NMR spectroscopy: A powerful tool for detecting the conformational features of symmetrical persubstituted mixed cyclomaltoheptaoses (β-cyclodextrins)

Article abstract of DOI:10.1016/j.carres.2004.11.022

The conformation in solution of exhaustively derivatized mixed cyclomaltooligosaccharides (cyclodextrins) has been defined by NMR spectroscopy. Both tilting of glucopyranose units about the glycosidic linkages and ring deviations from the ⁴C

Full text of DOI:10.1016/j.carres.2004.11.022

Carbohydrate
RESEARCH
Carbohydrate Research 340 (2005) 271–281
NMR spectroscopy: a powerful tool for detecting the
conformational features of symmetrical persubstituted mixed
cyclomaltoheptaoses (b-cyclodextrins)
*
Gloria Uccello-Barretta,Giuseppe Sicoli,Federica Balzano and Piero Salvadori
Dipartimento di Chimica e Chimica Industriale, Universit a` di Pisa, via Risorgimento 35, 56126 Pisa, Italy
Received 21 September 2004; accepted 23 November 2004
Abstract—The conformation in solution of exhaustively derivatized mixed cyclomaltooligosaccharides (cyclodextrins) has been
defined by NMR spectroscopy. Both tilting of glucopyranose units about the glycosidic linkages and ring deviations from the
4
C chair conformation are detected,the entities of which are strongly dependent on the nature of the derivatizing groups.
1
ꢀ
2004 Elsevier Ltd. All rights reserved.
Keywords: Cyclomaltooligosaccharides; Cyclodextrins; NMR spectroscopy; Stereochemistry; Conformational distortions
1. Introduction
dramatic changes of their complexing properties. A
1
recent detailed conformational analysis in solution of
3
Cyclomaltooligosaccharides (cyclodextrins) are cyclic
oligosaccharides having a truncated cone shape,the
structural preorganization of which,together with the
chirality of the single glucose units,is responsible for
their complexing and enantiodiscriminating properties,
which can be significantly affected by derivatization of
the peracetylated cyclomaltoheptaose (b-cyclodextrin)
revealed the presence of glucopyranose units having
4
strong deviations from the C conformation. In order
1
to gain better insight on the conformational features
of b-cyclodextrins,we have taken into consideration
the exhaustively derivatized cyclodextrins having mixed
carbamoyl/methyl or acetyl functions,heptakis[6- O-
methyl-2,3-di-O-(3,5-dimethylphenylcarbamoyl)]cyclo-
maltoheptaose (1),heptakis[6- O-acetyl-2,3-di-O-(3,5-
1
their secondary and/or primary hydroxyl groups.
During the last 10 years we have been paying atten-
tion to cyclodextrin derivatives to be used as Chiral Sol-
2
,3
vating Agents (CSAs) that are able to induce NMR
nonequivalence of the enantiotopic nuclei of enantio-
meric mixtures,making them distinguishable in the
NMR spectra. Alkylated or acetylated cyclodextrins,
dimethylphenylcarbamoyl)]cyclomaltoheptaose
(2),
heptakis[2,3-di-O-methyl-6-O-(3,5-dimethylphenylcarb-
amoyl)]cyclomaltoheptaose (3) and heptakis[2,3-di-O-
acetyl-6-O-(3,5-dimethylphenylcarbamoyl)]cyclomalto-
heptaose (4) (Fig. 1),describing their synthesis and
determining by NMR spectroscopy their stereo-
chemistry.
4
extensively employed in GC, have shown great potenti-
5
–8
alities as CSAs for apolar substrates, whereas benzoyl-
ated,benzylated and carbamoylated cyclodextrins have
been successfully employed as CDCl -soluble CSAs for
3
9
–11
enantiodiscriminating polar chiral substrates.
investigations also evidenced the distortions induced in
These
2. Results and discussion
2.1. Synthesis of 1–4
the cyclomaltohexaose (a-cyclodextrin) structure due
to the derivatization, which consequently produced
1
2
*
Corresponding author. Tel.: +39 050 2219273; fax: +39 050 2219260;
e-mail: psalva@dcci.unipi.it
The synthesis of 1 (Scheme 1) involves the protection of
the primary hydroxyl groups as tert-butyldimethylsilyl
0
008-6215/$ - see front matter ꢀ 2004 Elsevier Ltd. All rights reserved.
doi:10.1016/j.carres.2004.11.022

272
G. Uccello-Barretta et al. / Carbohydrate Research 340 (2005) 271–281
OR1
The removal of the TBDMS groups was carried out
1
5
by using BF ÆOEt (7a–b, Scheme 1). The methylation
3
2
O
of the primary sites in the presence of acetyl groups on
the secondary ones (8a) has been carried out by using
RO
OR
O
1
5
methyl triflate and 2,6-di-tert-butyl-4-methylpyridine.
7
When the benzyl groups were present on the secondary
sites,the primary ones could be methylated ( 8b) very
simply by using the procedure involving the addition
of an excess of MeI to the mixture in DMF containing
the cyclodextrin and NaH. However,complete methyl-
ation (>99%) was achieved only by adding the preformed
alcoholate to the DMF solution of MeI. The removal of
the acetyl groups (9a) was performed by using MeONa–
MeOH,whereas complete benzyl ether cleavage ( Fig.
2a) involved the catalytic hydrogenation (10% Pd/C,
R
R1
1
2
3
4
Ar Me
Ar Ac
Me Ar
Ac Ar
H
N
Ar =
O
Figure 1. Persubstituted b-cyclodextrins 1–4.
5 atm) of 8b in MeOH containing a minimum amount
of CH Cl needed to dissolve the cyclodextrin. In fact,
partial debenzylation or decomposition of cyclic malto-
2
2
1
TBDMS) ethers (5), followed by protection of the sec-
4
(
ondary ones as acetyl esters (6a) or benzyl ethers (6b).
1
heptaose derivatives occurred by using only MeOH or
6
OR
O
AcO
AcO
4
OTBDMS
O
OH
c
O
O
b
7
OMe
AcO
AcO
d
AcO
O
AcO
O
O
6
a
7a
AcO
7
7
AcO
e
h
OMe
OMe
O
a
8
a
O
O
c
7
HO
RO
OH
OR
O
O
9a
1
OMe
7
7
O
OTBDMS
O
BnO
BnO
OTBDMS
O
OH
g
a
BnO
8b
O
O
O
f
b
HO
7
7
BnO
BnO
OH
OAc
OAc
OAc
BnO
O
BnO
O
5
O
O
O
O
6
b
7b
7
h
c
7
7
HO
RO
BnO
8c
OH
OR
O
O
c
9b
2
7
7
OTBDMS
O
OH
O
b
b
a
g
RO
RO
OR
OR
6
c
O
7c
O
7
7
7
OTBDMS
O
OH
OR
R = CONH
O
O
c
MeO
MeO
MeO
MeO
O
MeO
MeO
O
O
6
d
7d
3
7
7
Scheme 1. Reagents and conditions: (a) Ac
CF SO CH ,CH Cl ,2 6, -di- tert-butyl-4-methylpyridine; (e) MeONa–MeOH; (f) NaH,DMF,BnBr; (g) NaH,DMF,MeI; (h) H
MeOH–CH Cl
2
O,pyridine,100 ꢁC; (b) BF
3
ÆOEt
2
,CH
2
Cl
2
; (c) 3,5-dimethylphenyl isocyanate, pyridine, 80 ꢁC; (d)
3
3
3
2
2
2
(5 atm),Pd/C,
2
2
.

G. Uccello-Barretta et al. / Carbohydrate Research 340 (2005) 271–281
273
duced a distribution of the cyclodextrin NMR signals
as shown for the H-1 proton in Figure 2c.
