Molecular recognition of carbohydrates with acyclic pyridine-based receptors

Article abstract of DOI:10.1021/jo048979k

The recognition capabilities of acyclic pyridine-based receptors toward monosaccharides were evaluated. Aminopyridine receptors based on the 2,4,6-trimethyl- or 2,4,6-triethylbenzene frame show high β vs α binding selectivity in the recognition of glucopyranosides. Amidopyridine receptors, which are sterically less hindered at nitrogen, display high efficiency and an inverse selectivity. The 2-aminopyridine group has been established as a highly effective recognition group in the binding of monosaccharides. The factors influencing the binding properties of receptors 1-15, which differ in the nature and number of binding and spacer subunits used as the buildings blocks, are discussed.

Full text of DOI:10.1021/jo048979k

Molecu la r Recogn ition of Ca r boh yd r a tes w ith Acyclic
P yr id in e-Ba sed Recep tor s
Monika Mazik,*,† Wolfgang Radunz,‡ and Roland Boese§
Institut fu¨r Organische Chemie der Technischen Universita¨t Braunschweig, Hagenring 30,
38106 Braunschweig, Institut fu¨r Organische Chemie and Institut fu¨r Anorganische Chemie der
Universita¨t Duisburg-Essen, Universita¨tsstrasse 5, 45117 Essen, Germany
m.mazik@tu-bs.de
Received J une 17, 2004
The recognition capabilities of acyclic pyridine-based receptors toward monosaccharides were
evaluated. Aminopyridine receptors based on the 2,4,6-trimethyl- or 2,4,6-triethylbenzene frame
show high â vs R binding selectivity in the recognition of glucopyranosides. Amidopyridine receptors,
which are sterically less hindered at nitrogen, display high efficiency and an inverse selectivity.
The 2-aminopyridine group has been established as a highly effective recognition group in the
binding of monosaccharides. The factors influencing the binding properties of receptors 1-15, which
differ in the nature and number of binding and spacer subunits used as the buildings blocks, are
discussed.
In tr od u ction
peutics or chemosensors. On the other hand, the artificial
carbohydrate receptors also provide model systems to
study the basic molecular features of carbohydrate
recognition.
A wide range of biological processes require the inter-
actions of proteins with carbohydrates;1 thus, the molec-
ular understanding of the features associated with these
interactions is of particular importance. The analyses of
the crystal structures of protein-carbohydrate complexes
give some insight into the molecular recognition phe-
nomena. According to these results, the protein-carbo-
hydrate interactions include hydrogen bonding, van der
Waals forces, interactions of sugar CHs with aromatic
residues of the protein (often one or two aromatic residues
stack on the sugar ring), and metal coordination. The
sugar OHs usually participate in cooperative hydrogen
bonds, which result from the simultaneous participation
of OHs as donor and acceptor of hydrogen bonds. The
hydrogen bonds are both direct and water mediated and
are the main factors determining the specificity of
protein-carbohydrate interactions. Carboxylate side
chains of the protein are especially important in binding
of anomers and epimers.2 The protein-carbohydrate
interactions inspire the development of artificial receptor
molecules,3-5 which in the feature may serve as thera-
Recently, we have developed an acyclic receptor system
for the recognition of monosaccharides in chloroform
solutions.5 The effective acyclic receptors (such as 1-3
and 5-7) are able to form both the multiple hydrogen
bonds and the stacking interactions with sugar rings. An
important design criterion is that the receptor structures
contain both donor and acceptor groups which are able
to participate in the cooperative hydrogen bonds with the
carbohydrate OHs, similar to those in the protein-
carbohydrate complexes. The presence of moieties which
are able to take part in additional secondary hydrogen
bonds with sugar hydroxyls (such as 1,8-naphthyridine
groups)5b causes a substantial enhancement of the bind-
ing properties. In addition, structural elements forming
additional van der Waals contacts with the carbohydrate
molecule favor the binding process. Our systematic
studies showed the suitability of the amide-NH/pyridine-
(4) Some recent examples: (a) Velasco, T.; Lecollinet, G.; Ryan, T.;
Davis, A. P. Org. Biol. Chem. 2004, 2, 645-647. (b) Fang, J .-M.; Selvi,
S.; Liao, J .-H.; Slanina, Z.; Chen, C.-T.; Chou, P.-T. J . Am. Chem. Soc.
2004, 2, 645-647. (c) Welti, R.; Abel, Y.; Gramlich, V.; Diederich, F.
Helv. Chim. Acta 2003, 86, 548-562. (d) Welti, R.; Diederich, F. Helv.
Chim. Acta 2003, 86, 494-503. (e) Dukh, M.; Saman, D.; Lang, K.;
Pouzar, V.; Cerny, I.; Drasar, P.; Kra´l, V. Org. Biomol. Chem. 2003, 1,
3458-3463. (f) Ishi-I, T.; Mateos-Timoneda, M. A.; Timmerman, P.;
Crego-Calama, M.; Reinhoudt, D. N.; Shinkai, S. Angew. Chem., Int.
Ed. 2003, 42, 2300-2305. (g) Lecollinet, G.; Dominey, A. P.; Velasco,
T.; Davis, A. P. Angew. Chem., Int. Ed. 2002, 41, 4093-4096. (h)
Tamaru, S.; Shinkai, S.; Khasanov, A. B.; Bell, T. W. Proc. Natl. Acad.
Sci. U.S.A. 2002, 99, 4972-4976. (i) Ladomenou, K.; Bonar-Law, R.
P. Chem. Commun. 2002, 2108-2109.
* To whom correspondence should be addressed. Phone: +49-531-
391-5266. Fax: +49-531-391-8185.
† Institut fu¨r Organische Chemie der Technischen Universita¨t
Braunschweig.
‡ Institut fu¨r Organische Chemie der Universita¨t Duisburg-Essen.
§ Institut fu¨r Anorganische Chemie der Universita¨t Duisburg-Essen.
(1) (a) Osborn, H.; Khan, T. Oligosaccharides. Their synthesis and
biological roles; Oxford University Press: New York, 2000. (b) Lis, H.;
Sharon, N. Chem. Rev. 1998, 98, 637-674.
(2) (a) Weiss, W. I.; Drickamer, K. Annu. Rev. Biochem. 1996, 65,
441-473. (b) Quiocho, F. A. Pure. Appl. Chem. 1989, 61, 1293-1306.
(c) Lemieux, R. U. Chem. Soc. Rev. 1989, 18, 347-374.
(3) (a) Davis, A. P.; Wareham, R. S. Angew. Chem., Int. Ed. 1999,
38, 2978-2996. (b) Hartley, J . H.; J ames, T. D.; Ward, C. J . J . Chem.
Soc., Perkin Trans. 1 2000, 3155-3184. (c) For a recent review on
boronic acid based receptors, see: J ames, T. D.; Shinkai, S. Top. Curr.
Chem. 2002, 218, 159-200.
(5) (a) Mazik, M.; Bandmann, H.; Sicking, W. Angew. Chem., Int.
Ed. 2000, 39, 551-554. (b) Mazik, M.; Sicking, W. Chem. Eur. J . 2001,
7, 664-670. (c) Mazik, M.; Radunz, W.; Sicking, W. Org. Lett. 2002, 4,
4579-4582. (d) Mazik, M.; Sicking, W. Tetrahedron Lett. 2004, 45,
3117-3121.
10.1021/jo048979k CCC: $27.50 © 2004 American Chemical Society
Published on Web 10/02/2004
7448
J . Org. Chem. 2004, 69, 7448-7462

