DIBAL-H mediated triple and quadruple debenzylations of perbenzylated cyclodextrins

Article abstract of DOI:10.1039/b915450g

Diisobutylaluminium hydride (DIBAL-H) mediated reductive removal of benzyl groups was investigated for perbenzylated α-, β- and γ-cyclodextrins using DIBAL-H in hexane as the reagent. It was found that under the new conditions, the debenzylation can be better controlled to provide sequentially tri- and tetra-debenzylated products in moderate yields and in a regioselective manner. In the case of α-cyclodextrin, the removal of the third and fourth benzyl groups took a different path involving the secondary rim, compared to β- and γ-cyclodextrins which both gave only 6-O-debenzylated products. The Royal Society of Chemistry 2010.

Full text of DOI:10.1039/b915450g

View Article Online / Journal Homepage / Table of Contents for this issue
PAPER
www.rsc.org/obc | Organic & Biomolecular Chemistry
DIBAL-H mediated triple and quadruple debenzylations
of perbenzylated cyclodextrins†
Girish K. Rawal, Shikha Rani, Sandra Ward and Chang-Chun Ling*
Received 29th July 2009, Accepted 30th September 2009
First published as an Advance Article on the web 12th November 2009
DOI: 10.1039/b915450g
Diisobutylaluminium hydride (DIBAL-H) mediated reductive removal of benzyl groups was
investigated for perbenzylated a-, b- and g-cyclodextrins using DIBAL-H in hexane as the reagent. It
was found that under the new conditions, the debenzylation can be better controlled to provide
sequentially tri- and tetra-debenzylated products in moderate yields and in a regioselective manner. In
the case of a-cyclodextrin, the removal of the third and fourth benzyl groups took a different path
involving the secondary rim, compared to b- and g-cyclodextrins which both gave only
6-O-debenzylated products.
and quadruple debenzylations of perbenzylated a-CD by slightly
Introduction
modified conditions using DIBAL-H in hexane as reagent instead
of the original DIBAL-H in toluene (Scheme 1). The simple switch
of reagent resulted in much slower debenzylation thus allowing for
a better monitoring and control of the progress of the reaction. In
addition to the known mono- (2) and didebenzylated a-CD (3), we
also observed the formation of two major debenzylated products,
namely a triol (4, 32% yield) and a tetraol (5, 10% yield).
Recently, there has been an increasing interest1–4 in developing
methodologies for the regioselective substitution of cyclodextrins
(CDs). These multi-functionalized CDs have great utility in many
areas such as for use as novel host molecules to complex bioactive
compounds with high affinities,5 as artificial enzymes6 and as
gene delivery vehicles.7 Among all the hydroxyls present in a
CD molecule, those attached to the C-6’s of the glucopyra-
nosyl units are sterically the most accessible and they are all
located at the primary rim of the hydrophobic cavity; the remaining
hydroxyl groups are all found at the secondary rim because
they are attached to either C-2’s or C-3’s of the sugar units;
they are sterically less accessible. It’s relatively straightforward
to persubstitute all hydroxyls or regioselectively differentiate one
group of hydroxyls from the other.8 However, it is extremely
challenging to regioselectively differentiate hydroxyls of the same
group due to their identical chemical properties. There have been
a few methods reported in literature involving the use of either
bulky groups such as trityl9 and its variant10 or geometrically
restricted capping reagents to react with close hydroxyl pairs at
the primary rim.11–13 Several years ago, Sinay¨’s group introduced a
distinct approach14 which involved the use of diisobutylaluminium
hydride (DIBAL-H) in toluene as the reagent to remove 1–2
benzyl groups attached to O-6 positions of perbenzylated a-, b-
and g-CDs. The reaction conditions could be tuned to produce
either mono- or didebenzylated CDs in moderate to excellent
yields. The most prominent feature of the method was the high
regioselectivity during the removal of the second benzyl group.
For example, for a/b-CDs, the second debenzylation occurs
exclusively at the O-6 of the D sugar unit of cyclic oligosaccharides
to form 6A,6D-O-didebenzylated a-/b-CD in high yields while
in the case of perbenzylated g-CD, a mixture of 6A,6D/6A,6E-
O-didebenzylated products was formed in moderate yield. More
recently, we reported our preliminary results15 on the one-pot triple
Scheme 1 Sequential debenzylations of perbenzylated a-CD.
As shown in the structures, both the third and fourth benzyl
removals occurred at the secondary rim of the a-CD cavity
and remarkably, the glucopyranosyl unit which had a prior
O-6 debenzylation also preferred to have its O-3 benzyl removed.
The structures of both 4 and 5 were determined via either a
sequence of acetylation/hydrogenation to remove all the benzyl
Alberta Ingenuity Center for Carbohydrate Science, Department of Chem-
istry, University of Calgary, Calgary Alberta, T2N 1N4, Canada. E-mail:
ccling@ucalgary.ca; Fax: +1 (403)289-9488; Tel: +1 (403) 220-2768
† Electronic supplementary information (ESI) available: 1H and 13C NMR
data of synthesized compounds. See DOI: 10.1039/b915450g
This journal is
The Royal Society of Chemistry 2010
Org. Biomol. Chem., 2010, 8, 171–180 | 171
©

View Article Online
groups (4 → 6 → 8) or simply an acetylation (5 → 7). It
is important to note that all acetylated products were stable
after the removal of benzyl groups and no acetate migration
was observed. For example, in compound 6, we observed that
a glucopyranosyl unit had respectively an acetate at O-3 and O-6
positions by a combination of 1D/2D (GCOSY and GTOCSY)
NMR experiments and these substitution patterns were further
confirmed by a selective 1D TOSCY experiment (see supporting
information†); the same acetate-substitution patterns were also
observed in the fully debenzylated polyol 8 by NMR experiments.
These results suggested that the catalytic hydrogenation step was
carried out under extremely mild conditions therefore no acetate
migration occurred. In this paper, we report further studies in
applying our reported conditions to perbenzylated b- and g-CDs.
A rational to explain the unusual debenzylation path is proposed.
again that the debenzylation process was sluggish. Even after 12 h,
we still isolated 15% starting material, and the previously known
6-O-monodebenzylated b-CD (10) and its 6A,6D-O-didebenzylated
analog (11) were isolated in only 45% and 25% yields respectively.
However, in the literature, it was reported14a that if DIBAL-H in
toluene was used at 0.5 M concentration, compound 11 could be
isolated in 87% yield after heating for just 2 h at 50◦ C. After 24 h,
the starting material was completely consumed and we started to
observe the formation of some more polar compounds besides the
monool 10 and diol 11 – both remained as the major components
of the reaction mixture until 48 h (entries 2 and 3). The monool
10 finally disappeared after 3 days and the amounts of the two
new spots gradually increased and eventually became the majority
after 96 h. The less polar spot (Rf 0.5, 7 : 3 hexane : ethyl acetate)
was isolated in 25% and the more polar spot (Rf 0.25, 1 : 1 ethyl
acetate : hexane) in 22% yield (entries 4 and 5).
The less polar spot was subjected to an acetylation and from 1H
NMR, it was found to be a inseparable 1 : 1 mixture containing
two triacetylated b-CD derivatives because 6 acetate peaks were
observed (15 and 16); interestingly none of the acetyl groups
were found to attach to a secondary hydroxyl group as after a
debenzylation, we did not observe any proton resonating above
5.0 ppm (see supporting information†) which was the case for a-
CD (compounds 5 and 7). This was further confirmed from the
1H NMR spectra of the fully debenzylated products (not shown).
The structures of the two triols were assigned to be 6A,6B,6D-b-
O-tridebenzylated b-CD 12 and 6A,6B,6E-b-O-tridebenzylated 13
(Scheme 2) respectively, according to our proposed rational (see
later).
Results and discussion
b-Cyclodextrin
The intriguing debenzylation path of 1 prompted us to investigate
the debenzylation of perbenzylated b-CD (9). We treated com-
pound 9 with DIBAL-H in hexane (40 equiv, 0.9 M) under similar
heating conditions (50 ◦C). As shown in Table 1; it was confirmed
Table 1 DIBAL-H in hexane mediated stepwise debenzylation of 9
Conditionsa
9
10
11
12 + 13 14
1
2
3
4
5
40 eq, 0.9 M, 50 ◦C, 12 h 15% 45%
25%
52%
42%
18%
8%
40 eq, 0.9 M, 50 ◦C, 24 h
40 eq, 0.9 M, 50 ◦C, 48 h
40 eq, 0.9 M, 50 ◦C, 72 h
40 eq, 0.9 M, 50 ◦C, 96 h
16%
5%
Trace
16%
30%
25%
Trace
Similarly, the more polar spot was acetylated and fortunately,
the 1H and 13C NMR spectra revealed that it constitutes a single
compound 17 with four acetates. Since we did not observe any
1H NMR signals with chemical shifts greater than 5.0 ppm in the
spectra of the fully debenzylated 18, we concluded that all the four
4%
15%
22%
a In all conditions reported, DIBAL-H in hexane (1.4 M) was used.
Scheme 2 Stepwise debenzylations of perbenzylated b-cyclodextrin.