Also for the synthesis of system 2 we followed two dif-
ferent pathways (Scheme 1). The first involved carbam-
oylation of the secondary sites (6c) of the derivative 5,
which had its primary rim protected by silyl groups,
the removal of its TBDMS groups (7c) and,finally,
the acetylation of the primary hydroxyl groups (2) with
acetic anhydride; the latter required the acetylation of
the primary hydroxyl groups (8c) of the benzylated
derivative 7b,the cleavage of the benzyl ether bonds
c
10.0 9.6
9.2
8.8
8.4
8.0
7.6
(
9b) and the final carbamoylation of the secondary rim
(2).
Cyclodextrins 3 and 4,having inverted substitution on
b
its primary and secondary sites with respect to 1 and 2,
respectively,were synthesized by carbamoylation of the
OH -3
OH -2
H
1
2,3-di-O-methyl- and 2,3-di-O-acetyl cyclodextrins 7d
and 7a (Scheme 1).
Products 1–4 and their intermediates were character-
a
1
ized by H NMR spectroscopy by comparison of the
1
3
homonuclear DQF-COSY and ROESY maps. The
C
6
.0
5.8
5.6
5.4
5.2
5.0
4.8
4.6
4.4
ppm
NMR resonances were assigned on the basis of the het-
eronuclear scalar correlations detected in the HETCOR
1
Figure 2. H NMR (300 MHz,DMSO- d
corresponding to the secondary hydroxyls and the anomeric H-1
proton of the debenzylation products obtained by (a) H ,Pd/C,
MeOH–CH Cl ,(b) NaBrO –Na in AcOEt–H O and (c) EtSH–
BF ÆOEt
6
,25 ꢁC) spectral regions
1
13
maps. The H and C NMR chemical shift data are re-
ported in Section 4.
2
2
2
3
2
S
2
O
4
2
3
2
.
2.2. Conformational analysis of 1–4
1
by applying the procedure that involved the use of
7
The stereochemistry of the derivatives 1–4 has been
investigated in CDCl solution both by ROESY analy-
dioxane–EtOH and one drop of HClO . Compound 1
4
3
was finally obtained by reacting 9a with 3,5-dimethyl-
phenyl isocyanate in pyridine at 80 ꢁC.
In regard to the removal of the benzyl protecting
ses and proton-selective relaxation measurements.
Indeed,as recently reported for the per- O-acetyl-b-
cyclodextrin, selected traces of the ROESY maps are
diagnostic in the detection of the macrocyclic distortion
induced by the derivatization.
1
3
groups of 8b,we also tried other methods reported in
1
8–20
the literature
as alternative approaches to catalytic
low-pressure hydrogenation. Of these,we took into con-
sideration debenzylation employing NaBrO –Na S O
4
In particular,the H-1 traces contain information con-
cerning the mutual rotation of adjacent glucose units
around the glycosidic linkages (Fig. 3),being the inter-
3
2
2
reagent under two-phase (ethyl acetate–water) condi-
1
8
0
tions, the deprotection in the presence of a hard acid,
boron trifluoride etherate,and a soft nucleophile,
unit H-1–H-4 NOE (the apex is referred to protons
belonging to the adjacent unit) more intense than the
intraunit H-1–H-2 NOE in the absence of such a rota-
1
9
EtSH, and ꢀcatalytic-transfer hydrogenationꢁ involving
the use of sodium hypophosphite associated with palla-
12
tion,as in the case of the parent b-cyclodextrin, where
the distance r14 is shorter than r12.
2
0
dium–charcoal.
An attempt to cleave benzyl ether C–O bonds in 8b
0
The analysis of the H-3 and/or H-2 traces allows one
to detect the glucopyranose ring conformational distor-
tions. In fact,the distances of the proton pairs H-4–H-3
with sodium hypophosphite in the presence of 10% Pd/
C in DMF–H O at 50 ꢁC (4 h) gave only 15% of the
2
1
O-debenzylated product as revealed by H NMR analy-
˚
and H-3–H-2 assume the highest values (>3 A) in the
4
sis. The treatment of 8b with NaBrO –Na S O in ethyl
3
C1 chair (Fig. 4a),but any ring distortion ( Fig. 4)
causes them to become shorter,as in the completely in-
2
2
4
acetate–H O at room temperature for 16 h resulted in
2
1
the formation of a complex mixture whose H NMR
1
verted C chair (Fig. 4b).
4
0
spectrum (Fig. 2b) indicated no removal of the benzyl
groups (Fig. 2b compared to Fig. 2a) as the OH-2 and
OH-3 signals of the fully debenzylated compound were
not detected. Analogous results (Fig. 2c) were obtained
in the case of the reaction of 8b with ethanethiol in the
presence of BF ÆOEt . The use of EtSH–BF ÆOEt pro-
Comparable H-1–H-4 and H-1–H-2 NOEs were de-
tected in the H-1 trace (Fig. 5) of derivative 4,which
was acetylated on the secondary rim,according to a de-
gree of mutual rotation of the adjacent units,which is
similar to that found for the peracetylated b-cyclodex-
1
3
trin. By contrast,in the corresponding ROESY trace
3
2
3
2

274
G. Uccello-Barretta et al. / Carbohydrate Research 340 (2005) 271–281
Figure 3. Two glucopyranose units of b-cyclodextrin.
Figure 5. 2D ROESY (300 MHz,CDCl
derivatives 1–4. Traces correspond to the H-1 proton.
3
m
,25 ꢁC, s = 0.3 s) analysis of
relaxation measurements,at least in the cases of the
derivatives 1, 2 and 4,which are characterized by well-
1
separated proton resonances in their H NMR spectra.
2
1
Indeed,in the initial rate approximation
the cross-
Figure 4. Representation of some glucopyranose ring conformation:
1
4
a) C
relaxation term r ,which describes the magnetization
ij
(
1
chair,(b)
4
C chair,(c) boat,(d) skew.
transfer between the proton pair ij,is a function of the
reorientational correlation time r of the vector connect-
ing the two spins i and j and of their distance r . In the
ij
c
(Fig. 5) of the cyclodextrin 2,having the acetyl moieties
on the primary rim and the O-carbamate functions on
0
fast and slow motion regimes,such parameters become a
simple function of s and r ,as reported in Eqs. 1 and 2:
the secondary one,a more intense H-1–H-4 NOE than
the H-1–H-2 one was observed as an evidence of a
c
ij
0
nearly retained coplanarity of the C-1–H-1 and C-4 –
0
2
4
À6
r ¼ 0:5c ꢀh r s fast motion
ð1Þ
ð2Þ
ij
ij
c
H-4 bonds,which is expected for the truncated cone
shape of the underivatized system.
4
2
À6
An analogous trend of the relative intensities of the H-
0
r
ij ¼ À0:1c ꢀh rij
s
c
slow motion
1
–H-4 and H-1–H-2 NOEs was observed in the traces
where c is the proton magnetogiric ratio and ⁄ is the re-
duced Planck constant.
In the hypothesis of isotropic motion,the same reori-
entational time can be attributed to all the molecule,and
of the anomeric protons of the corresponding O-methyl
systems 1 and 3 (Fig. 5). Indeed,once again the deriva-
tive 1,having the carbamoyl moieties on the cyclodex-
trin 2- and 3-positions,adopts an unperturbed
stereochemical arrangement. On the contrary,the mu-
tual rotation of adjacent rings around the glycosidic
linkages was observed for the derivative 3,which was
carbamated on the primary rim.
the ratio of the different r values are correlated to the
ij
ratios of the internuclear distances,as reported in Eq. 3:
6
ij
6
ik
r
ik
r
¼ r
ð3Þ
r
ij
The NOE data are in good agreement with the inter-
proton distances ratios calculated from the cross-relaxa-
1
tion terms obtained by H mono- and biselective
The r parameters can be very simply calculated as dif-
ij
ferences between the biselective and monoselective relax-

G. Uccello-Barretta et al. / Carbohydrate Research 340 (2005) 271–281
275
ation rates for the proton pair ij. The monoselective
ms
relaxation rates Ri are obtained by selective inversion
of the spin i leaving unperturbed the other protons,
bs
whereas the biselective ones Rij are measured following
the recovery of the proton i in the time under simulta-
neous inversion of the sole spin j.
