Molecular tweezers: Synthesis and formation of host-guest complexes

Article abstract of DOI:10.1002/ejoc.201000729

A chiral molecular tweezer obtained from (+)-usnic acid placed in solution in the presence of various aromatic compounds afforded complexes with low association constants. Thus, the X-ray structure of assembly 31 is presented, where the guest is sandwiched between the two pincers of the tweezer. The association constants for various guests were determined through different methods. Finally, other tweezers with electron-rich aromatic aldehydes and ketones were prepared from (1R,2R)-1,2-diaminocyclohexane. The most interesting complexes were also confirmed through structural analysis, and the best results were obtained with 10-hydroxyphenanthrene-9-carbaldehyde (51) as the aromatic moiety.

Full text of DOI:10.1002/ejoc.201000729

FULL PAPER
DOI: 10.1002/ejoc.201000729
Molecular Tweezers: Synthesis and Formation of Host–Guest Complexes
Béatrice Legouin,[a] Maud Gayral,[a] Philippe Uriac,[a] Jean-François Cupif,[a]
Nicolas Levoin,[b] Loic Toupet,[c] and Pierre van de Weghe*[a]
Keywords: Supramolecular chemistry / Host-guest systems / Pi interactions / Molecular recognition
A chiral molecular tweezer obtained from (+)-usnic acid
placed in solution in the presence of various aromatic com-
pounds afforded complexes with low association constants.
Thus, the X-ray structure of assembly 3i is presented, where
the guest is sandwiched between the two pincers of the
tweezer. The association constants for various guests were
determined through different methods. Finally, other tweez-
ers with electron-rich aromatic aldehydes and ketones were
prepared from (1R,2R)-1,2-diaminocyclohexane. The most
interesting complexes were also confirmed through struc-
tural analysis, and the best results were obtained with 10-
hydroxyphenanthrene-9-carbaldehyde (5i) as the aromatic
moiety.
variety of systems involved in π–π-stacked interactions is an
important step toward a better understanding of this type
of noncovalent interaction.
Introduction
The process of molecular recognition of guest molecules
by a synthetic host is of great interest in supramolecular
chemistry and depends on the nature of the intermolecular
interactions. For this reason the study and characterization
of these interactions have experienced enormous growth in
recent years. Among the different intermolecular forces,
noncovalent aromatic interactions are of particular rele-
vance. The π–π-stacked aromatic interactions have been
found to govern diverse molecular organizations in both
solution and the solid state. Furthermore, this is considered
as a very important component of ligand–receptor interac-
tions, particularly in medicinal chemistry.[1] However, in
comparison to more conventional interactions such as hy-
drogen bonds, ion pairs, and hydrophobic interactions, the
π–π-stacked interaction is not clear, and to date, no com-
monly useful model has been built to interpret properly the
experimental observations. This could be due to the compli-
cation of the π–π-stacked dependency on substituent
groups, variable geometries, and the cooperative effect of
various noncovalent interactions. Therefore, study of a
Whitlock first reported the synthesis of molecular tweez-
ers and their use as suitable models to bind aromatic
guests.[2] Molecular tweezers are simple molecular hosts
that can be described by the presence of two tethered pin-
cers by a more or less rigid spacer. With a distance of about
7 Å between electron-rich or electron-poor aromatic pin-
cers, molecular tweezers appear to be an ideal model to
study noncovalent aromatic interactions.[3] We recently syn-
thesized novel chiral molecular tweezer 1 by using (+)-usnic
acid as the tethered pincer and (1R,2R)-1,2-diaminocyclo-
hexane as the spacer (Figure 1) and showed its ability to
form a host–guest complex with 2,4,7-trinitro-9H-fluoren-
9-one (TNF; 2a).[4] Here we first reported the determination
of association constants of host–guest complexes from 1
with different common guest aromatic compounds substi-
tuted with strong electron-withdrawing groups. The associa-
tion constants K in molecular complexes in solution were
1
determined by H NMR[5] and UV/Vis spectroscopy.[6] We
then extended this approach by describing the synthesis of
new potential molecular tweezers from (1R,2R)-1,2-diami-
nocyclohexane and various carbonyl derivatives aiming to
obtain better tweezers.