172 | Org. Biomol. Chem., 2010, 8, 171–180 This journal is The Royal Society of Chemistry 2010
©

View Article Online
Table 2 1H Chemical shifts of compound 21
H-1
H-2
H-3
H-4
H-5
H-6(a)
H-6b
4.42
I
II
5.74
5.60
5.59
5.56
5.56
5.41
5.40
4.17
4.14
4.14
4.20
4.20
4.16
4.16
4.82
4.69
4.69
4.81
4.81
4.68
4.68
4.36
3.59
3.59
4.36 or 4.30
4.30 or 4.36
3.66
4.56
4.45
4.45
3.96
4.96
4.46
4.42
4.50
1.58
1.58
4.63–4.38
4.63–4.38
1.73
III
IV
3.64
1.68
debenzylations had taken place at the primary rim. In fact, in the
1H spectra of the 14, we also observed 4 sets of signals at 3.02,
2.79, 2.58, 2.48 ppm, each appeared as a doublet of doublets, they
indeed corresponded to 4 sets of primary hydroxyl group signals.
Thus the triple and quadruple debenzylations of perbenzylated
b-CD took a different path than a-CD. The four debenzylation
sites in tetraol 14 could not be unambiguously established even
with the help of advanced 2D NMR experiments for both 17
and 18 due to significant overlaps in 1H NMR spectra. Thus
we performed a series of chemical derivatizations to assist the
task. We decided to perform 6-deoxygenation to convert all the
6-O-debenzylated sites to 6-deoxygenated sites. This in principle
should generate 4 sets of doublets (H-6s) which should appear at
a distinct region (1.0–2.0 ppm), well separated from the rest of the
sugar protons (3.2–5.5 ppm). Indeed, the tetraol 14 was smoothly
converted to the tetramesylates (19, 84% yield) by treating 14
with methanesulfonyl chloride in anhydrous dichloromethane
using triethylamine as a base. The activated sulfonyl ester was
then reduced with excess lithium aluminium hydride (LAH)
at room temperature (→20); compound 20 was finally fully
debenzylated under standard hydrogenation conditions to give
the fully debenzylated 21. In the 1H NMR spectra of 21, we
observed four sets of doublets at the 1.50–1.80 ppm region (two of
them overlapped), correlating to the H-6’s of four 6-deoxygenated
glucopyranosyl units, which reconfirm our earlier observation that
the quadruple debenzylations had occurred only at the primary
rim of b-CD.
In order to determine the exact sites of debenzylations, we
performed a systematic NMR study of compound 21, recorded
in deuterated pyridine. As shown in Fig. 1, the 1H–1H COSY and
TOSCY spectra clearly showed 7 anomeric protons at 5.74, 5.60,
5.59, 5.56 (¥2), 5.41, 5.40 ppm, which can be divided into four
groups according to their chemical shift similarities (Table 2).
The group I consists of one sugar unit which has its anomeric
proton resonating at 5.74 ppm; the other three groups consist of
two sugar units each with their H-1’s at 5.60/5.59 ppm for the
group II, at 5.56/5.56 ppm for group III and at 5.41/5.40 ppm
for group IV. The GTOCSY experiment revealed that the four
sugar units of groups II and IV had 6-deoxygenations because
the H-1 proton of each of them correlated with a CH3 at the most
upfield region. All 6-deoxygenated sugar units had a more upfield-
shifted H-4; the H-4’s of group II were observed slightly upfield
than those of group IV. All the H-4’s of group I and III were found
to be more downfield at a very narrow 4.27–4.39 ppm region with
the H-4 of group I resonating slightly upfield than H-4’s of group
III.
Fig. 1 1H–1H GCOSY and GTOCSY spectra of compound 21 in
pyridine-d5 recorded at 400 MHz.
between neighboring H-1 and H-4 pairs, we were able to unam-
biguously establish the connectivities of the cyclic oligosaccharide.
For example, the H-1 of group I unit (non-deoxygenated) has a
strong correlation with one of the H-4 of the sugar unit of group III
(non-deoxygenated), while both H-1’s of group III units correlate
with H-4’s of group IV (deoxygenated) units; and similarly, the
H-1’s of the group IV units correlate with the H-4’s of group II units
(deoxygenated). Additionally, when analyzing the anomeric H-1’s
of group II units, it was found that one of them correlates with H-4
of the group I monosaccharide, and another one correlates with
the H-4 of one of the group III unit. These information suggested
that the anomeric O-1 of monosaccharide of group I is connected
1
The 2D H–1H ROESY experiment of compound 21 is shown
in Fig. 2. By analyzing the across glycosidic ROE correlations
This journal is
The Royal Society of Chemistry 2010
Org. Biomol. Chem., 2010, 8, 171–180 | 173
©

View Article Online
Fig. 2 Analysis of 2D 1H–1H ROESY spectrum of compound 21 in pyridine-d5 (Sugar units with carbons in blue have a 6-deoxy functionality).
to the C-4 of one of the unit in group III; both anomeric O-1’s of
group III are connected to C-4’s of group IV and their anomeric
O-1’s are in turn connected to C-4’s of group II; for the anomeric
O-1’s of group II, one of them is connected to C-4 of group I
and another one is connected to C-4 of one of group III unit.
The above information allowed us to conclude that the structure
of compound 21 is the 6A,6B,6D,6E-tetradeoxygenated b-CD, as
shown in Fig. 2. Consequently, the original compound 14 had
6A,6B,6D,6E-O-tetradebenzylations.
We noticed that a partially protected analog related to tetraol 14
was previously reported in the literature2b which was synthesized
in eight steps with an overall yield of ~8%. The present method
has clear advantage as it gives similarly tetrasubstituted b-CD in
22% yield in one step.
The reaction mixture became even more complex when left
reacting further (4 days); according to TLC, numerous small
components with similar intensities were detected with no visually
major spots produced. However, by following our previous
knowledge about the polarity of the tetraols, we were able to
isolate one product 23 (Rf 0.35, 9 : 1 dichloromethane : acetone)
in 8% yield after repeated column chromatographies. Compound
23 was characterized as the 6A,6B,6E,6F-O-tetradebenzylated g-CD
as both 23 and its subsequently acetylated form 24 had a C2 sym-
metry (Scheme 3). Clearly, the triple and quadruple debenzylation
of perbenzylated g-CD has little synthetic advantage due to the
extreme low yields and difficulty to isolate products in pure form.
c-Cyclodextrin
To further explore the potential of the current method, we at-
tempted to study the one-pot triple and quadruple debenzylations
of perbenzylated g-CD (22). Due to the presence of an additional
glucopyranosyl unit and three more benzyl groups than b-CD, it
was expected that the removal of benzyl groups should be much
more complex. Indeed, when we applied similar debenzylation
conditions as before, we did not obtain encouraging results. After
the treatment of 22 with DIBAL-H in hexane for one day as
before, we observed two major spots – which were determined
to be the known 6-O-mono and the inseparable 6A,6D/6A,6E-di-
debenzylated g-CDs. By prolonging the reaction time, we observed
several more polar spots which were all found to contain a mixture
of different triols from the acetylated products (not shown). The
Scheme 3 DIBAL-H mediated quadruple debenzylation of perbenzyl-
ated g-CD.
1
1D H and 2D GCOSY and GHSQC NMR spectra of all fully
Rational for the observed regioselectivities
acetylated products revealed an absence of signals above 5.0 ppm
(except benzyl protons), indicating that these mixtures were all
6-O-debenzylated.
Our results were consistent with the literature reports14a that the
DIBAL-H mediated removal of first and second benzyls in all
174 | Org. Biomol. Chem., 2010, 8, 171–180
This journal is
The Royal Society of Chemistry 2010
©

View Article Online
Fig. 3 Proposed rationale for the observed regioselectivities in triple and quadruple debenzylations of CDs.
perbenzylated CDs remained at the primary rim. However, our
results also revealed that there are significant differences at the
triple and quadruple debenzylation stages where in the case of
perbenzylated a-CD (1), the DIBAL-H preferred attacking a
benzyl group attached to a sterically more hindered secondary
hydroxyl, specifically the one attached to the O-3 position(s) of the
same glucopyranosyl unit which had a prior loss of O-6 benzyl;
while in the case of perbenzylated b- (9) and g-CDs (22), the
reducing reagent kept attacking the more accessible benzyl groups
at the primary rim. Considering that compound 1 has the smallest
cavity size among the three substrates, we think that our results
can be rationalized by the steric crowding existing at the primary
rim.
As depicted in Fig. 3, in the case of perbenzylated a-CD, because
of its smaller cavity and conical shape which bring the benzyl
groups at the primary rim closer through space, after the second
debenzylation (structure I), one of the OAl(i-Bu)2 groups was
pushed out to avoid steric crowding and orients itself towards the
outside of the cavity (structure II); consequently, the aluminium
of the OAl(i-Bu)2 group becomes spaciously close to the O-3 of
the same sugar unit which could dynamically coordinate to the
O-3 possibly after a ring flip which would bring O-3 and O-6 even
closer (structure III). The coordination triggers the polarization
of the corresponding benzylic C–O bond; a subsequent attack by
hydride on the benzylic methylene carbon from another DIBAL-H
molecule (present in excess) causes a regioselective removal of
the benzyl at the O-3 position of the same unit to produce triol
4. The other OAl(i-Bu)2 group can also bind to the O-3 of the
same ring but to a lesser extent, because of the already decreased
crowding, which results in the formation of tetraol 6 in lower
yield.
cavity because of the van der Waals attraction between isobutyl
and benzyl groups. This is why we observed debenzylations only
at the primary rim in both cases.
