Recognition of ionic guests by ionic β-cyclodextrin derivatives

Article abstract of DOI:10.1002/chem.200800295

Inclusion compounds of cationic, anionic, and neutral p-substituted derivatives of feri-butylbenzene complexed in β-cyclodextrin and its ionic 6-mono and 6-hepta derivatives were systematically investigated by isothermal titration calorimetry (ITC). All inclusion compounds showed 1:1 stoichiometry with binding constants ranging from 10 to 3 × 10⁶ M ^-1. The binding free energies could be subdivided into apolar and electrostatic contributions. The electrostatic interactions could be quantitatively described by Coulomb's law by taking into account the degree of protonation of hosts and guests, the orientations of the guests within the hosts, and ion shielding as described by the Debye-Hueckel-Onsager theory. The orientations of the guests within the cyclodextrin cavities were determined by ROESY NMR spectroscopy.

Full text of DOI:10.1002/chem.200800295

DOI: 10.1002/chem.200800295
Recognition of Ionic Guests by Ionic b-Cyclodextrin Derivatives
[a]
GerhardWenz,* Christian Strassnig,[a] Carolin Thiele,[a] Annegret Engelke,[a]
[b]
[b]
BerndMorgenstern, andKaspar Hegetschweiler
Abstract: Inclusion compounds of cat-
ionic, anionic, and neutral p-substituted
derivatives of tert-butylbenzene com-
plexed in b-cyclodextrin and its ionic 6-
mono and 6-hepta derivatives were sys-
tematically investigated by isothermal
titration calorimetry (ITC). All inclu-
sion compounds showed 1:1 stoichiom-
etry with binding constants ranging
from 10 to 3”106 mÀ1. The binding free
energies could be subdivided into
apolar and electrostatic contributions.
The electrostatic interactions could be
quantitatively described by Coulombꢀs
law by taking into account the degree
of protonation of hosts and guests, the
orientations of the guests within the
hosts, and ion shielding as described by
the Debye–Hückel–Onsager theory.
The orientations of the guests within
the cyclodextrin cavities were deter-
mined by ROESY NMR spectroscopy.
Keywords:
cyclodextrins · host–guest systems ·
molecular recognition
supramolecular chemistry
calorimetry
·
·
Introduction
sion of cosmetics,[15] catalysis,[16] and chromatographic sepa-
rations.[17,18]
Design, synthesis, and characterization of highly specific ar-
tificial receptors are among the main tasks of supramolec-
ular chemistry.[1–4] Among other hosts, cyclodextrins (CDs),
which are 1!4 a-linked cyclic oligomers of anhydrogluco-
pyranose, have attracted particular interest since they are
readily available and soluble in aqueous media.[5] All CDs
have an internal cavity that is mainly hydrophobic and
allows the binding of hydrophobic guest molecules. A large
number of hydrophobic or amphiphilic guests can be com-
plexed inside this cavity; indeed, binding data are available
for numerous host–guest complexes, known as inclusion
compounds, involving CDs.[6] Inclusion of guests in CDs and
CD derivatives facilitates many applications,[7–10] such as sol-
ubilization[11] and targeting of pharmaceuticals,[12–14] disper-
The driving forces for the formation of CD inclusion com-
pounds are mainly nondirectional interactions such as hy-
drophobic and van der Waals interactions.[19,20] The domi-
nance of solvatophobic interactions is evident in the fact
that the inclusion of guests in CDs occurs preferentially in
aqueous solution. The addition of small amounts of organic
solvents to aqueous solutions is enough to render inclusion
compounds significantly less stable.[19] The magnitude of hy-
drophobic interactions is determined mainly by the hydro-
phobic surface area of the guest. Binding free energies DG8
of b-CD inclusion compounds involving several homologous
series of guests become more negative by 2.8Æ0.6 kJmolÀ1
per methylene group.[6,21] We investigated the influence of
para substituents at benzoic acid on the stabilities of b-CD
inclusion compounds and found that binding free energy
DG8 decreased significantly from À6.9 kJmolÀ1 for benzoic
acid to À24.3 kJmolÀ1 for p-tert-butylbenzoic acid, that is,
4.3 kJmolÀ1 per C atom.[22] Thus, a benzene ring alone
seems to be too small to fill the b-CD cavity. The additional
tert-butyl group elongates the hydrophobic part of the guest
and makes it large enough to fill the b-CD cavity complete-
ly.
[a] Prof. Dr. G. Wenz, Dr. C. Strassnig, C. Thiele, A. Engelke
Organische Makromolekulare Chemie
Saarland University, Geb. C4.2
D-66123 Saarbrücken (Germany)
Fax : (+49)681-302-3909
E-mail: g.wenz@mx.uni-saarland.de
[b] Dr. B. Morgenstern, Prof. Dr. K. Hegetschweiler
Anorganische Chemie
Additional polar groups must be attached to the CD ring
to improve binding selectivity. A great variety of CD deriva-
tives carrying one or more substituents at the primary or
secondary positions has been synthesized.[23] Many of these
CD derivatives showed improved molecular recognition. For
Saarland University, Geb. C4.1
D-66123 Saarbrücken (Germany)
Supporting information for this article is available on the WWW
under http://dx.doi.org/10.1002/chem.200800295.
7202
ꢁ 2008 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2008, 14, 7202 – 7211

FULL PAPER
example, 6-monoamino-b-CD and 6-O-carboxymethyl-b-CD
show stronger chiral selectivity for amino acids than native
b-CD.[24–27] Disubstituted b-CD derivatives performed even
better.[28] The “three-point rule” states that at least three di-
rected interactions between host and guests are prerequi-
sites for noticeable chiral discrimination.[29] In addition,
hepta-6-,[24] hepta-2,3-di-,[30] hepta-2,6-di-,[31] and hepta-2,3,6-
trisubstituted[32–34] b-CD derivatives have been synthesized.
