Recognition of peptides by cyclodextrin trimers

Article abstract of DOI:10.1002/ejoc.201100671

The efficient synthesis of a new class of cyclodextrin trimers has been carried out by using a click chemistry strategy. The cyclodextrin trimers were subsequently investigated for molecular recognition of peptides with aromatic side chains. Binding affin

Full text of DOI:10.1002/ejoc.201100671

FULL PAPER
DOI: 10.1002/ejoc.201100671
Recognition of Peptides by Cyclodextrin Trimers
Helena Staunstrup Christensen,[a] Bent W. Sigurskjold,[a] Tobias Gylling Frihed,[a]
Lavinia G. Marinescu,[a] Christian M. Pedersen,[a] and Mikael Bols*[a]
Keywords: Cyclodextrins / Peptides / Synthesis design / Supramolecular chemistry / Click chemistry / Cycloaddition /
Binding affinity / Receptors / Recognition
The efficient synthesis of a new class of cyclodextrin trimers
has been carried out by using a click chemistry strategy. The
nance. Peptides were prepared and immobilized on the sen-
sor surface. The association constants were obtained by ti-
cyclodextrin trimers were subsequently investigated for mo- tration with different solutions of the cyclodextrin trimers and
lecular recognition of peptides with aromatic side chains.
Binding affinities for the self-assembly of different peptides
to cyclodextrin trimers were determined by using real-time
bimolecular interaction analyses with plasmon surface reso-
they were in the range of 103 –1. The selectivity of molecu-
M
lar recognition of nonapeptides favored cyclodextrin trimers
over unmodified cyclodextrin.
played by either α- or β-cyclodextrin, depending on the size
of the aromatic species.
Introduction
For chemists interested in the design and synthesis of
molecular receptors that are able to express their recogni-
tion features towards a range of different compounds, large
structurally constrained ring molecules are an excellent
tool.[1] Exploiting noncovalent bonding interactions, syn-
thetic macrocyclic substances, such as crown ethers,[2] ca-
lixarenes,[3,4] and cyclophanes,[5] are excellent molecular re-
ceptors. These designed receptors are inspired by nature’s
own large-sized ring compounds, cyclic peptides and pepto-
ids,[6] and carbohydrates (cyclodextrins).[7] Since they, like
other carbohydrates, are abundant and readily produced,
they have become popular building blocks in receptor mole-
cules.
A distinct drawback in the use of cyclodextrin building
blocks is their traditional difficult and low yielding modifi-
cation, requiring specialist competence. Because cyclodex-
trins are cyclic oligosaccharides, consisting of 6–8 d-gluco-
pyranose units linked by α-1,4-glycosidic bonds, they have
18–24 hydroxy groups and it is clear that selective modifica-
tion is nontrivial. On the other hand, it is a very attractive
feature that cyclodextrins are water soluble and are known
to form thermodynamically stable inclusion complexes with
a range of smaller organic compounds in aqueous solu-
tion.[8] Cyclodextrins in general show molecular recogni-
tionof aromatic moieties, and specific recognition can be dis-
DNA and RNA are biopolymers that, through their
structure and sequence-selective recognition of complimen-
tary strands, can store and replicate information. Nucleic
acids are unique both in structure and properties. They use
special base pairing as the tool for interstrand binding and
can undergo replication like no other molecule because one
strand can act as template for the synthesis of its comple-
mentary strand. However, perhaps the properties of DNA
can be emulated. Closely related analogues of nucleic acids
can indeed copy their properties of sequence selective bind-
ing; herein, we wish to go one step further and look at the
binding of cyclodextrin oligomers to a peptide strand of
aromatic amino acids. Previously, cyclodextrin oligomers
have been shown to exhibit characteristic binding specifi-
cities of aromatic moieties.[9]
The objective of the present study was to investigate
strandlike binding of aromatic functionalities by using tri-
meric cyclodextrin-based hosts. It was envisaged that a lin-
ear compound with phenyl and naphthyl substituents might
bind well to a linear assembly of cyclodextrins (Figure 1).
For synthetic convenience, a peptide was chosen as the lin-
ear attachment site for the aromatic groups. (For previous
work on recognition of peptides, see ref.[10]) Since different
aromatic groups bind with different affinity to cyclodex-
trins, selectivity between different peptides is also possible.
It is known that α- and β-cyclodextrin display different
binding efficiencies for phenyl and naphthyl groups, with
the larger β-cyclodextrin binding the naphthyl group
stronger than the α-cyclodextrin. While the pKD of a phenyl
group is 2–3 for both α- and β-cyclodextrin, it is 3 and 5,
respectively, for the dissociation of the naphthyl group from
the cyclodextrin.[8]
[a] Department of Chemistry, University of Aarhus and
Department of Chemistry, University of Copenhagen,
Universitetsparken 5, 2100 Copenhagen, Denmark
E-mail: bols@kemi.ku.dk
Supporting information for this article is available on the
WWW under http://dx.doi.org/10.1002/ejoc.201100671.
Eur. J. Org. Chem. 2011, 5279–5290
© 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
5279

