (20 mL) was added, and
the solution was extracted with CHCl3 (3ϫ20 mL). The combined
organic phases were washed with brine (20 mL) and dried with
MgSO4, and the solvent was removed under reduced pressure. The
crude product was purified by column chromatography (CHCl3/
MeOH, 7:1, v/v). Yield 0.49 g (72 %); m.p. 82–83 °C; 1H NMR
(400 MHz, CDCl3): δ = 1.22 (t, J = 7.3 Hz, 3 H), 1.23 (t, J =
7.4 Hz, 6 H), 2.23 (s, 6 H), 2.36 (s, 6 H), 2.72 (q, J = 7.3 Hz, 2 H),
2.80 (q, J = 7.4 Hz, 4 H), 3.92 (s, 2 H), 4.36 (s, 6 H), 6.10 (s, 2 H),
6.34 (s, 2 H) ppm. 13C NMR (CDCl3): δ = 16.76, 16.81, 21.15,
22.48, 22.76, 23.97, 39.47, 40.63, 103.61, 113.82, 133.15, 135.02,
142.07, 142.60, 155.22, 158.27 ppm. HR-MS calcd. for C29H41N5:
459.3356; found 459.3352. Rf = 0.17 (CHCl3/CH3OH, 7:1).
1,3-Bis[(4,6-dimethylpyridin-2-yl)aminomethyl]-2,4,6-triethylbenzene
(4): K2CO3 (316 mg, 2.30 mmol) was added to a solution of 1,3-
bis(bromomethyl)-2,4,6-triethylbenzene[10a] (400 mg, 1.15 mmol) in
THF (20 mL) and CH3CN (10 mL). 2-Amino-4,6-dimethylpyridine
(281 mg, 2.30 mmol) dissolved in CH3CN (10 mL) was added to
this suspension. The reaction mixture was stirred at room tempera-
ture for 48 h and was then filtered. After removal of the solvents
the crude product was purified by column chromatography (tolu-
ene/ethyl acetate, 3:1, v/v). Yield 0.38 g (76%); m.p. 157–158 °C. 1H
NMR (400 MHz, CDCl3): δ = 1.21 (t, J = 7.5 Hz, 3 H), 1.23 (t, J
= 7.5 Hz, 6 H), 2.22 (s, 6 H), 2.34 (s, 6 H), 2.70 (q, J = 7.5 Hz, 4
H), 2.76 (q, J = 7.5 Hz, 2 H), 4.15 (t, J = 4.0 Hz, 2 H), 4.36 (d, J
= 4.0 Hz, 4 H), 6.08 (s, 2 H), 6.33 (s, 2 H), 6.99 (s, 1 H) ppm. 13C
NMR (100 MHz, CDCl3): δ = 15.93, 16.92, 21.06, 22.57, 24.17,
26.11, 40.27, 103.39, 113.78, 127.36, 132.00, 143.31, 143.57, 148.67,
156.70, 158.33 ppm. HR-MS calcd. for C28H38N4: 430.3091; found
430.3083. Rf = 0.05 (CHCl3/CH3OH, 7:1).
3,5-Bis[(4,6-dimethylpyridin-2-yl)aminomethyl]-2,4,6-triethyl-1-
(hydroxymethyl)benzene (3): Aqueous NaOH (366.7 mg, 9.15 mmol
in 10 mL of H2O) was added to a solution of 7 (1.2 g, 2.29 mmol)
in THF (20 mL). The solution was heated at reflux for five hours.
After evaporation of the THF, the remaining water phase was ex-
tracted with chloroform (3ϫ20 mL). The combined organic phases
were dried with MgSO4. After evaporation of the solvent, the ob-
tained powder was purified by column chromatography (CHCl3/
CH3OH, 7:1). Yield 0.80 g (75 %); m.p. 75–76 °C. 1H NMR
(400 MHz, [D6]DMSO): δ = 1.13 (t, J = 7.5 Hz, 3 H), 1.16 (t, J =
7.4 Hz, 6 H), 2.08 (s, 6 H), 2.26 (s, 6 H) 2.69 (q, J = 7.5 Hz, 2 H),
2.79 (q, J = 7.4 Hz, 4 H), 4.38 (d, J = 4.1 Hz, 4 H), 4.56 (d, J =
4.2 Hz, 2 H), 4.78 (t, J = 4.2 Hz, 1 H), 5.93 (t, J = 4.1 Hz, 2 H),
6.19 (s, 2 H), 6.21 (s, 2 H) ppm. 13C NMR (100 MHz, [D6]DMSO):
δ = 16.52, 16.61, 20.48, 22.13, 22.50, 24.08, 38.86, 56.97, 105.30,
112.01, 132.86, 135.21, 142.59, 142.83, 146.65, 155.19, 158.29 ppm.