In particular, hepta-6-S-6-deoxy-b-CD derivatives,[35–38]
which are readily synthesized from heptakis-6-iodo-6-deoxy-
b-CD and functional thiols, show interesting binding proper-
ties. They can form supramolecular structures including mo-
lecular capsules[39,40] and nanotubes.[31] Recently, we demon-
strated the superior binding properties of hepta-6-S-6-
deoxy-b-CD derivatives towards the drug camptothecin,
which is used for cancer treatment.[38]
The contribution of Coulomb interactions to the binding
of charged drug molecules by statistically substituted sulfo-
butyl ether b-CD derivatives has already been demonstrat-
ed.[41] Although extensive binding data are available on
these and other b-CD derivatives, a quantitative understand-
ing of specific interactions between charged hosts and guests
is still lacking.[6,27,42] This understanding is very important
for the design of highly selective artificial receptor mole-
cules.[43,44]
reactions starting from 6-O-tosyl-b-CD[45] (Scheme 1). 6-
Amino-6-deoxy-b-CD (2) was synthesized via 6-azido-6-
deoxy-b-CD.[46] Reactions of 6-O-tosyl-b-CD with 2-amino-
A
deoxy-6-(2-aminoethylsulfanyl)]-b-CD (4) and mono-[6-
deoxy-6-(2-sulfanyl acetic acid)]-b-CD (6), as described re-
cently.[38]
In addition, both amino derivatives 2 and 4 were convert-
ed to the corresponding guanidinium salts 3 and 5 by reac-
tion with 1H-pyrazole-1-carboxamidine hydrochloride, in
analogy to a published procedure.[47,48] Analogous cationic
(7, 8), anionic (9–13), and neutral (14) heptasubstituted b-
CD derivatives were synthesized by nucleophilic displace-
ment reactions of heptakis(6-iodo-6-deoxy)-b-CD with NaN3
followed by Staudinger reduction,[49] and with functional
thiols or thiosulfate (Scheme 2) by a previously described
procedure.[38] We were unable to completely convert hepta-
A
dinium compounds.[48] For the synthesis of cationic guests
(Scheme 3), 4-tert-butyl-aniline was protonated with HCl to
give 15, which was converted by reaction with 1H-pyrazole-
1-carboxamidine hydrochloride to guanidinium derivative
16, and by reaction with methyl iodide to trimethylammoni-
um iodide 18. Anionic sulfonate 20 was obtained by sulfona-
tion of tert-butylbenzene, and neutral N-oxide 19 was syn-
We have now investigated the molecular recognition of
cationic, anionic, and neutral
tert-butylbenzene derivatives by
a series of 13 charged mono-6-
and hepta-6-substituted b-CD
derivatives. The tert-butylphen-
yl group was chosen because of
its high binding affinity towards
the b-CD cavity.[22] For a quan-
titative description of the host–
guest systems we collected a
broad array of binding data, in
addition to data on the proto-
nation equilibria of hosts and
guests and data on the orienta-
tions of the guests within the
hosts. These data allowed us to
distinguish between apolar
binding, due to hydrophobic
and van der Waals interactions
within the cavity, and electro-
static
interactions
between
functional groups on host and
guest.
Results
Synthesis of hosts andguests :
Monosubstituted derivatives 2–
6 of b-CD (1) were synthesized
Scheme 1. Synthesis of mono substituted b-CD derivatives. a) TosCl, NaOH, ACN/H2O; b) NaN3, H2O;
c) PPh3, DMF, NEt3; d) 1H-pyrazole-1-carboxamidine hydrochloride, DMF; e) 2-aminoethanthiol, NH4HCO3,
by nucleophilic displacement DMF/H2O; f) 1. methyl ester of 2-mercaptoacetic acid, NEt3, DMF; 2. NaOH.
Chem. Eur. J. 2008, 14, 7202 – 7211
ꢁ 2008 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
7203

G. Wenz et al.
groups of host 7, which leads to
significant decay of pKa values
with increasing degree of proto-
nation. This repulsive effect is
less pronounced for host 8,
since the cationic groups are
farther apart from each other
due to the SCH2CH2 spacers.
Guest 19 is neutral at pH 6.8
and cationic at pH 3.0.
Determination of binding ther-
modynamics between hosts and
guests: Since all hosts 1–14,
guests 15–21, and their inclu-
sion compounds are extremely
soluble in water, isothermal ti-
tration microcalorimetry was
used to determine all thermo-
dynamic binding parameters.
Titrations
were
performed
under standard conditions, nor-
mally at pH 6.8 with 50 mm
buffer. Hosts
7 and 8 with
seven amino groups were titrat-
ed with charged guests 16 and
20 under acidic conditions
(pH 3.0) to force complete pro-
tonation of all amino groups.
All titration curves were in
good agreement with a 1:1 stoi-
chiometry of host/guest in the
inclusion compounds. Binding
constants K, molar binding free
energies DG8, and molar bind-
ing enthalpies DH8 (Tables 2
and 3) were obtained by nonlin-
ear regression of the titration
curves. Although the hydropho-
bic parts of all guests 15–21 are
the same, binding constants
varied over five orders of mag-
nitude, from 20 to 3”
Scheme 2. Synthesis of heptasubstituted CD derivatives. a) 1. PPh3, I2, DMF; 2. CH3ONa, CH3OH; b) 1. NaN3,
DMF; 2. PPh3/DMF, NH3 or Raney Ni/H2/H2O; c) 2-aminoethanethiol, NH4HCO3, DMF/H2O; d) 1. methyl
ester of 2-mercaptoacetic acid, NEt3, DMF; 2. NaOH; e) 1. methyl ester of 2-mercaptopropanoic acid, NEt3,
DMF; 2. NaOH; f) 1. methyl ester of 2-mercaptoacetic acid, NEt3, DMF; 2. NaOH; g) 1. 2-mercaptoethanesul-
fonate, NEt3, DMSO; 2. NaOH; h) 1. Na2S2O3, NEt3, DMSO; 2. NaOH. i) 3-mercaptopropane-1,2-diol, NEt3,
DMF.
Scheme 3. Amphiphilic guests for b-CD and its derivatives.
thesized by H2O2 oxidation of commercially available terti-
ary amine 17.
106 LmolÀ1. In each case, conditions for the measurement
were optimized to achieve maximum accuracy. The DG8
values showed only a slight dependence on pH for those
cases in which the protonation states of host and guest were
The protonation constants of polyamines 7 and 8, poly-
carboxylates 9 and 11, and amine oxide 19 were determined
in aqueous solution by potentiometric titration. The pKa
values are listed in Table 1, and a representative species dis-
tribution diagram for polyamine 7 is shown in Figure 1.
From these distribution curves the net charge numbers of a
host zh or guest zg were calculated for our standard pH of
6.8 by summation of the contributions of the relevant
charged species. The charge numbers zh are listed in Table 4
below. Interestingly, host 7 is not fully protonated at pH 6.8.
The low charge number of zh =6.2 was attributed to Cou-
lomb repulsion between adjacent cationic ammonium
Table 1. pKa values derived from potentiometric titration.
[a]
Compound
pKa
7
8
9
11
19
9.50(1), 8.89(1), 8.33(1), 8.07(1), 7.57(1), 7.35(1), 6.75(1)
9.99(1), 9.45(1), 9.05(1), 8.72(1), 8.32(1), 7.96(1), 7.37(1)
6.10(2), 5.32(2), 4.91(3), 4.37(3), 4.04(3), 3.50(2), 3.01(2)
6.49(1), 5.69(1), 5.25(1), 4.75(1), 4.31(1), 3.78(1), <3
4.30(1)
[a] pKa,i (=ÀlogKa,i, Ka,i =[HnÀiL]”[H]”[HnÀi+1L]À1, 25.08C, 0.10m KCl).
Uncertainties (3s) are given in parentheses.
7204
www.chemeurj.org
ꢁ 2008 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2008, 14, 7202 – 7211

Guest Recognition by Ionic b-Cyclodextrins
FULL PAPER
N-oxide 19 (pKa =4.3) showed a significantly higher com-
plex stability at pH 6.8 than at pH 3: DG8=À22.27 kJmolÀ1
versus À21.06 kJmolÀ1, respectively (Table 2). This reflects
the fact that it changes from neutral at pH 6.8 to cationic at
pH 3.