M. Bols et al.
FULL PAPER
The nucleophilic substitution[16] of 1α and 1β to the az-
ides 2α and 2β proceeded in excellent yields (Scheme 1). The
substitution was carried at 75 °C in dimethyl formamide
overnight, affording 80–95% yield after flash chromatog-
raphy. Notably, the reaction could not be monitored by
TLC because compounds 1 and 2 have identical Rf values.
The modification of the 6A-alcohols 3α and 3β into the
corresponding propargyl ethers 4α and 4β, respectively, was
conducted as described in Scheme 1. Treatment of the
alcohol with sodium hydride in dimethyl formamide at 0 °C,
and subsequent addition of freshly distilled propargyl bro-
Figure 1. A trimer constructed from cyclodextrins binding a pept-
ide with three aromatic side chains. The β-cyclodextrin preferably mide resulted in full conversion after 4 h to the propargyl
binds naphthyl groups.
ether 4 in 88–93% after flash chromatography.
Results and Discussion
Model Experiments for the Click Reaction
The host–guest complex was analyzed by modeling to
establish the optimal spacing between aromatic functionali-
To investigate the viability of the click reaction, we first
ties to design peptides that could interact with the cyclodex-
carried out some model experiments with simpler coupling
trin trimers. The cyclodextrin oligomers were synthesized
partners. While examples of successful application of click
by a click chemistry strategy[11,12] from building blocks
chemistry to cyclodextrins have been reported,[17,18] these
readily accessible by standard synthetic procedures used in
have involved unprotected cyclodextrins for which the usual
our group. Experimental validation of the calculated obser-
“click protocol” with water and tert-butyl alcohol was em-
ployed. Since the benzylated cyclodextrin derivatives 2 and
4 were very poorly soluble in water, a protocol utilizing cop-
vations was conducted by determining binding affinities for
the self-assembly of peptides to the cyclodextrin trimers.
per(I) iodide in an organic solvent with an organic base was
necessary. Of the different nonaqueous protocols that have
Synthesis of Cyclodextrin Building Blocks
been employed for click reactions in the literature,[19,20,21,22]
The cyclodextrin building blocks intended for preparing the procedure by Hotha and Kashyap,[19] with the addition
the trimers were synthesized as described in Scheme 1. of ethyl acetate as a co-solvent, was the successful method.
First, the monomer building blocks were prepared: di-
First, we treated 2β with phenylacetylene in the presence
iodides 1α and 1β (Scheme 1) were obtained by perbenz- of CuI and diisopropylethylamine (DIPEA) in MeCN/
ylation of α- and β-cyclodextrin, respectively, using sodium EtOAc. This reaction gave, as expected, only one product 5
hydride and benzyl bromide in dimethyl sulfoxide,[13] fol- in 81% yield after 3 d at 40 °C (Scheme 2). However, if CuI
lowed by selective debenzylation of the 6A-, 6D-positions was omitted then the starting material was unchanged even
with diisobutylaluminum hydride,[14] and finally conversion after 5 d. This experiment showed that the reaction could
of the 6A-, 6D-alcohols to iodides 1α and 1β.[15]
only occur with these substrates when catalyzed by copper.
Scheme 1. Preparation of diazide and propargyl ether building blocks. The cyclodextrins were converted into diiodides 1α or 1β,[15] or
monools 3α and 3β[14] by known methods and transformed into 2α–β and 4α–β (Bn = benzyl).
5280
www.eurjoc.org
© 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Org. Chem. 2011, 5279–5290

