Evidence of a self-inclusion phenomenon for a new class of mono-substituted alkylammonium-β-cyclodextrins

Article abstract of DOI:10.1039/b416018e

A new class of mono-substituted N-alkyl-N, N-dimethylammonium-β- cyclodextrins has been synthesized in a three step procedure from the native β-cyclodextrin. The structural analysis of these compounds undertaken by combined use of 1D and 2D NMR spectra indicate that the two methyl groups bound on the nitrogen are magnetically inequivalent due to a self-inclusion phenomenon of the alkyl chain inside the CD cavity. A variable-temperature ¹H NMR study showed that these mono-substituted CD derivatives formed temperature-independent intramolecular complexes with their own alkylammonium substituent. The strength of the interaction between the alkyl moiety and the cyclodextrin cavity has been evaluated by a competitive method using an adamantane derivative. Finally, surface tension measurements demonstrated the surface active character of these compounds and confirmed their self-inclusion ability. The Royal Society of Chemistry 2005.

Full text of DOI:10.1039/b416018e

View Article Online / Journal Homepage / Table of Contents for this issue
A R T I C L E
Evidence of a self-inclusion phenomenon for a new class of
mono-substituted alkylammonium-b-cyclodextrins
Ce´cile Binkowski,a Fre´de´ric Hapiot,*a Vincent Lequart,b Patrick Martinb and Eric Monfliera
a Laboratoire de Physico-chimie des Interfaces et Applications-FRE CNRS 2485, Faculte´ Jean
Perrin, Universite´ d’Artois, rue Jean Souvraz, S. P. 18, 62307 Lens Cedex, France.
E-mail: hapiot@univ-artois.fr; Fax: (33) 03 21 79 17 55; Tel: (33) 03 21 79 17 73
b Laboratoire de la Barrie`re He´mato-Ence´phalique, Faculte´ Jean Perrin, Universite´ d’Artois,
62307 Lens Cedex, France
Received 19th October 2004, Accepted 2nd February 2005
First published as an Advance Article on the web 22nd February 2005
A new class of mono-substituted N-alkyl-N,N-dimethylammonium-b-cyclodextrins has been synthesized in a three
step procedure from the native b-cyclodextrin. The structural analysis of these compounds undertaken by combined
use of 1D and 2D NMR spectra indicate that the two methyl groups bound on the nitrogen are magnetically
inequivalent due to a self-inclusion phenomenon of the alkyl chain inside the CD cavity. A variable-temperature 1H
NMR study showed that these mono-substituted CD derivatives formed temperature-independent intramolecular
complexes with their own alkylammonium substituent. The strength of the interaction between the alkyl moiety and
the cyclodextrin cavity has been evaluated by a competitive method using an adamantane derivative. Finally, surface
tension measurements demonstrated the surface active character of these compounds and confirmed their
self-inclusion ability.
Introduction
The great variety of available modified cyclodextrins (CD)
and their ability to recognise a wide range of lipophilic guest
molecules has made them useful tools in supramolecular
chemistry.1 Actually, thanks to their toroidal shape and strong
binding affinity for hydrophobic drugs, CDs constitute potential
carriers for transport through biological barriers.2 In particular,
it is thought that amphiphilic b-CD derivatives interact with
biological membranes due to their lipophilic character and
many modified cyclodextrins have been synthesized for that
purpose.3 Concurrently, studies have also shown that cationic
entities may be anchored into membranes due to the electronic
attraction force between the positive charge of these species and
the negative charge of the lipid bilayer of the membrane.4
Scheme 1 Three step synthesis of DMA–Cn–CDs.
In order to take benefits from both cationic and lipophilic
It is noteworthy that, contrary to amphiphilic CDs described
properties, we have recently focused our attention on modified
in the literature whose use in large quantities proved inconve-
CDs containing an ammonium salt and a long-alkyl chain. We
describe here a straightforward three step synthesis of N-alkyl-
N,N-dimethylammonium cyclodextrins (DMA–Cn–CD, with n
nient because of their poor solubility in water,7 the DMA–Cn–
CDs of this study showed great solubility in water (>60 mM).
the number of carbons on the alkyl chain). This new family of
mono-substituted b-CDs is constituted of a b-cyclodextrin on
which a quaternary ammonium group is grafted on its primary
face so that the wider opening of the CD (secondary face)
remains accessible for organic guest species. A detailed NMR
analysis has been carried out to determine the conformation
of the ammonium moiety with respect to the CD. In addition,
their behaviour in water was examined as a function of their
concentration and the length of the alkyl chain.