Influence of the functional groups on guests 15–21 on bind-
ing free energy DG8: Although native b-CD 1 is uncharged,
it can recognize the charge of a guest, as shown in Table 2
and Figure 2. The binding free energies of the anionic guests
Figure 1. Distribution of protonated species for host 7 (L) as a function
of pH, as determined by potentiometric titration. Calculations were car-
ried out with the deprotonation constants listed in Table 1.
Table 2. Molar binding free energies DG8 for the inclusion of various
guests in b-CD 1 and in 6-amino-6-deoxy-b-CD 2 as determined by ITC
at T=258C.
Guest
pH
DG8 [kJmolÀ1] in host 1
DG8 [kJmolÀ1] in host 2
15
16
17
18
19
19
20
20
21
3.0
6.8
3.0
6.8
6.8
3.0
6.8
3.0
6.8
À22.03Æ0.03
À21.71Æ0.03
À22.08Æ0.04
À22.25Æ0.03
À22.27Æ0.04
À21.06Æ0.02
À23.19Æ0.03
À23.12Æ0.05
À24.03Æ0.07
À19.29Æ0.05
À19.75Æ0.04
À19.30Æ0.20
À20.11Æ0.05
À19.92Æ0.03
À18.64Æ0.02
À22.99Æ0.02
À23.25Æ0.11
À23.41Æ0.15
Figure 2. Molar binding free energies for b-CD (light gray) 1 and mono-
6-amino-b-CD 2 (dark gray) for various guest molecules 15–21; pH
values as specified in Table 2.
Table 3. Molar binding free energies DG8 for the inclusion of cationic
guest 16, anionic guest 20, and neutral guest 19 in various CD derivatives,
as determined by ITC at T=258C.
21 (DG8=À24.03 kJmolÀ1) and 20 (DG8 À23.19 kJmolÀ1)
were more negative than those of cationic guests 16 (DG8
À21.71 kJmolÀ1) and 15 (À22.03 kJmolÀ1), even though the
strength of hydrophobic interactions between host and guest
should be very similar in both cases. The value of the bind-
ing free energy of neutral guest 19 in host 1 fell in the
middle of the range between the values of the cationic and
anionic guests.
Charge recognition was even more pronounced for cation-
ic b-CD derivative 2, as shown in Table 2. Again, all cationic
guests 15–18 showed similar binding free energies for host 2.
The DG8 values of the anionic guests 21 and 20 were similar
as well, but much more negative (by DDG8~4 kJmolÀ1)
than the values for the cationic guests. Since the DG8 values
were rather similar for various guests of the same charge,
further ITC measurements were performed with three rep-
resentative guests rather than the entire panel. Because of
their greater solubility in water, guests 16, 19, and 20 with
guanidinium, amine N-oxide and sulfonate groups, respec-
tively, were chosen as cationic, neutral, and anionic refer-
ence compounds for investigation of polar interactions be-
tween hosts and guests (see Table 3). The DG8 values of the
Host
pH
DG8
[kJmolÀ1
DG8
[kJmolÀ1
DG88
[kJmolÀ1
]
G
]
G
]
N
with guest 16
with guest 20
with guest 19
1
2
3
4
5
6
7
8
6.8
6.8
6.8
6.8
6.8
6.8
3.0
3.0
6.8
6.8
6.8
6.8
6.8
6.8
À21.71Æ0.03
À19.75Æ0.04
À20.80Æ0.03
À22.30Æ0.03
À22.60Æ0.12
À23.05Æ0.05
À8.93Æ0.08
À23.10Æ0.04
À34.57Æ0.24
À32.39Æ0.24
À32.73Æ0.08
À32.89Æ0.21
À26.10Æ0.08
À26.57Æ0.05
À23.19Æ0.03
À19.4Æ0.04
À24.34Æ0.03
À24.86Æ0.05
À24.42Æ0.02
À22.23Æ0.13
À22.55Æ0.06
À37.27Æ0.23
À21.65Æ0.05
À20.17Æ0.03
À21.68Æ0.04
À21.00Æ0.07
À14.86Æ0.06
À25.00Æ0.03
À22.27Æ0.04
À19.92Æ0.04
À21.21Æ0.04
À22.28Æ0.05
À22.61Æ0.05
À22.36Æ0.04
À10.66Æ0.17
À25.57Æ0.09
À25.61Æ0.04
À23.23Æ0.03
À24.60Æ0.09
À25.61Æ0.04
À18.40Æ0.11
À25.61Æ0.07
9
10
11
12
13
14
not influenced by pH. The DG8 values of the inclusion com-
plex of tert-butylbenzenesulfonic acid 20 and 1 were the
same for pH 6.8 and pH 3. On the other hand, when the
protonation state of the guest was influenced by pH, DG8
varied considerably with pH. tert-Butyl-N,N-dimethylaniline
Chem. Eur. J. 2008, 14, 7202 – 7211
ꢁ 2008 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
7205

G. Wenz et al.
Table 4. Selectivities of hosts 1–14 as a function of spacer distance r and
net charge zh.
charged guests 16 and 20 have a much broader range (À9 to
À37 kJmolÀ1) than those of neutral guest 19 (À11 to
À26 kJmolÀ1).
[c]
Host
DDG8
[kJmolÀ1
DDG8
[kJmolÀ1
rS
zh
[a]
[a]
G
]
A
]
[nm][b]
with guest 16
with guest 20
Influence of functional groups on hosts 1–14 on apolar bind-
ing: The binding free energies DG88 for neutral guest 19 at
pH 6.8 was mainly attributed to apolar binding due to hy-
drophobic and van der Waals interactions between the guest
and the b-CD cavity. Apolar binding free energies DG88 of
hosts 1–14 are summarized in Table 3. Heptakis-6-deoxy-6-
thioether derivatives 8–12 and 14 showed the highest DG88
values of around À25 kJmolÀ1, followed by mono thioethers
4–6 (DG88ꢀÀ22 kJmolÀ1), which have values close to that
of native b-CD (DG88=À22.3 kJmolÀ1). Much smaller
values were observed for the mono-6-deoxy-6-ammonium 2
1
2
3
4
5
6
7
8
À0.92
À3.07
À3.13
À2.58
À1.82
À0.69
À11.89
À11.70
À8.96
À9.17
À8.14
À7.28
À7.70
À0.96
0.56
0.17
0.42
0.00
0.00
0.13
1.73
2.47
3.96
3.06
2.92
4.62
3.54
0.62
–
0
0.147
0.250
0.544
0.657
0.537
0.233
0.544
0.537
0.537
0.663
0.716
0.459
–
1
1
1
1
À1
6.2
6.8
À6.8
À7
9
10
11
12
13
14
À6.6
À7
À7
and mono-6-deoxy-6-guanidinium derivatives
3
(DG88ꢀ
0
À21 kJmolÀ1), and especially for hepta-6-deoxy-6-amino de-
rivative 7 (DG88 ꢀÀ11 kJmolÀ1). These different binding
free energies can be correlated with the hydrophobicity of
the atoms attached to the C-6 positions on the b-CD scaf-
fold of the hosts. The atomic increments for the calculation
of logP values (where P is the octanol/water partition coef-
ficient) were taken as measures of these hydrophobicities:
OH, À1.4; S, À0.4; NH3+, À4.6.[50,51] They clearly reflect the
greater hydrophobicity of the sulfur linker and the much
lower hydrophobicity of the ammonium group compared to
the hydroxyl group. The influence of hydrophilic end groups
on the host seems to level off with increasing spacer length,
since binding free energies of derivatives 8, 11, 12, and 14
with C2 spacers are very similar despite their different end
groups. Consequently, only atoms close to the hydrophobic
part of the guest contribute to apolar binding.