[a] EA Substances Lichéniques et Photoprotection, Faculté des
Sciences Biologiques et Pharmaceutiques,
Université de Rennes 1,
2 av. du Prof. Léon Bernard, 35043 Rennes Cedex, France
Fax: +33-2-23234425
E-mail: pierre.van-de-weghe@univ-rennes1.fr
[b] Bioprojet-Biotech,
4 rue du Chenay Beauregard, BP 96205, 35762 Saint-Grégoire,
France
[c] GM CM, CNRS UMR 6509, Université de Rennes 1, Campus
de Beaulieu,
35042 Rennes Cedex, France
Supporting information for this article is available on the
WWW under http://dx.doi.org/10.1002/ejoc.201000729.
Figure 1. Structure of chiral molecular tweezer 1.
View this journal online at
wileyonlinelibrary.com
Eur. J. Org. Chem. 2010, 5503–5508
© 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
5503

P. van de Weghe et al.
FULL PAPER
CH2Cl2 on the basis of the UV/Vis Job’s plot[7] as shown in
Figure 2. The association constants of 3 were determined
by a nonlinear curve-fitting analysis from H NMR spec-
Results and Discussion
1
Determination of Association Constants of Host–Guest
Complexes from 1 with Electron-Poor Aromatic Guests
troscopy and are reported in Table 1.
Due to its electron-rich pincers, molecular tweezer 1
could form stable complexes 3 with different electron-de-
manding aromatic derivatives 2 in solution involving π–π
interactions (Scheme 1). The binding ability of 1 with re-
gard to a variety of aromatic derivatives was first evaluated
by 1H NMR titration in CD2Cl2 at 298 K. No signal of the
free species in the spectrum was observed, suggesting that
the rate of complexation and decomplexation are fast under
these conditions (time-average signals with 500 MHz 1H
NMR). In most cases, significant shifts of the peaks ascrib-
able to 1 and aromatic derivatives 2 were observed. Relying
on our previous study, the 1:1 stoichiometry of the com-
plexes was considered and confirmed for complex 3a in
Figure 2. Plot of Job’s method for a mixture of 1 and 2a ([1] + [2a]
= 20 m) at 450 nm (solvent: CH2Cl2).
Weak complexes 3a–g and 3i were obtained by mixing
molecular tweezer 1 with electron-deficient aromatic deriva-
tives 2a–g and 2i; the association constants determined by
NMR spectroscopy were found to be between 7 and 53 –1.
These results allow the following conclusions: Complex 3a
obtained from TNF (2a) is substantially more stable than
corresponding complexes 3b–g by a factor of at least 2. We
believed we could improve the affinity of 1 for electron-
poor aromatic guests by combining H-bonding with π–π
interactions as reported sometimes for cucurbituril host–
guest systems.[8] However, carboxylic acids 2b and 2e and
phenol 2d only led to weak complexes 3 (15 Ͻ K Ͻ 26 –1).
On the other hand, the data indicate the influence of the
steric hindrance of guests 2 to the stability of complexes
3. As observed, the addition of a methyl ester group or a
carboxylic acid function to starting fluorenone derivative
2a decreased the strength of the association for complexes
3. A similar comparison could be done between assembly
3e and 3f, acid 2e affording a more stable complex. Al-
though the association constants of complexes 3a–g and 3i
are weak, the use of the 1H NMR titration method allowed
us to have a better idea of the affinity of the guests with
the tweezer. For compound 3h, the signal intensity de-
creased but no chemical shift change was observed during
Scheme 1. Formation of host–guest complexes 3.
Table 1. Values of association constants K (500 MHz NMR in CD2Cl2, UV/Vis in CH2Cl2).