HR-MS calcd. for C29H40N4O: 460.3196; found 460.3187. Rf =
0.17 (CHCl3/CH3OH, 7:1).
Supporting Information (see also the footnote on the first page of
1
this article): H and 13C NMR spectra of compounds 1–4; repre-
1
sentative mole ratio plots; further examples of H NMR titrations.
Acknowledgments
This work was supported by the Deutsche Forschungsgemein-
schaft. We thank Prof. C. S. Wilcox for providing access to the
HOSTEST program.
1-{[Bis(ethoxycarbonylmethyl)]aminomethyl}-3,5-bis[(4,6-dimethyl-
pyridin-2-yl)aminomethyl]-2,4,6-triethylbenzene (8): K2CO3
(212 mg, 1.52 mmol) was added to a solution of compound 7
(800 mg, 1.52 mmol) in THF (15 mL) and CH3CN (15 mL). Di-
ethyl iminodiacetate (0.29 g, 0.34 mL, 1.52 mmol) was added to
this suspension. The reaction mixture was stirred at room tempera-
ture for 24 h and was then filtered. After removal of the solvents
the crude product was purified by column chromatography
(CHCl3/CH3OH, 7:1). Yield 0.58 g (60%); m.p. 52–53 °C; 1H NMR
(400 MHz, CDCl3): δ = 1.18 (t, J = 7.5 Hz, 6 H), 1.23 (t, J =
7.6 Hz, 3 H), 1.26 (t, J = 7.1 Hz, 6 H), 2.23 (s, 6 H), 2.34 (s, 6 H),
2.72 (q, J = 7.6 Hz, 2 H), 2.89 (q, J = 7.5 Hz, 4 H), 3.53 (s, 4 H),
4.01 (s, 2 H), 4.13 (q, J = 7.1 Hz, 4 H), 4.24 (brs, 2 H), 4.35 (d, J =
4.2 Hz, 4 H), 6.31 (s, 2 H), 6.33 (s, 2 H) ppm. 13C NMR (100 MHz,
CDCl3): δ = 14.14, 16.41, 16.68, 21.04, 22.48, 22.80, 24.07, 40.56,
50.98, 53.13, 60.31, 103.20, 113.72, 131.69, 132.77, 143.45, 144.78,
148.70, 156.60, 158.22, 171.36 ppm. HR-MS calcd. for
C37H53N5O4: 631.4092; found 631.4098. Rf = 0.22 (CHCl3/CH3OH
7:1).
[1] For reviews, see: a) A. P. Davis, T. D. James, in Functional Syn-
thetic Receptors (Eds.: T. Schrader, A. D. Hamilton), Wiley-
VCH, Weinheim, Germany, 2005, p. 45–109; b) A. P. Davis,
R. S. Wareham, Angew. Chem. Int. Ed. 1999, 38, 2979–2996; c)
S. Penadés (Ed.), Host–Guest Chemistry – Mimetic Approaches
to Study Carbohydrate Recognition, Top. Curr. Chem., vol. 218,
Springer-Verlag, Berlin, 2002.
[2] For some recent examples of receptors operating through non-
covalent interactions, see: a) Y. Ferrand, M. P. Crump, A. P.
Davis, Science 2007, 318, 619–622; b) E. Klein, Y. Ferrand,
E. K. Auty, A. P. Davis, Chem. Commun. 2007, 2390–2392; c)
M. Mazik, A. C. Buthe, J. Org. Chem. 2007, 72, 8319–8326; d)
M. Mazik, H. Cavga, J. Org. Chem. 2007, 72, 831–838; e) M.
Mazik, A. König, Eur. J. Org. Chem. 2007, 3271–3276; f) M.