The preferred formation of 14 and 23 could be explained by
involving the respective clockwise coordination by the aluminium
of the O6-Al(i-Bu)2 to the neighboring O6 (Fig. 3). In the case
of b-CD, after the second removal of 6-O-benzyl at the D-unit,
the aluminium attached at O-6’s of A and D units coordinate
to the O-6’s of B and E units (Structure IV), resulting in the
formation of 6A,6B,6D,6E-O-tetradebenzylated b-CD 14; at the
stage of the third benzyl removal, a mixture of compounds:
6A,6B,6D-O-tridebenzylated (12) and 6A,6B,6E-O-tridebenzylated
b-CD 13 was formed. In the case of g-CD, the isolated product
6A,6B,6E,6F-O-tetradebenzylated g-CD 23 originated from one of
the didebenzylated intermediates–the 6A,6E-O-didebenzylated-g-
CD, after a clockwise activation of the neighboring 6-O-benzyl
groups (Structure V). In the case of a-CD, because of its much
reduced ring size at the primary face compared to b- and g-CDs, if
a clockwise coordination by the aluminium of the O6-Al(i-Bu)2
to the neighboring O6 occurred after the removal of second
benzyl, the van der Waals repulsion between the benzyl groups and
i-Bu groups might be too significant; therefore, debenzylation path
via structure V became favored. Sollogoub et al. were the first to
investigate the clockwise debenzylation mechanism2b,16 using some
bridge-capped and 6-deoxygenated CD derivatives; however, their
mechanism did not invoke the clockwise aluminium coordination
to the neighboring O6, but using the diminished steric crowding
argument. We think that in our cases, the clockwise aluminium
coordination is more plausible which explains the observed regios-
electivity. A molecular model is constructed (Fig. 4) to show the
probable intermediate. It is noteworthy that previously, we also ob-
served that in the 1H NMR spectra of a 6A,6B,6D,6E-tetratritylated
a-CD compound, the trityl groups shield the anomeric re-
gion of the neighboring glucopyranosyl group in a clockwise
fashion.9c
Since the cavity sizes of CD derivatives 9 and 22 were larger, the
steric crowding is much less prominent in these cases; thus after
the second debenzylation, there was less steric crowding therefore
no need for the OAl(i-Bu)2 to reorientate toward the outside of the
This journal is
The Royal Society of Chemistry 2010
Org. Biomol. Chem., 2010, 8, 171–180 | 175
©

View Article Online
spectrometer using 2,5-dihydroxybenzoic acid (DHP) as matrix by
the analytical services of the Department of Chemistry, University
of Calgary.
2A,2B,2C,2D,2E,2F,3A,3C,3D,3E,3F,6A,6C,6D,6F-Hexadeca-O-benzyl-
a-cyclodextrin (4) and 2A,2B,2C,2D,2E,2F,3A,3C,3D,3F,6A,6C,6D,6F-
hexadeca-O-benzyl-a-cyclodextrin (5)
To a solution of perbenzylated a-CD (1, 6.0 g, 2.32 mmol) in
anhydrous toluene (36.6 mL), was added a solution of DIBAL-H
in hexane (66 mL, 1.4 M solution, 92.4 mmol) with stirring under
argon atmosphere at room temperature. The reaction mixture
was heated to 50 ◦C for 4 days and then quenched by pouring
the reaction mixture carefully into ice-cold water (40 mL) with
vigorous stirring. The reaction mixture was subsequently diluted
with more EtOAc (100 mL) and filtered off by passing through
a thin bed of celite. This celite was then washed thoroughly with
more hot EtOAc (3 ¥ 30 mL). More H2O (100 mL) was added to
the filtrate and the organic layer was separated. After washing
with brine (100 mL) and drying over anhydrous Na2SO4, the
organic phase was evaporated. The crude mixture was purified by
column chromatography on silica gel using a mixture of AcOEt
in hexane (gradient 15–40%) as the eluent to afford sequentially
diol 3 (278 mg, 5% yield), triol 4 (1.72 g, 32% yield) and tetraol 5
(517 mg, 10% yield) as colorless solids.
Fig. 4 The molecular model of a possible b-CD intermediate that shows
the clockwise coordination by aluminium from the O6–Al(i-Bu)2 group
in unit A and B, to the neighboring benzylated O6 (unit B and E). The
coordination activates the related benzyl group, which facilitates the attack
by another DIBAL-H from the exterior of the cavity. (For clarity, all H
atoms and benzyl groups attached to the secondary face were removed.)
Data for a-CD diol 3. Rf 0.50 (hexane : AcOEt 7 : 3); [a]D25
+32.6◦ (c 0.9, CHCl3); H and 13C NMR data are identical to
1
In conclusion, we have developed a one-pot synthesis of some
multi-substituted CD derivatives, previously unable or difficult
to obtain, using DIBAL-H mediated debenzylation. Although
the yields for individual CD derivatives were low, the method
allowed access to triply and quadruply functionalized a- and b-
CD with ease; the short reaction sequence makes the syntheses
of these novel CD derivatives feasible and we demonstrated
that some compounds could be obtained in gram quantities.
The unique topology of these CD derivatives should make them
ideal candidates for use as scaffolds in the preparation of diverse
supramolecular assemblies.
the literature.1b
Data for a-CD triol 4. Rf 0.45 (hexane : AcOEt 6 : 4); [a]D25
+42.2◦ (c 0.9, CHCl3); dH (400 MHz, CDCl3) 7.46–7.07 (m, 75H,
Aromatic), 5.62 (d, 1H, J = 4.0 Hz, H-1), 5.52 (d, 1H, J = 11.0 Hz,
PhCH), 5.32 (d, 1H, J = 10.6 Hz, PhCH), 5.21 (d, 1H, J = 10.8 Hz,
PhCH), 5.15 (d, 1H, J = 11.2 Hz, PhCH), 5.09 (d, 1H, J = 3.8 Hz,
H-1), 5.01–4.67 (m, 13H, 4 ¥ H-1, 9 ¥ PhCH), 4.59–4.36 (m,
17H, 17 ¥ PhCH), 4.29–4.19 (m, 2H), 4.16–3.53 (m, 30H), 3.48–
3.36 (m, 4H), 3.28 (dd, 1H, J = 3.5, 9.9 Hz, H-2), 3.10 (brs, 1H,
OH), 2.76 (brs, 1H, OH); dC (100 MHz, CDCl3) 139.66, 139.63,
139.27, 139.19, 139.16, 138.66, 138.41, 138.19, 138.09, 138.01,
137.93, 137.85, 137.56, 128.50, 128.46, 128.34, 128.30, 128.28,
128.21, 128.15, 128.10, 128.08, 128.06, 127.96, 127.93, 127.89,
127.82, 127.73, 127.68, 127.63, 127.61, 127.58, 127.50, 127.45,
127.33, 127.30, 127.25, 127.03, 126.90, 126.85, 126.54, 100.59,
100.33, 99.16, 98.43, 97.88, 97.80 (6 ¥ C-1), 82.54, 82.17, 81.60,
81.52, 81.23, 81.14, 80.72, 80.68, 80.56, 79.81, 79.64, 78.67, 78.08,
77.65, 76.43, 76.34, 75.93, 75.81, 75.75, 74.38, 74.05, 73.38, 73.34,
73.34, 73.29, 72.80, 72.71, 72.46, 72.33 (CBn, C-2, C-3, C-4), 72.17,
72.12, 71.83, 71.61, 71.47, 71.29 (6 ¥ C-5), 70.03, 69.92, 69.63,
69.33, 61.90, 61.85 (6 ¥ C-6); m/z (MS MALDI-TOF) calcd for
[C141H150O30 + Na]+ 2346.0, found 2346.4.
Experimental section
General methods
Optical rotations were determined in a 5 cm cell at 25
2
◦C.
[a]D values are given in units of 10-1 deg cm2 g-1. Analytical TLC
was performed on Silica Gel 60-F254 (Merck, Darmstadt) with
detection by quenching of fluorescence and/or by charring with
5% sulfuric acid in water or with a ceric ammonium molybdate dip.