[a] Contributions of electrostatic interactions to binding free energy.
[b] Spacer length. [c] Charge of the host.
secondary rim of b-CD. As a result, the increased distance
between the charged groups reduces the electrostatic repul-
sion energy. Measurements of the NOE were performed to
test this hypothesis.
DDGꢁ ¼ DGꢁÀDGꢁꢁ
ð1Þ
Influence of functional groups on hosts 1–14 on the orienta-
tion of the guest molecules: The orientation of the guest
molecules in the CD host were determined by ROESY
NMR spectroscopy. In all cases, strong ROESY cross-peaks
were observed between the internal protons of the b-CD
scaffold (H-3, H-5, H-6) and the two aromatic protons of
guests 15–21, that is, the hydrophobic portion of all guests
was completely included within the CD host, except for in-
clusion compound 7·15. Assignments of the NMR signals of
the inclusion compounds of monosubstituted b-CD deriva-
tives 2–6 were difficult in many cases because of the com-
plexity of the host spectra caused by the missing C7 symme-
try. In the case of the cationic guest 15 in b-CD, the signals
of H-3 and H-5 overlapped because they were shifted due to
complexation; as a result, it was impossible to determine the
molecular orientation. However, unambiguous results were
obtained for homologous guest 15a, which should be similar
to guest 15.
The relative ROESY intensities of six inclusion com-
pounds as measures of the interatomic distances are shown
in Figure 3. The orientation of a guest was defined on the
basis of the ROESY intensities between outer protons H-6
and H-3 of the host and the outer protons of the guest, H-
tert-butyl, and the protons ortho to the functional group, H-
o. As a convention, the CD cone was always depicted with
the primary side up to facilitate comparison of the guest ori-
entations. If the functional group of the guest was pointing
towards the primary side of CD, it was said to be oriented
upwards, and downwards in the opposite case. Upward ori-
entations were found for inclusion compounds 1·20, 4·20,
Quantification of electrostatic interactions between func-
tional groups: Binding data for charged guests 16 and 20
with various hosts 1–14 (see Table 3) were collected to in-
vestigate electrostatic interactions. Binding constants and
binding free energies are strongly dependent on the combi-
nation of functional groups on host and guest. For example,
the binding constant K of cationic guest 16 in cationic host 7
is as low as 21mÀ1, while that of anionic guest 20 in cationic
host 8 is as high as 3.6”106 mÀ1, close to binding constants of
natural receptors.[4]
For a more detailed data analysis, electrostatic interac-
tions DDG8 were estimated for all hosts 1–14 from the dif-
ferences between the binding free energies of the charged
guests (16, 20) and the binding free energy of neutral guest
19 according to Equation (1); they are summarized in
Table 4. Both attractive interactions (negative values) and
repulsive interactions (positive values) were found. Neither
interaction was symmetrical. For each host, for example, 9,
the attractive interaction with DDG8=À8.96 had a higher
absolute value than the repulsive interaction with DDG8=
3.96. This finding may be explained by the fact that the ori-
entation of the guest inside the CD cavity is influenced by
interactions of functional groups. The guest may avoid re-
pulsive forces and orient its functional group towards the
7206
www.chemeurj.org
ꢁ 2008 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2008, 14, 7202 – 7211

Guest Recognition by Ionic b-Cyclodextrins
FULL PAPER
Figure 4. Influence of ionic strength I on a) apolar binding free energy of
neutral guest 19 DG88, and b) contribution of electrostatic interactions to
binding free energy DDG8 of anionic guest 20 for heptacationic host 8.
The curve was calculated according to Debye–Hückel–Onsager theory
[Eqs. (2)–(4) with reff =0.23 nm].
Figure 3. Relative ROESY NMR intensities [%] for various host–guest
complexes. The strongest signal was set to 100%.
7·20, and downward orientations were determined for 1·15a,
12·20, and 7·15. These findings clearly support our previous
hypothesis that attractive interactions lead to upward orien-
tation and repulsive ones to downward orientations.
Discussion
The potential of native b-CD to recognize charged guests
was already suggested in the literature.[52] For instance, the
b-CD inclusion compound of adamantane-1-carboxylate
(DG8À25.7 kJmolÀ1)[53] is more stable than that of adaman-
tane-1-amine·HCl (DG8=À22.5 kJmolÀ1).[54] The preference
of b-CD for anionic guests can be rationalized by electro-
static interactions between the intrinsic dipole moment of b-
CD, parallel to the C7 symmetry axis, and the charge of the
guest. The dipole moment of b-CD was estimated to be 14–
20 D in the crystalline state,[55] and 2.9–3.7 D in solution,[56]
with the partial positive charge situated on the primary side.
This dipole moment should control the orientation of an in-
cluded dipolar guest. Indeed, anionic guests like benzoic
acid, phenol,[57] and tert-butylbenzoic acid are oriented with
their electronegative functional group upwards towards the
primary side of b-CD, whereby dipole–dipole interactions
are maximized.[58] Our ROESY results for b-CD inclusion
compounds support an antiparallel alignment of the dipoles.
Anionic guest 20 is oriented upwards in b-CD, while cationic
guest 15a is oriented downwards. The downward orientation
seems to lead to weaker binding, possibly due to only partial
space filling of the cavity. Nevertheless, these dipole–dipole
Dependence of apolar binding and electrostatic interactions
on salt concentration: To provide an additional test of our
approach of dividing binding free energy into apolar binding
and electrostatic interactions, we investigated the effect of
salt. Differential salt effects on the two types of binding are
conceivable: apolar binding could be strengthened by an in-
crease in salt concentration, while electrostatic interactions
should level off. Indeed, these contrasting effects have been
reported in investigations of the effect of salt on the binding
behavior of statistical b-CD sulfonates.[41] The dependence
of the binding free energies on the ionic strength I of NaCl
was measured for the heptacationic host 8 with neutral
guest 19 and anionic guest 20. As expected, increasing ionic
strength I made the apolar binding free energy DG88 more
negative but the electrostatic part DDG8 less negative
(Figure 4). The influence of ionic strength I on electrostatic
interactions was stronger than on the apolar binding.