Recognition of Peptides by Cyclodextrin Trimers
Scheme 2. Reaction of 2β with simple phenylacetylene.
Scheme 3. Cu-catalyzed Huisgen reaction with bulky acetylene (TIPS = triisopropylsilyl).
Scheme 4. Cu-catalyzed Huisgen reaction with α-d-glucopyranoside derivatives.
Next, the coupling of diazide 2β with more bulky substi- reaction was monitored by TLC until the starting material
tuted alkynes (Scheme 3) was attempted. Reaction of 2β disappeared. The product was isolated in 64% yield and
with triisopropylsilyl acetylene under the same conditions was used in the copper-catalyzed Huisgen reaction with az-
afforded only one compound, 6, in 58% yield. The presence ide 11 (Scheme 4). This reaction also worked well to give a
of a secondary compound was observed during chromatog- single triazole product 12 in 58% yield.
raphy, which proved to be the Glaser coupling product of
terminal alkynes. Additional debenzylation of compound 6
certified the presence of only one isomer.
As an even better model of the intended reaction, azide
2β was treated with 6-O-propargyl ether 7 in the presence of
CuI and DIPEA in MeCN/EtOAc (Scheme 4). The desired
Synthesis of the Trimeric Cyclodextrins
Having established a working click chemistry procedure,
product 8 was isolated in quantitative yield. The Glaser- we proceeded to assemble building blocks 2 and 4 into tri-
coupled byproduct 9 (alkyne–alkyne coupling) was ob- mers (Scheme 5). The preparation of the cyclodextrin tri-
served as a side product due to the use of a larger excess of mers proceeded smoothly in ethyl acetate/acetonitrile (1:2)
alkyne and the presence of oxygen during the reaction.
with copper(I) iodide and Hünig’s base by using 2.1 equiva-
The alternative click coupling reaction was also tested lents of 4α or 4β per equivalent of diazide 2α or 2β
with building blocks 10 and 11. Dipropargyl ether cyclodex- (Scheme 2). Removal of the inorganic copper(I) species was
trin (10) was prepared by treatment of β-cyclodextrin diol achieved by washing the organic phase with an aqueous
with tBuOK in dimethylformamide at –78 °C and dropwise solution of ammonium chloride, which eased purification
addition of propargyl bromide. The reaction mixture was to a great extent. Since the reaction is copper(I) catalyzed,
allowed to reach room temperature and the progress of the and it is clear from the above model experiment that it can-
Eur. J. Org. Chem. 2011, 5279–5290
© 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjoc.org
5281

M. Bols et al.
FULL PAPER
Scheme 5. Synthesis of cyclodextrin trimers by click chemistry. From the combination of 4α or 4β (2 equiv.) and diazide 2α and 2β
(1 equiv.) the four possible combination of ditriazoles were prepared.
not proceed otherwise, the structure of the trimeric cyclo- tation on the catalyst. There was also the dilemma that acid
dextrins must have the 1,4-regioisomeric configuration, as favored the reaction, but could decompose the glycosidic
shown in Scheme 2. The four possible combinations of tri- bonds. Possible traces of sulfur-based catalyst poisons were
meric cyclodextrins, 13–16, were obtained in 73–85% yield eliminated by desulfurization with Raney nickel prior to re-
after flash chromatography. The click reaction strategy action, which had a positive but by no means dramatic ef-
could be performed even on a gram scale without loss of fect (Table 1, entries 5–6). The solution was found when
efficiency (e.g., preparation of 1.60 g of 14 occurred in 81% Pearlman’s catalystϪpalladium hydroxide on charcoalϪwas
yield). Nevertheless, the yield can be regarded as compara- applied (Table 1, entries 7–10), with 2-methoxyethanol as
tively low for click reactions, which may be explained by the solvent. The solvent was perfectly suited for dissolving
pronounced steric hindrance in the benzylated β-cyclodex- all stages of the debenzylation process; a task that was in-
trin derivatives.
sufficiently achieved by the three-phase solvent system of
The remaining step in the synthetic pathway to form tri- ethyl acetate, methanol, and water. The best results arose
mer cyclodextrins was global deprotection of the benzyl when the progress of the deprotection reaction was carefully
ethers by hydrogenolysis (Scheme 2), which was more diffi- monitoring, and in some cases extra palladium catalyst was
cult than anticipated at first and various strategies were car- added.
ried out. In Table 1 the screening of reaction conditions for
Upon successful deprotection, compounds 10 and 12
deprotection of trimers 13–16 is summarized. At first a pro- (and not 9 and 11) gave rise to an apparent isomerism that
cedure that had been applied to deprotection of a benzyl- was visible by NMR spectroscopy as doubling of some of
ated oligosaccharide click product by Gin’s group,[23] utiliz- the signals and two peaks were also obtained by HPLC. We
ing transfer hydrogenolysis with ammonium formate as the believe that this is conformational isomerism of the same
hydrogen donor and palladium on charcoal, was attempted type as the β-cyclodextrin inversion phenomena recently de-
(Table 1, entry 1), but after two days all benzyl ethers re- scribed by the Monflier group.[24] They observed that a sin-
mained intact on the cyclodextrins. More conventional hy- gle 6-triazole-substituted glucose residue in the β-cyclodex-
drogenolysis under a hydrogen atmosphere was likewise un- trin ring could rotate 360°, thereby embedding the triazole
successful (Table 1, entry 2). With trifluoroacetic acid deeper in the cavity. Arguments that this is a similar type
(TFA) present and a hydrogen pressure of 10–20 bar, hydro- of isomerization as that in the Monflier case are that doub-
genolysis proceeded at a slow pace (Table 1, entries 3–6); ling of the signals is only observed when the triazoles are
nevertheless, yields were low and one compound, 13, could substituted on a central β-cyclodextrin and not on α-cyclo-
not be deprotected. The causes of these problems may be dextrin. α-Cyclodextrin is presumably to small to undergo
associated with catalyst poisoning from possible impurities such inversion. Second, the ratio of conformational isomers
and/or the difficulties of finding a solvent mixture that dis- appeared to vary, depending on preparation conditions and
solved both starting material and product, avoiding precipi- solvents. In one preparation of 12, when comparatively ra-
5282
www.eurjoc.org
© 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Org. Chem. 2011, 5279–5290