Entry
3
Association constant K [–1, 298 K]
1H NMR (∆δmax, Hz)
UV/Vis, Method A[a] (ε, Lmol–1 cm–1) UV/Vis, Method B[b] (ε, Lmol–1 cm–1)
1
2
3
4
5
6
7
8
9
3a
3b
3c
3d
3e
3f
3g
3h
3i
53Ϯ1 (272Ϯ1)
26Ϯ2 (252Ϯ8)
16Ϯ1 (194Ϯ5)
15Ϯ1 (271Ϯ11)
22Ϯ1 (326Ϯ5)
7Ϯ1 (308Ϯ25)
10Ϯ1 (103Ϯ3)
n.d.[d]
27 (1250)
45 (1500)
23[c] (1500)
12 (13400)
18 (750)
21 (750)
8 (1600)
n.d.[d]
35Ϯ4 (1100Ϯ100)
31Ϯ5 (2000Ϯ200)
22Ϯ3[c] (1600Ϯ170)
14Ϯ3 (11800Ϯ2200)
20Ϯ1 (750Ϯ100)
24Ϯ5 (700Ϯ100)
11Ϯ3 (1300Ϯ150)
24Ϯ7 (2800Ϯ500)
n.d.[d]
9Ϯ0.3 (233Ϯ4)
n.d.[d]
[a] Determined by the Benesi–Hildebrand method. [b] Determined by a curve-fitting procedure. [c] Determined at 440 nm, otherwise
determined at 450 nm. [d] n.d. = not determined.
5504
www.eurjoc.org
© 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Org. Chem. 2010, 5503–5508

Molecular Tweezers
the NMR titration; the latter did not then allow the deter- plex 3a.[4] The presence of a guest in the molecular tweezer
mination of the association constant. Therefore, we also de- reduces the distance between both usnic acid units and ap-
cided to evaluate the association constants by using UV/Vis pears to be proportional to the value of the association con-
spectroscopy and to compare those results with the pre- stant of the complex: 2.6Ϯ0.2 –1 at 298 K for molecular
1
vious one obtained by H NMR titration.
tweezer 1 alone,[13] weaker for assembly 3i (9Ϯ0.3 –1 at
Molecular tweezer 1 and guests 2 lead to colorless or 298 K), and better for complex 3a (53Ϯ1 –1 at 298 K).
pale yellow solutions. This color changes when guest mole-
cule 2 is added to a solution of 1 in methylene chloride; the
appearance of a more or less intense color during the pro-
cess of association results from charge-transfer bands. Be-
cause of its simplicity of implementation, the graphical Be-
nesi–Hildebrand method[9] (method A) was first used to de-
termine the association constants. The spectroscopic data
were obtained from solutions in methylene chloride by
varying the concentration of tweezer 1 while the concentra-
tion of guest 2 was held constant. The equation of the plot
1/[1] vs. [2]/A gives 1/Kε as the intercept and 1/ε as the slope,
where ε is the molar extinction coefficient of complex 3.
Due to the weak values of constants and to avoid negative
value of the intercept, the concentrations of 1 and 2 were
carefully defined.[10,11] Unfortunately, we were not able to
determine the association constant of 3h due to the pres-
Figure 3. X-ray structure of complex 3i (molecules of CH2Cl2 that
are present in the crystals were voluntarily omitted to simplify the
ence of absorption bands of 2h in the same area as those figure, Ͼ50% probability level).
of charge transfer (450 nm). It is also important to note
During this work we also investigated the possibility of
that due to the low values of the association constants, this
graphical method is too vague and is a source of errors. In
order to have more precise values of K and finally to evalu-
ate the association constant of complex 3h, we decided to
determine the association constant by treating the experi-
mental data with nonlinear analysis (method B).[12]
forming complexes through noncovalent aromatic interac-
tions between usnic acid and TNF. Whatever the method of
analysis used, NMR or UV/Vis spectroscopy, no evidence
suggests the presence of these π–π stacked aromatic interac-
tions between usnic acid and TNF. This observation indi-
cates that the presence of a tether such as trans-1,2-di-
aminocyclohexane is essential, probably to limit entropy
loss associated with the binding event.