In this way we calculated the r12 and r14 cross-relax-
0
ation terms by means of the biselective and monoselec-
bs
bs
14
ms
tive relaxation rates R , R
0
and R1 indicated in
12
Table 1,and,hence,by using Eq. 3,we determined the
ratios between the interproton distances r14 and r12
Table 1).
For cyclodextrins 1 and 2,having carbamoylated sec-
ondary sites,these ratios are,respectively,0.81 and 0.76,
to indicate that the r14 distance is significantly shorter
with respect to that of r . By contrast,the value 0.98
0
(
0
1
2
was calculated for 4,having acetylated secondary func-
tions,which is very close to the one obtained for the per-
1
3
acetylated cyclodextrin, according to the presence of
relative rotation of the glucosidic units around the gly-
cosidic linkages.
We could not perform the same kind of analysis for
the cyclodextrin 3 due to partial resonance superimposi-
tions,which prevented us from selectively inverting the
Figure 6. 2D ROESY (300 MHz,CDCl
derivatives 1–4. Traces correspond to the H-2 proton.
3
,25 ꢁC, s
m
= 0.3 s) analysis of
0
proton H-4 .
Regarding to the C chair distortion,the ROESY
4
1
traces of the H-2 protons (Fig. 6) for 1–4 and also of
H-3 protons (Fig. 7) for the systems 1, 2 and 4,which
gave rise to well-resolved proton resonances,were
analyzed.
In the H-2 traces (Fig. 6),together with the usual H-
4
–H-1 and H-2–H-4 NOEs typical for the C chair,H-
–H-3 dipolar interactions were detected for all the
2
2
1
derivatives. These last NOEs are unexpected for 1,2-
4
diaxial protons in the C chairs and can be diagnostic
1
of the presence of distorted units,having the two pro-
tons in pseudodiequatorial arrangements. However,
the most interesting point is the comparison between
the relative magnitudes of the H-2–H-4 and H-2–H-3
NOEs. In fact,for the cyclodextrins with the carbamoyl
functions on the secondary rim and acetylated or meth-
ylated groups on the primary positions (1–2),the inter-
NOE H-2–H-4 was largely prevailing on the H-2–H-3
effect,and,hence,only few distorted units are present.
ms
i
bs
ij
Table 1. Measured mono- and biselective relaxation rates (R , R ,
s
À1
À1
),calculated cross-relaxation parameters ( rij, s ) and interproton
0
/r12 ratios for 1–2 and 4
14
Figure 7. 2D ROESY (300 MHz,CDCl
derivatives 1, 2 and 4. Traces correspond to the H-3 proton.
3
m
,25 ꢁC, s = 0.3 s) analysis of
r
ms
bs
ij
CD
ij
Ri
R
r
ij
14
r /r12
0
1
12
2.49
2.49
2.32
1.88
À0.17
À0.61
0.81
0.76
0.98
By contrast,for the two cyclodextrins 3 and 4,which
are carbamoylated on the primary rim,these NOEs
are comparable. This result indicates the presence of a
significant population of distorted units and can be the
consequence of two different phenomena: (1) the short-
ening of the H-2–H-3 distance and/or (2) the lengthening
0
0
0
14
2
4
12
2.20
2.20
2.12
1.80
À0.08
14
À0.40
12
2.26
2.26
2.11
2.09
À0.15
À0.17
14

276
G. Uccello-Barretta et al. / Carbohydrate Research 340 (2005) 271–281
of the H-2–H-4 distance,which could be concomitant in
a distorted ring. This last conclusion is supported by the
fact that in the two cyclodextrins 3 and 4,the inter-
NOEs H-2–H-4 become smaller with respect to the H-
2–H-1 one,and,taking into account that any ring dis-
tortion cannot change the H-2–H-1 distance signifi-
cantly,the H-2–H-4 distance must be larger in the
4
distorted units relative to the C chairs.
The traces of the H-3 proton showed the expected H-
1
4
–H-5 NOE corresponding to the undistorted C con-
3
1
formation and significant H-3–H-4 and H-3–H-2 NOEs
Fig. 7),supporting the presence of some units in a ste-
(
reochemical arrangement,which brings H-3 pseudoequ-
atorial. The relative contribution of the distorted units
with respect to the undistorted ones is well reflected in
the relative intensities of the H-3–H-5 and H-3-H-4 or
H-3–H-2 NOEs. As a matter of fact,the last ones are
more significant for the cyclodextrin 4,which is carba-
moylated on the primary rim,than they are for 1 and 2.
It is noteworthy that the traces of the protons H-1,H-
2
distortions. In fact the completely inverted C chair
and H-3 also allow us to exclude some kinds of ring
1
Figure 8. 2D ROESY (300 MHz,CDCl
of 3. Traces correspond to the ortho protons and the methyl groups of
3
m
,25 ꢁC, s = 0.3 s) analysis
4
(
Fig. 4b) should give rise to a strong contribution to
the 3,5-dimethylphenyl moiety.
the H-1–H-6 NOE,which is not detected. The boat con-
formation indicated in Figure 4c could keep the H-3 and
H-4 protons diaxial as well. This is in disagreement with
the relevant H-3–H-4 inter-NOE detected in all the H-3
traces. Therefore,our data support the presence of some
units in a skew type conformation having pseudoequa-
torial the H-2,H-3 and H-4 protons,as shown in Figure
carbamate functions (Fig. 10),which is in agreement
with a nearly torus shape in which the carbamate moie-
ties contribute to extend the cavity bringing them far
away from the methoxy groups on the primary rim
(Fig. 11).
4
d.
In the case of the peracetylated b-cyclodextrin,an
Me
Me
analogous kind of ring distortion was evidenced,not
only by NOE measurements,but also by determining
the averaged interproton distances by means of pro-
ton-selective relaxation rates; however,for cyclodex-
trins 1–4,this approach was not possible as the
resonances of the two diastereotopic protons H-6a and
H-6b at fixed distance were partially superimposed on
other signals. Hence,we could not determine their
cross-relaxation term to be used to calculate other un-
known proton distances by means of Eq. 3.
Me
Me
1
3
N
H
O
H
N
O
OR
6
O
O
O
2
3
O
3
O
6
RO
OR
2
4C
O
O
1
skew
Finally,the relative positions of the substituents on the
primary and secondary rims have been determined for
the methylated–carbamoylated cyclodextrins 1 and 3.
For 3,which is carbamoylated on the primary hydroxyl
groups,inter-NOEs between the 35 ,- dimethylphenyl
and the methoxy groups in the 2- and 3-positions were
observed (Fig. 8),which can be attributed to the tilting
of adjacent units and to the relevant degree of glucopyr-
anose distortions,which leads to the spatial proximity of
the aromatic and methoxy moieties,as shown in Figure 9.
In the case of 1,which is methylated on the primary
Me
RO
NOE
Me
Me
Me
H
O
N
O
N
H
O
6
O
O
O
6
2
2
O
O
RO
3
OR
3
OR
4C
O
4C
1
1
rim,and for which a lesser degree of tilting and devia-
RO
R = Me
4
tion from the C chair were found,no dipolar interac-
tions were observed between the OMe-6 and the
1
Figure 9. Stereochemical representation of 3.

G. Uccello-Barretta et al. / Carbohydrate Research 340 (2005) 271–281
277
able to stabilize interunit attractive interactions are pres-
ent on the secondary rim,the cyclodextrin derivatives
show a low degree of ring deformations and negligible
tilting of adjacent units,which is a behaviour that is ex-
pected for an underivatized b-cyclodextrin. In these
cases,the derivatizing groups can produce an extension
of the truncated cone cavity,which,in principle,could
retain its inclusive properties. By contrast,when the sec-
ondary sites are devoid of hydrogen-bond donor/accep-
tor groups,the truncated cone shape of cyclodextrins is
strongly perturbed. This is due mainly to the glucopyra-
nose ring deformations with a small,but significant,
contribution from the tilting of the units. It is worth not-
ing that analogous phenomena were determined by the
Figure 10. 2D ROESY (300 MHz,CDCl
of 1. Trace correspond to the methoxy protons on the primary rim.