Chem. Eur. J. 2008, 14, 7202 – 7211
ꢁ 2008 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
7207

G. Wenz et al.
interactions are small (ca. 1 kJmolÀ1) and are neglected in
the following discussion for the sake of simplicity.
deviation (SD=1.3 kJmolÀ1). The slope of 1.0 within the ex-
perimental error verified our model. The negligible intercept
Orientations of charged guests in charged hosts are pre-
dominantly controlled by Coulomb forces. An attractive in-
teraction (DDG8<0) forces the guest to adopt an upward
orientation, while a repulsive one (DDG8 >0) leads to a
downward orientation. This rule was generalized to all the
inclusion compounds tested when calculating binding ener-
gies.
for Coulomb energy ECoul =0 at DDG8=À
(0.5Æ0.3) kJmolÀ1
G
demonstrates that the specificities DDG8 are due solely to
Coulomb interactions.
NAe2
4pe0er
zhzg
r
eff=lD
ð2Þ
ð3Þ
ð4Þ
ECoul
¼
eÀr
r ¼ r0 þ rs
The measured DDG8 were correlated with Coulomb ener-
gies, calculated according to Equation (2). In these calcula-
tions, zh and zg are the charge numbers of host and guest, r
is the intercharge distance, and er =78.5 the relative dielec-
tric constant of water. The intercharge distance r was esti-
mated by Equation (3) with r0 =0.3 nm equal to the internal
radius of the b-CD cavity[59] for an upwards-oriented guest,
and r0 =1.0 nm equal to the height of the CD cavity for a
downward-oriented guest, and the length of the spacer
group rs (see Table 4) between C-6 and the ionic atom of
hosts 2–13 was estimated by summation of increments for
all bonds. The increments were calculated from the projec-
tions of the bond lengths parallel to the main direction of
the spacer, assuming all-trans conformations (see Supporting
Information). The shielding effect exerted by clouds of op-
positely charged ions surrounding the interacting functional
groups was quantitatively described by Debye–Hückel–Ons-
ager theory.[60] We used a simplified formulation that has al-
ready been used to describe Coulomb interactions of pro-
teins in aqueous salt solution, with the Debye length lD pro-
portional to I1/2 according to Equation (4) at T=298 K.[61] It
was assumed that the buffer ions cannot intrude into the
CD cavity and that formation of an ion cloud is partially
sterically hindered. Therefore we defined an effective length
reff within which ion shielding occurs. This effective length
was approximated by the spacer length rs diminished by one
bond length (reff =rsÀ0.15 nm), which led to the best fit. This
calculated Coulomb energy ECoul was plotted against the ex-
perimental electrostatic interactions DDG8 (Figure 5). Con-
sidering an ionic strength of I=0.2m for the buffer solution,
the correlation was very good (R=0.97) with a low standard
rffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
e0erkBT
2NAe2I
lD
¼
The decay of the electrostatic interaction ÀDDG8 with in-
creasing ion strength I (Figure 4b) was also quantitatively
described by Debye–Hückel–Onsager theory. The best fit
was found for reff =0.27 nm, somewhat less than expected.
The small value of reff was attributed to steric hindrance of
ion-cloud formation within the constricted environment of
the host. On the other hand, apolar binding increases linear-
ly with the square root of ionic strength I. This is, to our
knowledge, the first time this relationship has been report-
ed. It may be due to an increase of the hydrophobic interac-
tion caused by depletion of ions down to a thin layer close
to the hydrophobic surface.[62,63]
In addition, electrostatic interaction enthalpies DDH8
were calculated by taking the binding enthalpies DH8 for
the cationic and anionic guests (16 and 20) and subtracting
the binding enthalpies DH88 of the neutral guest (19). Sur-
prisingly, only a very bad correlation (r<0.5) was found
with the Coulomb energies, which were calculated according
to Equations (2)–(4). We have no explanation for this find-
ing at present.
The well-known enthalpy–entropy compensation is valid
only for the apolar binding free energies for neutral guest
19, as shown in Figure 6. A straight line with reasonable cor-
relation (R=0.93) and standard deviation (SD=
0.8 kJmolÀ1) was found. The slope A=0.5 was smaller than
reported elsewhere.[6] On the other hand, correlation was in-
Figure 5. Coulomb energies ECoul, calculated according to Equations (2)–
(4), as functions of the measured electrostatic interactions DDG8.
Figure 6. Plot of binding enthalpy DH88 versus entropy TDS88 for hosts
1–14 and neutral guest 19.
7208
www.chemeurj.org
ꢁ 2008 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2008, 14, 7202 – 7211

Guest Recognition by Ionic b-Cyclodextrins
FULL PAPER
calcd (%) for C43H73N3O28·HCl·16H2O: C 36.77, H 7.50, N 3.00; found: C
36.50, H 6.48, N 4.47.
significant (R<0.5) for both ionic guests 16 and 20, in agree-
ment with previously reported results.[64] This finding leads
us to conclude that the enthalpy–entropy compensation
principle mainly holds for apolar interactions. Coulomb in-
teractions may influence solvation effects, especially by ions
of the solvent, and thereby lead to more complicated situa-
tions.
N-(6-deoxy-6-b-cyclodextrinylsulfanylethyl)guanidinium chloride (5):
Mono-[6-deoxy-6-(2-aminoethylsulfanyl)]-b-CD 4 (1.48 g, 1.24 mmol) was
dissolved in anhydrous DMF (15 mL). The mixture was stirred for 24 h
at room temperature after addition of Hünigꢀs base (0.21 mL, 1.24 mmol)
and 1H-pyrazole-1-carboxamidine hydrochloride (0.18 g, 1.24 mmol). The
solvent was evaporated and the product precipitated in ethanol to give a
1
white material (1.29 g, 1.04 mmol, 84%). H NMR (500 MHz, D2O/DSS):
d=2.85 (m, 2H; H-7’), 3.29 (m, 2H; H-8’), 3.40–3.51 (m, 21H; H-2/2’/4/
4’/5/5’), 3.63–3.83 (m, 21H; H-3/3’/6a/6b/6a’/6b’), 4.88–4.90 ppm (m, 7H;
H-1/1’); 13C NMR (125.71 MHz, D2O/DSS): d=37.98 (C-7’), 38.50 (C-6’),
41.69 (C-8’), 61.24 (C-6), 73.19 (C-2/C-2’/C-5/C-5’), 74.28 (C-3), 82.38 (C-
Conclusion
4), 85.51 (C-4’), 103.21 (C-1/C-1’), 157.91 ppm (C-9’); IR: n˜ =1020, 1149
Attachment of seven charged substituents at the primary
rim of b-CD gives rise to highly potent host molecules.
Strikingly, electrostatic repulsion between the charged sub-
stituents does not diminish apolar binding due to deforma-
tion of the CD cavity as described elsewhere.[30] This can be
explained by the fact that stabilization by intramolecular hy-
drogen bonds at the secondary rim is retained for our com-
pounds. These so-called flip-flop hydrogen bonds are known
to stabilize the CD scaffold.[65,66]
Electrostatic interactions can be described in good ap-
proximation by Coulombꢀs law combined with Debye–
Hückel–Onsager theory. These simple theories would fail
unless the CD hosts of this study would not be quite rigid.