Mazik, H. Cavga, Eur. J. Org. Chem. 2007, 3633–3638; g) C.
Schmuck, M. Heller, Org. Biomol. Chem. 2007, 5, 787–791; h)
M. Mazik, A. König, J. Org. Chem. 2006, 71, 7854–7857; i) M.
Mazik, H. Cavga, J. Org. Chem. 2006, 71, 2957–2963; j) M.
Mazik, M. Kuschel, W. Sicking, Org. Lett. 2006, 8, 855–858;
k) O. Francesconi, A. Ienco, G. Moneti, C. Nativi, S. Roelens,
Angew. Chem. 2006, 118, 6845–6848; l) M. Waki, H. Abe, M.
Inouye, Chem. Eur. J. 2006, 12, 7839–7847; m) H. Abe, Y. Aoy-
agi, M. Inouye, Org. Lett. 2005, 7, 59–61; n) E. Klein, M. P.
1-{[Bis(carbamoylmethyl)]aminomethyl}-3,5-bis[(4,6-dimethyl-
pyridin-2-yl)aminomethyl]-2,4,6-triethylbenzene (1): Ammonia solu-
Eur. J. Org. Chem. 2008, 1517–1526
© 2008 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjoc.org
1525

M. Mazik, M. Kuschel
FULL PAPER
Crump, A. P. Davis, Angew. Chem. Int. Ed. 2005, 44, 298–302;
o) M. G. J. Ten Cate, D. N. Reinhoudt, M. Crego-Calama, J.
Org. Chem. 2005, 70, 8443–8453; p) H. Abe, N. Masuda, M.
Waki, M. Inouye, J. Am. Chem. Soc. 2005, 127, 16189–16196;
q) M. Mazik, H. Cavga, P. G. Jones, J. Am. Chem. Soc. 2005,
127, 9045–9052; r) J. M. Benito, M. Meldal, QSAR Comb. Sci.
2004, 23, 117–129; s) G. Gupta, C. R. Lowe, J. Mol. Recog.
2004, 17, 218–235; t) T. Velasco, G. Lecollinet, T. Ryan, A. P.
Davis, Org. Biol. Chem. 2004, 2, 645–647; u) J.-M. Fang, S.
Selvi, J.-H. Liao, Z. Slanina, C.-T. Chen, P.-T. Chou, J. Am.
Chem. Soc. 2004, 126, 645–647; v) R. Welti, Y. Abel, V. Gram-
lich, F. Diederich, Helv. Chim. Acta 2003, 86, 548–562; w) K.
Wada, T. Mizutani, S. Kitagawa, J. Org. Chem. 2003, 68, 5123–
5131; x) M. Segura, B. Bricoli, A. Casnati, E. M. Muñoz, F.
Sansone, R. Ungaro, C. Vicent, J. Org. Chem. 2003, 68, 6296–
6303; y) R. Welti, F. Diederich, Helv. Chim. Acta 2003, 86,
ridge, A. Bernardi, J. Am. Chem. Soc. 2007, 129, 2890–2900;
b) M. I. Chávez, C. Andreu, P. Vidal, N. Aboitiz, F. Freire, P.
Groves, J. L. Asensio, G. Asensio, M. Muraki, F. J. Canˇada, J.
Jiménez-Barbero, Chem. Eur. J. 2005, 11, 7060–7074.
[7] For examples of CH–π interactions in the crystal structures of
the complexes formed between artificial receptors and carbo-
hydrates, see ref.[2q]
.
[8] Many biological interactions occur in enzyme pockets or in
membranes, meaning that they occur in an environment with a
lower dielectric constant than the bulk solvent. For this reason,
many theoretical studies on different enzyme model systems
have been performed in a medium with a lower dielectric con-
stant (mostly ε = 5.7, corresponding to chlorobenzene). See,
for example: a) K.-B. Cho, Y. Moreau, D. Kumar, D. A. Rock,
J. P. Jones, S. Shaik, Chem. Eur. J. 2007, 13, 4103–4115; b) S. P.
de Visser, S. Shaik, P. K. Sharma, D. Kumar, W. Thiel, J. Am.