3
m
,25 ꢁC, s = 0.3 s) analysis
9
,12
derivatization of a-cyclodextrins. In some cases,the
tilting of the units led to a limit situation where reduc-
9
,22
4C
tion of the symmetry from C-6 to C-3 occurred,
OR
6
1
4C
1
and in the H NMR spectra two different kinds of glu-
copyranose rings were observed,some of which deviated
4
themselves from the C structure. Therefore,in general,
OR
6
1
O
O
2
O
3
O
2
3
1
O
O
O
O
O
when these two effects,tilting and unit deformations,
give strong contributions to the stereochemical arrange-
ments of the cyclodextrins,then they can generate a flat-
tening of the structure where the polarity differences
between internal and external surfaces are made less
pronounced,and,hence,the complexing properties
could change dramatically. In these cases,inclusion pro-
cesses could become less favoured relatively to superfi-
cial interactions,mainly driven by the nature of the
derivatizing groups independently from the cyclic struc-
ture of the systems.
O
H
O
H
O
N
H
N
O
H
N
Me
N
Me
Me
Me
R = Me
Me
Me
Me
Me
Figure 11. Stereochemical representation of 1.
For the corresponding acetylated systems, 2 and 4,the
superimposition of the methyl resonances of the acetyl
and 3,5-dimethylphenyl moieties precludes the analysis.
4. Experimental
4.1. General methods
3. Conclusions
NMR measurements were performed on a spectrometer
1
operating at 300 and 75 MHz for H and C,respec-
13
NMR spectroscopy is confirmed to constitute one of the
most powerful tools of detection of conformational dis-
tortions originated by derivatization of the glucopyra-
nose rings of cyclodextrins,in spite of their symmetry.
In fact,intraunit and interunit dipolar interactions,
due to interproton spatial proximity,can be efficiently
detected by mono- and bidimensional techniques,allow-
ing us to impose valid restraints in conformational anal-
ysis. In the case of symmetrical perderivatized b-
systems,two kinds of distortions are detected: (1) the
tively,and the temperature was controlled to ±0.1 ꢁC.
All H and C NMR chemical shifts are referenced to
1
13
Me Si as external standard. The 2D NMR spectra were
4
obtained using standard sequences. The double-quan-
tum-filtered (DQF) COSY experiments were recorded
with the minimum spectral width required; 512 incre-
ments of eight scans and 2K data points were acquired.
The relaxation delay was 5 s. The data were zero-filled
to 2K · 1K and a Gaussian function was applied for
processing in both dimensions. The HETCOR spectra
were acquired with the minimum spectral width required
in F and in F in 2K data points using 64 scans of the
tilting of glucopyranose rings about the glycosidic link-
4
ages and (2) the deviation of units from the expected C
1
chair,leading to skew conformations. Both of the above
effects give contributions to the molecular conforma-
tion,which strongly depend on the nature of the deriv-
atizing groups and mainly on the nature of the groups
introduced on the secondary sites. In particular,when
hydrogen-bond acceptor/donor substituents that are
2
1
512 increments. The relaxation delay was 1 s. The data
were zero-filled to 2K · 1K,and a Gaussian function
was applied for processing in both dimensions. The
ROESY spectra were recorded in the phase-sensitive
mode,by employing a mixing time ranging from 0.2 to

278
G. Uccello-Barretta et al. / Carbohydrate Research 340 (2005) 271–281
0
.6 s. Within the above mixing time range any significant
4.4. Heptakis(2,3-di-O-benzyl-6-O-tert-butyldimethyl-
silyl)cyclomaltoheptaose (6b)
variation of NOE intensities were not observed. The
spectral width used was the minimum required in both
dimensions. The pulse delay was maintained at 5 or
To a solution of 5 (7.50 g,3.88 mmol) in dry DMF
(120 mL) was added,at 0 ꢁC,NaH (4.51 g,188 mmol).
The suspension was allowed to stir at 0 ꢁC for 2 h,and
then benzyl bromide (22.4 mL,32.2 g,188 mmol) was
added. The reaction mixture was stirred for further 2 h
at 0 ꢁC and 12 h at room temperature and then
quenched with MeOH. The organic layer was extracted
with pentane,washed with H O,dried (Na SO ) and fi-
1
0 s; 512 increments of eight scans and 2K data points
each were collected. The data matrix was zero-filled to
K · 1K,and a Gaussian function was applied for pro-
2
cessing in both dimensions. The selective relaxation
rates were measured in the initial rate approximation
by employing a selective p pulse with the proton decou-
pler at the selected frequency for 25 ms. After the delay
s,a nonselective p/2 pulse was employed to detect the
longitudinal magnetization. For the biselective measure-
ments,the two protons were inverted consecutively.
Thin-layer chromatography (TLC) was carried out on
silica gel plates (E. Merck,Silica G-60 0.2 mm),and
compounds were visualized with iodine vapour or by
examination under UV light. Chromatography was car-
ried out using Silica Gel 60 (70–230 mesh ASTM). Melt-
ing points were determined using a Koffler hot-stage
apparatus.
2
2
4
1
nally concentrated to give 6b (87%): H NMR (CDCl ,
3
25 ꢁC): d 0.01 (MeSi,s,21H),0.02 (MeSi,s,21H),
t
0.88 ( Bu,s,63H),3.39 (H-2,dd,
J
9.3 Hz, J2,1
,3
2
0
3.2 Hz,7H),3.71 (H-6 ,br d, J6
0
11.5 Hz,7H),3.74
,6
(H-5,br d,
J5,4 9.0 Hz,7H),4.0 (H-4,dd,
J4,5 = J4,3 = 9.0 Hz,7H),4.06 (H-3,dd, J3,2 9.3 Hz, J3,4
9.0 Hz,7H),4.26 (H-6,br d,
and 4.53 (CH ,Bn-2,d,
0
J6,6 11.5 Hz,7H),4.49
J 12.0 Hz,14H),4.71 and
2
5.08 (CH ,Bn-3,d, J 10.8 Hz,14H),5.32 (H-1,d,
2
J1,2
3.2 Hz,7H),6.96–7.45 (Ar,Bn-2 and Bn-3,m,70H).
1
3
C NMR (CDCl ,25 ꢁC): d À5.2 (MeSi), À4.8 (CH Si),
3
3
1
7
8.3 [(CH ) C],25.9 [( CH ) C],62.4 (C-6),72.5 (C-5),
3 3 3 3
4
.2. Materials
2.6 (CH ,Bn-2),75.5 (CH 2,Bn-3),77.7 (C-4),79.3
2
(
1
C-2),80.9 (C-3),97.9 (C-1),126.8,127.4,127.5,127.7,
27.9,128.0 (CH,Bn-2 and Bn-3); 138.3,139.3 (quater-
Cyclomaltoheptaose (b-cyclodextrin) and tert-butyl-
chlorodimethylsilane (TBDMSCl) were purchased from
Fluka Chemical Co. Boron trifluoride diethyl etherate
nary C,Bn-2 and Bn-3). Anal. Calcd for
C182H252O Si : C,68.39; H,7.95. Found: C,68.10; H,
7
35
7
(
BF ÆEt O),3 5, -dimethylphenyl isocyanate,Pd/C,meth-
3 2
.81.
yl triflate and 2,6-di-tert-butyl-4-methylpyridine were
obtained from Aldrich Chemical Co. b-Cyclodextrin
was dried (8 h) at 110 ꢁC/0.1 Torr,in the presence of
P O . Heptakis(6-O-tert-butyldimethylsilyl)cyclomalto-
4.5. Heptakis(6-O-tert-butyldimethylsilyl-2,3-di-O-
methyl)cyclomaltoheptaose (6d)
2
5
1
4
heptaose (5), heptakis[6-O-tert-butyldimethylsilyl-2,3-
di-O-(3,5-dimethylphenylcarbamoyl)]cyclomaltoheptaose
To a suspension of NaH (4.30 g,0.18 mmol) in dry THF
(30 mL) was added at 0 ꢁC under stirring a solution of 5
(4.95 g,2.56 mmol) in dry THF and iodomethane
1
6c) and heptakis[2,3-di-O-(3,5-dimethylphenylcarba-
1
(
moyl)]cyclomaltoheptaose (7c)
described elsewhere.