Within the heptasubstituted CD derivatives, spacer groups
are forced to point radially outwards because of the intra-
molecular repulsive forces between the charged end groups.
These results may help to design new cyclodextrin recep-
tors with even higher binding constants. Furthermore, these
binding data may also prove useful for parameterization of
force fields to allow more accurate molecular dynamics cal-
culations of CD inclusion compounds in water boxes.
À1
À
À
À
À
(s, C C),1353 (w), 1650 (w, N H), 2916 (m, C H), 3269 cm (s, O H);
MS (ESI): m/z: 1236.4 [M+H+]; elemental analysis calcd (%) for
C45H77N3O28·HCl·10H2O: C 40.77, H 7.47, N 3.17; found: C 40.28, H 6.69,
N 2.70.
1-(4-tert-Butylphenyl)guanidine
(16):
4-tert-Butylaniline
(2.0 g,
13.4 mmol) was dissolved in anhydrous DMF (5 mL). The mixture was
stirred for 48 h at room temperature after addition of 1H-pyrazole-1-car-
boxamide hydrochloride (2.17 g, 14.7 mmol). The solvent was evaporated
and the residue was washed with diethyl ether, dissolved in water, and
extracted with diethyl ether. The aqueous phase was lyophilized to give
the white product (2.14 g, 11.2 mmol, 70%). 1H NMR (500 MHz, D2O/
DSS): d=1.32 (s, 9H; H-6), 7.28 (d, 3J=8.83 Hz, 2H; H-2), 7.85 ppm (d,
3J=8.51 Hz, 2H; H-3); 13C NMR (125.71 MHz, D2O/DSS): d=32.84 (C-
6), 36.57 (C-5), 128.35 (C-2), 129.52 (C-3), 133.30 (C-1), 149.22 (C-4),
154.40 ppm (C-7); MS (ESI): m/z: 192.2 [M+H+].
4-tert-Butyl-N,N-dimethylaniline N-oxide (19): 4-tert-Butyldimethylani-
line (0.5 g, 2.82 mmol) was dissolved in methanol (5 mL), hydrogen per-
oxide (30%, 1.15 mL) was added dropwise, and the mixture was stirred
for seven days at room temperature. After addition of a catalytic amount
of Pd/C, the mixture was filtered through Celite to give the colorless
product (0.23 g, 1.18 mmol, 42%) after evaporation of the solvent.
1H NMR (500 MHz, D2O/DSS): d=1.31 (s, 9H; H-6), 3.60 (s, 6H; H-7)
7.62 (d, 3J=8.51 Hz, 2H; H-3), 7.80 ppm (d, 3J=8.51, 2H; H-2);
13C NMR (125.71 MHz, D2O/DSS): d=32.98 (C-6), 36.62 (C-5), 64.21 (C-
7), 121.65 (C-2), 129.10 (C-3), 152.36 (C-1), 155.71 ppm (C-4); MS (ESI):
m/z: 194.2 [M+H+]; elemental analysis calcd (%) for C12H19NO·H2O: C
68.21, H 10.02, N 6.63; found: C 68.72, H 9.91, N 6.25.
Experimental Section
4-tert-Butylbenzenesulfonic acid(20)
:
4-tert-Butylbenzene (5.37 g,
40.0 mmol) was cooled to 08C and oleum (1.62 mL, 40.0 mmol) was
added dropwise. The mixture was stirred for 3 h at room temperature,
water was added (20 mL), and the solution was neutralized with Amber-
lite IRA-402. The solution was lyophilized to give the white product
(1.21 g, 5.68 mmol, 14%). 1H NMR (500 MHz, D2O/DSS): d=1.32 (s,
9H; H-6), 7.62 (d, 3J=8.83, 2H; H-2), 7.74 ppm (d, 3J=8.51 Hz, 2H; H-
3); 13C NMR (125.71 MHz, D2O/DSS): d=32.97 (C-6), 36.91 (C-5),
121.65 (C-2), 129.10 (C-3), 152.36 (C-1), 155.71 ppm (C-4); MS (ESI): m/
z: 213.0 [M+H+]; elemental analysis calcd (%) for C10H13O3SNa·H2O: C
47.23, H 5.95; found: C 46.63, H 6.25.
Materials: Compounds 15, 17, and 21 were purchased from Aldrich. b-
CD was donated by Wacker. Synthesized host compounds were purified
by nanofiltration by using a Berghof BM-5 membrane (molecular weight
cutoff 500 Da) and Milli-Q water.
Synthetic procedures: Monosubstituted b-CD derivatives 2–6 were syn-
thesized starting from 6-O-tosyl-b-CD[45] by displacement reactions with
nitrogen and sulfur nucleophiles, as depicted in Scheme 1, by using stan-
dard procedures described previously.[38] Heptasubstituted b-CD deriva-
tives 7–14 were synthesized starting from heptakis(6-iodo-6-deoxy)-b-
CD[67] by displacement reactions with nitrogen and sulfur nucleophiles, as
depicted in Scheme 2, by using standard procedures described previous-
ly.[38]
Potentiometric measurements andcalculation of p Ka values: The titration
experiments were performed as described previously[68] (25.08C, 0.1m
KCl) by using a Metrohm 665 piston burette, a Metrohm 6.0262.100 glass
electrode with an incorporated Ag/AgCl reference, and a Metrohm 713
pH/mV meter. Data acquisition and addition of the titrant (0.1m KOH)
were controlled by a PC.[69] The total concentration of the various poly-
bases was always 0.5 mm. The two polyamines 7 and 8 were used as hy-
drochlorides, and the polycarboxylates 9 and 11 as sodium salts together
with eight equivalents of HCl to ensure complete protonation at the be-
ginning of the experiment. Potentiometric data were evaluated with the
computer program HYPERQUAD.[70] All protonation constants were
calculated as concentration constants, pH was defined as Àlog[H+], and
the fixed value pKw =13.78 was used.[71] For each system, at least four ti-
tration experiments were performed, and in the final evaluation two
curves of each system were combined into one data set and evaluated to-
gether.
N-(6-deoxy-6-b-cyclodextrinyl)guanidinium chloride (3): Mono-[6-deoxy-
6-amino]-b-CD 2 (1.13 g, 1.00 mmol) was dissolved in anhydrous DMF
(15 mL). The mixture was stirred for 24 h at room temperature after ad-
dition of Hünigꢀs base (0.17 mL, 1.0 mmol) and 1H-pyrazole-1-carbox-
A
and the product precipitated in acetone to give a white material (1.05 g,
0.89 mmol, 89%). 1H NMR (500 MHz, D2O/DSS): d=3.40–3.59 (m,
15H; H-2/2’/4/4’/6a’), 3.59–3.74 (m, 8H; H-5/5’/6b’), 3.74–3.90 (m, 19H;
H-3/3’/6a/6b), 4.95–5.00 ppm (m, 7H; H-1/1’); 13C NMR (125.71 MHz,
D2O/DSS): d=44.07 (C-6’), 61.86 (C-6), 73.93 (C-2/C-2’/C-5/C-5’), 75.09
(C-3), 83.04 (C-4), 84.83 (C-4’), 104.08 (C-1/C-1’), 159.58 ppm (C-7’); IR:
À
À
À
n˜ =1020, 1149 (s, C C), 1359 (w), 1649 (w, N H), 2916 (m, C H),
+
3272 cmÀ1 (s, O H); MS (ESI) m/z: 1175.4 [M+H ]; elemental analysis
À
Chem. Eur. J. 2008, 14, 7202 – 7211
ꢁ 2008 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
7209

G. Wenz et al.
ITC measurements: Isothermal microcalorimetric titrations were per-
formed at 25.08C with an AutoITC isothermal titration calorimeter (Mi-
croCal Inc., Northampton, USA) with 1.414 mL of sample and reference
cells. The reference cell was filled with distilled water. Prior to the meas-
urements, samples of hosts and guests were brought to pH 6.8 by titration
with small amounts of aqueous HCl or NaOH and lyophilized afterwards.