Molecular Recognition of Carbohydrates
CHART 1
N, pyrimidine-N, or naphthyridine-N recognition units
as hydrogen bonding groups for monosaccharides.5a,b,d
Furthermore, the replacement of the pyridine-N/amide-
NH groups by pyridine-N/amine-NH moieties (receptors
5-7) in the receptor structure leads to favorable changes
of the binding affinity and selectivity.5c Interestingly, the
addition of small amounts of water to the chloroform
solutions enhances in some cases the binding affinity of
the receptors, indicating the formation of water-mediated
hydrogen bonds,5c in line with the observation in protein-
carbohydrate complexes.2 The three-dimensional arrange-
ment of the binding sites was shown to be of great
importance.
based receptors 4, 8-12, 14, and 15 toward monosacchar-
ides. We compare their recognition properties with previ-
ously studied receptors and discuss the factors, which
influence the recognition of monosaccharides. Well-chosen
systematic changes in the receptor structure provide
invaluable insights into molecular recognition phenomena
and can lead to an easier development of new carbohy-
drate receptors with favorable recognition properties.
Resu lts a n d Discu ssion
In this study, the binding properties of four series of
pyridine-based receptors were explored and compared:
receptors including three heterocyclic groups attached to
the central phenyl ring via the -NHCO- unit (1-4), re-
ceptors based on the 2,4,6-trimethyl- or triethylbenzene
In this paper we describe the syntheses, crystal struc-
tures, and binding properties of new acyclic pyridine-
J . Org. Chem, Vol. 69, No. 22, 2004 7449

Mazik et al.
a
TABLE 1. Associa tion Con sta n ts Ka (M-1) a n d Cor r esp on d in g F r ee En er gy Ch a n ges ∆G° (k J m ol-1) for Recep tor s 2, 3,
a n d 6-9 a n d Glu cop yr a n osid es 16-18
d
host-guest complex
Ka1 (∆G°)
Ka2 (∆G°)
∆δmax (∆δobs)e (ppm)
2‚16i
2‚17i
2‚18
3‚16i
3‚17i
3‚18
6‚16j
6‚17j
6‚18
7‚16j
7‚17j
8‚16
8‚17
8‚18
9‚16
9‚17
9‚18
660 (-16.1)
3640 (-20.3)
420 (-14.9)
440 (-15.1)
2100 (-18.9)
300 (-14.1)
20950 (-24.7)
800 (-16.6)
1360 (-17.9)
9500 (-22.7)
620 (-15.9)
48630 (-26.7)
1310 (-17.8)
3070 (-19.9)
19590 (-24.5)
1100 (-17.4)
8410 (-22.4)
24200 (-25.0)b
82450 (-28.0)b
50770 (-26.8)b
22600 (-24.8)b
47600 (-26.7)b
29350 (-25.5)b
790 (-16.5)c
NH:f 0.85 (0.85); CH:g 0.18 (0.18)
NH:f 1.10(1.09); CH:g 0.56 (0.55)
NH:f 0.95(0.92); CH:g 0.15 (0.15)
NH:f 0.80(0.80); CH:g 0.17 (0.15)
NH:f 1.10(1.08); CH:g 0.50 (0.48)
NH:f 0.80(0.75); CH:g 0.15 (0.12)
NH:h 1.28 (1.23)
NH:h 1.45 (1.14)
211 (-13.3)c
NH:h 1.20 (0.97)
4800 (-21.0)b
NH:h 1.20 (1.10)
NH:h 1.50 (0.98)
1320 (-17.8)c
NH:h 1.35 (1.31)
NH:h 1.44 (1.16)
470 (-15.2)c
NH:h 1.33 (1.15)
14490 (-23.7)b
NH:h 1.28 (1.16)
NH:h 1.35 (1.02)
8680 (-22.5)b
NH:h 1.29 (1.04)
a
Average Ka values from multiple titrations (CDCl3, stored over activated molecular sieves and deacidified with Al2O3). The
reproducibility of the Ka values was (10-20%. Uncertainty in a single Ka estimation was (2-10%. Dilution experiments show that
b
receptors do not self-aggregate in the used concentration range. 2:1 Receptor/pyranoside complex. c 1:2 Receptor/pyranoside complex.
d
Change in chemical shift at saturation binding, values provided by HOSTEST. e Largest change in chemical shift observed during the
titration. f Complexation-induced shifts observed for the amide-NH of receptor (the concentration of receptor was kept constant and that
of sugar varied). Complexation-induced shifts observed for the anomeric CH of sugar (the concentration of sugar was kept constant and
the host concentration varied). Complexation-induced shifts observed for the amine-NH (CH2NH) of receptor (the concentration of receptor
was kept constant and that of sugar varied). Results from ref 5d. Results from ref 5c.
g
h
i
j
a
TABLE 2. Associa tion Con sta n ts Ka (M-1) a n d
Cor r esp on d in g F r ee En er gy Ch a n ges ∆G° (k J m ol-1) for
Recep tor s 1, 4, 10-12, 14, a n d 15 a n d
frame with three pyridine moieties attached via
-NHCH2- (5-9), or -NHCOCH2- spacer (10 and 11)
and receptors incorporating only two heterocyclic groups
(12-15). The interactions of the receptors and monosac-
charides were investigated by 1H NMR spectroscopy,
microcalorimetry, and extraction experiments. Octyl â-D-
and R-D-glucopyranosides (16 and 17), octyl â-D-galacto-
pyranoside (18), as well as methyl R- and â-glucopyra-
nosides (16a , 17a ) were selected as substrates for the
Octyl-â-D-glu cop yr a n osid e (16)
receptor
Ka1 (∆G°)
Ka2 (∆G°)
∆δmaxc (∆δobs)d (ppm)
1e
4
10
11
12
14
15
8700 (-22.5)
990 (-17.1)
650 (-16.0)
1230 (-17.6)
0.70 (0.65)
0.60 (0.39)
0.47 (0.14)
0.25 (0.19)
0.69 (0.66)
0.77 (0.50)
0.36 (0.15)
1420 (-18.0) 3890 (-20.5)b
950 (-17.0)
1
receptors. The H NMR titration was the main method
190 (-12.9)
used in this work for the determination of the association
constants. The titration data were analyzed by nonlinear
regression analysis using the HOSTEST 5.6 program.6a
The stoichiometry of receptor-sugar complexes was
established by the curve-fitting analysis of the titration
data and by mole ratio plots.7 The binding constants6b
are given in Tables 1 and 2.
a
See Table 1. b 2:1 Receptor/glucopyranoside complex. c Change
in chemical shift at saturation binding, values provided by
d
HOSTEST. Largest change in chemical shift observed for the
amide-NH of receptor during the titration. e Results from ref 5a.
dramatically. The pyridine units in 2 and 3 are sterically
less hindered at nitrogen (free pyridine R-position) and
can be involved more easily and effectively in the binding
interactions. Consequently, very strong 2:1 receptor/
sugar complexes are formed between the receptors 2 and
3 and glucopyranosides 16 and 17.5d Compared to 1,
receptors 2 and 3 show not only higher efficiency and
different binding modes but also an inverse selectivity
since they bind the R-glucopyranoside better than the
â-anomer. These R/â-anomer selectivities differ from
those observed for the hydrogen-bonding host molecules
described so far, which usually show higher affinity
toward the â-anomer. One reason for this tendency is the
formation of different intramolecular hydrogen bonds in
the two anomers.4h,8 The axial 1-alkoxy group in R-ano-
mer 17 can form intramolecular hydrogen bonds with the
2-OH group more easily than the equatorial 1-alkoxy
substituent in the â-anomer can do. Therefore, the 2-OH
in â-glucopyranoside 16 is relatively free from intramo-
Recep tor s 1-4. The binding studies with receptors
1-4 showed that the steric interactions involving the
pyridine substituents significantly affect the binding
properties. Even a minimal structural variation can lead
to remarkable changes of the receptor properties, as
shown with receptors 2 and 3. These two receptors,
compared to the receptor 1, reflect only the change from
R,R- to the R,â- or R,γ-disubstituted pyridine groups;
however, the binding properties of 2 and 3 change
(6) (a) Wilcox, C. S.; Glagovich, N. M. Program HOSTEST 5.6,
University of Pittsburgh. This program is designed to fit data to
different binding models, which include both “pure” binding models,
taking into consideration the formation of only one type of complex in
solution, and “mixed” binding models containing more than one type
of complex in solution. (b) Ka1 corresponds to the 1:1 association
constant. Ka2 corresponds to the 1:2 or 2:1 receptor/sugar association
constant (for details see the text and Tables 1-3).
(7) Schneider, H.-J .; Yatsimirsky, A. Principles and Methods in
Supramolecular Chemistry; J ohn Wiley & Sons: Chichester, 2000; p
148. (b) Tsukube, H.; Furuta, H.; Odani, A.; Takeda, Y.; Kudo, Y.;
Inoue, Y.; Liu, Y.; Sakamoto, H.; Kimura, K. In Comprehensive
Supramolecular Chemistry; Atwood, J . L., Davis, J . E. D., MacNicol,
D. D., Vo¨gtle, F., Eds.; Pergamon: Oxford, UK, 1996; Vol. 8, p 425.
(8) Cuntze, J .; Owens, L.; Alcazar, V.; Seiler, P.; Diederich, F. Helv.
Chim. Acta 1995, 78, 367-390.
7450 J . Org. Chem., Vol. 69, No. 22, 2004