Recognition of Peptides by Cyclodextrin Trimers
Table 1. Hydrogenation experiments with different catalysts.
Entry Starting material Pd cat.
Raney Ni Additive
Solvents
Time [d]
H2 pressure [bar]
Yield [%]
1
2
3
4
5
6
7
8
9
13
13
15
14
14
13
13
15
14
16
Pd/C
Pd/C
Pd/C
Pd/C
Pd/C
Pd/C
Pd(OH)2/C
Pd(OH)2/C
Pd(OH)2/C
Pd(OH)2/C
–
–
–
NH4HCO3 EtOAc/EtOH/H2O
2
2
2
3
2
0
1
10
15
20
20
1
1
1
1
–
–
81
–
38
–
EtOAc/EtOH
TFA
TFA
TFA
TFA
TFA
TFA
TFA
TFA
EtOAc/EtOH/H2O
EtOAc/EtOH/H2O
EtOAc/EtOH/H2O
EtOAc/EtOH/H2O
2-methoxyethanol
2-methoxyethanol
2-methoxyethanol
2-methoxyethanol
–
+
+
+
+
+
+
2
–
16
10
8
quant.
91
quant.
quant.
10
3
pid debenzylation was achieved, only one set of signals were lized on the sensor surface, while the other interacts with it
initially observed; however, after some manipulation of the from solution in constant flow over the surface. The bind-
compound another set of signals started to emerge.
ing of the soluble component to the sensor by noncovalent
interactions can then be monitored and binding constants
determined. To circumvent poor solubility of the peptides,
they were immobilized on the sensor.
Peptide Synthesis
There are several methods for covalent attachment of a
ligand to the surface: the most commonly utilized strategies
involve amine coupling, thiol coupling, and aldehyde cou-
pling to attach a ligand to the carboxymethylated dextran
matrix covering the gold layer. Based on the fact that all
peptides contain a free amino group at the N-terminus, the
amine coupling protocol was employed with respect to an
existing study. Amine coupling consisted of four successive
steps, in which the attachment of all six peptides was
achieved after activation of the carboxylic acids of the carb-
oxymethylated dextran matrix of the BIAcore CM5®
chip.[25]
With the six chips in hand, titrations were carried out
with trimers 17–20 and α- and β-cyclodextrin for compari-
son. In each of the 24 cases, a binding curve was obtained
from adding increasing concentrations of the cyclodextrin
ligand to the immobilized peptide and recording the maxi-
mum response at equilibrium. From these results, apparent
association constants were determined and they are shown
Six peptides were synthesized for this study with
the
structures
FGGGFGGGF,
FGGGYGGGF,
FGGGNalGGGF, FGGGWGGGF, FGGFGGF, and
GGGF, in which F, G, Y, and W are phenylalanine, glycine,
tyrosine, and tryptophan, respectively, and Nal is (S)-2-(2-
naphthylmethyl)glycine. The synthesis was conducted by
automated microwave-assisted solid-phase peptide synthesis
and subsequent purification was achieved by conventional
reverse-phase HPLC. The peptides were synthesized on
Rink amide resin by applying a 9-fluorenylmethyl (Fmoc)
protecting group strategy and O-(benzotriazol-1-yl)tetra-
methyluronium hexafluorophosphate (HBTU) activation
for coupling. Cleavage was achieved by treating the resin
with TFA, with triisopropylsilane (TIS) and water as scav-
engers (95:2.5:2.5 v/v). The yields ranged from 40 to 75%.
Binding Affinity Determination
Initially it was anticipated that isothermal titration calo- in Table 2. The results give rise to several observations: (1)
rimetry (ITC) could be used for experimental determination α-CD binds with a very different affinity to the different
of binding affinities between nonapeptides and cyclodex- peptides, and does not, as one might anticipate, have a sim-
trins, but it could not accomplished due to difficulties with ilar KA value to all compounds that reflects the typical
the solubility of the peptide samples. Therefore, an alterna- binding for a cyclodextrin to phenyl group. (2) β-CD fol-
tive methodology using surface plasmon resonance was lows the same trend as α-CD, but binds 1.2 to 1.4 times
used.[25] The surface plasmon resonance (SPR) technology more tightly with a single exception. (3) With a single excep-
follows the real-time formation and dissociation of bimolec- tion (compound 17), all compounds bind GGGF with sim-
ular complexes on a sensor surface. One of the two compo- ilar affinity. (4) Trimers 18 and 20 bind all hepta- and non-
nents in the binding study under consideration is immobi- apeptides more strongly than they bind GGGF; trimer 17
Table 2. KA values (in m–1) determined by SPR.
Cyclodextrin
GGGF
FGGFGGF
FGGGFGGGF
FGGGYGGGF
FGGGNalGGGF
FGGGWGGGF
α-CD
β-CD
17 (α-α-α)
18 (α-β-α)
19 (β-α-β)
20 (β-β-β)
1060
1320
3280
1150
720
186
288
3790
5940
1030
4660
441
593
1210
6210
1770
3430
960
13
162
6100
5900
1170
4290
524
751
5480
5790
834
1010
4690
5700
1000
5920
1510
5990
Eur. J. Org. Chem. 2011, 5279–5290
© 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjoc.org
5283