The sample cell was degassed and filled with a 1 mm solution of the re-
spective host in 50 mm phosphate buffer (pH 6.8) and constantly stirred
at 450 rpm. A 13 mm solution of the guest in the same buffer was auto-
matically added by syringe in 20 separate injections of 12.5 mL. The re-
sulting 20 heat signals were integrated to yield the mixing heats, which
were corrected by the corresponding dilution enthalpies of the guest,
which had been measured separately. The host–guest titration curve
(molar heats vs. molar ratio guest/host) was fitted by nonlinear regression
with the program Origin 7.0 for ITC. For each inclusion compound, the
host–guest stoichiometry was found to be 1:1. The binding constant K
and the molar binding enthalpy DH8 were obtained as fitting parameters,
from which the binding free energy DG8 and binding entropy DS8 were
derived. The titration was repeated 2–3 times with concentrations of
[host]=10/K and [guest]=130/K to achieve optimal accuracy. For small
binding constants (K<2500 LmolÀ1), concentrations were limited to
[host]=4 mm and [guest]=40 mm.
[16] K. Kano, H. Kitagishi, T. Mabuchi, M. Kodera, S. Hirota, Chem.
Asian J. 2006, 1, 358–366.
[17] C. Desiderio, S. Fanali, J. Chromatogr. A 1995, 716, 183–196.
[18] P. Mikus, D. Kaniansky, S. Fanali, Electrophoresis 2001, 22, 470–477.
[19] K. A. Connors, Chem. Rev. 1997, 97, 1325–1357.
[20] L. Liu, Q.-X. Guo, J. Inclusion Phenom. Macrocyclic Chem. 2002,
42, 1–14.
[21] M. R. Eftink, M. L. Andy, K. Bystrom, H. D. Perlmutter, D. S. Kris-
tol, J. Am. Chem. Soc. 1989, 111, 6765–6772.
[22] T. Hçfler, G. Wenz, J. Inclusion Phenom. Mol. Recognit. Chem.
1996, 25, 81–84.
[23] C. J. Easton, S. F. Lincoln, Modified Cyclodextrins: Scaffolds and
Templates for Supramolecular Chemistry, World Scientific Publishing
Company, London, 1999.
[24] T. Kitae, T. Nakayama, K. Kano, J. Chem. Soc. Perkin Trans. 2 1998,
207–212.
[25] M. Rekharsky, H. Yamamura, M. Kawai, Y. Inoue, J. Am. Chem.
Soc. 2001, 123, 5360–5361.
[26] M. V. Rekharsky, Y. Inoue, J. Am. Chem. Soc. 2002, 124, 813–826.
[27] F. Hacket, S. Simova, H. Schneider, J. Phys. Org. Chem. 2001, 14,
159–170.
[28] H. Yamamura, M. V. Rekharsky, Y. Ishihara, M. Kawai, Y. Inoue, J.
Am. Chem. Soc. 2004, 126, 14224–14233.
NMR measurements: All 1H NMR experiments were carried out in deu-
[29] K. Kano, J. Phys. Org. Chem. 1997, 10, 286–291.
[30] K. Kano, Y. Horiki, T. Mabuchi, H. Kitagishi, Chem. Lett. 2004, 33,
1086–1087.
[31] B. J. Ravoo, R. Darcy, A. Mazzaglia, D. Nolan, K. Gaffney, J. Chem.
Soc. Chem. Commun. 2001, 827–828.
[32] T. Kraus, M. Budesinsky, J. Zavada, Carbohydr. Res. 1997, 304, 81–
84.
[33] T. Kraus, M. Budesinsky, I. Cisarova, J. Zavada, Eur. J. Org. Chem.
2004, 4060–4069.
terium oxide at 258C on
a Bruker Avance 500 spectrometer at
500.00 MHz. ROESY experiments were performed on 15 mm solutions of
hosts and guests in 0.05m phosphate buffer solutions by using the Bruker
standard routine “roesyph.” The data consisted of NS=32 scans collected
over 2048 complex points with a spectral width (SW) of 4005 Hz. A
mixing time (P15) of 300 ms, a repetition delay (D1) of 2 s, an acquisition
time of 0.167 s, and a 908 pulse width (P1) of 7.95 ms at PL1=À2 dB
power attenuation were used.
[34] T. Kraus, M. Budesinsky, J. Zavada, Tetrahedron Lett. 2006, 47, 679–
681.
[35] F. Guillo, B. Hamelin, L. Jullien, J. Canceill, J. M. Lehn, L. Derober-
tis, H. Driguez, Bull. Soc. Chim. Fr. 1995, 132, 857–866.
[36] J. M. Adam, D. J. Bennett, A. Bom, J. K. Clark, H. Feilden, E. J.
Hutchinson, R. Palin, A. Prosser, D. C. Rees, G. M. Rosair, D. Ste-
venson, G. J. Tarver, M.-Q. Zhang, J. Med. Chem. 2002, 45, 1806–
1816.
Acknowledgements
The authors thank Ronny Heisel for the synthesis of host 7 and guests
18–20, Dr. Josef Zapp (Department of Pharmacy, Saarland University)
for performing the ROESY measurements, and Andreas Steffen (Astra-
Zeneca, Sweden) for helpful discussions.
[37] N. Mourtzis, K. Eliadou, K. Yannakopoulou, Supramol. Chem. 2004,
16, 587–593.
[38] A. Steffen, C. Thiele, S. Tietze, C. Strassnig, A. Kꢂmper, T. Lenga-
uer, G. Wenz, J. Apostolakis, Chem. Eur. J. 2007, 13, 6801–6809.
[39] B. Hamelin, L. Jullien, C. Derouet, C. Herve du Penhoat, P. Ber-
thault, J. Am. Chem. Soc. 1998, 120, 8438–8447.
[40] L. Jullien, H. Cottet, B. Hamelin, A. Jardy, J. Phys. Chem. B 1999,
103, 10866–10875.
[41] V. Zia, R. A. Rajewski, V. J. Stella, Pharm. Res. 2001, 18, 667–673.