1
1
were prepared as
(16.2 mL,36.9 g,0.26 mol). The mixture was allowed
to stir at 0 ꢁC for 1 h and at room temperature for fur-
ther 18 h. MeOH was added to quench the reaction. The
organic solvents were removed under reduced pressure,
and the residue was dissolved in CH Cl and washed
4
lyl)cyclomaltoheptaose (6a)
.3. Heptakis(2,3-di-O-acetyl-6-O-tert-butyldimethylsi-
1
5
2
2
with H O. The organic layer was dried (Na SO ),and
2
2
4
1
Yield,97%. H NMR (CDCl ,25 ꢁC): d 0.017 (MeSi,s,
the solvent was evaporated to give 6d: mp 89–91 ꢁC.
3
t
1H),0.021 (MeSi,s,21H),0.86 ( Bu,s,63H),2.02 (Me-
1
2
3
1
H NMR (CDCl ,25 ꢁC): d À0.03 (Me Si,s,42H),
3
2
0
t
0.82 ( Bu,s,63H),3.01 (H-2,dd,
,s,21H),2.03 (Me-2,s,21H),3.70 (H-6
1.9 Hz,7H),3.84 (H-5,br d,
H-4,dd, J4,3 9.8 Hz, J4,5 6.6 Hz,7H),4.02 (H-6,br d,
11.9 Hz,7H),4.68 (H-2,dd,
.6 Hz,7H),5.14 (H-1,d,
dd, J3,2 10.1 Hz, J3,4 9.8 Hz,7H).
,d, J6,6
J5,4 6.6 Hz,7H),3.86
0
J
3.6 Hz,7H),3.46 (OMe-2,s,21H),3.50 (H-3,dd,
9.7 Hz, J
2,3 2,1
J3,2
J5,4 9.3 Hz, J5,6
(
J6,6
9.7 Hz, J3,4 9.3 Hz,7H),3.55 (H-5,m,
2.5 Hz,7H),3.60 and 4.07 (H-6–H-6
0
0
J2,3 10.1 Hz, J2,1
J1,2 3.6 Hz,7H),5.33 (H-3,
,dd,
11.7 Hz, J6,5 2.5 Hz,14H),3.62 (OMe-3,s,21H),3.68
J6,6
0
3
1
3
C NMR (CDCl3,
(H-4,t, J4,3 = J = 9.3 Hz,7H),5.15 (H-1,d,
4,5
J1,2
1
3
2
2
7
1
5 ꢁC): d À5.4 (MeSi), À5.1 (MeSi),18.2 [(CH ) C],
3.6 Hz,7H). C NMR (CDCl ,25 ꢁC): d À4.9 (MeSi),
3
3
3
0.7 (Me-3),20.8 (Me-2),25.8 [( CH ) C],61.8 (C-6),
3
À5.2 (MeSi),18.2 [(CH ) C],25.9 [( CH ) C],58.6
3
3 3
3 3
1.2 (C-2),71.5 (C-3),71.8 (C-5),75.2 (C-4),96.4 (C-
),169.4 (CO,Ac-3),170.8 (CO,Ac-2). Anal. Calcd
(OMe-2),61.4 (OMe-3),62.3 (C-6),72.1 (C-5),78.6
(C-4),82.0 (C-3),82.2 (C-2),98.1 (C-1). Anal. Calcd
for C H O Si : C,55.23; H,9.27. Found: C,55.15;
for C112H196O Si : C,53.3; H,7.83. Found: C,53.91;
49
7
98 196 35
7
H,7.04.
H,9.11.

G. Uccello-Barretta et al. / Carbohydrate Research 340 (2005) 271–281
279
1
5
4
.6. Heptakis(2,3-di-O-acetyl)cyclomaltoheptaose (7a)
7H),4.01 (H-5,br d, J5,4 8.7 Hz,7H),4.80 (H-2,dd,
J2,3 10.1 Hz, J2,1 3.7 Hz,7H),5.12 (H-1,d, J1,2 3.7 Hz,
1
13
Yield,98%. H NMR (Me SO-d ,50 ꢁC): d 1.99 and
7H),5.33 (H-3,dd, J3,2 10.1 Hz, J3,4 8.4 Hz,7H).
C
2
6
0
2
1
1
7
.01 (Me-2 and Me-3,s,42H),3.63 (H-6
,dd,
J
0
NMR (CDCl ,25 ꢁC): d 20.7 and 20.8 (Me-3 and Me-
6
,6
3
1.8 Hz, J6 4.6 Hz,7H),3.79 (H-6,dd,
0
J6,6
J5,4 8.7 Hz,
H),3.84 (H-4,dd, J4,5 8.7 Hz, J4,3 8.6 Hz,7H),4.62
0
2),59.1 (OMe),70.6 (C-5),70.8 (C-2),71.0 (C-6),71.2
(C-3),75.7 (C-4),96.5 (C-1),169.4 and 170.6 (CO,Ac-
3 and Ac-2). Anal. Calcd for C H O : C,50.77; H,
,OH
1.8 Hz, J6,OH 6.3 Hz,7H),3.80 (H-5,d,
77
112 49
(
H-2,dd, J2,3 9.7 Hz, J2,1 3.4 Hz,7H),4.65 (OH,br
dd, JOH,6 6.3 Hz, JOH,6 4.6 Hz,7H),5.09 (H-1,d,
.4 Hz,7H),5.26 (H-3,dd, J3,2 9.7 Hz, J3,4 8.6 Hz,
H). C NMR (Me SO-d 50 ꢁC): d 20.2 and 20.3
6.20. Found: C,51.25; H,6.34.
0
J2,1
3
7
4.10. Heptakis(2,3-di-O-benzyl-6-O-methyl)cyclomalto-
heptaose (8b)
1
3
2
6
(
Me-2 and Me-3),59.5,70.4,70.5,71.8,75.1,95.8 (C-
1
Calcd for C H O : C,48.78; H,5.73. Found: C,
–C-6),169.1 and 169.9 (CO,Ac-2 and Ac-3). Anal.
To a solution of 7b (4.31 g,1.80 mmol) in dry DMF
7
0
98 49
(50 mL) was added,at 0 ꢁC,NaH (2.81 g,117 mmol).
4
9.11; H,5.08.
The mixture was stirred at 0 ꢁC for 1 h. The resulting
suspension was added at 0 ꢁC to a solution of iodome-
thane (0.9 mL,2.05 g,14.4 mmol) in dry DMF
4
.7. Heptakis(2,3-di-O-benzyl)cyclomaltoheptaose (7b)
(
14 mL). The reaction mixture was stirred at 0 ꢁC for a
O-Desilylation of 6b was carried out as described for the
1
further 1 h,then at room temperature for 3 h. The reac-
tion was quenched at 0 ꢁC by adding MeOH,and the
mixture was concentrated. The residue was dissolved
in CH Cl ,washed with H O,Na S O (10%),brine
5 1
preparation of 7a (94%): H NMR (CDCl ,25 ꢁC): d
3
3
dd, J4,5 8.8 Hz, J4,3 8.1 Hz,7H),3.77 (H-6
.46 (H-2,dd, J2,3 9.3 Hz, J2,1 3.7 Hz,7H),3.62 (H-4,
0
, J
0
6 ,6
2
2
2
2
2
3
1
2.3 Hz,7H),3.91 (H-5,d, J5,4 8.8 Hz,7H),3.91 (H-6,
d, J6,6 12.3 Hz,7H),3.93 (H-3,dd, J3,2 9.3 Hz, J3,4
.1 Hz,7H),4.43 and 4.52 (CH 2,Bn-2,d, J 12.3 Hz,
4H),4.67 and 4.87 (CH ,Bn-3,d, J 12.3 Hz,14H),
and dried (Na SO ). Compound 8b was obtained
2
4
0
(
1
92%) by removing the solvent under reduced pressure.