NMR structural analysis
First, to get information about the position of the ammonium
moiety towards the CD cavity, 2D T-ROESY experiments have
been performed. These experiments were preferred to classical
2D ROESY as it was shown that this sequence provides reliable
dipolar cross-peaks with a minimal contribution of scalar
transfer.8 As an example, the region of the 2D T-ROESY contour
maps concerning the interactions through space between the
protons of the C4 and C12 aliphatic chains and the inner protons
of the cavity is shown in Fig. 1.
Results and discussion
Intense cross-peaks were detected from DMA–C4–CD to
DMA–C16–CD between the inner protons of the cavity (H-3 and
H-5) and the alkyl region which confirmed the inclusion between
the CD and the alkyl chain. The contacts detected for the DMA–
C4–CD derivative were informative of the depth of the inclusion:
indeed, the intensity of the correlation between the methyl and
the CD was much more important than those of the methylenes,
suggesting that an alkyl chain length of four carbons constitutes
a minimum for a recognition process to occur. By contrast, no
contact was observed for DMA–C2–CD which clearly showed
Synthesis
Starting from the native b-CD, mono-tosylated5 and mono-
iodide6 b-CDs were successively obtained according to pro-
cedures already described in the literature (Scheme 1). The
mono-iodide b-CD then reacted on a ternary N-alkyl-N,N-
dimethylamine whose linear alkyl moiety contained from 2 to
16 carbons (even number) to give the expected N-alkyl-N,N-
dimethylammonium cyclodextrins in an average yield of 40%.
T h i s j o u r n a l i s
T h e R o y a l S o c i e t y o f C h e m i s t r y 2 0 0 5
O r g . B i o m o l . C h e m . , 2 0 0 5 , 3 , 1 1 2 9 – 1 1 3 3
1 1 2 9
©

View Article Online
Fig. 1 2D T-ROESY data (300 ms mixing time) at 298 K in D2O of DMA–C4–CD (10 mM) and DMA–C12–CD (10 mM).
that the ethyl group did not penetrate the b-CD cavity and was
closely located in the outer layer of the CD.
Nevertheless, the question of intra- or inter-inclusion was
not elucidated at this stage of the study. Thus, we then tried
to get more information about whether the alkyl arm was
included in the cavity of the CD on which it is grafted or in
the cavity of another DMA–Cn–CD. It is then customary to
assess the accurate position of the substituent with regard to
the cavity of the CD by investigating the effect of the dilution
1
of the compound in water on the H NMR chemical shifts.7a,9
Thus, to evaluate the influence of the alkyl chain length as a
function of the concentration, three DMA–Cn–CD derivatives
were chosen: DMA–C4–CD, DMA–C10–CD and DMA–C14–
Fig. 2 Partial 1H NMR spectra of DMA–C2–CD and DMA–C12–CD
(10 mM) at 298 K in D2O.
CD. Their concentrations were varied from 10−4 to 10−2
M
at 25 ◦C. Whatever the concentration or length of the alkyl
chain, no difference in the chemical shifts was detected on the
1H NMR spectra, neither in the CD pattern nor in the alkyl
region. Given the low concentrations throughout the study,
we favour the view that the obtained 1H NMR spectra are
relative to intramolecular inclusion processes. Moreover, the
high hydrophobicity of long alkyl chains is not a favourable
argument for their spreading in water and their inclusion in
another DMA–Cn–CD cavity. Nevertheless, the hypothesis of
an intermolecular recognition could not be definitely excluded,
especially at high concentrations of DMA–Cn–CDs, since the
corresponding spectrum could be very similar to the previous
one. When increasing the concentrations, the system favours
equilibrium between intra- and intermolecular complexes.