[42] H.-J. Schneider, A. K. Yatsimirsky, Chem. Soc. Rev. 2008, 37, 263–
277.
[1] J.-M. Lehn, Angew. Chem. 1988, 100, 91–116; Angew. Chem. Int.
Ed. Engl. 1988, 27, 89–112.
[2] D. J. Cram, J. Inclusion Phenom. Macrocyclic Chem. 1988, 6, 397–
413.
[3] S. M. Biros, J. Rebek, Jr., Chem. Soc. Rev. 2007, 36, 93–104.
[4] K. N. Houk, A. G. Leach, S. P. Kim, X. Y. Zhang, Angew. Chem. Int.
Ed. 2003, 42, 4872–4897.
[5] G. Wenz, Angew. Chem. 1994, 106, 851–870; Angew. Chem. Int. Ed.
Engl. 1994, 33, 803–822.
[43] W. Chen, C.-E. Chang, M. K. Gilson, J. Am. Chem. Soc. 2006, 128,
4675–4684.
[6] M. V. Rekharsky, Y. Inoue, Chem. Rev. 1998, 98, 1875–1917.
[7] J. Szejtli, Chem. Rev. 1998, 98, 1743–1753.
[44] A. Ashnagar, P. Tin Culnane, C. J. Easton, J. B. Harper, F. Lincoln.
Stephen, Aust. J. Chem. 1997, 50, 447–450.
[8] M. Singh, R. Sharma, U. C. Banerjee, Biotechnol. Adv. 2002, 20,
341–359.
[45] J. Defaye, S. Crouzy, N. Evrard, H. Law (Centre National De La Re-
cherche Scientifique, Fr.), WO 9961483, 1999; [Chem. Abstr. 1999,
132, 24077].
[9] M. Del Valle, Process Biochem. (Amsterdam, Netherlands) 2004, 39,
1033–1046.
[10] J. Szejtli, Pure Appl. Chem. 2004, 76, 1825–1845.
[11] G. Wenz, Clin. Drug Invest. 2000, 19, Suppl. 2, 21–25.
[12] C. Pean, A. Wijkhuisen, F. Djedaini-Pilard, J. Fischer, S. Doly, M.
Conrath, J.-Y. Couraud, J. Grassi, B. Perly, C. Creminon, Biochim.
Biophys. Acta, Molecular Cell Res. 2001, 1541, 150–160.
[13] C. Pean, C. Creminon, A. Wijkhuisen, J. Grassi, P. Guenot, P. Jehan,
J.-P. Dalbiez, B. Perly, F. Djedaini-Pilard, J. Chem. Soc. Perkin
Trans. 2 2000, 853–863.
[14] M. E. Davis, M. E. Brewster, Nat. Rev. Drug Discovery 2004, 3,
1023–1035.
[15] L. Holland, G. Rizzi, P. Malton (The Procter & Gamble Company),
WO 2000067717, 2000; [Chem. Abstr. 2000, 133, 339996].
[46] L. D. Melton, K. N. Slessor, Carbohydr. Res. 1971, 18, 29–37.
[47] E. Mavridou, K. Giannakopoulou, K. Iliadou, N. Mourtzis, C. An-
gelidou (Ethniko Kentro Erevnas Fysikon Epistimon “Dimokritos”,
Greece), GR 1004952, 2005; [Chem. Abstr. 2005, 144, 53296].
[48] N. Mourtzis, K. Eliadou, C. Aggelidou, V. Sophianopoulou, I. M.
Mavridis, K. Yannakopoulou, Org. Biomol. Chem. 2007, 5, 125–131.
[49] P. R. Ashton, R. Kçniger, J. F. Stoddart, J. Org. Chem. 1996, 61,
903–908.
[50] W. M. Meylan, P. H. Howard, J. Pharm. Sci. 1995, 84, 83–92.
[51] Y. Sakuratani, K. Kasai, Y. Noguchi, J. Yamada, QSAR Comb. Sci.
2007, 26, 109–116.
7210
www.chemeurj.org
ꢁ 2008 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2008, 14, 7202 – 7211

Guest Recognition by Ionic b-Cyclodextrins
FULL PAPER
[52] K. Kano, Y. Ishida, K. Kitagawa, Chem. Asian J. 2007, 2, 1305–
1313.
[63] D. Harries, D. C. Rau, V. A. Parsegian, J. Am. Chem. Soc. 2005, 127,
2184–2190.
[64] D. M. Ford, J. Am. Chem. Soc. 2005, 127, 16167–16170.
[65] V. Zabel, W. Saenger, S. A. Mason, J. Am. Chem. Soc. 1986, 108,
3664–3673.
[66] J. E. H. Koehler, W. Saenger, W. F. Van Gunsteren, Eur. Biophys. J.
1988, 16, 153–168.
[67] K. Chmurski, J. Defaye, Supramol. Chem. 2000, 12, 221–224.
[68] A. Zimmer, D. Kuppert, T. Weyhermüller, I. Müller, K. Hegetsch-
weiler, Chem. Eur. J. 2001, 7, 917–931.
[53] L. E. Briggner, X.-R. Ni, F. Tempesti, I. Wadsç, Thermochim. Acta
1986, 109, 139–143.
[54] R. I. Gelb, L. M. Schwartz, J. Inclusion Phenom. Mol. Recognition
Chem. 1989, 7, 537–543.
[55] M. Sakurai, M. Kitagawa, Y. Inoue, R. Chujo, Carbohydr. Res. 1990,
198, 181–191.
[56] X. S. Li, L. Liu, T. W. Mu, Q. X. Guo, Monatsh. Chem. 2000, 131,
849–855.
[69] S. Steinhauser, U. Heinz, M. Bartholomꢂ, T. Weyhermüller, H. Nick,
K. Hegetschweiler, Eur. J. Inorg. Chem. 2004, 4177–4192.
[70] P. Gans, A. Sabatini, A. Vacca, Talanta 1996, 43, 1739–1753.
[71] R. M. Smith, A. E. Martell, R. J. Motekaitis, Critically Selected Sta-
bility Constants of Metal Complexes, NIST Standard Reference Da-
tabase 46, Version 8.0, Gaithersburg, MD, USA 2004.
[57] M.-J. Huang, J. D. Watts, N. Bodor, Int. J. Quantum Chem. 1997, 65,
1135–1152.
[58] A. Rontoyianni, I. M. Mavridis, E. Hadjoudis, A. J. M. Duisenberg,
Carbohydr. Res. 1994, 252, 19–32.
[59] A. Müller, G. Wenz, Chem. Eur. J. 2007, 13, 2218–2223.
[60] L. L. Onsager, Physik. Z. 1926, 27, 388–392.
[61] K. K. Lee, C. A. Fitch, E. Bertrand Garcia-Moreno, Protein Sci.
2002, 11, 1004–1016.
Received: February 17, 2008
Revised: May 13, 2008
Published online: July 9, 2008
[62] L. Onsager, N. N. T. Samaras, J. Chem. Phys. 1934, 2, 528–536.
Chem. Eur. J. 2008, 14, 7202 – 7211
ꢁ 2008 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
7211