Chem. Soc. 2003, 125, 15779–15788; c) S. P. de Visser, J. Phys.
Chem. A 2005, 109, 11050–11057.
[9] Recognition of neutral sugars in aqueous solution through
noncovalent interactions remains an important challenge in ar-
tificial receptor chemistry; for some examples, see refs.[2a,2h,2n]
and: a) V. Král, O. Rusin, F. P. Schmidtchen, Org. Lett. 2001,
3, 873–876; b) R. D. Hubbard, S. R. Horner, B. L. Miller, J.
Am. Chem. Soc. 2001, 123, 5810–5811; c) R. Yanagihara, Y.
Aoyama, Tetrahedron Lett. 1994, 35, 9725–9728.
ˇ
ˇ
494–503; z) M. Dukh, D. Saman, K. Lang, V. Pouzar, I. Cerny,
P. Dra sˆar, V. Král, Org. Biomol. Chem. 2003, 1, 3458–3463.
[3] a) H. Lis, N. Sharon, Lectins, Kluwer Academic Publishers,
Dordrecht, The Netherlands, 2003; b) F. A. Quiocho, Pure
Appl. Chem. 1989, 61, 1293–1306; c) W. I. Weiss, K. Drickamer,
Annu. Rev. Biochem. 1996, 65, 441–473; d) R. U. Lemieux,
Chem. Soc. Rev. 1989, 18, 347–374; e) H. Lis, N. Sharon, Chem.
Rev. 1998, 98, 637–674; f) G. Hester, H. Kaku, I. J. Goldstein,
C. S. Wright, Nat. Struct. Biol. 1995, 2, 472–479.
[4] Our previous binding studies showed that 2-aminopyridines
provide an excellent structural motif for binding carbohydrates,
associated with the ability to form cooperative and bidentate
hydrogen bonds with the sugar OH groups. See: a) M. Mazik,
W. Radunz, R. Boese, J. Org. Chem. 2004, 69, 7448–7462; b)
M. Mazik, W. Sicking, Tetrahedron Lett. 2004, 45, 3117–3121;
c) M. Mazik, W. Radunz, W. Sicking, Org. Lett. 2002, 4, 4579–
4582; d) M. Mazik, W. Sicking, Chem. Eur. J. 2001, 7, 664–
670; e) M. Mazik, H. Bandmann, W. Sicking, Angew. Chem.
Int. Ed. 2000, 39, 551–554.
[5] Anslyn and coworkers have exploited the 2-aminopyridine unit
for binding of cyclohexanediols and -triols. See: C. Y. Huang,
L. A. Cabell, E. V. Anslyn, J. Am. Chem. Soc. 1994, 116, 2778–
2792.
[10] a) L. A. Cabell, M. D. Best, J. J. Lavigne, S. E. Schneider, D. M.
Perreault, M.-K. Monahan, E. V. Anslyn, J. Chem. Soc. Perkin
Trans. 2 2001, 315–323; b) K. J. Wallace, R. Hanes, E. Anslyn,
J. Morey, K. V. Kilway, J. Siegel, Synthesis 2005, 2080–2083.
[11] a) The binding constants were determined in chloroform at
25 °C by titration experiments. Dilution experiments show that
receptors do not self-aggregate in the concentration range used.
1
For each system at least three H NMR titrations were carried
out; for each titration 15–20 samples were prepared; b) Error
in a single Ka estimation was less than 10%; c) K11 corresponds
to the 1:1 association constant, K21 corresponds to the 2:1 re-
ceptor–sugar association constant, K12 corresponds to the 1:2
receptor–sugar association constant. β21 = K11 ϫK21, β21
11 ϫK12.
=
K
[6] The character of carbohydrate–aromatic interaction is still a [12] C. S. Wilcox, N. M. Glagovich, Program HOSTEST 5.6, Uni-
subject of controversy; for recent discussions on the impor-
tance of carbohydrate–aromatic interactions, see: a) G. Ter-
raneo, D. Potenza, A. Canales, J. Jiménez-Barbero, K. K. Bald-
versity of Pittsburgh, Pittsburgh, PA, 1994.
Received: November 20, 2007
Published Online: February 5, 2008
1526
www.eurjoc.org
© 2008 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Org. Chem. 2008, 1517–1526