All commercial reagents were used as supplied unless otherwise
stated. Column chromatography was performed on Silica Gel 60
(Silicycle, Ontario). Organic solutions from extractions were dried
with anhydrous Na2SO4 prior to concentration under vacuum
Data for a-CD tetraol 5. Rf 0.25 (hexane:AcOEt 6 : 4); [a]D25
+50.8◦ (c 1.0, CHCl3); dH (400 MHz, CDCl3) 7.51–7.06 (m, 35H,
aromatic), 5.44 (d, 1H, J = 11.2 Hz, PhCH), 5.24 (d, 1H, J =
11.2 Hz, PhCH), 4.94–4.84 (m, 3H, 2 ¥ PhCH + H-1), 4.83–4.70
(m, 4H, 2 ¥ PhCH, H-1), 4.61 (d, 1H, J = 13.0 Hz, PhCH),
4.53 (m, 2H, PhCH), 4.46–4.38 (m, 3H, PhCH), 4.35 (d, 1H, J =
13.0 Hz, PhCH), 4.29 (dd, 1H, J = 9.33, 9.70 Hz, H-3), 4.23 (d,
1H, J = 13.0 Hz, PhCH), 4.17 (dd, 1H, J = 9.16, 8.56 Hz, H-3),
4.04–3.94 (m, 2H, H-3, H-6), 3.90–3.72 (m, 6H, 3 ¥ H-5, 2 ¥ H-4,
◦
1
at < 40 C (bath). H NMR spectra were recorded at 300, and
400 MHz on Bruker spectrometers. The first order proton chemical
shifts dH and dC are reported in d (ppm) and referenced to either
residual CHCl3 (dH 7.24, dC 77.0, CDCl3) or residual CD2HOD (dH
1
3.30, dC 49.5, CD3OD). H and 13C NMR spectra were assigned
with the assistance of GCOSY, GHSQC. Matrix-assisted laser
desorption/ionization and time-of-flight mass spectra (MALDI-
TOF MS) were recorded on a Bruker Daltonics AUTOFLEX III
176 | Org. Biomol. Chem., 2010, 8, 171–180
This journal is
The Royal Society of Chemistry 2010
©

View Article Online
1 ¥ H-6), 3.71–3.56 (m, 5H, 4 ¥ H-6, 1 ¥ H-2), 3.45–3.35 (m, 2H,
H-4, H-2), 3.29 (dd, 1H, J = 3.4, 10.0 Hz, H-2); dC (100 MHz,
CDCl3) 139.71, 139.46, 138.57, 138.46, 138.08, 138.06, 137.24,
128.78, 128.56, 128.33, 128.31, 128.22, 128.19, 128.14, 128.10,
127.93, 127.87, 127.82, 127.70, 127.61, 127.57, 127.46, 127.26,
127.15, 127.01, 126.83, 101.69 (C-1), 100.75 (C-1), 99.92 (C-1),
83.46 (C-4), 82.41 (C-4), 82.10 (C-4), 81.11 (C-3), 80.42 (C-3),
79.10 (C-2), 78.56 (C-2), 77.12 (C-2), 76.29 (PhCH), 75.73 (PhCH),
74.45 (PhCH), 73.44 (PhCH), 73.34 (PhCH), 72.92 (C-3), 72.45
(PhCH), 72.28 (PhCH), 72.13 (C-5), 71.93 (C-5), 71.04 (C-5), 69.73
(C-6), 69.25 (C-6), 61.78 (C-6); m/z (MS MALDI-TOF) calcd for
[C134H145O30 + H]+ 2234.0, found 2234.3.
dimethylaminopyridine, and the reaction mixture was stirred for
6 h at room temperature. The reaction was monitored and worked
up as 6. Compound 7 was obtained by column chromatography on
silica gel using a mixture of AcOEt in hexane (gradient 10–25%)
as eluent to give compound 7 as colorless solid (108 mg, yield
90%). Rf 0.65 (hexane : AcOEt 7 : 3); [a]25 +39.3◦ (c 0.6, CHCl3);
D
dH (400 MHz, CDCl3) 7.39–7.11 (m, 35H, aromatic), 5.80 (t, 1H,
J = 9.9 Hz, H-3), 5.28 (d, 1H, J = 11.9 Hz, PhCH), 5.17 (d, 1H,
J = 11.5 Hz, PhCH), 4.96 (d, 1H, J = 3.1 Hz, H-1), 4.94 (d, 1H,
J = 11.3 Hz, PhCH), 4.90 (d, 1H, J = 3.1 Hz, H-1), 4.85 (d, 1H,
J = 11.7 Hz, PhCH), 4.67 (d, 1H, J = 13.0 Hz, PhCH), 4.58 (d,
1H, J = 13.2 Hz, PhCH), 4.57 (d, 1H, J = 3.1 Hz, H-1), 4.55–4.46
(m, 2H, PhCH, H-6), 4.46–4.38 (m, 3H, 3 ¥ PhCH), 4.34 (d, 1H,
J = 11.7 Hz, PhCH), 4.30–4.18 (m, 3H, 2 ¥ PhCH, H-6), 4.17–
4.02 (m, 4H, 2 ¥ H-3, H-5, PhCH), 3.98–3.74 (m, 6H, 2 ¥ H-4,
2 ¥ H-5, 2 ¥ H-6), 3.66 (br d, 1H, J = 9.7 Hz, H-6), 3.60–3.51
(m, 2H, H-4, H-6), 3.39 (dd, 1H, J = 3.1, 9.8 Hz, H-2), 3.36 (dd,
1H, J = 2.9, 10.2 Hz, H-2), 3.32 (dd, 1H, J = 3.1, 10.4 Hz, H-2),
1.92 (s, 3H, COCH3), 1.88 (s, 3H, COCH3); dC (100 MHz, CDCl3)
3A,6A,6D-Tri-O-acetyl-2A,2B,2C,2D,2E,2F,3B,3C,3D,3E,3F,6B,
6C,6E,6F-hexadeca-O-benzyl-a-cyclodextrin (6)
To the solution of triol
4 (232 mg, 0.1 mmol) in dry
dichloromethane (2 mL) was added Ac2O (0.5 mL), anhy-
drous pyridine (0.5 mL) and a catalytic amount of 4-N,N-
dimethylaminopyridine, and the reaction mixture was stirred for
6 h at room temperature. The reaction was monitored by TLC, and
when the reaction was complete, the mixture was concentrated
under reduced pressure to afford syrup which was purified by
column chromatography on silica gel using a mixture of AcOEt in
hexane (gradient 10–25%) as eluent to give compound 6 (225 mg,
92% yield) as a white solid. Rf 0.65 (hexane:AcOEt 7 : 3); [a]D25
+23.7◦ (c 1.7, CHCl3); dH (400 MHz, CDCl3) 7.45–7.10 (m, 75H),
5.74 (dd, 1H, J = 10.0, 9.6 Hz, H-3), 5.35 (d, 1H, J = 11.3 Hz,
PhCH), 5.32 (d, 1H, J = 11.0 Hz, PhCH), 5.26 (d, 1H, J = 11.0 Hz,
PhCH), 5.22 (d, 1H, J = 11.2 Hz, PhCH), 5.11 (d, 1H, J = 10.9 Hz,
PhCH), 5.10 (d, 1H, J = 3.5 Hz, H-1), 5.02–4.84 (m, 9H, 5 ¥
PhCH, 4 ¥ H-1), 4.73 (d, 1H, J = 13.0 Hz, PhCH), 4.63–4.55
(m, 2H, PhCH, H-1), 4.55–4.20 (m, 25H, 18 ¥ PhCH, 4 ¥ H-6,
3 ¥ H-3), 4.18–3.95 (m, 8H, 4 ¥ H-5, H-6, 3 ¥ H-3), 3.87 (m, 7H,
2 ¥ H-6, 2 ¥ H-5, 3 ¥ H-4), 3.78–3.63 (m, 4H, 3 ¥ H-6, H-4),
3.62–3.53 (m, 3H, 2 ¥ H-6, H-4), 3.49–3.36 (m, 6H, 6 ¥ H-2), 1.97
(s, 3H, COCH3), 1.92 (s, 3H, COCH3), 1.92 (s, 3H, COCH3); dC
=
170.19, 169.97 (2 ¥ C O), 140.17, 140.12, 138.83, 138.57, 138.31,
138.29, 138.17, 128.38 - 126.47, 101.74 (C-1), 101.04 (C-1), 100.45
(C-1), 82.47 (C-4), 80.43 (C-4), 80.37 (C-2), 80.05 (C-3), 79.80 (C-
3), 78.77 (C-2), 76.50 (C-2), 75.91 (PhCH), 75.49 (PhCH), 74.00
(PhCH), 73.29 (PhCH), 73.20 (PhCH), 72.19 (C-5), 72.16 (PhCH),
71.81 (PhCH), 71.44 (C-5), 71.26 (C-3), 69.94 (C-5), 69.17 (C-
6), 68.71 (C-6), 63.11 (C-6), 21.16 (CH3), 20.89 (CH3); m/z (MS
MALDI-TOF) calcd for [C142H152O34 + Na]+ calcd 2424.0, found
2424.0.