In this first part, we showed that chiral molecular
tweezer 1 prepared from (+)-usnic acid could form host–
guest complexes with a variety of electron-poor aromatic
derivatives. The binding strength of these complexes varies
between 7 and 53 –1 after evaluating by 1H NMR ti-
trations and between 11 and 35 –1 from spectrophotomet-
ric data (method B). The correlation of these association
constants with the help of spectrophotometric and NMR
methods shows a quite similar order of the guests. However,
the values of the association constants are too weak to ob-
tain a good correspondence between both methods.
The results of the UV/Vis determination of K are given
in Table 1. The association constants calculated by both
methods A and B are self-consistent considering the pre-
cision of the results. The use of method B allowed us to
determine the association constant of complex 3h. There-
fore, method B appears more satisfactory than method A,
as it can overcome the problem of recovery of the character-
istic charge-transfer bands of the complex with one of the
starting materials. A comparison of the results obtained
1
from H NMR spectroscopy with the spectrophotometric
data shows a greater incertitude with the latter. However,
conclusions on the behavior of guests is quite similar what-
ever the method concerned; the association constant of 3a
remained the strongest, provided that the inversion of the
K values of 3a and 3b by using the Benesi–Hildebrand
method was ignored.
Crystals of complex 3i suitable for X-ray diffraction
analysis were obtained by slow evaporation of dichloro-
methane at room temperature from an ethanol/dichloro-
Synthesis of New Molecular Tweezers from (1R,2R)-1,2-
Cyclohexanediamine
With a centroid–centroid distance around 7.4Å and
methane (1:1) solution (Figure 3). X-ray crystal analysis of 14.23° as angle formed between the planes of the two aro-
3i shows that 4-chloro-7-nitro-2,1,3-benzoxadiazole (NBD; matic rings, molecular tweezer 1 has good characteristics
2i) is sandwiched between the two usnic acid units, as re- for the construction of molecular assemblies. However, we
ported for complex 3a.[4] The centroid distances in 3i be- wanted to determine whether tweezer 1 can be simplified
tween adjacent usnic acids and 2i are 3.701 and 3.619 Å. by replacing the usnic acid by a residue such as acetophen-
These values are consistent with those expected for π–π- one or benzaldehyde derivatives. Consequently, several im-
stacking interactions. While the centroid distance between portant points have to be checked (Figure 4): (1) By chang-
each aromatic ring of pincers is 7.486 Å in 1,[4] the distance ing the mode of bonding between the spacer and the pincer
in complex 3i decreases to 7.247 Å and to 6.900 Å in com- (an imine function instead of the keto enamine system in
Eur. J. Org. Chem. 2010, 5503–5508
© 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjoc.org
5505

P. van de Weghe et al.
FULL PAPER
the case of usnic acid) the stability of the resulting deriva- Table 2. Synthesis of 6 and values of association constants K with
TNF (2a).
tives could be different. Is the presence of an o-hydroxy
function to stabilize the imine function by a hydrogen bond
essential? (2) The use of acetophenone and benzaldehyde
derivatives as pincers can lead to compounds that have sim-
ilar structures to the salen ligands.[14] Which elements favor
spatial arrangement of the aromatic groups that lead to a
tweezer and not a salen-type system?
Entry
6
Yield [%]
Tweezer[a] K [–1][b] (∆δmax, Hz)
1
2
3
4
5
6
7
8
9
6a
6b
6c
6d
6e
6f
6g
6h
6i
61
quant.
85
no
no
no
no
yes
–
–
–
–
–
85
78
7Ϯ0.4 (206Ϯ8)
–
5.9Ϯ0.5 (392Ϯ28)
8Ϯ0.4 (502Ϯ18)
24Ϯ1 (592Ϯ15)
n.r.[c]
45
yes
yes
yes
87
86
[a] Evaluation by 1H NMR spectroscopy. [b] Determined by 1H
NMR spectroscopy (500 MHz) at 298 K, 1:1 stoichiometry of the
complexes in CD2Cl2. [c] No reaction.
In order to determine if products 6 can act as tweezers
in the presence of aromatic derivatives we developed a rapid
test by using H NMR spectroscopy. By adding 1 equiv. of
Figure 4. Conception of potential molecular tweezers.