H NMR (CDCl ,25 ꢁC): d 3.26 (OMe,s,21H),3.45
8
1
4
7
2
3
2
0
(H-2,dd, J 9.5 Hz, J 3.6 Hz,7H),3.47 (H-6 ,d, J
2
,3
2,1
.75 (OH,br s,7H),5.01 (H-1,d,
.43 (Ar,Bn-2 and Bn-3,m,70H).
J
3.7 Hz),7.01–
,2
1
1
3
8
1
5
0.0 Hz,7H),3.85 (H-4,m,7H),3.87 (H-6,m,7H),
.87 (H-5,m,7H),3.99 (H-3,dd, J3,2 9.5 Hz, J3,4
.5 Hz,7H),4.41 and 4.50 (CH 2,Bn-2,d, J 12.5 Hz,
1
3
C NMR (CDCl3,
5 ꢁC): d 61.6 (C-6),72.8 (CH ,Bn-2),73.0 (C-5),74.9
2
(
(
CH ,Bn-3),78.4 (C-2),78.8 (C-4),80.4 (C-3),98.3
C-1),127.0,127.3,127.5,127.8,128.0,128.2 (CH,Bn-
2
4H),4.76 and 5.06 (CH ,Bn-3,d, J 10.9 Hz,14H),
2
.04 (H-1,d, J1,2 3.6 Hz,7H),7.10–7.30 (Ar,Bn-2 and
13
2
3
and Bn-3),138.3,139.0 (quaternary C,Bn-2 and Bn-
). Anal. Calcd for C140H154O : C,70.16; H,6.48.
Bn-3,m,70H).
C NMR (CDCl ,25 ꢁC): d 58.9
3
35
(
(
(
MeO),71.1 (C-5),71.4 (C-6),73.0 (CH
CH ,Bn-3),77.4 (C-4),78.8 (C-2),80.9 (C-3),98.5
,Bn-2),75.4
2
Found: C,70.61; H,6.15.
.8. Heptakis(2,3-di-O-methyl)cyclomaltoheptaose (7d)
O-Desilylation of 6d was carried out as described for the
2
C-1); 126.9,127.2,127.5,128.0,128.2 (CH,Bn-2 and
4
Bn-3); 138.3,139.3 (quaternary C,Bn-2 and Bn-3).
Anal. Calcd for C147 H168O : C,70.82; H,6.79. Found:
C,71.11; H,6.97.
35
1
5
1
preparation of 7a (80%): mp 158–160 ꢁC. H NMR
CDCl ,25 ꢁC): d 3.16 (H-2,dd, 8.9 Hz, J
(
J
2,3
3
2,1
3
3
5
7
2
.7 Hz,7H),3.44 (H-3,m,7H),3.47 (H-4,m,7H),
.47 (OMe-2,s,21H),3.59 (OMe-3,s,21H),3.74 (H-
,m,7H),3.79 and 3.80 (H-6,m,14H),4.78 (OH-6,s,
H),5.03 (H-1,d,
5 ꢁC): d 58.6 (OMe-2),61.3 (OMe-3),61.4 (C-6),72.4
4.11. Heptakis(6-O-acetyl-2,3-di-O-benzyl)cyclomalto-
3
heptaose (8c)
2
1
J1,2 3.7 Hz). C NMR (CDCl3,
3
To a solution of 7b (7.53 g,3.14 mmol) in dry pyridine
(80 mL) was added Ac O (26.0 mL,28.1 g,276 mmol).
2
(C-5),80.2 (C-4),81.6 (C-2),81.9 (C-3),98.8 (C-1).
Anal. Calcd for C H O : C,50.52; H,7.42. Found:
The solution was stirred at 100 ꢁC for 14 h and at room
temperature for a further 83 h. The pyridine was re-
moved under reduced pressure. The residue was dis-
solved in toluene,dried (Na SO ) and concentrated to
5
6
98 35
C,50.43; H,7.49.
2
4
1
4
heptaose (8a) (92%)
.9. Heptakis(2,3-di-O-acetyl-6-O-methyl)cyclomalto-
1
give 8c (94%): H NMR (CDCl ,25 ꢁC): d 2.17 (Me,
3
5
s,21H),3.63 (H-2,dd,
.85 (H-4,dd, J4,3 = J4,5 = 8.7 Hz,7H),4.18 (H-3,dd,
J3,2 9.1 Hz, J3,4 8.7 Hz,7H),4.25 (H-5,d, J5,4 8.7 Hz,
7H),4.56 (H-6,m,14H),4.55 and 4.71 (CH 2,Bn-2,d,
J 12.3 Hz,14H),4.90 and 5.17 (CH 2,Bn-3,d,
J 11.3 Hz,14H),5.11 (H-1,d, J1,2 3.2 Hz,7H),
J
9.1 Hz, J 3.2 Hz,7H),
,3 2,1
2
3
1
H NMR (CDCl ,25 ꢁC): d 2.02 and 2.05 (Me-2 and
3
0
Me-3,s,42H),3.38 (OMe,s,21H),3.55 (H-6
0
,br d,
J4,5 8.7 Hz, J4,3
10.9 Hz, J6,5 5.1 Hz,
J6
10.9 Hz,7H),3.82 (H-4,dd,
J6,6
,6
8
.4 Hz,7H),3.88 (H-6,dd,
0

280
G. Uccello-Barretta et al. / Carbohydrate Research 340 (2005) 271–281
1
3
7
2
7
9
.15–7.55 (Ar,m,70H).
C NMR (CDCl ,25 ꢁC): d
4.14. Heptakis[6-O-methyl-2,3-di-O-(3,5-dimethylphen-
ylcarbamoyl)]cyclomaltoheptaose (1)
3
0.5 (Me),63.3 (C-6),69.7 (C-5),72.9 (CH
5.1 (CH ,Bn-3),78.5 (C-2),79.6 (C-4),80.2 (C-3),
8.6 (C-1),126.9,127.3,127.5,127.8,128.0,128.1 (C,
,Bn-2),
2
2
A solution of 9a (1.12 g,0.91 mmol) and 3 5, -dimethyl-
phenyl isocyanate (2.7 mL,2.82 g,19.1 mmol) in dry
pyridine (20 mL) was stirred at 80 ꢁC for 20 h. The pyr-
idine was removed under reduced pressure,and the res-
Bn-2 and Bn-3),138.1 and 138.9 (quaternary C);
1
70.2 (CO). Anal. Calcd for C154H168O : C,68.74; H,
42
6
.29. Found: C,67.88; H,6.03.
idue,dissolved in CHCl ,was filtered through Celite and
3
PTFE filters (0.45 lm) to eliminate the insoluble frac-
4
.12. Heptakis(6-O-methyl)cyclomaltoheptaose (9a)
1
tion to give 1 (75%): H NMR (CDCl ,25 ꢁC): d 1.88
3
1
Method A. To a solution of 8a (1.66 g,0.91 mmol) in
5
(Me-2,s,21H),2.02 (Me-3,s,21H),3.45 (MeO,s,
0
2
1H),3.70 (H-6 ,br d, J
0
10.8 Hz,7H),3.97 (H-6,br
,6
6
anhyd MeOH (70 mL) was added MeONa in MeOH
(0.07 mL,1 M). The mixture was stirred at room tem-
perature for 3 h,and then H O (70 mL) and Amberlite
+
IR-120 [H ] were added. The mixture was filtered,and
d, J6,6
.4 Hz,7H),4.20 (H-5,br d, J5,4 9.6 Hz,7H),5.10 (H-
2,dd, J2,3 10.4 Hz, J2,1 3.0 Hz,7H),5.18 (H-1,d, J
1,2
0
10.8 Hz,7H),4.02 (H-4,dd,
J4,5 9.6 Hz, J4,3
9
2
3
7
3
.0 Hz,7H),5.22 (H-3,dd,
H),6.39 (H -2,s,7H),6.42 (H o-3,s,14H),6.46 (H p-
J3,2 10.4 Hz, J3,4 9.4 Hz,
the filtrate was dried (Na SO ) and then concentrated
2
4
p
to give 9a (94%).