3A,6A,6D-Tri-O-acetyl-a-cyclodextrin (8)
The perbenzylated derivative 6 (122 mg, 0.05 mmol) was dissolved
in a mixture of THF–H2O (3 : 1, 5 mL) and was stirred with
10% Pd–C (50 mg) under a hydrogen atmosphere overnight. The
insoluble Pd–C was removed by filtration through celite and
evaporation of the solvent under reduced pressure gave pure
compound 8 (52 mg, quantitative yield). [a]2D5 +134.7◦ (c 1.4,
CH3OH); dH (400 MHz, CD3OD) 5.38 (dd, 1H, J = 9.2, 9.7 Hz,
H-3), 5.01 (d, 1H, J = 3.5 Hz, H-1), 4.98–4.91 (m, 4H, 4 ¥ H-1),
4.79 (d, 1H, J = 3.3 Hz, H-1), 4.66 (br d, 1H, J = 10.8 Hz, H-6),
4.58 (br d, 1H, J = 10.8 Hz, 1H, H-6), 4.36 (dd, 1H, J = 6.0,
12.1 Hz, H-6), 4.22 (dd, 1H, J = 6.6, 11.9 Hz, H-6), 4.16–4.09
(m, 1H, H-5), 4.09–4.02 (m, 1H, H-5), 4.02–3.78 (m, 17H), 3.67–
3.38 (m, 12H), 2.11 (s, 3H, COCH3), 2.10 (s, 3H, COCH3), 2.09
(s, 3H, COCH3); dC (100 MHz, CD3OD) 172.36, 171.44, 171.42
(3 ¥ CO), 102.49, 102.46, 102.43, 102.11, 102.02, 101.68 (6 ¥ C-1),
82.45, 82.11, 82.08, 82.01, 81.28, 79.98 (6 ¥ C-2), 73.90, 73.81,
73.77, 73.74, 72.62, 72.50, 72.44, 72.33, 72.25, 72.23, 72.14, 71.49,
70.97, 70.24, 69.67 (C-3’s, C-4’s, C-5’s), 63.83 (C-6_OAc), 63.67
(C-6_OAc), 60.51, 60.46, 60.39, 60.17 (4 ¥ C-6_OH), 20.40 (CH3),
=
(100 MHz, CDCl3) 170.72, 170.35, 170.16 (3 ¥ C O), 140.07,
140.04, 139.77, 139.61, 138.78, 138.69, 138.56, 138.51, 138.36,
138.31, 138.23, 138.22, 138.18, 138.13, 128.43, 128.33, 128.30,
128.29, 128.18, 128.13, 128.06, 128.05, 128.01, 127.96, 127.91,
127.82, 127.75, 127.72, 127.70, 127.65, 127.56, 127.53, 127.47,
127.43, 127.38, 127.18, 127.13, 127.07, 126.83, 126.78, 126.62,
126.53, 101.36, 101.08, 100.67, 100.24, 100.21, 99.89 (6 ¥ C-1),
82.07, 81.94, 81.74, 81.62, 80.50, 80.37, 80.32, 80.13, 80.08, 79.36,
78.99, 78.81, 78.73, (5 ¥ C-2, 5 ¥ C-3, 6 ¥ C-4), 76.41, 76.06,
75.96, 75.76, 75.02, 74.21, 73.34, 73.25, 72.95, 72.39, 72.22, 72.18,
72.14 (CBn, 1 ¥ C-5, 1 ¥ C-2), 71.78, 71.59, 71.53, 71.38, 69.96,
69.82 (5 ¥ C-5, 1 ¥ C-3), 69.16 (2 ¥ C-6), 68.96 (C-6), 68.86 (C-
6), 63.74 (C-6), 62.94 (C-6), 21.32, 20.94, 20.89 (3 ¥ CH3); m/z
(MS MALDI-TOF) calcd for [C147H156O33 + Na]+ 2472.0, found
2472.1.
19.40 (2 ¥ CH3); m/z (MS MALDI-TOF) calcd for [C42H66O33
+
Na]+ 1121.3, found 1121.3; calcd for [C42H66O33 + K]+ 1137.3,
found 1137.2.
3A,3D,6A,6D-Tetra-O-acetyl-2A,2B,2C,2D,2E,2F,3B,3C,3D,3E,
3F,6B,6C,6E,6F-hexadeca-O-benzyl-a-cyclodextrin (7)
2A,2B,2C,2D,2E,2F,2G,3A,3B,3C,3D,3E,3F,3G,6A,6B,6E-Hexadeca-O-
benzyl-b-cyclodextrin (14)
To a solution of tetraol 5 (112 mg, 0.05 mmol) in dry
dichloromethane (1.0 mL) was added Ac2O (0.3 mL), anhy-
drous pyridine (0.3 mL) and a catalytic amount of 4-N,N-
To a solution of perbenzylated b-cyclodextrin 9 (5.0 g, 1.653 mmol)
in anhydrous toluene (26.5 mL), was added a solution of
This journal is
The Royal Society of Chemistry 2010
Org. Biomol. Chem., 2010, 8, 171–180 | 177
©

View Article Online
DIBAL-H in hexane (47 mL, 1.4 M solution, 66.1 mmol) with
stirring under an atmosphere of argon at room temperature.
The reaction mixture was heated to 50 ◦C for 4 days and then
quenched by slowly pouring the reaction mixture into ice-cold
water (35 mL) with vigorous stirring. The reaction mixture was
diluted with EtOAc (80 mL) and filtered through celite. This
celite was then washed thoroughly with hot EtOAc (3 ¥ 25 mL).
More water (100 mL) was added to the filtrate and the organic
layer was separated and washed with brine (100 mL), dried over
anhydrous Na2SO4 and concentrated under reduced pressure.
The crude mixture was purified by column chromatography on
silica gel using a mixture of AcOEt in hexane (gradient 10–
50%) as eluent to afford sequentially the diol 11 (376 mg,
8% yield) and two major fractions. The less polar fraction (Rf
0.45, dichloromethane : acetone 85 : 15), possibly 12 and 13, as
confirmed from the acetylated products (see below) contained an
inseparable mixture of two triols (1.138 g, 25% yield); the more
polar fraction is the tetraol 14 (969 mg, 22% yield).
6A,6B,6D,6E-Tetra-O-acetyl-2A,2B,2C,2D,2E,2F,2G,3A,3C,3D,
3E,3F,3G-6C,6F,6G-heptadeca-O-benzyl-b-cyclodextrin (17)
The pure b-CD tetraol 14 (200 mg, 0.075 mmol) was acetylated
with Ac2O (0.5 mL) in a mixture of anhydrous dichloromethane
(2 mL) and anhydrous pyridine (0.5 mL) in the presence of
a catalytic amount of 4-N,N-dimethylaminopyridine as before.
The tetraacetate 17 (182 mg, yield 86%) was obtained as a
white solid after column chromatography on silica gel using a
mixture of AcOEt in hexane (gradient 15–35%) as eluent. Rf 0.40
(EtOAc : hexane 4 : 6); [a]2D5 +20.4◦ (c 0.4, CHCl3); dH (400 MHz,
CDCl3) 7.36–7.01 (m, 85H, Ar-H), 5.37–5.23 (m, 4H, 2 ¥ H-1, 2 ¥
PhCH), 5.21 (d, 1H, J = 10.6 Hz, PhCH), 5.07 (d, 1H, J = 3.7 Hz,
H-1), 5.05 (d, 1H, J = 3.5 Hz, H-1), 5.04–4.93 (m, 5H, 3 ¥ H-1,
2 ¥ PhCH), 4.84–4.73 (m, 9H, 9 ¥ PhCH), 4.69–4.33 (m, 26H, 20 ¥
PhCH, 6 ¥ H-6), 4.31–4.25 (m, 2H, 2 ¥ H-6), 4.23–3.97 (m, 17H,
3 ¥ H-4, 3 ¥ H-6, 4 ¥ H-5, 7 ¥ H-3), 3.92 (br d, 1H, J = 8.4 Hz,
H-5), 3.88–3.68 (m, 6H, 4 ¥ H-4, 2 ¥ H-5), 3.66–3.58 (m, 3H, 3 ¥
H-6), 3.57–3.45 (m, 5H, 5 ¥ H-2), 3.45–3.39 (m, 2H, 2 ¥ H-2), 2.01
(s, 3H, COCH3), 1.99 (s, 3H, COCH3), 1.99 (s, 3H, COCH3), 1.94
(s, 3H, COCH3); dC (100 MHz, CDCl3) 170.39 (2 ¥ CO), 170.37
(CO), 170.36 (CO), 139.43, 139.28, 139.26, 139.22, 139.17, 138.99,
138.95, 138.63, 138.55, 138.49, 138.34, 138.28, 138.26, 138.17,
138.12, 128.32, 128.28, 128.25, 128.22, 128.17, 128.16, 128.13,
128.11, 128.06, 128.02, 127.98, 127.94, 127.92, 127.87, 127.82,
127.79, 127.73, 127.64, 127.61, 127.51, 127.49, 127.42, 127.15,
127.09, 126.83, 99.19 (C-1), 98.93 (C-1), 98.81 (C-1), 98.63 (C-1),
98.37 (C-1), 98.30 (C-1), 98.23 (C-1), 80.95, 80.87, 80.58, 80.48,
79.80, 79.70, 79.22, 79.03, 78.60, 76.29 (PhCH), 76.16 (PhCH),
75.67 (PhCH), 74.90 (PhCH), 73.43 (PhCH), 73.36 (PhCH),
73.24 (PhCH), 73.20 (PhCH), 72.85 (PhCH), 72.82 (PhCH), 72.75
(PhCH), 72.43 (PhCH), 72.22 (PhCH), 71.79 (C-5), 71.72 (C-5),
71.57 (C-5), 69.83 (C-5), 69.59 (C-5), 69.48 (C-5), 69.32 (C-5),
68.84 (C-6), 68.75 (C-6), 68.58 (C-6), 63.68 (C-6_OAc), 63.59 (C-
6_OAc), 63.48 (C-6_OAc), 63.42 (C-6_OAc), 20.91 (CH3), 20.86
(CH3), 20.83 (CH3), 20.72 (CH3); m/z (MS MALDI-TOF) calcd
for [C169H180O39 + Na]+ 2856.2, found 2856.0.