1
TNF (2a; corresponding to a concentration of 2 m) to
30 equiv. of 6, the formation of a molecular assembly could
be monitored by upfield or downfield shifts of the proton
signals of 2a.
In order to prepare these new potential tweezers, several
aromatic ketones and aldehydes bearing a phenol function
or a methyl group in the ortho position were selected.
(1R,2R)-1,2-Diaminocyclohexane (4) in the presence of
these carbonyl derivatives 5 (2 equiv.) in methanol at room
temperature or reflux afforded expected imines 6 in moder-
ate to excellent yields (Scheme 2 and Table 2). The presence
of the hydroxy group in the ortho position of the carbonyl
seems to be essential for the formation and for the stability
of 6, as aldehyde 5f[15] bearing a methoxy group instead of
a phenol function did not lead to substrate 6f. 2-Acetyl-1,3-
indanedione (5c) was also engaged in this study due to the
formation of a keto enamine bond with the diaminocyclo-
hexane that is quite similar to the tweezer obtained from
usnic acid.
We first prepared compounds 6a and 6b that have struc-
tures very close to those of the previously reported Schiff
bases obtained from 4 and 2-hydroxyacetophenone and sal-
icylaldehyde.[16] The X-ray structures obtained from these
described compounds showed a centroid–centroid distance
of 7.20 and 6.51 Å, respectively, but the angle formed be-
tween the planes of the two aromatic rings is 83.13° for the
first substrate and 56.46° for the second. These very open
angles are probably the reason[17] why compounds 6a and
6b do not seem to give molecular assemblies with 2a
(Table 2, Entries 1 and 2). 2-Acetyl-1,3-indanedione (5c)
combined with 4 afforded in good yield expected product
6c, for which crystals suitable for X-ray diffraction were ob-
tained. Compound 6c did not seem to form a host–guest
assembly, as evidenced by the NMR test and foreshadowed
by the data obtained from X-ray analysis of 6c (the
centroid–centroid distance is 8.569 Å and the angle formed
between the planes of the two aromatic rings is 71.40°). This
was confirmed by the isolation of crystals formed from an
equimolar mixture of 6c/2a, whose X-ray structure shows
that 2a is not included in 6c (Figure 5).
Figure 5. X-ray structures of compound 6c (A) and mixture of 6c/
2a (B).
Scheme 2. Synthesis of new potential molecular tweezers 6.
5506
www.eurjoc.org
© 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Org. Chem. 2010, 5503–5508

Molecular Tweezers
We were also interested in forming potential tweezers sembly 2i@6i with 4Ϯ1 –1 (∆δmax = 986Ϯ129 Hz) as as-
1
from naphthyl derivatives to increase the surface of the pin- sociation constant at 298 K determined after H NMR ti-
cers so that they could better interact with 2a to improve tration.
the formation of the molecular assemblies. 2-Hydroxy-
naphthalene-1-carbaldehyde (5d), in the presence of 4, led
to 6d in good yield and afforded crystals suitable for X-
ray analysis (Figure 6). The aromatic moieties are properly
oriented, and the centroid–centroid distance is 7.728 Å
(view A), but the angle between the planes of the two aro-
matic rings is wide open (84.48°) and the aromatic overlap-
ping is limited to only one ring (view B), which probably
results in a lack of interaction with 2a as shown by 1H
NMR spectroscopy. To obtain maximum overlap of the
aromatic rings, we prepared a series of 3-hydroxynaphth-
Figure 7. X-ray structure of tweezer 6i.
alene-2-carbaldehyde derivatives. Schiff bases 6e, 6g, and 6h
were obtained from 5e,[15] 5g,[18] and 5h,[19] respectively, af-
ter reaction with 4 in methanol. The 1H NMR test re-
Conclusions
sponded positively, suggesting the formation of molecular
assemblies of type “host–guest” between 6e, 6g, and 6h and
2a. The association constants were determined by 1H NMR
titration experiments and were found to be around 7 –1
(Table 2, Entries 5, 7, and 8). These results confirmed the
importance of the spatial arrangement of the aromatic rings
and the maximization of their overlap to obtain a molecular
tweezer. Moreover, we learnt, by comparing 6e and 6g, that
the tweezer can be prepared either from an aromatic deriva-
tive bearing a 2-hydroxy carbaldehyde or a methyl ketone
without any alteration of the spatial arrangement of the
aromatic rings.