Molecular Recognition of Carbohydrates
hydrogen bonds. The evidence for the particularly strong
complexation of glucopyranosides with the receptors 2
and 3 was obtained by NMR spectroscopy5d and extrac-
tion experiments. The extraction experiments with meth-
yl glucopyranosides 16a and 17a also provided an
indication for the formation of 2:1 receptor/glucopyrano-
side complexes. For example, the receptor 2 is able to
extract 0.5 equiv of methyl R-D-glucopyranoside (17a )
from the solid state into a CDCl3 solution (Figure 1), the
anomeric CH and the OCH3 protons were integrated with
respect to the host’s proton signals to give the sugar/
receptor ratio.
Furthermore, the titration experiments performed with
receptors 2/3 and octyl â-D-galactopyranoside (18) also
revealed the formation of strong 2:1 receptor/sugar
complexes. The 1H NMR titration experiments were
carried out by adding increasing amounts of the sugar
18 to a solution of the host 2 or 3. The complexation was
evidenced by the downfield shifts of the receptor amide
protons (∆δmax ) 0.95 and 0.80 ppm for 2 and 3,
respectively), reflecting the formation of hydrogen-bonded
complexes, and by the upfield shifts of the phenyl protons
(∆δmax ) -0.17 ppm for both receptors), as illustrated in
Figure 2a for 2‚18. The plot of the observed and calcu-
lated downfield chemical shifts of the NH resonances of
2 as a function of added sugar 18 is shown in Figure 2b.
Both the fitting the binding isotherms and the molar ratio
plots indicated that hosts 2 and 3 form 2:1 receptor/sugar
complexes with 18. Moreover, additional inverse titra-
tions were performed in which the concentration of
galactopyranoside 18 was held constant and that of
receptor 2 or 3 varied. The 1H NMR spectra obtained
during binding experiments showed large downfield
shifting of the sugar hydroxyl resonances, indicating
hydrogen bond formation; however, the strong broaden-
ing of these resonances prevented their use in the
estimation of the binding constants. For this reason, the
motion of the signal due to anomeric CH proton of 18,
which is shifted downfield by 0.15 ppm (Figure 3a), was
F IGURE 1. 1H NMR spectra showing the NH and the
aromatic protons of receptor 2 before (bottom) and after (top)
the extraction of solid methyl-R-glucopyranoside (17a ) by a
CDCl3 solution of receptor 2 (1.38 × 10-3 mol/l). The anomeric
CH of sugar 17a is marked (top).
lecular hydrogen bonding and can interact with receptor
molecules more strongly. The particularly strong binding
of R-anomer 17 with 2 or 3 indicates that the intermo-
lecular host-guest interactions compete effectively with
the intramolecular H-bonding network in carbohydrates.
As indicated by molecular modeling the R-anomer is
particularly favorable positioned in the cavity between
the two receptors in 2:1 receptor/sugar complexes. These
complexes are stabilized by hydrogen bonds between
sugar OHs and the amide-NH/pyridine-N of 2 or 3
(including cooperative hydrogen bonds OH‚‚‚N-pyr,
HO‚‚‚HN-amide), interactions of sugar CH moieties with
the phenyl and pyridine groups of the receptors (both
sides of the pyranoside ring are involved in the stacking
interactions with aromatic residues of the two receptor
molecules) as well as by weak CH‚‚‚N-pyr and CH‚‚‚OdC
1
F IGURE 2. Titration of receptor 2 (1.32 mM) with octyl â-D-galactopyranoside (18). (a) H NMR spectra (CDCl3, 25 °C) of the
receptor 2 (NH and CHPh resonances are shown) after addition of (from bottom to top) 0, 0.15, 0.23, 0.31, 0.39, 0.47, 0.63, 0.78,
1.10, 1.41, 1.96, and 2.35 equiv of 18. (b) Plot of the observed (+) and calculated (s) downfield chemical shifts of the δN-H resonances
of 2 as a function of added 18. The [receptor]/[galactopyranoside] ratio is marked.
J . Org. Chem, Vol. 69, No. 22, 2004 7451

Mazik et al.
1
F IGURE 3. Titration of 18 (0.57 mM) with receptor 2. (a) H NMR spectra (CDCl3, 25 °C) of 18 (anomeric CH resonances are
shown) after addition of (from bottom to top) 0, 0.19, 0.38, 0.58, 0.77, 0.96, 1.16, 1.35, 1.54, 1.74, 1.93, 2.22, 2.51, 2.80, 3.09, 3.38,
3.87, 4.35, 4.84, and 5.80 equiv of 2. (b) The corresponding binding isotherm. The [galactopyranoside]/[receptor] ratio is marked.
monitored. The chemical shift of the anomeric CH proton
of â-galactopyranoside 18 during the titration with 2 or
3 is comparable to that of â-glucopyranoside 16, which
was affected significantly weakly (∆δmax ) 0.15-0.18
ppm) by complexation than the CH-(1) of R-glucopyra-
noside 17 (∆δmax ) 0.50-0.56 ppm). These results
indicate an important contribution of the anomeric CH
of 17 to the stability of the complexes 2‚17 and 3‚17. The
addition of 2 equiv of receptor 2 or 3 led to practically
complete complexation of 18. The typical titration curve
is shown in Figure 3b. The best fit of the titration data
was obtained with the “mixed” 1:1 and 2:1 receptor/sugar
binding models. The association constants of 420 (Ka1)
and 50770 M-1 (Ka2) were determined for 2‚18 (Table 1),
whereas the binding constants for 3‚18 amount to 300
(Ka1) and 29340 M-1 (Ka2). Thus, among the tested sugars,
the receptors 2 and 3 showed the highest affinity toward
R-glucopyranoside 17. The stronger binding of 2 in
comparison to 3 can be attributed both to the different
basicity of the pyridine units9 and to different sterical
effects.
When the R-methyl group in 1 was replaced by an ethyl
unit (compound 4), the expected decrease in the binding
affinity toward pyranosides was observed due to the
stronger steric hindrance from the R-position of the
pyridine ring. Compound 4 was synthesized from benzene-
1,3,5-tricarbonyl chloride and 2-amino-6-ethylpyridine.
The â-glucopyranoside 16 acted as probe for the binding
studies. In contrast to 1, the amide proton of 4 showed
lower displacement on complexation with the glucopy-
ranoside (after the addition of 5 equiv of sugar the NH
protons shifted downfield by only 0.39 ppm; see Table
2). The binding constant was found to be 990 M-1; i.e.,
the receptor 4 exhibited a 9-fold lower affinity for 16 than
the receptor 1. Thus, the higher the degree of steric
hindrance, the less available is the pyridine nitrogen
atom for hydrogen bond formation.
R-position of the pyridine ring. The pyridine nitrogen
atom does not participate in the typical NH‚‚N-pyr
hydrogen bonds,11 but the molecules of 4 prefer the
formation of CH‚‚‚O hydrogen bonds between the 5-CH
of the pyridine ring and the carbonyl oxygen of the
neighboring molecule (Figure 4a). In contrast to 4, the
molecules of 2 interact via NH‚‚N-pyr hydrogen bonds,
as shown in the Figure 4b (the crystals were obtained
from THF/hexane solutions). The crystals of 3 were grown
from ethanol solutions, and the ethanol molecules are
hydrogen bonded to the receptor molecules. As illustrated
in Figure 4c, the hydroxyl groups of the ethanol molecules
participate in cooperative hydrogen bonds with the
amide-NH and the pyridine nitrogen atom of the receptor
molecule. Between the receptor molecules the CH‚‚‚O
hydrogen bonds are observed.
Recep tor s 5-9. The replacement of the amidopyridine
groups by aminopyridine moieties and the incorporation
of methyl groups into the central phenyl ring provides
receptors (formulas 5-7) which show remarkable â/R
binding selectivity in the recognition of glucopyranosides.5c
The aminomethyl group attached to the pyridine unit as
well as a second methyl group at the 4-position of the
pyridine ring in host 6 or an additional R-amino group
in 7 favorably increased the basicity of the pyridine
moieties.9 In addition, the incorporation of a second
amino group in 7 provides an additional binding site for
carbohydrates. The substituted central phenyl ring in
5-7 can participate in effective CH-π interactions with
sugar CHs.
The variations in the determined Ka values (signifi-
cantly higher affinity of 5-7 toward â-anomer) cor-
respond qualitatively to expected differences in hydrogen-
bonding abilities of sugar hydroxyl groups (in contrast
(10) Crystallographic data for the structures reported in this paper
have been deposited with the Cambridge Crystallographic Data Centre
as supplementary publication nos. CCDC-229363 (2), 229364 (3),
235794 (4), 235792 (6), and 235793 (8).
(11) For hydrogen-bonding motifs in the crystals of secondary
diamides with 2-amino-6-methyl- and 2,6-diaminopyridine subunits,
see: Mazik, M.; Bla¨ser, D.; Boese, R. Tetrahedron 1999, 55, 12771-
12781
Also the hydrogen-bonding motifs in the crystal struc-
ture of 410 illustrate the steric hindrance from the
(9) Katritzky, A. R.; Pozharski, A. F. Handbook of Heterocyclic
Chemistry; Pergamon: Amsterdam, The Netherlands, 2000; p 178.
7452 J . Org. Chem., Vol. 69, No. 22, 2004