Data for b-CD diol 11. Rf 0.50 (hexane : AcOEt 7 : 3); [a]D25
+34.8◦ (c 1.2, CHCl3); 1H and 13C NMR data are identical to the
literature.1b
Data for b-CD tetraol 14. Rf 0.25 (ethyl acetate : hexane 1 : 1);
Rf 0.30 (dichloromethane : acetone 85 : 15); [a]25 +25.4◦ (c 0.3,
D
CHCl3); dH (400 MHz, CDCl3) 7.37–7.01 (m, 85H, Ar-H), 5.42
(d, 1H, J = 3.7 Hz, H-1), 5.34 (d, 1H, J = 3.8 Hz, H-1), 5.17 (m,
4H, H-1, 3 ¥ PhCH), 5.03 (d, 1H, J = 3.5 Hz, H-2), 4.98–4.87 (m,
3H, H-1, 2 ¥ PhCH), 4.83–4.34 (m, 31H, PhCH), 4.08–3.59 (m,
35H, H-3, 7 ¥ H-4, 7 ¥ H-5, 7 ¥ H-6, 7 ¥ H-6¢), 3.51 (m, 4H, 4 ¥
H-2), 3.42 (dd, 1H, J = 9.7, 3.5 Hz, H-2), 3.39–3.32 (m, 2H, 2 ¥
H-2), 3.03 (dd, 1H, J = 5.3, 5.3 Hz, OH-6), 2.79 (dd, 1H, J = 5.6,
5.6 Hz, OH-6), 2.68 (dd, 1H, J = 6.2, 6.2 Hz, OH-6), 2.48 (dd, 1H,
J = 6.5, 6.5 Hz, OH-6); dC (100 MHz, CDCl3) 139.39, 139.25,
139.20, 139.08, 138.93, 138.81, 138.58, 138.49, 138.36, 138.22,
138.17, 138.08, 138.03, 137.78, 128.37, 128.33, 128.31, 128.25,
128.22, 128.15, 128.11, 128.03, 127.98, 127.90, 127.82, 127.74,
127.69, 127.58, 127.53, 127.26, 127.21, 127.04, 127.00, 126.89,
98.71, 98.61, 98.24, 98.21, 97.98, 97.82, 97.68 (7 ¥ C-1) 81.12,
80.85, 80.78, 80.69, 80.61, 80.47, 80.39, 80.23, 79.34, 79.26, 79.20,
78.78, 78.66, 78.19, 77.83, 76.65, 76.34, 76.21, 76.09, 75.63, 75.59,
75.40, 74.97, 74.78, 74.37, 73.46, 73.36, 73.07, 72.79, 72.75, 72.64,
72.61, 72.39, 72.00, 71.84, 71.77, 71.73 (PhCH, C-3’s, C-4’s, C-5’s,
C-6’s), 69.41, 69.35, 69.29, 62.15, 62.10, 61.96, 61.82 (7 ¥ C-2); m/z
(MS MALDI-TOF) calcd for [C161H172O35 + Na]+ 2688.2, found
2688.0.
6A,6B,6D,6E-Tetra-O-acetyl-b-cyclodextrin (18)
Compound 18 was obtained in a similar fashion as 8 (yield:
quantitative). dH (400 MHz, CD3OD) 5.04–4.92 (m, 7H, 7 ¥ H-1),
4.67–4.40 (m, 4H, 4 ¥ H-6a_OAc), 4.30–4.08 (m, 4H, 4 ¥ H-
6b_OAc), 4.05–3.69 (m, 20H, 7 ¥ H-5, 7 ¥ H-3, 3 ¥ H-6a_OH, 3 ¥
H-6b), 3.63–3.37 (m, 14H, 7 ¥ H-2, 7 ¥ H-4), 2.10 (s, 6H, 2 ¥ Ac),
2.07 (s, 6H, 2 ¥ Ac); dC (100 MHz, CD3OD) 171.45 (CO), 171.39
(CO), 171.29 (CO), 171.26 (CO), 103.01, 103.00, 102.72, 102.71,
102.40, 102.21, 102.04 (7 ¥ C-1), 82.76, 82.73, 82.36, 82.35, 81.89,
81.27, 81.24, 73.46, 73.40, 73.38, 73.32, 73.26, 73.17, 72.89, 72.77,
72.63, 72.38, 72.29, 69.65, 69.60, 69.55, 63.53 (¥ 3, C-6_OAc),
63.44 (C-6_OAc), 60.45 (C-6_OH), 60.07 (C-6_OH), 59.95 (C-
6_OH), 19.40 (¥ 2, Ac), 19.36 (¥ 2, Ac); m/z (MS MALDI-TOF)
calcd for [C50H78O39 + Na]+ 1325.4, found 1325.1.
Acetylation of the less polar fraction. The 1 : 1 mixture (200 mg)
was acetylated for 6 h as 4 to afford an inseparable mixture
(presumably 15 and 16) by column chromatography on silica gel
using a mixture of AcOEt in hexane (gradient 10–30%) (167 mg,
yield 80%). Rf 0.45 (EtOAc : hexane 4 : 6); characteristic NMR
data: dH (400 MHz, CDCl3) 2.01 (s, 3H, COCH3), 1.99 (s, 3H,
COCH3), 1.99 (s, 3H, COCH3), 1.98 (s, 3H, COCH3), 1.95 (s, 3H,
COCH3), 1.94 (s, 3H, COCH3); dC (100 MHz, CDCl3) 170.39,
170.37, 170.34, 170.31, 170.25 (2 ¥ C) (CO), 20.95, 20.89, 20.86,
20.80; m/z (MS MALDI-TOF) calcd for [C174H184O38 + Na]+
2904.2, found 2903.8.
6A,6B,6D,6E-Tetra-O-methanesulfonyl-2A,2B,2C,2D,2E,2F, 2G,3A,3C,
3D,3E,3F,3G,6C,6F,6G-heptadeca-O-benzyl-b-cyclodextrin (19)
To the clear solution of tetraol 14 (156 mg, 0.0585 mmol) in
anhydrous dichloromethane (4 ml) and triethylamine (0.0652 ml,
0.469 mmol, 8 eq), was added methanesulfonyl chloride (0.024 ml,
178 | Org. Biomol. Chem., 2010, 8, 171–180
This journal is
The Royal Society of Chemistry 2010
©

View Article Online
0.305 mmol, 5.2 eq) at 0 ◦C, and the reaction was continued at
15 ◦C for 4 h. The reaction was quenched with a few drops of cold
water. More dichloromethane (20 mL) was added, and the organic
solution was washed with water (10 mL) and brine (10 mL), dried
over anhydrous Na2SO4, and evaporated under reduced pressure.
The residue was purified by column chromatography on silica gel
using a mixture of AcOEt in hexane (gradient 20–45%) as the
eluent to afford pure tetramesylate 19 as a colorless solid (146 mg,
yield 84%). Rf 0.30 (hexane : AcOEt 6 : 4); dH (400 MHz, CDCl3)
7.37–6.92 (m, 85H, 17¥ Ph), 5.45 (d, 1H, J = 3.8 Hz, H-1), 5.31 (d,
1H, J = 10.5 Hz, CHPh), 5.25 (d, 1H, J = 11.3 Hz, CHPh), 5.22
(d, 1H, J = 4.7 Hz, H-1), 5.20–5.16 (m, 2H, H-1, CHPh), 5.02 (d,
1H, J = 10.9 Hz, CHPh), 4.99 (d, 1H, J = 3.7 Hz, H-1), 4.98 (d,
1H, J = 3.6 Hz, H-1), 4.93 (d, 1H, J = 3.4 Hz, H-1), 4.91 (d, 1H,
J = 3.4 Hz, H-1), 4.92–4.87 (m, 1H, CHPh), 4.81–4.27 (m, 29H,
CHPh), 4.21–3.54 (m, 35H, 7 ¥ H-3, 7 ¥ H-4, 7 ¥ H-5, 7 ¥ H-6,
7 ¥ H-6¢), 3.52 (d, 1H, J = 3.7 Hz, H-2), 3.49 (dd, 1H, J = 5.5,
3.8 Hz, H-2), 3.47–3.44 (m, 2H, 2 ¥ H-2), 3.44–3.41 (m, 1H, H-2),
3.38 (dd, 1H, J = 3.1, 2.5 Hz, H-2), 3.36 (dd, 1H, J = 3.2, 2.4 Hz,
H-2), 2.93 (s, 3H, SO2CH3), 2.90 (s, 3H, SO2CH3), 2.87 (s, 6H, 2 ¥
SO2CH3); dC (100 MHz, CDCl3) 139.36, 139.20, 139.16, 139.15,
139.04, 138.81, 138.78, 138.59, 138.42, 138.38, 138.26, 138.23,
138.19, 138.01, 137.90, 137.77, 128.38, 128.35, 128.33, 128.24,
128.18, 128.15, 128.12, 128.05, 128.03, 127.98, 127.92, 127.86,
127.77, 127.73, 127.67, 127.58, 127.54, 127.48, 127.43, 127.33,
127.17, 127.16, 127.13, 127.09, 127.06, 126.94, 126.86, 126.66,
98.94, 98.84, 98.76, 98.60, 98.28, 98.21, 97.94 (7 ¥ C-1), 80.91,
80.74, 80.67, 80.60, 80.47, 80.14, 79.74, 79.11, 78.97, 78.20, 77.88,
76.66, 76.33, 76.31, 76.08, 75.66, 75.30, 74.66, 73.48, 73.43, 73.36,
73.26, 72.89, 72.73, 72.43, 72.02, 71.76, 71.46, 69.59, 69.28, 69.16,
69.06, 68.98, 37.25, 37.13, 37.07, 37.03 (4 ¥ SO2CH3); m/z (HRMS
MALDI-TOF) calcd for [C165H180O43S4 + Na]+ 3000.06734, found
3000.06823.