In summary, we demonstrated that chiral molecular
tweezer 1 prepared from (+)-usnic acid and (1R,2R)-1,2-
diaminocyclohexane (4) forms host–guest complexes with a
variety of aromatic compounds. The association constants
of these molecular assemblies were determined by two dif-
ferent methods (1H NMR and UV/Vis spectroscopy) and
showed good correlation between the values. We then aimed
to design new chiral molecular tweezers by using 4 as a
spacer and various ortho-hydroxybenzaldehyde and aryl
ketone derivatives. This work enabled us to determine the
elements necessary for the formation of tweezers: (1) ortho-
Hydroxybenzaldehyde and arylketone derivatives are re-
quired to have a stable imine function. (2) The centroid–
centroid distance should be around 7–8 Å. (3) To maximize
aromatic overlap, the angle formed between the planes of
the two aromatic pincers should be close to 0° (in compari-
son, this angle is 14.2° in tweezer 1 and 4.7° in molecular
assembly 3a).
Although the association constants are low, the possibi-
lity of using functionalized guests as esters allows the use
of this chiral molecular tweezer as a new tool for chiral
recognition of chiral guests to be foreseen. Investigations in
the field of chiral discrimination are currently in progress[21]
as is the synthesis of other chiral molecular tweezers by
varying either the spacer or the pincers and their use. To
have an explanation of aromatic host–guest interactions
with our system, electrostatic potential surfaces (EPS)[22]
are underway and will be reported in due course.
Figure 6. X-ray structure of compound 6d.
These last results are particularly encouraging, and to
reinforce these observations, 10-hydroxyphenanthrene-9-
carbaldehyde (5i)[20] was prepared and converted into chiral
tweezer 6i after stirring with 4. Crystals suitable for X-ray
diffraction analysis were obtained. Although the X-ray
structure shows that the aromatic rings are not perfectly
parallel, 6i meets the criteria for giving an efficient molecu-
lar tweezer (Figure 7). A complex 2a@6i was obtained by
mixing molecular tweezer 6i with 2a, and the association
constant determined by 1H NMR spectroscopy was
24Ϯ1 –1 (Table 2, Entry 9). So far, crystals of 2a@6i suit-
able for X-ray diffraction analysis have not been obtained
to confirm that 2a is sandwiched between the two aromatic
moieties. The same tweezer 6i was also placed in the pres-
Experimental Section
General Procedure for the Synthesis of Potential Tweezers: To a
solution of 1,2-diaminocyclohexane in methanol (3 mL/0.5 mmol
diamine) was added the carbonyl derivative (2 equiv.). The mixture
was stirred at room temperature or heated to reflux until complete
disappearance of the starting material was observed. The product
of reaction was isolated either by filtration of the reaction mixture
and crystallization (for 6c, 6e, 6g, 6h, and 6i) or after evaporation
of the solvent and purification on silica gel with CH2Cl2/MeOH
ence of electron-deficient NBD (2i) to afford molecular as- (95:5) (for 6a, 6b, and 6d).
Eur. J. Org. Chem. 2010, 5503–5508
© 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjoc.org
5507

P. van de Weghe et al.
FULL PAPER
[8] A. E. Rowan, J. A. A. A. Elemans, R. J. M. Nolte, Acc. Chem.
Res. 1999, 32, 995–1006.
CCDC-755089 (for 3i), -731096 (for 6c), -711952 (for 6d), and
-765948 (for 6i) contain the supplementary crystallographic data
for this paper. These data can be obtained free of charge from the
Cambridge Crystallographic Data Center via www.ccdc.cam.ac.uk/
data_request/cif.