Molecular Recognition of Carbohydrates
F IGURE 4. Packing of 4 (a), 2 (b, CH hydrogens are ommitted for clarity), and 3 (c, hydrogen-bonded ethanol molecules are
shown).
TABLE 3. Th er m od yn a m ic P a r a m eter s Eva lu a ted by ITC for 6‚16 a n d 6‚17 a t 25 °C
Ka1 (∆G°)a
Ka2 (∆G°)b
∆H1
∆H2
T∆S1
T∆S1
host-guest complex
(M-1, kJ mol-1
)
(M-1, kJ mol-1
)
(kJ mol-1
)
(kJ mol-1
)
(kJ mol-1
)
(kJ mol-1
)
6‚16
6‚17
12800 (-23.4)
1830 (-18.6)
1300 (-17.8)
-51.3
-20.1
-34.6
-27.9
-1.5
-16.8
a
b
1:1 Receptor/glucopyranoside complex. 1:2 Receptor/glucopyranoside complex.
to receptors 2 and 3, see above). The evidence for the
preferred complexation of the â anomer was obtained by
the NMR spectroscopy,5c extraction experiments,5c and
microcalorimetry (isothermal titration calorimetry, ITC).
The microcalorimetric data revealed that the receptor-
glucopyranoside interactions are enthalpy driven (see, for
example, the thermodynamic parameters evaluated for
6‚16 and 6‚17, Table 3 and Figures 5 and 6) and the
enthalpy of binding is more negative than, or equal to,
the free energy of binding (similar to the lectine-
carbohydrate complexes1b). Thus, the unfavorable loss in
entropy is compensated by an advantageous change in
enthalpy. The values of the binding constants obtained
by the NMR spectroscopy and the microcalorimetry are
of the same magnitude but differ by a factor of about 1.6-
2. Such differences are typical for the results obtained
by these two methods.12
The binding affinity obtained for receptor 6 and â-ga-
lactopyranoside 18 is comparable to that obtained with
R-glucopyranoside 17, which is significantly lower than
J . Org. Chem, Vol. 69, No. 22, 2004 7453

Mazik et al.
F IGURE 5. ITC-titration of receptor 6 (1.09 mM) with 10 µL aliquots of â-glucopyranoside 16 (11.62 mM) in chloroform (25 °C).
Left: thermogram. Right: the integrated curve showing experimental points (×) and the best fit (s) for 1:1 and 1:2 receptor:
sugar binding. The molar ratio of the sugar to the receptor is given (for the thermodynamic parameters see Table 3).
F IGURE 6. ITC-titration of receptor 6 (1.09 mM) with 12 µL aliquots R-glucopyranoside 17 (7.55 mM) in chloroform (25 °C).
Left: thermogram. Right: the integrated curve showing experimental points (×) and the best fit (s) for 1:1 receptor-sugar binding.
The molar ratio of the sugar to the receptor is given (for the thermodynamic parameters see Table 3).
the previously determined for â-glucopyranoside 16. The
complexation between 6 and 18 was again evidenced by
several changes in the H NMR spectra, particularly by
Table 1). The crystal structure of 6,10 for example, reveal-
ed that two pyridine-based recognition units are pointing
to the same face of the central aryl ring, while the third
recognition unit points in the opposite direction (Figure
7). In such cases a significant conformational change of
the receptor structure is required for complexation.
1
the downfield shifts of the receptor NH protons (∆δmax
)
1.20 ppm) and upfield shifted resonances for the meth-
ylene protons of 6 (∆δmax ) -0.12 ppm). The curve fitting
of the titration data suggested the existence of both 1:1
and 1:2 receptor:sugar complexes, with 1:1 association
constant of 1360 M-1 (Ka1) and a weaker association
constant for an 1:2 receptor:sugar complex (Ka2 ) 211
M-1).
To achieve an improved degree of preorganization of
the heterocyclic recognition units the 2,4,6-triethylben-
zene frame was used for the construction of the receptor
molecules 8 and 9. Ethyl groups should stabilize a
conformation in which the three recognition groups are
preferably directed to one side of the benzene plane.13
The compounds 8 and 9 were synthesized from 1,3,5-tris-
(bromomethyl)-2,4,6-triethylbenzene14 and 2-amino-4,6-
dimethylpyridine or 2,6-diaminopyridine, respectively. As
expected, this structural variation resulted in a substan-
tial enhancement of the binding affinity toward monosac-
charides. In the crystal, however, the compound 810 is
Although the methylene groups in 5-7 give conforma-
tional mobility to the whole molecule and the receptors
are far from being perfectly preorganized, the binding
affinity is high (particularly for â-glucopyranoside, see
(12) Connors, K. A. In Comprehensive Supramolecular Chemistry;
Atwood, J . L., Davis, J . E. D., MacNicol, D. D., Vo¨gtle, F., Eds.;
Pergamon: Oxford, U.K., 1996; Vol. 3, p 233.
7454 J . Org. Chem., Vol. 69, No. 22, 2004

Molecular Recognition of Carbohydrates
F IGURE 7. Crystal structure of 6, top and side views (hydrogen-bonded ethanol molecules are shown).
F IGURE 8. Crystal structure of 8, top and side views (hydrogen-bonded ethanol molecule is shown).
still not perfectly preorganized and adopts a similar
conformation as 6. Two pyridine units are on one side of
the central phenyl ring and one on the other (Figure 8).
The packing of the molecules of 6 and 8 in the crystals
is shown in Figure 9. The crystals were obtained from
ethanol solutions, and in both crystal structures the
ethanol molecules are hydrogen bonded to the receptor
molecules. In the case of 6, two ethanol molecules are
placed between two receptors (Figure 9a).
The noncovalent interactions between the ethanol and
receptor molecules involve OH‚‚‚N-pyr and HO‚‚‚HN
hydrogen bonds as well as interactions between the ethyl
groups and the central benzene rings of the two receptors.
The hydroxyl groups of the ethanol molecules form
cooperative hydrogen bonds through simultaneous par-
ticipation as donor and acceptor of hydrogen bonds. Such
cooperative hydrogen bonds are also present in the
crystal structure of 8 (Figure 9b). The participation of
the amide-NH/N-pyr units in the formation of hydrogen
bonds with the hydroxy groups of the solvent molecules
shows the potential of this type of recognition units in
the formation of complexes with substrates involving
hydroxy substituents.
The complexation between the receptors 8/9 and pyra-
nosides 16-18 was evidenced by the downfield shift of
the receptor amine protons and upfield shift of the CH2
resonances. During the titration of 8 with â-D-glucopy-
ranoside 16 the signal due to the amine NH moved
downfield by about 1.3 ppm and the methylene CH2
moved upfield by 0.16 ppm (Figure 10a). Similar to 6,
the curve fitting of the titration data for receptor 8 and
16 suggested the existence of both 1:1 and 1:2 receptor/
sugar complexes in the chloroform solution (Figure 10b),
(13) For examples on the use of other systems based on the
triethylbenzene frame, see: (a) Stack, T. D. P.; Hou, Z.; Raymond, K.
N. J . Am. Chem. Soc. 1993, 115, 6466-6467. (b) Niikura, K.; Metzger,
A.; Anslyn, E. V. J . Am. Chem. Soc. 1998, 120, 8533-8534. (c) Metzger,
A.; Lynch, V. M.; Anslyn, E. V. J . Angew. Chem., Int. Ed. Engl. 1997,
36, 862-865. (d) Kim S.-G.; Ahn, K. H. Chem. Eur. J . 2000, 6, 3399-
3403. (e) Chin, J .; Walsdorff C.; Stranix, B.; Oh, J .; Chung, H.-J .; Park,
S.-M.; Kim, K. Angew. Chem., Int. Ed. 1999, 38, 2756-2759.
(14) Cabell, L. A.; Best, M. D.; Lavigne, J . J .; Schneider, S. E.;
Perreault, D. M.; Monahan, M.-K.; Anslyn, E. V. J . Chem. Soc., Perkin
Trans. 2 2001, 315-323.
with a strong 1:1 association constant (Ka1 ) 48630 M-1
)
J . Org. Chem, Vol. 69, No. 22, 2004 7455