To allow a better understanding of the inclusion process, an
in-depth NMR analysis was necessary. A partial assignment of
two magnetically different methyl groups. At first sight, the
magnetic inequivalence between these methyl groups indicates
that the motion of the ammonium residue is restrained, at
least according to the NMR timescale, as soon as the alkyl
chain lengthens. The ethyl group of the DMA–C2–CD is too
small to hinder the rotation of the ammonium around the bond
linking the nitrogen to the primary carbon C-6 of the substituted
glucoside unit in a consistent way. Therefore, the two methyl
groups are magnetically equivalent in that case. On the other
hand, the presence of longer alkyl chains (>C4) may influence
the respective orientation of the methyl groups or protons of the
alkyl chain and their relative motion due to the self-inclusion
process. Consequently, the geometry of the molecule might be
modified to such an extent that the methyl groups bound to the
nitrogen might be magnetically inequivalent.7 For comparison,
the 1H NMR spectrum of the non-covalent complex b-CD–N-
hexadecyl-N,N,N-trimethylammonium chloride (1 : 1 mixture)
has been performed and exhibited for the host and guest signals a
non-perturbed symmetry. Actually, in that case, the three methyl
groups bound to the nitrogen were magnetically equivalent since
only one resonance was detected at 3.14 ppm. Another proof of
the structural inequivalence of the two methyls of DMA–Cn–
CD in water was given by 1H NMR spectra in DMSO-d6 which
is a well known dissociating solvent.7a,10 As an example, the 1H
NMR spectra of DMA–C4–CD obtained in D2O and DMSO-d6
are displayed in Fig. 3.
1
the resonances of the H spectra with D2O as the solvent gave
us more details on the geometry of the ammonium moiety with
respect to the CD cavity. A total assignment of the 1D spectrum
was limited due to the extreme spectral overcrowding in the CD
region, giving very broad and shouldered peaks.
Interestingly, a splitting into two of the resonance of some
protons of the alkyl moieties has been detected for DMA–Cn–
CDs with n ≥ 4. As an example, the partial 1H NMR spectra of
DMA–C2–CD and DMA–C12–CD are displayed in Fig. 2.
The methyl groups appeared as only one singlet for DMA–
C2–CD and as two singlets for DMA–Cn–CD (n ≥ 4). This
magnetic inequivalence has also been detected for the protons
in the b-position to the nitrogen for DMA–C4–CD. The same
In that case, only one resonance was detected for both methyl
groups indicating that the alkyl chain is not confined in the
cavity but is able to spread in the solvent. The protons of the
alkyl chain in the b-position to the nitrogen were also indicative
of a non-included substituent since the two resonances observed
in D2O were merged in a single broad peak in DMSO-d6.
1
splittings into two were observed on the 13C{ H} NMR spectra
and HMQC experiments confirmed the assignments made
through 1H homonuclear NMR, especially the presence of
1 1 3 0
O r g . B i o m o l . C h e m . , 2 0 0 5 , 3 , 1 1 2 9 – 1 1 3 3

View Article Online
space between the adamantyl part and the inner protons of the
CD cavity (Fig. 4).
Fig. 3 Partial 1H NMR spectra of DMA–C4–CD (10 mM) in D2O and
DMSO-d6 at 298 K.
In order to get further clarification on the dynamic of the
system, the conformational change of DMA–Cn–CD deriva-
tives, which may be induced by temperature variations, has
1
been studied by a variable-temperature H NMR experiment.
The 1H NMR spectrum of the alkyl or CD regions was
virtually unaffected by temperature variations in the range 25–
80 ◦C. Consequently, the self-inclusion of the alkyl arm in a CD
cavity is not a function of the temperature certainly due to the
high hydrophobic character of the former and its desire to stay
closely coiled in the cavity. This led us to think that the associa-
tion constant between the alkyl moiety and the CD is high.
Fig. 4 2D T-ROESY data (300 ms mixing time) at 298 K in D2O of a
1 : 1 mixture of DMA–C12–CD (10 mM) and Adac (10 mM).