[9] H. Benesi, H. Hildebrand, J. Am. Chem. Soc. 1949, 71, 2703–
2707.
[10] B. K. Seal, A. K. Mukherjee, D. C. Mukherjee, P. G. Farrell,
J. V. Westwood, J. Magn. Res. 1983, 51, 318–322.
Supporting Information (see also the footnote on the first page of
this article): Details of the calculation of association constants. De-
tailed description of all experimental procedures, synthesis, and an-
alytical data for all compounds.
[11]
W. Person, J. Am. Chem. Soc. 1965, 87, 167–170.
[12] F. H. Stootman, D. M. Fisher, A. Rodger, J. R. Aldrich-Wright,
Analyst 2006, 131, 1145–1151.
[13]
F.-G. Klärner, B. Kahlert, A. Nellesen, J. Ziemau, C. Oschen-
feld, T. Schrader, J. Am. Chem. Soc. 2006, 128, 4831–4841 and
Supporting Information for calculations.
[14]
[15]
P. G. Cozzi, Chem. Soc. Rev. 2004, 33, 410–421.
K.-C. Wu, Y.-S. Lin, Y.-S. Yu, C.-Y. Chen, M. O. Ahmed, P.-
T. Chou, Y.-S. Hon, Tetrahedron 2004, 60, 11861–11868.
Acknowledgments
Financial support by the Université de Rennes 1, the Région Bret-
agne, and Rennes Métropole are gratefully acknowledged.
[16] J. C. Cannadine, J. P. Corden, W. Errington, P. Moore,
M. G. H. Wallbridge, Acta Crystallogr., Sect. C 1996, 52, 1014–
1017.
[1] a) C. A. Hunter, K. R. Lawson, J. Perkins, C. Urch, J. Chem.
Soc. Perkin Trans. 2 2001, 651–669; b) E. A. Meyer, R. K. Cas-
tellano, F. Diederich, Angew. Chem. Int. Ed. 2003, 42, 1210–
1250; c) A. Tewari, R. Dubey, Bioorg. Med. Chem. 2008, 16,
126–143; d) C. Bissantz, B. Kuhn, M. Stahl, J. Med. Chem.
2010, 53, 5061–5084.
[17] The conformational organization of the compounds obtained
by X-ray (the distances between the aromatic rings, angles)
does not necessarily reflect the conformation in solution; how-
ever, it can give valuable insights into the design of new tweez-
ers.
[18] J. Einhorn, P. Demerseman, R. Royer, R. Cavier, Eur. J. Med.
Chem. 1983, 18, 175–180.
[19] R. Royer, J. P. Buisson, Eur. J. Med. Chem. 1980, 15, 275–278.
[20] S. H. Alarcon, A. C. Olivieri, G. R. Guillermo, R. M. Cravero,
M. Gonzalez-Sierra, Tetrahedron 1995, 51, 4619–4626.
[21] B. Legouin, M. Gayral, P. Uriac, S. Tomasi, P. van de Weghe,
Tetrahedron: Asymmetry 2010, 21, 1307–1310.
[22] S. E. Wheeler, K. N. Houk, J. Chem. Theory Comput. 2009, 5,
2301–2312 and references cites therein.
[2] C.-W. Chen, H. W. Whitlock, J. Am. Chem. Soc. 1978, 100,
4921–4922.
[3] F.-G. Kläner, B. Kahlert, Acc. Chem. Res. 2003, 36, 919; M.
Harmata, Acc. Chem. Res. 2004, 37, 862–873.
[4] B. Legouin, P. Uriac, S. Tomasi, L. Toupet, A. Bondon, P.
van de Weghe, Org. Lett. 2009, 11, 745–748.
[5] L. Fielding, Tetrahedron 2000, 56, 6151–6170.
[6] K. A. Connors, Binding Constants – The Measurement of Mo-
lecular Complex Stability, Wiley, New York, 1987.
[7] P. Job, Ann. Chim. App. 1928, 9, 113–203.
Received: May 20, 2010
Published Online: August 16, 2010
5508
www.eurjoc.org
© 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Org. Chem. 2010, 5503–5508