Mazik et al.
F IGURE 9. Packing of 6 (a) and 8 (b) in the crystal (hydrogen-bonded ethanol molecules are shown, CH hydrogens are ommitted
for clarity).
and a weaker association constant for 1:2 receptor/sugar
complex (Ka2 ) 1320 M-1). Thus, the determined binding
constants for complex of 8 and 16 were about 2-fold
higher than the previously estimated binding constants
for complex of 6 and 16.5c
sugar complexes are formed. According to molecular
modeling the 2:1 complexes display hydrogen-bonding
between receptor and sugar, as well as between the two
receptor molecules. The titration data with R-glucopyra-
noside 17 showed that 1:1 binding is taking place. Several
changes in the NMR spectra of 9 after the addition of
pyranosides 16-18 were observed. The NMR signals of
the NH and NH2 groups of 9 were substantially shifted
An 1H NMR titration of 8 with R-glucopyranoside 17
or â-galactopyranoside 18 produced similar spectral
changes. In particular, the signal due to the amine NH
moved downfield by about 1.15 ppm, and the methylene
CH2 moved upfield by 0.15 ppm (Figure 10c,e). The fit of
NMR shift changes of the NH as well as the CH2 group
of 8 during the titration with 17 agreed with a 1:1 binding
model (Figure 10d), yielding an association constant of
1310 M-1. The motions of the signals 8 observed during
the titration with 18 were consistent with 1:1 and 1:2
receptor/sugar binding (Figure 10f) and could be analyzed
to give association constants of 3070 M-1 (Ka1) and 470
M-1 (Ka2). Thus, among the examined monosaccharides,
â-glucopyranoside 16 exhibited the highest affinity to-
ward 8. Comparison of Figure 10b and d/f clearly shows
that the process of complexation is not reflected by these
chemical shifts upon binding with these sugars in the
same way. Whereas after the addition of 2 equiv of
â-glucopyranoside almost no change was observed in the
chemical shift of receptor signals, with the R-glucopyra-
noside or â-galactopyranoside chemical shift changes
continue to higher sugar/receptor ratios. Thus, similar
to 6, the compound 8 displays high â/R anomer selectivity
in the recognition of glucopyranosides. The receptor 8
exhibited about 2-fold higher affinity for sugars 16-18
than the receptor 6. These results again demonstrate the
importance of preorganization in the molecular recogni-
tion process.
downfield (ΝΗ: ∆δmax ) 1.20-1.30 ppm, NH2: ∆δmax
)
0.20-0.30 ppm) and the signal of the CH2 group moved
upfield (∆δmax ) -0.10-0.20 ppm). The binding constants
of â-glucopyranoside 16 and receptor 9 were found to be
19590 M-1 (Ka1) and 14490 (Ka2), the one for R-glucopy-
ranoside 17 and 9 amounts to 1100 M-1, and the binding
constants for the complex with â-galactopyranoside 18
were determined to 8410 M-1 (Ka1) and 8680 (Ka2). Thus,
significantly weaker binding for the R-anomer was de-
terminated (similar to the previously described receptor
75c).
Recep tor s 10 a n d 11. The receptors 10 and 11,
bearing three R,γ-dimethyl-substituted pyridine groups
covalently attached to the 2,4,6-trimethyl- or -triethyl-
substituted phenyl ring via -NHCOCH2- units, show
significantly lower affinity toward glucopyranosides in
comparison to both groups of receptors, 1-3 and 5-9.
According to the molecular modeling, the receptors 10
and 11 should be able to fully encapsulate a pyranoside
molecule (better than the more rigid 1) in 1:1 receptor/
sugar complexes. On the other hand, these receptors have
a higher conformational freedom and the unfavorable
entropy hinders the effective recognition process. On
titration of 10 or 11 with â-glucopyranoside 16 no
substantial shifts for receptor protons were observed (see
Table 2), indicating a weak binding. The pattern of shifts
was consistent with 1:1 stoichiometry. Whereas the
binding constant of 10 and 16 was calculated to be 650
M-1, the value for 11 and 16 with Ka ) 1230 M-1 is about
2-fold higher (receptor based on 2,4,6-triethylbenzene
frame). Thus, the unfavorable loss in entropy is not
compensated by an advantageous change in enthalpy.
In the case of receptor 9, the interactions with the
additional R-amino group in the pyridine ring play a
significant role and influence the binding model (similar
to the receptor 7). The R-amino group increases the
basicity and provides additional hydrogen bonding sites.
The titration data revealed that with â-glucopyranoside
16 and â-galactopyranoside 18 the 1:1 and 2:1 receptor/
7456 J . Org. Chem., Vol. 69, No. 22, 2004

Molecular Recognition of Carbohydrates
F IGURE 10. Titration of receptor 8 with â-glucopyranoside 16, R-glucopyranoside 17, and â-galactopyranoside 18 ([8] ) 1.23 or
0.93 mM). 1H NMR spectra (CDCl3, 25 °C) of the receptor 8 (NH and CH2 resonances of 8 and the anomeric CH of sugar are
shown) after addition of 0-3.50 equiv (from bottom to top) of 16 (a), 0-7.20 equiv of 17 (c), and 0-5.05 equiv of 18 (e). Plot of the
observed (+) and calculated (s) downfield chemical shifts of the δN-H resonances of 8 as a function of added 16 (b), 17 (d), and 18
(f). The [receptor]/[pyranoside] ratio is marked.
The results indicate the difficulty of stable complex
formation when the arms of the amidopyridine receptor
are longer (in comparison to 1).
The compounds 10 and 11 were synthesized from
2-amino-4,6-dimethylpyridine and acyl chloride 23a or
23b, respectively. The synthesis of 23a ,b is shown in
Scheme 1 (the preparation of the acids 22a and 22b is
described in refs 16 and 17, respectively).
no group are reflected by receptor 12, including only two
heterocyclic subunits interconnected by a phenyl spacer
(12 was prepared by the reaction of 2,6-diaminopyridine
with isophthaloyl dichloride11,15). The strong complexation
(15) Chang, S.-K.; Van Engen, D.; Fan, E.; Hamilton, A. D. J . Am.
Chem. Soc. 1991, 113, 7640-7645.
(16) Podlaha, J .; Cisarova, I.; Alexander, D.; Holy, P.; Kraus, T.;
Zavada, J . Collect. Czech. Chem. Commun. 2000, 65, 1587-1596.
(17) (a) J eong, K.-S.; Shin, K. H.; Kim, S.-H. Chem. Lett. 2002, 1166-
1167. (b) O’Leary, B. M.; Szabo, T.; Schalley, C. A.; Luezen, A.; Schaeer,
M.; Rebek, J . J . Am. Chem. Soc. 2001, 123, 11519-11533.
Recep tor s 12-15. The favorable properties of the
amidopyridine receptors possessing an additional R-ami-
J . Org. Chem, Vol. 69, No. 22, 2004 7457