Complexing properties with 1-adamantanecarboxylic acid
In fact, the DMA–Cn–CD intramolecular complex is only
weakened on interaction of the cavity with Adac. As a matter
of fact, Adac proved unable to completely extract the alkyl
chain from the CD-cavity to form a binary inclusion complex
since cross-peaks were always visible between the protons of
the aliphatic chain and H-3 and H-5. Moreover, the two methyl
groups still appeared magnetically inequivalent, proving that
the ammonium moiety closely remains around the cavity. The
adamantane group pushed out the alkyl chain from the CD
cavity but the high hydrophobicity of the alkyl chain drives it
to stay in the neighbourhood of the cavity, as if it covers the
primary face of the CD (Scheme 2(b)).
The determination of the association constant between the alkyl
moiety and the CD on which it is grafted by classical methods
proved inconvenient as the concentrations of the guest and the
host could not be varied for evident reasons.11 Therefore, we
attempted to approach the value of the association constant
between the alkyl chain and the CD on which it is grafted by a
competitive method using 1-adamantanecarboxylic acid (Adac)
as a guest (Scheme 2(a)).
The stoichiometry of the DMA–Cn–CD–Adac complexes
(n = 2 and 4) has been established by the method of continuous
variation (Job’s method)13, using various ratios of DMA–Cn–
CD to Adac, while keeping the total moles of DMA–Cn–CD
plus Adac constant. The Job diagrams for the protons of Adac
showed a maximum at a guest/(host + guest) ratio of 0.5
indicating that there is one molecule of Adac for each molecule
of DMA–Cn–CD. As a example, the Job plot of the DMA–C4–
CD–Adac couple is depicted in Fig. 5.
Scheme 2 (a) 1-Adamantanecarboxylic acid (Adac); b) inclusion
complex between DMA–Cn–CD (n >2) and Adac.
Actually, adamantyl derivatives are well known to form stable
inclusion complexes with many cyclodextrins (104 < KA
<
105 M−1).12 The inclusion of Adac in DMA–Cn–CDs has been
1
evidenced by H and T-ROESY NMR experiments. Actually,
since H-3 and H-5 are located inside the cavity of the CD, they
are sensitive to changes in the microenvironment upon inclusion.
Competitive inclusion complexation of Adac is supported by
several lines of evidence: (i) a 1 : 1 mixture of Adac and
DMA–Cn–CD clearly showed downfield induced chemical shifts
for some of the protons of the adamantyl moiety (0.2 ppm
downfield for protons b (Scheme 2(a)). The shifts were smaller
for the protons assumed to be exposed more to the solvent
than to the cavity (protons a, Scheme 2(a)); (ii) chemical shifts
Fig. 5 Job plot upon mixing DMA–C4–CD and Adac. The total
concentration of the two components was 1 lM in D2O, with mole
fractions varying from 0 to 1.
1
were also detected on the 13C{ H} NMR spectrum, here again
proving a through-space interaction; (iii) 2D T-ROESY NMR
experiments clearly pointed out spin–spin interactions through
A significant variation in the apparent association constant
between Adac and the cavity could be measured when varying
O r g . B i o m o l . C h e m . , 2 0 0 5 , 3 , 1 1 2 9 – 1 1 3 3
1 1 3 1

View Article Online
Table 1 Association constants (K) of 1-adamantanecarboxylic acid (Adac) with the native b-CD and DMA–Cn–CDs (n = 2 and 4)
Host–guest complexes
b-CD–Adac
DMA–C2–CD–Adac
39 000
DMA–C4–CD–Adac
16 100
K/M−1
42 000
the length of the alkyl chain. Indeed, 1H NMR titration
experiments have been performed with mixtures of DMA–Cn–
CD and Adac. The concentration of Adac was kept constant
(1 mM) and that of the modified cyclodextrin was varied from
0.3 to 3 mM. Association constants between those compounds
could be deduced from a fitting programme14 and are gathered
in Table 1.
cavity, the amphiphilic character was partially hidden which
prevented any classical micelle aggregation from occurring.