6A,6B,6D,6E-Tetradeoxy-b-cyclodextrin (21)
Compound 20 (90 mg, 0.0346 mmol) in
a mixture of
tetrahydrofuran–H2O (3 : 1, 4 mL), was hydrogenated in the
presence of 10% Pd–C (35 mg) as 8 to give pure compound
21 (34 mg, quantitative yield). [a]2D5 +85.3◦ (c 0.2, CH3OH); dH
(400 MHz, pyridine-d5) 5.74 (d, 1H, J = 3.5 Hz, H-1), 5.60 (d, 1H,
J = 3.5 Hz, H-1), 5.58 (d, 1H, J = 3.6 Hz, H-1), 5.55 (d, 2H, J =
3.1 Hz, 2 ¥ H-1), 5.41 (d, 1H, J = 3.8 Hz, H-1), 5.40 (d, 1H, J =
3.7 Hz, H-1), 4.86–4.78 (m, 3H, 3 ¥ H-3), 4.73–4.65 (m, 4H, 4 ¥
H-3), 4.62–4.39 (m, 13H, 7 ¥ H-5, 3 ¥ H-6, 3 ¥ H-6¢), 4.35 (dd, 1H,
J = 9.2 Hz, H-4), 4.31 (dd, 2H, J = 9.3 Hz, 2 ¥ H-4), 4.23–4.12 (m,
7H, 7 ¥ H-2), 3.66 (dd 1H, J = 8.6 Hz, H-4), 3.64 (dd, J = 8.6 Hz,
1H, H-4), 3.60 (overlap two dd, J = 9.0 Hz, 2H, 2 ¥ H-4), 1.73
(d, 3H, J = 6.2 Hz, -CH3), 1.68 (d, 3H, J = 6.2 Hz, -CH3), 1.58
(d (two overlapping), 6H, J = 6.2 Hz, 2 ¥ -CH3); dC (100 MHz,
pyridine-d5) 104.62, 104.60, 104.31, 104.28, 104.01, 103.91, (7 ¥
C-1), 90.25, 90.22, 90.13, 90.05 (4 ¥ C-4), 83.67, 83.52, 83.38 (3 ¥
C-4), 75.12, 75.07, 74.81, 74.79, 74.76, 74.69 (7 ¥ C-2, 7 ¥ C-3),
74.43, 74.40, 74.34 (3 ¥ C-5), 68.58, 68.46, 68.39, 68.36 (4 ¥ C-5),
62.02, 61.92, 61.83 (3 ¥ C-6), 18.72, 18.62, 18.59 (4 ¥ CH3); m/z
(MS MALDI-TOF) calcd for [C42H70O31 + Na]+ 1093.4, found
1093.0.
2A,2B,2C,2D,2E,2F,2G,2H,3A,3B,3C,3D,3E,3F,3G,3H,6A,6B,6E,6F-
Eicosa-O-benzyl-c-cyclodextrin (23)
To a solution of perbenzylated g-cyclodextrin 22 (3.0 g,
0.868 mmol) in anhydrous toluene (13.5 mL) was added a solution
of DIBAL-H (25 mL, 1.4 M solution in hexane, 34.7 mmol) with
stirring under argon atmosphere, and the mixture was heated to
50 ◦C for 4 days and then quenched by pouring the reaction
mixture into ice-cold water (40 mL). The spot with an Rf 0.35
(dichloromethane : acetone, 9 : 1) was isolated by repeated column
chromatographies on silica gel using a mixture of AcOEt in hexane
(gradient 15–50%, and then repurification using a mixture of
dichloromethane : acetone 9 : 1) as eluent to afford 23 (215 mg,
6A,6B,6D,6E-Tetradeoxy-2A,2B,2C,2D,2E,2F,2G,3A,3C,3D,3E,
3F,3G,6C,6F,6G-heptadeca-O-benzyl-b-cyclodextrin (20)
8% yield). Rf 0.35 (dichloromethane : acetone 9 : 1); [a]25 +40.0◦
D
To the clear solution of the tetramesylate 19 (140 mg, 0.0486 mmol)
in anhydrous tetrahydrofuran (4 ml), was added lithium alu-
minium hydride (14.8 mg, 0.389 mmol, 8 eq), and the reaction
was continued at room temperature for 15 h. The excess amount
of hydride was destroyed by AcOEt (0.5 mL) and the mixture
was evaporated to dryness, and the residue was re-dissolved in
dichloromethane (15 mL) and the solution was washed with water
(5 mL) and brine (5 mL), dried over anhydrous Na2SO4 and evapo-
rated. The crude material was purified by column chromatography
on silica gel using a mixture of AcOEt in hexane (gradient 10–20%)
as the eluent to afford pure tetradeoxygenated 20 (98.5 mg, yield
78%). Rf 0.4 (hexane : AcOEt 7 : 3); dH (400 MHz, CDCl3) 7.23–
7.03 (m, 85H, 17 ¥ Ph), 5.27–5.22 (m, 2H, CHPh), 5.21 (d, 1H,
J = 3.7 Hz, H-1), 5.18 (d, 1H, J = 3.8 Hz, H-1), 5.13 (d, 1H, J =
10.9 Hz, CHPh), 5.08 (d, 1H, J = 11.1 Hz, CHPh), 5.04 (d, 2H,
J = 3.7 Hz, 2 ¥ H-1), 4.99 (d, 1H, J = 3.5 Hz, H-1), 4.94–4.88 (m,
3H, H-1, 2 ¥ CHPh), 4.85 (d, 1H, J = 3.4 Hz, H-1), 4.83–4.36 (m,
28H, CHPh), 4.16–3.59 (m, 27H, 7 ¥ H-3, 7 ¥ H-4, 7 ¥ H-5, 3 ¥
H-6, 3 ¥ H-6¢), 3.56–2.72 (m, 7H, 7 ¥ H-2), 1.33 (d, 1H, J = 6.2 Hz,
CH3), 1.28 (d, 1H, J = 5.8 Hz, CH3), 1.27 (d, 1H, J = 6.0 Hz,
CH3), 1.25 (d, 1H, J = 6.4 Hz, CH3); m/z (HRMS MALDI-TOF)
calcd for [C161H172O31 + Na]+ 2624.1780, found 2624.1775.
(c 0.4, CHCl3); dH (400 MHz, CDCl3) 7.38–7.07 (m, 50H, Ar-
H), 5.36 (d, 1H, J = 3.5, H-1), 5.25–5.20 (m, 2H, H-1, PhCH),
5.12–5.03 (m, 1H, PhCH), 4.91 (d, 1H, J = 3.5, H-1), 4.87 (d,
1H, J = 3.5, H-1), 4.84–4.37 (m, 23H), 4.09–3.61 (m, 26H), 3.58–
3.51 (m, 2H), 3.49 (dd, 1H, J = 9.1, 3.5, H-2), 3.36 (dd, 1H, J =
9.6, 3.4, H-2), 3.26 (s, 1H, OH), 2.59 (s, 1H, OH); dC (100 MHz,
CDCl3) 139.53, 139.14, 138.70, 138.63, 138.34, 138.20, 138.14,
137.90, 128.43, 128.38, 128.32, 128.29, 128.27, 128.22, 128.15,
128.12, 128.09, 128.00, 127.87, 127.79, 127.74, 127.69, 127.63,
127.55, 127.43, 127.18, 127.02, 126.85, 98.44, 98.26, 97.93, 97.53
(C-1), 81.19, 80.87, 80.35, 80.24, 79.53, 78.95, 78.10, 76.15, 76.01,
75.93, 75.27, 74.78, 74.12, 73.46, 73.35, 73.30, 72.85, 72.78, 72.62,
72.45, 72.14, 71.69, 71.35, 68.92, 68.82, 62.07, 61.64 (C-6); m/z
(MS MALDI-TOF) calcd for [C188H200O40 + Na]+ 3120.3, found
3120.0.
6A,6B,6E,6F-Tetra-O-acetyl-2A,2B,2C,2D,2E,2F,2G,2H,3A,3C,
3D,3E,3F,3G,3H,6C,6D,6G,6H-eicosa-O-benzyl-c-cyclodextrin (24)
The tetraol 23 (155 mg, 0.05 mmol) was acetylated in a
similar fashion as before to afford the desired tetraacetate 24
after column chromatography on silica gel using a mixture of
This journal is
The Royal Society of Chemistry 2010
Org. Biomol. Chem., 2010, 8, 171–180 | 179
©

View Article Online
AcOEt–hexane (gradient 10–20%) as eluent (140 mg, 86% yield).
R. B. C. Jagt and M. Nitz, Org. Biomol. Chem., 2008, 6, 4622–4626;
(f) O. Bistri, P. Sinay¨ and M. Sollogoub, Tetrahedron Lett., 2006, 47,
4137–4139; (g) A. W. Coleman, C. C. Ling and M. Miocque, Angew.