Mazik et al.
1
F IGURE 11. Titration of receptor 12 (0.92 mM) with octyl â-D-glucopyranoside (16). (a) H NMR spectra (CDCl3, 25 °C) of the
receptor 12 (NH, 2-CHPh and CHpyr resonances are shown) after addition of 0.00-3.95 equiv (from bottom to top) of 16. (b) Plot
of the observed (+) and calculated (s) downfield chemical shifts of the δN-H resonances of 12 as a function of added 16. The
[receptor]/[glucopyranoside] ratio is marked.
F IGURE 12. Energy-minimized structure of the 2:1 complex formed between receptor 12 and â-glucopyranoside 16 (MacroModel
V.6.5, Amber* force field, Monte Carlo conformational searches, 50000 steps).
SCHEME 1
of sugars was clearly indicated by 1H NMR spectroscopy.
The molecules of 12 possess four potential groups of
proton donors (two amide-NH and two NH2 groups)
available for hydrogen bonding. During the titration of
12 by 16 substantial downfield motions were observed
for both the amide-NH (Figure 11a) and NH2 signals
(∆δmax ) 0.66 and 0.43 ppm for NH and NH2, respec-
tively), indicating that these groups are engaged in
hydrogen bonds. Furthermore, the aromatic CH reso-
nances (both the CHPh and CHpyr) shifted upfield by 0.18-
0.35 ppm. The addition of only 1 equiv of sugar 16 led to
practically complete complexation of the hosts 12. The
7458 J . Org. Chem., Vol. 69, No. 22, 2004

Molecular Recognition of Carbohydrates
SCHEME 2
SCHEME 3 a
a
Key: (i) 2 equiv of 2-amino-4,6-dimethylpyridine, 2 equiv of NEt3, CH2Cl2, 25 °C, 40 min; (ii) 1 equiv of HN(CH2COOCH2CH3)2, 1
equiv of NEt3, CH2Cl2, 25 °C, 24 h.
measurements indicated the formation of complexes with
receptor:sugar stoichiometry of 2:1 (similar to receptors
7 and 9 having also an R-NH2 group). The motions of the
NH of 12 were consistent with 1:1 and 2:1 receptor/sugar
binding (Figure 11b) and could be analyzed to give
association constants of 1420 M-1 (Ka1) and 3890 M-1
(Ka2). Thus, the receptor 12 exhibits more favorable
binding properties than the previously described bipy-
ridyl host 135a as well as the triarmed receptors 4, 10,
and 11. These results reflect clearly the particular
suitability of the 2-aminopyridine group for the recogni-
tion of carbohydrates. The potential of the 2-aminopyri-
dine group in the recognition processes has been already
discussed by other groups. For example, Anslyn and co-
workers have noted the potential of this motif for binding
vicinal diols.18 Anslyn has exploited the 2-aminopyridine
unit as a heterocyclic analogue of the asparagine/
glutamine primary amide side chain.3a,18
potentially be stabilized by several hydrogen bonds
between the OHs of sugar and the amide-NH, amine-
NH, and pyr-N of the two receptors (pyr-N‚‚‚HO, amide-
NH‚‚‚OH, amine-NH‚‚‚OH) as well as by O‚‚‚HN hydro-
gen bond between the ring oxygen and the amide-NH.
Typical hydrogen-bonding motifs found by molecular
modeling studies are shown in Scheme 2. Furthermore,
the complex is stabilized by interactions of sugar CHs
with the aromatic groups of the two receptor molecules
and weak CH‚‚‚N-pyr hydrogen bonds. Moreover, the
interactions between the two receptor molecules contrib-
ute to the stabilization of the 2:1 receptor/sugar complex.
Receptor 12 shows also significantly higer affinity than
the host 14 incorporating not only two heterocyclic units,
but also two ester groups. 1H NMR binding studies with
â-glucoside 16 as substrate yielded rather low association
constant of 950 M-1 for 14‚16 (Table 2). The 1H NMR
spectra obtained during binding experiments showed only
moderate downfield shifting of the NH resonances. Even
after the addition of 5 equiv of 16 the saturation has not
been achieved. However, further improvement of the
complexation properties of 14 can be expected after the
hydrolysis of the ester groups. This structural change will
provide compound including both the carboxylate groups
Molecular modeling indicated the suitable multipoint
hydrogen bonding arrangement between 12 and 16, as
shown in Figure 12. A 2:1 receptor/sugar complex can
(18) Huang, C.-Y.; Cabell, L. A.; Anslyn, E. V. J . Am. Chem. Soc.
1994, 116, 2778-2792.
J . Org. Chem, Vol. 69, No. 22, 2004 7459

Mazik et al.
1
F IGURE 13. Titration of 15 (0.88 mM) with octyl â-D-glucopyranoside (16). (a) H NMR spectra (CDCl3, 25 °C) of the receptor
12 (NH, CHPh and CHpyr resonances are shown) after addition of 0-5.80 (from bottom to top) equivalents of 16. (b) Plot of the
observed (+) and calculated (s) downfield chemical shifts of the δN-H resonances of 15 as a function of added 16. The [receptor]/
[glucopyranoside] ratio is marked.
and the heterocyclic units and may lead to favorable bind-
ing properties both in water and organic solution. The
synthesis of 14 involves the reaction of benzene-1,3,5-tri-
carbonyl chloride with 2 equiv of 2-amino-4,6-dimethyl-
pyridine, followed by the reaction with 1 equiv of imino-
diacetic acid diethyl ester. The separation of the product
14 and the byproducts 24 and 25 (Scheme 3) was carried
out by crystallization (see the Experimental Section).
Additionally, comparative complexation studies with
host 15 were carried out (15 was prepared by the reaction
of 2,6-diaminopyridine with m-phenylenedioxydiacetyl
chloride11). Receptors 12 and 15 have the same recogni-
tion groups, but differ in the flexibility, since the OCH2-
unit in 15 gives high conformational mobility to the whole
molecule. As expected, the binding studies with â-glu-
copyranoside 16 as probe verified that the interactions
with host 15 are less favorable than with host 12. The
lower binding affinity of 15 can be also explained by the
formation of the intramolecular hydrogen bonding be-
tween amide-NH and oxygen atom of the OCH2 group.
Such intramolecular hydrogen bonds have been previ-
ously observed in the crystal structure of 15.11 In contrast
to 12, the amide proton of 15 showed minimal displace-
ment on complexation with 16 (∆δmax ) 0.15 ppm, see
Table 2), confirming the weak binding. Figure 13 shows
the small chemical shifts of the NH and aromatic
resonances during the titration with â-glucopyranoside
16. The pattern of shifts was consistent with 1:1 stoichi-
ometry. The receptor 15 revealed a Ka value of only 190
M-1 for glucopyranoside 16. Comparison of Figures 11
and 13 clearly reflects the strong differences in the com-
plexation ability of 12 and 15. These studies showed, that
the flexibility and the formation of intramolecular hydro-
gen bonding can partially or wholly cancel the increase in
the affinity caused by R-amino group in the pyridine ring.
buildings blocks, a variety of structures with different
binding properties could be obtained.
The 2-aminopyridine group has been established as a
highly effective functional group in the binding of monosac-
charides. The aminopyridine receptors 5-9 show high
affinity for â-D-glucopyranoside and marked â vs R
selectivity. Receptors based on 2,4,6-triethylbenzene
frame (8 and 9) display about 2-fold higher binding
affinity than those based on 2,4,6-trimethylbenzene unit
(6 and 7), indicating more favorable preorganization of
8 and 9. The binding constants for receptors 6-9 decrease
in the sequence â-glucoside > â-galactoside > R-glucoside.
The comparison of the properties of 1-4 shows that
the steric interactions involving the pyridine substituents
in pyridine-based receptors significantly affect the bind-
ing properties. Remarkable changes in the binding af-
finity and selectivity have been observed when the degree
of steric hindrance at the pyridine nitrogen atom de-
creases. Receptors 2 and 3, which are sterically less
hindered at nitrogen, show high efficiency and an inverse
selectivity (in contrast to receptors 5-9 and 1) since they
bind the R-glucopyranoside better than the â-anomer.
Compounds 2 and 3 have a tendency to form strong 2:1
receptor/sugar complexes; the binding constants Ka2
ranged between 22600 and 82450 M-1 in the sequence
R-glucoside > â-galactoside g â-glucoside. Similar ten-
dencies to the formation of 2:1 receptor/sugar complexes
show the effective receptors possessing R-NH2 group as
a building block (hosts 7, 9, and 12).
High conformational mobility of the receptor structure
and the formation of intramolecular hydrogen bonding
drastically decrease the binding affinity of receptors
possessing effective recognition units, as reflected by the
properties of 10, 11, and 15.
Exp er im en ta l Section
Con clu sion
1H and 13C NMR spectra were measured using a 500 MHz
spectrometer; chemical shifts are reported in ppm downfield
to TMS as internal standard. Analytical TLC was carried out
on Kieselgel 60 F254 plates employing a methanol-chloroform
1:7 (v/v) or ethyl acetate-toluene 3:1 (v/v) as the mobile phase.
Melting points are uncorrected.
The results showed that acyclic receptors containing
amino- or amidopyridine binding subunits perform ef-
fective recognition of carbohydrates through multiple
interactions. Depending on the nature and number of
recognition subunits and connecting bridges used as the
7460 J . Org. Chem., Vol. 69, No. 22, 2004