Method B. To a solution of 8b (4.07 g,1.63 mmol) in
,s,7H),6.64 (H o-2,s,14H),6.64 (NH-3,s,7H),6.74
1
3
(NH-2,s,7H).
C NMR (CDCl ,25 ꢁC): d 20.8 (Me-
3
anhyd MeOH (180 mL) and CH Cl (5 mL) was added
2
2
3
),21.0 (Me-2),59.1 (MeO),71.0 (C-6),71.2 (C-2 and
o-3),
17.3 (C -2),124.7 (C p-3),125.3 (C p-2),136.9 and
Pd/C (10%) (3.0 g). The reaction was carried out under
H (5 atm) for 24 h at room temperature. The mixture
C-5),73.8 (C-3),77.7 (C-4),98.7 (C-1),116.6 (C
1
2
o
was filtered and the organic layer was dried (Na SO )
4
1
and then concentrated to give 9a (60%): H NMR
2
137.0 (CNH),137.8 and 138.5 (quaternary C, meta),
52.5 and 153.0 (CO). Anal. Calcd for C175H210
N O : C,63.82; H,6.43; N,5.95. Found: C,63.11;
1
(Me SO-d ,25 ꢁC): d 3.24 (OMe,s,21H),3.26 (H-4,
dd, J4,5 9.9 Hz, J 9.6 Hz,7H),3.31 (H-2,ddd,
2
6
1
4
49
J2,3
4,3
H,6.89; N,6.03.
8.8 Hz, J 3.6 Hz, J2,OH-2 7.1 Hz,7H),3.56 (H-6 and
2,1
H-6 ,m,14H),3.59 (H-3,ddd,
0
J3,4 9.6 Hz, J3,2 8.8 Hz,
J5,4 9.9 Hz,
J1,2 3.6 Hz,
4.15. Heptakis[6-O-acetyl-2,3-di-O-(3,5-dimethylphen-
ylcarbamoyl)]cyclomaltoheptaose (2)
J3,OH-3 1.9 Hz,7H),3.71 (H-5,ddd,
0
J5,6 = J5,6
= 6.8 Hz,7H),4.77 (H-1,d,
H),5.72 (OH-3,d, J 1.9 Hz,7H),5.80 (OH-2,d,
.1 Hz,7H).
7
7
J
1
3
Method A. A solution of 9b (1.70 g,1.19 mmol) and 3 ,5 -
dimethylphenyl isocyanate (3.5 mL,3.66 g,25 mmol) in
dry pyridine (25 mL) was stirred at 80 ꢁC for 20 h. The
pyridine was evaporated under reduced pressure to give
C NMR (Me SO-d ,25 ꢁC): d 53.3
2
6
(
7
OMe),65.6 (C-5),66.2 (C-6),67.6 (C-2),68.3 (C-3),
7.6 (C-4),97.5 (C-1). Anal. Calcd for C 49H O : C,
84 35
4
7.72; H,6.87. Found: C,47.12; H,6.09.
2
(64%). Method B. To a solution of 7c (1.95 g,
.61 mmol) in dry pyridine (40 mL) was added Ac O
0
(5 mL,5.41 g,53 mmol). The mixture was stirred at
2
4
.13. Heptakis(6-O-acetyl)cyclomaltoheptaose (9b)
1
ual solvent was coevaporated with toluene. The residue
00 ꢁC for 7 h. The pyridine was removed,and the resid-
To a solution of 8c (3.36 g,1.25 mmol) in anhyd MeOH
175 mL) and CH Cl (5 mL) was added Pd/C (10%)
(
was dissolved in CH Cl ,and 2 was obtained (94%) by
2
2
2
2
1
(
was carried out under H (5 atm) for 24 h at room tem-
2.4 g). (Caution: extreme fire hazard!) The reaction
precipitation with hexane. H NMR (CDCl ,25 ꢁC): d
3
2
1.89 and 2.02 (Me-2 and Me-3,s,84H),2.15 (Me-6,s,
21H),3.93 (H-4,dd, J4,3 9.4 Hz, J4,5 9.3 Hz,7H),4.33
perature. The mixture was filtered,and the organic layer
was dried (Na SO ) and then concentrated to give 9b
0
0
5.0 Hz,7H),4.39 (H-6 ,dd,
2
4
(H-5,dd, J5,4 9.3 Hz, J5,6
1
(
95%): H NMR (Me SO-d ,25 ꢁC): d 1.97 (Me,s,
J
0
12.0 Hz, J 5.0 Hz,7H),4.67 (H-6,d,
0
J6,6
J2,3 10.5 Hz, J2,1 3.7 Hz,
7H),5.15 (H-1,d, J1,2 3.7 Hz,7H),5.54 (H-3,dd, J3,2
-2 and H -3,s,
0
2
6
6 ,6
6 ,5
2
3
7
1H),3.35 (H-2,ddd,
J
.4 Hz,7H),3.36 (H-4,dd,
9.2 Hz, J 6.4 Hz, J2,1
2,OH
12.0 Hz,7H),5.05 (H-2,dd,
2
,3
J4,3 9.2 Hz, J4,5 8.8 Hz,
H),3.62 (H-3,dd, J3,2 = J3,4 = 9.2 Hz,7H),3.83 (H-5,
10.5 Hz, J3,4 9.4 Hz,7H),6.42 (H
21H),6.48 (H p-2,s,7H),6.66 (H
(NH-2,s,7H),6.74 (NH-3,s,7H).
o
p
0
dd, J5,4 8.8 Hz, J5,6
0
6.4 Hz,7H),4.11 (H-6 ,dd, J
0
0
-3,s,14H),6.68
o
6 ,6
1
3
1
7
7
1.0 Hz, J6
0
6.4 Hz,7H),4.34 (H-6,d, J6,6
H),4.87 (H-1,d, J1,2 3.4 Hz,7H),5.80 (OH-3,br s,
11.0 Hz,
C NMR (CDCl3,
,5
25 ꢁC): d 20.7 (Me-6),20.8 (Me-3),21.0 (Me-2),62.5
(C-6),70.0 (C-5),71.3 (C-2),73.3 (C-3),78.1 (C-4),
98.6 (C-1),116.5 (C o-2),117.2 (C o-3),124.9 (C p-2),
1
H),5.88 (OH-2,d, J 6.4 Hz,7H). C NMR (Me SO-
3
2
d ,25 ꢁC): d 20.3 (Me),63.1 (C-6),69.1 (C-5),72.1 (C-
6
2
Calcd for C H O : C,47.06; H,5.92. Found: C,
),72.9 (C-3),82.2 (C-4),102.1 (C-1),170.1 (CO). Anal.
125.5 (C -3),136.8 and 137.0 (CNH),137.8 and 138.6
p
5
6
84 42
(quaternary C, meta),152.5 and 153.0 (CO,Carb),
170.3 (CO,Ac-6). Anal. Calcd for C 182H210N O : C,
4
6.82; H,6.04.
14
56

G. Uccello-Barretta et al. / Carbohydrate Research 340 (2005) 271–281
281
6
5
2.64; H,6.07; N,5.62. Found: C,61.92; H,6.15; N,
.84.