Chem., Int. Ed. Engl., 1992, 31, 1381–1383.
Rf 0.40 (EtOAc : hexane 4 : 6); [a]25 +26.6◦ (c 0.2, CHCl3); dH
D
(400 MHz, CDCl3) 7.44–7.02 (m, 50H, Ar-H), 5.46 (d, 1H, J =
10.6, PhCH), 5.37 (d, 1H, J = 10.6, PhCH), 5.24 (d, 1H, J = 3.7,
H-1), 5.11 (d, 1H, J = 3.5, H-1), 5.08 (d, 1H, J = 3.1, H-1), 5.04 (d,
1H, J = 3.5, H-1), 4.97 (2 ¥ d, 2H, J = 11.0, PhCH), 4.90–4.77 (m,
4H, PhCH), 4.61–4.41 (m, 12H), 4.31–3.91 (m, 14H), 3.85–3.71
(m, 3H), 3.69–3.57 (m, 3H), 3.53–3.42 (m, 4H, 4 ¥ H-2), 1.94 (s, 3H,
Ac), 1.92 (s, 3H, Ac); dC (100 MHz, CDCl3) 170.27, 170.25 (CO),
139.58, 139.52, 139.16, 138.92, 138.65, 138.36, 138.33, 138.22,
138.07, 128.33, 128.29, 128.24, 128.14, 128.09, 127.96, 127.95,
127.81, 127.73, 127.61, 127.58, 127.50, 127.47, 127.10, 126.95,
99.23, 98.35, 98.25, 98.16 (C-1), 81.07, 80.99, 80.95, 80.80, 80.66,
80.57, 79.37, 79.04, 78.95, 78.20, 76.47, 76.37, 75.28, 74.98, 73.58,
73.31, 73.17, 72.74, 72.40, 71.77, 71.58, 69.44, 69.24, 68.86, 68.79,
63.35, 20.89 (Ac), 20.78 (Ac); m/z (HRMS MALDI-TOF) calcd
for [C196H208O44 + Na]+ 3288.39363, found 3288.39382.
6 (a) J. Bjerre, C. Rousseau, L. Marinescu and M. Bols, Appl. Microbiol.
Biotechnol., 2008, 81, 1–11; (b) R. Breslow, Acc. Chem. Res., 1995, 28,
146–153; (c) J. Bjerre, E. H. Nielsen and M. Bols, Eur. J. Org. Chem.,
2008, 745–752; (d) J. Bjerre, T. H. Fenger, L. G. Marinescu and M. Bols,
Eur. J. Org. Chem., 2007, 704–710; (e) F. Ortega-Caballero, J. Bjerre,
L. S. Laustsen and M. Bols, J. Org. Chem., 2005, 70, 7217–7226; (f) C.
Rousseau, F. Ortega-Caballero, L. U. Nordstrom, B. Christensen, T. E.
Petersen and M. Bols, Chem.–Eur. J., 2005, 11, 5094–5101.
7 (a) F. Ortega-Caballero, C. Ortiz-Mellet, L. Le Gourrierec, N. Guil-
loteau, C. Di Giorgio, P. Vierling, J. Defaye and J. M. G. Ferna´ndez,
Org. Lett., 2008, 10, 5143–5146; (b) A. D´ıaz-Moscoso, P. Balbuena, M.
Go´mez-Garc´ıa, C. Ortiz-Mellet, J. M. Benito, L. Le Gourrie´rec, C. Di
Giorgio, P. Vierling, A. Mazzaglia, N. Micali, J. Defaye and J. M. G.
Ferna´ndez, Chem. Commun., 2008, 2001; (c) A. Me´ndez-Ardoy, M.
Go´mez-Garc´ıa, C. Ortiz-Mellet, N. Sevillano, M. D. Giro´n, R. Salto,
F. Santoyo-Gonza´lez and J. M. G. Ferna´ndez, Org. Biomol. Chem.,
2009, 7, 2681–268; (d) S. Menuel, S. Fontanay, I. Clarot, R. E. Duval,
L. Diez and A. Marsura, Bioconjugate Chem., 2008, 19, 2357–2362;
(e) A. McMahon, E. Gomez, R. Donohue, D. Forde, R. Darcy and
C. M. O’Driscoll, J. Drug Delivery Sci. Technol., 2008, 18, 303–307;
(f) N. Mourtzis, M. Paravatou, I. M. Mavridis, M. L. Roberts and K.
Yannakopoulou, Chem.–Eur. J., 2008, 14, 4188–4200.
Acknowledgements
We wish to acknowledge the Alberta Ingenuity, the Natural
Sciences and Engineering Research Council of Canada (NSERC),
and the University of Calgary for supporting the current project,
and Dr. Ping Zhang for technical assistance.
8 (a) K. Uekama, F. Hirayama and T. Irie, Chem. Rev., 1998, 98, 2045–
2076; (b) A. R. Khan, P. Forgo, K. J. Stine and V. T. D’Souza, Chem.
Rev., 1998, 98, 1977–1996.
9 (a) J. Boger, D. G. Brenner and J. R. Knowles, J. Am. Chem. Soc., 1979,
101, 7630–7631; (b) C.-C. Ling, A. W. Coleman and M. Miocque,
Carbohydr. Res., 1992, 223, 287–291; (c) S. Ward, P. Zhang and C. C.
Ling, Carbohydr. Res., 2009, 344, 808–814.
10 (a) L. Poorters, D. Armspach and D. Matt, Eur. J. Org. Chem., 2003,
1377–1381; (b) D. Armspach, L. Poorters, D. Matt, B. Benmerad, F.
Balegroune and L. Toupet, Org. Biomol. Chem., 2005, 3, 2588–2592.
11 E. Engeldinger, D. Armspach and D. Matt, Chem. Rev., 2003, 103,
4147–4173.
12 (a) I. Tabushi, K. Shimokawa, N. Shimizu, H. Shirakata and K. Fujita,
J. Am. Chem. Soc., 1976, 98, 7855–7856; (b) I. Tabushi, Y. Kuroda, K.
Yokota and L. C. Yuan, J. Am. Chem. Soc., 1981, 103, 711–712; (c) I.
Tabushi, L. C. Yuan, K. Shimokawa, K. I. Yokota, T. Mizutani and
Y. Kuroda, Tetrahedron Lett., 1981, 22, 2273–2276; (d) I. Tabushi, T.
Nabeshima, H. Kitaguchi and K. Yamamura, J. Am. Chem. Soc., 1982,
104, 2017–2019; (e) I. Tabushi, T. Nabeshima, K. Fujita, A. Matsunaga
and T. Imoto, J. Org. Chem., 1985, 50, 2638–2643.
Notes and references
1 (a) T. Lecourt, A. Herault, A. J. Pearce, M. Sollogoub and P. Sinay¨,
Chem.–Eur. J., 2004, 10, 2960–2971; (b) M. Sollogoub, Eur. J. Org.
Chem., 2009, 1295–1303; (c) S. Guieu and M. Sollogoub, Angew. Chem.,
Int. Ed., 2008, 47, 7060–7063; (d) N. Masurier, O. Lafont, R. Le Provost,
D. Lesur, P. Masson, F. Djedaini-Pilard and F. Estour, Chem. Commun.,
2009, 589–591.
2 (a) S. Guieu and M. Sollogoub, J. Org. Chem., 2008, 73, 2819–2828;
(b) O. Bistri, P. Sinay¨, J. J. Barbero and M. Sollogoub, Chem.–Eur. J.,
2007, 13, 9757–9774.
3 P. Balbuena, D. Lesur, M. J. G. Alvarez, F. Mendicuti, C. Ortiz-Mellet
and J. M. G. Fernandez, Chem. Commun., 2007, 3270–3272.
4 M. Petrillo, L. Marinescu, C. Rousseau and M. Bols, Org. Lett., 2009,
11, 1983–1985.
5 (a) W. H. Tang, I. W. Muderawan, T. T. Ong and S. C. Ng, Tetrahedron:
Asymmetry, 2007, 18, 1548–1553; (b) S. Riela, F. D’Anna, P. Lo Meo,
M. Gruttadauria, R. Giacalone and R. Noto, Tetrahedron, 2006, 62,
4323–4330; (c) Y. Liu, L. Li, X. Y. Li, H. Y. Zhang, T. Wada and Y.
Inoue, J. Org. Chem., 2003, 68, 3646–3657; (d) M. V. Rekharsky and Y.
Inoue, J. Am. Chem. Soc., 2002, 124, 813–826; (e) R. F. Gomez-Biagi,
13 D. Armspach and D. Matt, Carbohydr. Res., 1998, 310, 129–133.
14 (a) A. J. Pearce and P. Sinay¨, Angew. Chem., Int. Ed., 2000, 39, 3610–
3612; (b) B. du Roizel, J. P. Baltaze and P. Sinay¨, Tetrahedron Lett.,
2002, 43, 2371–2373.
15 G. K. Rawal, S. Rani and C.-C. Ling, Tetrahedron Lett., 2009, 50,
4633–4636.
16 O. Bistri, P. Sinay¨ and M. Sollogoub, Tetrahedron Lett., 2005, 46, 7757–
7760.
180 | Org. Biomol. Chem., 2010, 8, 171–180
This journal is
The Royal Society of Chemistry 2010
©