Molecular Recognition of Carbohydrates
Gen er a l P r oced u r e for th e Syn th esis of 2-4. To a
solution of 2-aminopyridine derivative (2-amino-4-methylpy-
ridine, 2-amino-5-methylpyridine, or 2-amino-4-ethylpyridine;
0.033 mol) and triethylamine (2.3 mL) in dry CH2Cl2 (50 mL)
was added dropwise a CH2Cl2 (25 mL) solution of benzene-
1,3,5-tricarbonyl chloride (1.35 g, 0.005 mol). The reaction
mixture was stirred at room temperature for 24 h, and then
100 mL of water was added. The suspension was stirred for
30 min, and then CH2Cl2 was removed under reduced pressure.
The resulting precipitate was filtered, washed several times
with water, and dried. The crude product was crystallized from
THF/hexane and ethanol (compounds 2 and 3 were crystallized
two to three times from ethanol).
N,N′,N′′-Tr is(4-m et h ylp yr id in -2-yl)b en zen e-1,3,5-t r i-
ca r bon a m id e (2). Yield: 30%. Mp: 260-261 °C. 1H NMR
(500 MHz, CDCl3): δ ) 2.40 (s, 9H), 6.93 (d, 3H, J ) 5.2 Hz),
8.17 (d, 3H, J ) 5.2 Hz), 8.20 (s, 3H), 8.68 (s, 3H), 8.82 (s,
3H). 13C NMR (125 MHz, CDCl3): δ ) 21.4, 114.9, 121.5, 129.4,
135.7, 147.6, 150.1, 151.3, 163.9. HR-MS: calcd for C27H24N6O3
480.1910, found 480.1903. Anal. Calcd for C27H24N6O3: C,
67.49; H, 5.03; N, 17.49. Found: C, 67.24; H, 5.15; N, 17.66.
N,N′,N′′-Tr is(5-m et h ylp yr id in -2-yl)b en zen e-1,3,5-t r i-
ca r bon a m id e (3). Yield: 38%. Mp: 225-226 °C. 1H NMR
(500 MHz, CDCl3): δ ) 2.32 (s, 9H), 7.58 (dd, 3H, J ) 8.6/2.2
Hz), 8.13 (d, 3H, J ) 2.2 Hz), 8.25 (d, 3H, J ) 8.6 Hz), 8.67 (s,
3H), 8.73 (s, 3H). 13C NMR (125 MHz, CDCl3): δ ) 17.9, 113.8,
129.2, 129.8, 135.7, 139.1, 147.9, 148.9, 163.4. HR-MS: calcd
for C27H24N6O3 480.1910, found 480.1907. Anal. Calcd for
1,3,5-Tr is[(6-a m in op yr id in -2-yl)a m in om eth yl]-2,4,6-tr i-
eth ylben zen e (9). Yield: 40%. Mp: 163-165 °C. 1H NMR
(500 MHz, CDCl3): δ ) 1.22 (t, 9H, J ) 7.5 Hz), 2.75 (q, 6H,
J ) 7.5 Hz), 4.05 (t, 3H, J ) 4.2 Hz), 4.20 (s, 6H), 4.35 (d, 6H,
J ) 4.2 Hz), 5.83-5.86 (m, 6H), 7.24 (m, 3H). 13C NMR (125
MHz, CDCl3): δ ) 16.9, 23.0, 40.6, 95.9, 96.9, 133.3, 139.4,
143.8, 157.8, 157.9. HR-MS: calcd for C30H39N9 525.3328,
found 525.3322.
Gen er a l P r oced u r e for th e Syn th esis of 10 a n d 11. A
solution of acyl chloride 23a or 23b (0.61 mmol) in dry CH2Cl2
(5 mL) was added dropwise to a solution of 2-amino-4,6-dimeth-
ylpyridine (0.34 g, 2.80 mmol) and triethylamine (0.3 mL) in
dry CH2Cl2 (5 mL). After complete addition, the mixture was
stirred at room temperature for 24 h. The resulting precipitate
was filtered and crystallized from THF/hexane or ethanol.
2-[3,5-Bis[(4,6-dim eth ylpyr idin -2-ylcar bam oyl)m eth yl]-
2,4,6-tr im eth ylp h en yl]-N-(4,6-d im eth ylp yr id in -2-yl)a ce-
ta m id e (10). Yield: 65%. Mp: 273-275 °C. 1H NMR (500
MHz, CDCl3): δ ) 2.26 (s, 9H), 2.29 (s, 9H), 2.34 (s, 9H), 3.89
(s, 6H), 6.67 (s, 3H), 7.83 (s, 3H), 8.30 (s, 3H). 13C NMR (125
MHz, CDCl3): δ ) 17.4, 21.1, 23.6, 40.1, 111.7, 120.5, 130.6,
136.9, 149.9, 150.5, 156.1, 169.1. HR-MS: calcd for C36H42N6O3
606.3318, found 606.3309. Anal. Calcd for C36H42N6O3: C,
71.26; H, 6.98; N, 13.85. Found: C, 71.09; H, 7.14; N, 14.03.
2-[3,5-Bis[(4,6-dim eth ylpyr idin -2-ylcar bam oyl)m eth yl]-
2,4,6-tr ieth ylp h en yl]-N-(4,6-d im eth ylp yr id in -2-yl)a ceta -
m id e (11). Yield: 60%. Mp: 235-237 °C. 1H NMR (500 MHz,
CDCl3) δ ) 1.17 (t, 9H, J ) 7.6 Hz), 2.20 (s, 9H), 2.24 (s, 9H),
2.70 (q, 6H, J ) 7.6 Hz), 3.89 (s, 6H), 6.65 (s, 3H), 7.77 (s,
3H), 8.38 (s, 3H). 13C NMR (125 MHz, CDCl3): δ ) 14.2, 21.2,
23.6, 24.1, 38.8, 112.5, 120.5, 129.6, 143.6, 149.9, 150.5, 159.9,
170.1. HR-MS: calcd for C39H48N6O3 648.3788, found 648.3785.
Anal. Calcd for C39H48N6O3: C, 72.19; H, 7.46; N, 12.95.
Found: C, 71.98; H, 7.41; N, 13.13.
N-Di(eth oxycar bon ylm eth yl)-N′,N′′-bis(4,6-dim eth ylpy-
r id in -2-yl)ben zen e-1,3,5-tr ica r bon a m id e (14). A solution
of 2-amino-4,6-dimethylpyridine (1.38 g, 11.24 mmol) and
triethylamine (1.6 mL) in dry CH2Cl2 (30 mL) was added
dropwise (during 10-15 min) to a solution of 1,3,5-benzene-
tricarbonyl trichloride (1.49 g, 5.62 mmol) in dry CH2Cl2 (20
mL). After complete addition, the mixture was stirred at room
temperature for 40 min, and then a solution of iminodiacetic
acid diethyl ester (1.06 g, 5.62 mmol) and triethylamine (0.8
mL) in dry CH2Cl2 (20 mL) was added dropwise. After complete
addition, the mixture was stirred at room temperature for 24
h, the CH2Cl2 was removed under reduced pressure, and the
crude product was suspended in Et2O. After filtration and
evaporation of Et2O, the obtained powder was crystallized from
ethanol and diethyl ether (the repeated crystallization is
necessary for the complete separation of the byproducts 24 and
25). Yield: 30%. Mp: 165-166 °C. 1H NMR (500 MHz,
CDCl3): δ ) 1.22-1.32 (m, 6H),), 2.33 (s, 6H), 2.37 (s, 6H),
4.08-4.33 (m, 8H), 6.76 (s, 2H), 7.96 (s, 2H), 8.18 (s, 2H), 8.53
(s, 1H), 8.58 (s, 2H). 13C NMR (125 MHz, CDCl3): δ ) 14.1,
21.3, 23.7, 47.8, 51.8, 61.5, 62.1, 111.6, 120.9, 127.6, 128.9,
135.8, 136.1, 150.1, 150.3, 156.6, 163.4, 168.6, 168.8, 170.2.
HR-MS: calcd for C31H35N5O7 589.2536, found 589.2545. Rf )
0.49 (ethyl acetate/toluene 3:1 v/v).
Bin d in g Stu d ies. 1H NMR titrations were performed at
298 K in CDCl3 stored over activated molecular sieves and
deacidified with Al2O3 (for each titration 12-21 samples were
prepared). The titration data were analyzed by nonlinear
regression analysis using the HOSTEST 5.6 program.6 Ex-
amples: (a) [2] ) 1.32 mM, [18] ) 0.15-3.26 mM; (b) [4] )
1.01 mM, [16] ) 0.22-5.47 mM; (c) [8] ) 1.23 mM, [16] )
0.25-5.11 mM; (d) [8] ) 0.67 mM, [17] ) 0.16-4.85 mM; (e)
[8] ) 0.93 mM, [18] ) 0.18-4.73 mM; (f) [12] ) 0.94 mM, [16]
) 0.12-3.73 mM; (g) [14] ) 1.05 mM, [16] ) 0.35-7.02 mM;
(h) [18] ) 0.57 mM, [2] ) 0.11-2.49 mM. For each system at
least three titrations were carried out.
C
27H24N6O3: C, 67.49; H, 5.03; N, 17.49. Found: C, 67.36; H,
5.08; N, 17.58.
N,N′,N′′-Tr is(6-eth ylp yr id in -2-yl)ben zen e-1,3,5-tr ica r -
1
bon a m id e (4). Yield: 70%. Mp: 215-216 °C. H NMR (500
MHz, CDCl3): δ ) 1.27 (t, 9H, J ) 7.6 Hz), 2.72 (q, 6H, J )
7.6 Hz), 6.95 (d, 3H, J ) 7.6 Hz), 7.66 (t, 3H, J ) 7.9 Hz), 8.15
(d, 3H, J ) 8.2 Hz), 8.69 (s, 3H), 8.81 (s, 3H). 13C NMR (125
MHz, CDCl3): δ ) 13.7, 30.9, 111.3, 118.6, 129.2, 135.8, 138.9,
150.4, 162.3, 163.4. HR-MS: calcd for C30H30N6O3 522.2379,
found 522.2360.
Gen er a l P r oced u r e for th e Syn th esis of 6, 8, a n d 9. A
mixture of 1,3,5-tris(bromomethyl)-2,4,6-triethyl-14 or -2,4,6-
trimethylbenzene19 (3.50 mmol), 2-amino-4,6-dimethylpyridine
(2.40 g, 19.67 mmol), or 2,6-diaminopyridine (2.40 g, 22.02
mmol) and K2CO3 (1.50 g) in CH3CN (120 mL) was stirred at
room temperature for 24 h (in the case of 6) or heated under
reflux for 6 h and then stirred at room temperature for 34 h
(in the case of 8 and 9). After filtration of the reaction mixture
and evaporation of CH3CN, the obtained powder was sus-
pended in CHCl3. The suspension was filtrated (in the case of
9 the chloroform solution was first washed several times with
water and then dried), and CHCl3 was removed under reduced
pressure. The crude product was crystallized from THF/
hexane, ethanol, or diethyl ether.
1,3,5-Tr is[(4,6-dim eth ylpyr idin -2-yl)am in om eth yl]-2,4,6-
tr im eth ylben zen e (6): Yield: 60%. Mp: 193-195 °C. 1H
NMR (500 MHz, CDCl3): δ ) 2.21 (s, 9H), 2.33 (s, 9H), 2.38
(s, 9H), 4.11 (t, 3H, J ) 4.2 Hz), 4.37 (d, 6H, J ) 4.2 Hz), 6.08
(s, 3H), 6.32 (s, 3H). 13C NMR (125 MHz, CDCl3): δ ) 15.9,
21.1, 24.2, 41.8, 103.4, 113.9, 133.7, 136.8, 148.8, 156.7, 158.4.
HR-MS: calcd for C33H42N6 522.3471, found 522.3477. Anal.
Calcd for C33H42N6: C, 75.82; H, 8.10; N, 16.08. Found: C,
76.01; H, 7.97; N, 15.89.
1,3,5-Tr is[(4,6-dim eth ylpyr idin -2-yl)am in om eth yl]-2,4,6-
tr ieth ylben zen e (8): Yield: 57%. Mp: 183-185 °C. 1H NMR
(500 MHz, CDCl3): δ ) 1.21 (t, 9H, J ) 7.6 Hz), 2.21 (s, 9H)
2.33 (s, 9H), 2.72 (q, 6H, J ) 7.6 Hz), 4.09 (t, 3H, J ) 4.3 Hz),
4.36 (d, 6H, J ) 4.3 Hz), 6.06 (s, 3H), 6.32 (d, 3H). 13C NMR
(125 MHz, CDCl3): δ ) 17.0, 21.2, 23.0, 24.3, 40.8, 103.7, 114.0,
133.3, 143.7, 148.8, 156.8, 158.3. HR-MS: calcd for C36H48N6
564.3940, found 564.3931. Anal. Calcd for C36H48N6: C, 76.55;
H, 8.57; N,14.88. Found: C, 76.38; H, 8.64; N, 14.99.
The microcalorimetric titrations were carried out at 298 K
using a Thermometric titration calorimetric system. The
(19) Van der Made, A. W.; van der Made, R. H. J . Org. Chem. 1993,
58, 1262-1263.
J . Org. Chem, Vol. 69, No. 22, 2004 7461