Surface tension measurements have also been carried out
in water with an increasing amount of a DMA–C12–CD–
Adac complex in a stoichiometric ratio. The obtained curve
was superimposable on that of DMA–C12–CD alone (Fig. 6).
Therefore, the absence of CMC in that case clearly showed
that the amphiphilic character of DMA–C12–CD could not be
restored upon addition of Adac. This corroborates the above
NMR results according to which the alkyl chain and Adac
coexist in the CD cavity.
Firstly, the apparent association constant of the native b-
CD–Adac couple has been measured to serve as a reference
(KA = 42 000 M−1). Secondly, the similar value obtained
for the association constant between Adac and DMA–C2–
CD (KA = 39 000 M−1) logically showed that the penetration
of the adamantane part in the CD cavity by the secondary
face was barely hindered by the non-included alkylammonium
substituent grafted on the primary face. The determination of
association constants between Adac and DMA–Cn–CD with
a longer alkyl chain (n > 4) proved inconvenient because of
the intramolecular inclusion process. One can reasonably think
that the penetration of Adac was much more difficult with a
lengthening of the aliphatic part and consequently the value of
the apparent association constants decrease (KA < 16 000 M−1).
Finally, a variable-temperature 1H NMR experiment has also
been carried out with a 1 : 1 mixture of DMA–C4–CD and
Adac. This mixture was heated from 25 to 80 ◦C. Contrary
to what was observed in the absence of an external guest, the
1H NMR spectra obtained in the presence of Adac were slightly
different when increasing the temperature. The induced chemical
shifts at 60 and 80 ◦C were lower than those measured at 25 ◦C
(0.20 and 0.18 ppm vs. 0.23 ppm respectively), suggesting that
the association constant was lower at high temperature. This is
consistent with the general trend for which the strength of the
host–guest interactions decrease with the temperature.
Conclusion
As a conclusion, a direct route to a new and attractive class
of amphiphilic alkylammonium b-CDs has been found. The
compounds could be easily purified by a classical work-up and
isolated in good yields (≈40%) from the last step of the synthesis
whatever the length of the alkyl chain. A self-inclusion process
was proved by surface tension measurements and a detailed
NMR study. The use of Adac unambiguously demonstrated that
an external guest may at least partially, if not totally, reduce the
motion of the alkyl chain and confine it to the primary side of the
CD cavity. As a consequence, the alkyl chain is supramolecularly
trapped between a wall of water and a bulky guest as a spring
jammed in a goblet with a heavy ball on it. A study is in progress
to evaluate the behaviour of these compounds towards biological
membranes.
Experimental
General
Organic compounds were purchased from Sigma-Aldrich and
Acros Organics in their highest purity and used without further
purification. b-CD was a generous gift of Roquette Fre`res
(France). The Bio-Rex 70 resin was purchased from Bio-Rad
and the G-25 Sephadex resin from Sigma-Aldrich.
Surface tension
Surface tension measurements have been carried out to evaluate
the amphiphilic behaviour of these modified CDs. The results are
gathered in Fig. 6. Due to the self-inclusion phenomenon, these
compounds did not show similar curves to those of usual surfac-
tants such as N-alkyl-N,N,N-trimethylammonium chloride.15 In
particular, no critical micellar concentration (CMC) could be
defined and a lengthening of the alkyl chain did not lead to a
decrease in the surface tension. Indeed, all DMA–Cn–CD deriva-
tives showed similar behaviour since their surface tensions were
all gathered in the same cluster. The surface active behaviour of
these molecules was probably the result of a competition between
an intra- or inter-inclusion phenomenon and a micellization
process. Actually, when the alkyl chains were included in a CD
The surface tensions were measured by the pendant drop
method using a DSA 10 Mk2 drop shape analysis system (Kru¨ss
GmbH Germany). The DMA–Cn–CD solutions were prepared
by using double-distilled deionized water (Milli-Q, Millipore) as
a solvent. All measurements were carried out at 20.0 0.1 ◦C.