Acknowledgements
The work was supported by the University of Pisa,
MIUR (Project ꢀEngineering of separation systems,sen-
sors and arrays based on chemo- and stereoselective
molecular recognitionꢁ,Grant no 2003039537) and IC-
COM–CNR.
4
.16. Heptakis[2,3-di-O-methyl-6-O-(3,5-dimethylphen-
ylcarbamoyl)]cyclomaltoheptaose (3)
To a solution of 7d (2.81 g,2.11 mmol) in dry pyridine
(
(
90 mL) was added 3,5-dimethylphenyl isocyanate
3.1 mL,3.24 g,22.2 mmol). The mixture was stirred
at 80 ꢁC for 20 h,the solvent was evaporated under re-
References
duced pressure,and the residue was dissolved in CHCl
and filtered over Celite. The solid obtained was crystal-
3
1
. Atwood,J. L.; Davies,J. E. D.; Macnicol,D. D.; Vo ¨gtle,
F. In Comprehensive Supramolecular Chemistry; Perg-
amon,1996; Vol. 3.
. Rothchild,R. Enantiomer 2000, 5,457–471.
. Parker,D. Chem. Rev. 1991, 91,1441–1457.
. Schurig,V. TrAC Trends Anal. Chem. 2002, 21,647–
661.
5. Uccello-Barretta,G.; Balzano,F.; Caporusso,A. M.;
Salvadori,P. J. Org. Chem. 1994, 59,836–839.
6. Uccello-Barretta,G.; Balzano,F.; Caporusso,A. M.;
Iodice,A.; Salvadori,P. J. Org. Chem. 1995, 60,2227–
lized from CHCl –hexane to give 3 (93%): mp 155–
3
1
1
4
57 ꢁC. H NMR (CDCl ,25 ꢁC): d 2.05 (Me–Ar,s,
3
2
3
4
2H),3.19 (H-2,dd,
J
9.7 Hz, J 3.5 Hz,7H),3.42
,3 2,1
2
(
2
(
H-4,dd, J4,3 = J4,5 = 8.6 Hz,7H),3.51 (OMe-2,s,
1H),3.62 (H-3,dd, J3,2 9.7 Hz, J3,4 8.6 Hz,7H),3.65
OMe-3,s,21H),4.09 (H-5,m,7H),4.49 (H-6,m,
1
7
4H),5.03 (H-1,d,
H),6.93 (H ,br s,14H),7.78 (NH,s,7H).
J1,2 3.5 Hz,7H),6.53 (H p,br s,
1
3
C NMR
o
(
(
8
CDCl ,25 ꢁC): d 21.1 (Me–Ar),58.7 (OMe-2),61.2
OMe-3),64.7 (C-6),69.9 (C-5),81.2 (C-3),81.6 (C-4),
2.2 (C-2),99.1 (C-1),117.0 (C ),125.0 (C ),137.6,
o p
3
2
231.
7
8
. Uccello-Barretta,G.; Balzano,F.; Menicagli,R.; Salva-
dori,P. J. Org. Chem. 1996, 61,363–365.
. Uccello-Barretta,G.; Balzano,F.; Salvadori,P.; Lazza-
roni,R.; Caporusso,A. M.; Menicagli,R.
996, 1,365–375.
. Uccello-Barretta,G.; Cuzzola,A.; Balzano,F.; Menicagli,
1
C119H161O N : C,60.52; H,6.87; N,4.15. Found: C,
38.2 (quaternary C),154.0 (CO). Anal. Calcd for
Enantiomer
42
7
1
6
0.07; H,6.62; N,4.03.
9
R.; Iuliano,A.; Salvadori,P. J. Org. Chem. 1997, 62,827–
4
.17. Heptakis[2,3-di-O-acetyl-6-O-(3,5-dimethylphen-
ylcarbamoyl)]-cyclomaltoheptaose (4)
8
35.
10. Uccello-Barretta,G.; Cuzzola,A.; Balzano,F.; Menicagli,
R.; Salvadori,P. Eur. J. Org. Chem. 1998,2009–
012.
2
A solution of 7a (5.34 g,3.1 mmol) and 3 5, -dimethyl-
phenyl isocyanate (4.5 mL,4.70 g,32 mmol) in dry pyr-
idine (63 mL) was stirred at 80 ꢁC for 20 h. The pyridine
was removed under reduced pressure,and the residue
1
1. Uccello-Barretta,G.; Ferri,L.; Balzano,F.; Salvadori,P.
Eur. J. Org. Chem. 2003,1741–1748.
1
2. Uccello-Barretta,G.; Balzano,F.; Cuzzola,A.; Menicagli,
R.; Salvadori,P.
53.
Eur. J. Org. Chem. 2000,449–
4
was dissolved in CHCl ,filtered through Celite and then
3
1
1
3. Uccello-Barretta,G.; Sicoli,G.; Balzano,F.; Salvadori,P.
Carbohydr. Res. 2003, 338,1103–1107.
4. Ashton,P. R.; Ko¨niger,R.; Stoddart,J. F.; Alker,D.;
Harding,V. D. J. Org. Chem. 1996, 61,903–908.
concentrated. The crude residue was dissolved again in
THF and centrifuged to eliminate the insoluble fraction.
Finally,the residue was again filtered on PTFE filters
(
4
0.45 lm) to give,after evaporation of the solvent,pure
15. Takeo,K.; Mitoh,H.; Uemura,K. Carbohydr. Res. 1989,
87,203–221.
1
1
(87%): H NMR (CDCl ,25 ꢁC): d 2.05 (Me-2,s,
3
1
1
6. Angibeaud,P.; Utille,J.-P. Synthesis 1991,737–738.
7. Jullien,L.; Canceill,J.; Lacombe,L.; Lehn,J.-M. J. Chem.
Soc., Perkin Trans. 2 1994,989–1002.
2
1H),2.08 (MeAr,s,42H),2.11 (Me-3,s,21H),3.64
H-4,dd, J4,3 8.6 Hz, J4,5 4.3 Hz,7H),4.30 (H-5,br d,
(
J5,4 4.3 Hz,7H),4.57 (H-6 and H-6 ,br s,14H),4.83
0
18. Adinolfi,M.; Barone,G.; Guariniello,L.; Iadonisi,A.
Tetrahedron Lett. 1999, 40,8439–8441.
(
J1,2 3.7 Hz,7H),5.30 (H-3,dd, J3,2 9.0 Hz, J3,4 8.6 Hz,
7
7
H-2,dd, J2,3 9.0 Hz, J2,1 3.7 Hz,7H),5.09 (H-1,d,
1
2
2
2
2
9. Cipolla,L.; La Ferla,B.; Lay,L.; Peri,F.; Nicotra,F.
Tetrahedron: Asymmetry 2000, 11,295–303.
0. Sala,R.; Doria,G.; Passarotti,C. Tetrahedron Lett. 1984,
H),6.54 (H ,s,7H),6.90 (H o,s,14H),7.55 (NH,s,
p
1
3
H). C NMR (CDCl ,25 ꢁC): d 20.7 and 21.1 (Me-2
3
2
5,4565–4568.
and Me-3),63.5 (C-6),69.7 (C-5),70.2 (C-2),70.8 (C-
),77.7 (C-4),97.0 (C-1),117.3 (C ),125.2 (C ),137.3
1. Freeman,R.; Wittekoek,S. J. Magn. Reson. 1969, 1,238–
276.
2. Boger,J.; Corcoran,R. J.; Lehn,J. M. Helv. Chim. Acta
1978, 61,2190–2218.
3. Branch,S. K.; Holzgrabe,U.; Jefferies,T. M.; Mallwitz,
3
o
p
(
(
CNH),138.2 (C m),153.6 (CO,carb),169.4 and 170.6
CO,Ac-2 and Ac-3). Anal. Calcd for C 133H161N O :
7
56
C,58.01; H,5.89; N,3.56. Found: C,57.61; H,5.15;
N,3.87.
H.; Oxley,F. J. R. J. Chromatogr. A 1997, 758,277–
2
92.