Mazik et al.
reproducibility of the calorimeter was checked using the
complexation of Ba2+ by 18-crown-6. A solution of octyl-R-D-
glucopyranoside (6.33-7.55 mM) or octyl-â-D-glucopyranoside
(7.78-11.62 mM) in chloroform was titrated into a 1.01-1.09
mM solution of receptor in chloroform. The data obtained were
analyzed using the Digitam 4.1 software provided by Ther-
mometric (heat of dilution was corrected). Examples: (a) [6]
) 1.07 mM, [16] ) 11.62 mM, 30 automatic injections, 10 µL
each; (b) [6] ) 1.09 mM, [17] ) 6.33 mM, 40 injections, 12 µL
each; (c) [8] ) 0.97 mM, [16] ) 7.78 mM, 40 injections, 12 µL
each; (d) [8] ) 1.03 mM, [17] ) 7.78 mM, 40 injections, 12 µL
each.
Dipl.-Ing. Dieter Bla¨ser (Institut fu¨r Anorganische
Chemie der Universita¨t Duisburg-Essen) for performing
the X-ray measurements and Kerstin Antepoth for
technical assistance. We thank also Prof. Craig Wilcox
for giving access to the HOSTEST program.
Su p p or tin g In for m a tion Ava ila ble: X-ray data for com-
1
pounds 2-4, 6, and 8. H and 13C NMR spectra of compounds
2-4, 6, 8, 9-11, and 14. 1H NMR Titration of receptor 11 with
octyl â-D-glucopyranoside (16). Representative mole ratio plot
(complexation of receptor 3 with R-glucopyranoside 17). This
material is available free of charge via the Internet at
http://pubs.acs.org.
Ack n ow led gm en t. This work was supported by the
Deutsche Forschungsgemeinschaft (SFB 452). We thank
J O048979K
7462 J . Org. Chem., Vol. 69, No. 22, 2004