1
The H and 13C NMR spectra were recorded at 300.13 and
75.476 MHz respectively on a Bruker Avance DRX spectrome-
1
ter. 1H and 13C{ H} chemical shifts are given in ppm relative to
the external reference sodium [D4]3-(trimethylsilyl)propionate
(98% atom D) in D2O. The 2D T-ROESY experiments were
run using the software supplied by Bruker. Mixing times for
T-ROESY experiments were set at 300 ms. The data matrix
for the T-ROESY was made of 512 free induction decays, 1 K
points each, resulting from the co-addition of 32 scans. The
real resolution was 1.5–6.0 Hz/point in F2 and F1 dimension,
respectively. They were transformed in the non phase-sensitive
mode after QSINE window processing.
Synthesis of DMA–Cn –CD
To a stirred solution of mono-iodo-b-CD (1 g, 0.8 mmol)
in dry DMF (40 mL) was added N-alkyl-N,N-dimethylamine
(16 mmol). The reaction mixture was stirred for 24 h at
80 ◦C. After removal of the solvent, the residue was washed
under stirring in a mixture of acetone–chloroform. The insoluble
powder was recovered by filtration. After dissolution in water,
Fig. 6 Pendant drop surface tension curves of a DMA–Cn–CD series
(n = 4, 8, 12, 14, 16).
1 1 3 2
O r g . B i o m o l . C h e m . , 2 0 0 5 , 3 , 1 1 2 9 – 1 1 3 3

View Article Online
the crude product was subjected to chromatography on a
n > 4, an accurate standard deviation could not be properly
determined because the concentration of the intramolecular
CD–Cn complex is no more negligible than the concentration
of the CD–Adac complex.
R
Bio-Rexꢀ 70 ion-exchange column and eluted with a water
to 0.1 M aqueous NaCl gradient. The appropriate fractions
were partially evaporated and then desalted on a Sephadex
G-25 column with water as an eluent. The CD-containing
aliquots were then lyophilisized to give the expected N-alkyl-
N,N-dimethylammonium-b-CD as a white powder in an average
yield of 40%. The purity of each compound was checked by
NMR, ES-MS and elemental analysis.
Acknowledgements
We thank the Ministe`re de l’Education Nationale, de
l’Enseignement Supe´rieur et de la Recherche for financial sup-
port (C. Binkowski), and Prof. J.-M. Aubry for the tensiometry
facilities.
Selected spectral data for mono-6-(N,N-dimethylethylamino)-b-
cyclodextrin hydrochloride (DMA–C2–CD)
1H NMR (300 MHz, D2O) d 5.41 (d, 1H, H-1 ammonium–
glucose), 5.14–5.02 (m, 6H, H-1 b-CD), 4.49 (t, 1H, H-5
ammonium–glucose), 4.09–3.48 (m, 43H, H-2, H-3, H-4, H-
5, H-6, H-6ꢀ b-CD, H-2, H-3, H-4, H-6, H-6ꢀ ammonium–
glucose and CH2a), 3.16 (s, 6H, N+(CH3)2), 1.41 (t, 3H,
CH3b). 13C NMR (75 MHz, D2O) d 102.49–101.94 (C-1 b-
CD), 100.74 (C-1 ammonium–glucose), 83.41–81.17 (C-4 b-
CD), 79.64 (C-4 ammonium–glucose), 73.57–72.02 (C-2, C-
3, C-5 b-CD and C-2, C-3 ammonium–glucose), 67.20 (C-5
ammonium–glucose), 62.17 (CH2a), 61.19–60.53 (C-6), 51.52
(N+(CH3)2), 8.01 (CH3b). ES-MS (M–Cl)+ m/z calc 1190.4,
found 1190.5. C46H80O34NCl, 3H2O %N = 1.09; %C = 43.14;
%H = 6.77; found %N = 1.05; %C = 43.43; %H = 6.66.
References
1 (a) J. Szejtli, Inclusion of Guest Molecules, Selectivity and Molecular
Recognition by Cyclodextrins, in Comprehensive Supramolecular
Chemistry, ed. J. L. Atwood, J. E. D. Davies, D. D. MacNicol,
F. Vo¨gtle, J.-M. Lehn, Pergamon, Oxford, 1996, vol. 3, p 189.
2 (a) K. Uekama, F. Hirayama and T. Irie, Chem. Rev., 1998, 98,
2045; (b) D. Ducheˆne, D. Wouessidjewe and G. Ponchel, J. Controlled
Release, 1999, 62, 263.
3 (a) F. Sallas, K. Niikura and S.-I. Nishimura, Chem. Commun., 2004,
596; (b) R. Auze´ly-Velty, F. Djeda¨ıni-Pilard, S. De´sert, B. Perly and T.
Zemb, Langmuir, 2000, 16, 3727; (c) S. Moutard, B. Perly, P. Gode´,
G. Demailly and F. Djeda¨ıni-Pilard, J. Inclusion Phenom., 2002, 44,
317; (d) I. Javierre, M. Nedyalkov, V. Petkova, J.-J. Benattar, S. Weisse,
R. Auze´ly-Velty, F. Djeda¨ıni-Pilard and B. Perly, J. Colloid Interface
Sci., 2002, 254, 120; (e) M. Eddaoudi, A. Baszkin, H. Parrot-Lopez,
M. Boissonade and A. W. Coleman, Langmuir, 1995, 11, 13; (f) C. E.
Granger, C. P. Fe´lix, H. Parrot-Lopez and B. R. Langlois, Tetrahedron
Lett., 2000, 41, 9257; A. Dubes, D. Bouchu, R. Lamartine and H.
Parrot-Lopez, Tetrahedron Lett., 2001, 42, 9147; (g) C. C. Ling, R.
Darcy and W. Risse, J. Chem. Soc., Chem. Commun., 1993, 438; K.
Chmurski, A. W. Coleman and J. J. Jurczak, Carbohydr. Chem., 1996,
15, 787; (h) Y. Kawabata, M. Matsumoto, M. Tanaka, H. Takahashi,
Y. Irinatsu, S. Tamura, W. Tagaki, H. Nakahara and K. Fukuda,
Chem. Lett., 1986, 1933; (i) A. Mazzaglia, R. Donohue, B. J. Ravoo
and R. Darcy, Eur. J. Org. Chem., 2001, 1715; (j) T. Sukegawa, T.
Furuike, K. Niikura, A. Yamagishi, K. Monde and S. I. Nishimura,
Chem. Commun., 2002, 430; (k) M. J. Pregel, L. Jullien and J.-M.
Lehn, Angew. Chem., Int. Ed., 1992, 31, 1637.
4 (a) K. Romøren, B. J. Thu, N. C. Bols and Ø Evensen, Biochim.
Biophys. Acta, 2004, 1663, 127; (b) R. Donohue, A. Mazzaglia, B. J.
Ravoo, B. J. Darcy and R. Darcy, Chem. Commun., 2002, 2864;
(c) S. A. Cryan, A. Holohan, R. Donohue, R. Darcy and C. M.
O’Driscoll, Eur. J. Pharm. Sci., 2004, 21, 625.
5 R. C. Petter and J. S. Salek, J. Am. Chem. Soc., 1987, 109, 7897.
6 R. P. Bonomo, V. Cucinotta, F. D’Alessandro, G. Impellizzeri, G.
Maccarrone, G. Vecchio and E. Rizzarelli, Inorg. Chem., 1991, 30,
2708.
Selected spectral data for mono-6-(N,N-dimethylbutylamino)-b-
cyclodextrin hydrochloride (DMA–C4–CD)
1H NMR (300 MHz, D2O) d 5.41 (d, 1H, H-1 ammonium–
glucose), 5.25–5.07 (m, 6H, H-1 b-CD), 4,39 (t, 1H, H-5
ammonium–glucose), 4.02–3.54 (m, 43H, H-2, H-3, H-4, H-5,
H-6, H-6ꢀ b-CD, H-2, H-3, H-4, H-6, H-6ꢀ ammonium–glucose
and CH2a), 3.25 (s, 3H, CH3–N+), 3.17 (s, 3H, CH3b–N+),
1.87–1.65 (m, 2H, CH2b), 1.44–1.39 (m, 2H, CH2c ), 1.01 (t,
3H, CH3d). 13C NMR (75 MHz, D2O) d 102.40–101.90 (C-
1 b-CD), 100.20 (C-1 ammonium–glucose), 82.97–81.05 (C-4
b-CD), 79.18 (C-4 ammonium–glucose), 73.68–71.69 (C-2, C-
3, C-5 b-CD and C-2, C-3 ammonium–glucose), 67.14 (C-5
ammonium–glucose), 63.66 (Ca), 61.54–60.13 (C-6), 54.18 and
53.25 (N+(CH3)2), 24.73 (CH2b), 19.55 (CH2c ), 13.43 (CH3).
ES-MS (M–Cl)+ m/z calc 1218.5, found 1218.9. C48H84O34NCl,
2H2O %N = 1.09; %C = 44.67; %H = 6.87; found %N = 1.06;
%C = 44.69; %H = 7.06.
Compounds DMA–C6–CD, DMA–C8–CD, DMA–C10–CD,
DMA–C12–CD, DMA–C14–CD and DMA–C16–CD gave sim-
ilar NMR data to that of DMA–C4–CD except for the signal
at 1.50–1.00 ppm on the 1H NMR spectrum whose integration
depended on the number of methylene groups on the alkyl chain.
7 (a) N. Bellanger and N. B. Perly, J. Mol. Struct., 1992, 273, 215; (b) J.
Lin, C. Creminon, B. Perly and F. Djeda¨ıni-Pilard, J. Chem. Soc.,
Perkin Trans. 2, 1998, 2639.
8 (a) H.-J. Schneider, F. Hacket, V. Ru¨diger and H. Ikeda, Chem. Rev.,
1998, 98, 1755; (b) S. Simova and H. J. Schneider, J. Chem. Soc,
Perkin Trans. 2, 2000, 1717; (c) T. L. Hwang and A. J. Shaka, J. Am.
Chem. Soc, 1992, 114, 3157.
9 (a) M. Miyauchi and A. Harada, J. Am. Chem. Soc, 2004, 126, 11418;
(b) Y. Liu, Z. Fan, H.-Y. Zhang and C.-H. Diao, Org. Lett., 2003, 5,
251.
Calculation of association constants by 1H NMR spectroscopy
The Hb proton of Adac was chosen for evaluating the apparent
association constant. Assuming a 1 : 1 inclusion mechanism, the
observed chemical shift of the Hb proton (dOBS) and the complex
concentration [COMP] are described as follows:
10 Y. Sueishi, M. Kasahara, M. Inoue and K. Matsueda, J. Inclusion
Phenom., 2003, 46, 71.
11 L. Fielding, Tetrahedron, 2000, 56, 6151.
12 (a) W. Cromwell, K. Bystrom and M. Eftink, J. Phys. Chem., 1985,
89, 326; (b) M. V. Rekharsky and Y. Inoue, Chem. Rev., 1998, 95,
1875; (c) M. R. Eftink, M. L. Andy, K. Bystrom, H. D. Perlmutter
and D. S. Kristol, J. Am. Chem. Soc., 1989, 111, 6765; (d) P. M.
Ivanov, D. Salvatierra and C. Jaime, J. Org. Chem., 1996, 61, 7012.
13 (a) P. Job, Ann. Chim., 1928, 9, 113; (b) V. Gil and N. Oliveira, J. Chem.
Educ., 1990, 67, 473.
14 D. Landy, S. Fourmentin, M. Salome and G. Surpateanou, J. Inclu-
sion Phenom., 2000, 38, 187.
15 M. J. Rosen, Surfactants and Interfacial Phenomena, John Wiley &
Sons, New York, 2nd edn., 1989, p. 108.
(1)
where K and [ ]tot stand for the association constant and total
concentration, respectively.14 For a given value of K, [COMP]
is known and dCOMP may be calculated from eqn. (1) for each
[CD]tot. Standard deviation over dCOMP is minimized relative to K
to obtain the 1 : 1 association constant. For DMA–Cn–CD with
O r g . B i o m o l . C h e m . , 2 0 0 5 , 3 , 1 1 2 9 – 1 1 3 3
1 1 3 3