M. Bols et al.
FULL PAPER
behaves very similarly except that it binds GGGF unusually a lower affinity for most of the peptides compared with 17,
strongly and one of the nonapeptides poorly. (5) Trimer 19 18 and 20. The binding of 19 is mostly similar to binding
binds all peptides with the essentially the same affinity.
by a single cyclodextrin, which suggests that this compound
To explain these observations, which may appear confus- is only able to bind with one cyclodextrin in many cases.
ing, it is necessary to reflect on the properties and possible Yet 19 must uncoil the larger peptides apparently because
binding modes of the chip-bound peptides. First, binding all are bound, even if the affinity is lower than for the other
with the cyclodextrin trimers may involve one, two, and/or trimers (observation 5).
three cyclodextrins, as shown in Figure 2. The actual bind-
ing observed may therefore contain contributions from all
Conclusions
three binding modes. These different binding modes cannot
be discerned by SPR. Second, the larger peptides are very
poorly soluble in water, so they are presumably also poorly
solvated in chip-attached mode and may coil up upon them-
selves. It is possible that a strong ligand can uncoil the pept-
ide. By this “coiling” hypothesis, we can explain observa-
tions (1) and (2). The high affinity of α- and β-CD for
GGGF (KA ≈ 103 m–1) shows that this peptide has a single
exposed phenyl group. (For comparison,[26] the KA values
of phenyl alanine and phenyl alanine amide are 101–
102 m–1). The hepta- and nonapeptides are “coiled” and
therefore do not have as readily accessible a phenyl group.
A series of cyclodextrin trimers were successfully pre-
pared by a modified click chemistry methodology for com-
bining benzylated cyclodextrin building blocks. The synthe-
sis of trimers was achieved within 6 synthetic steps and re-
sulted in overall yields of the trimers ranging from 34 (β-β-
β) to 52% (α-α-α), starting from unmodified α- or β-cyclo-
dextrin. The copper-catalyzed Huisgen reaction was versa-
tile for all types of alkynes, including small organic alkynes
and bulky cyclodextrin–alkyne derivatives.
The cyclodextrin trimers were subsequently investigated
for molecular recognition of peptides. The molecular re-
cognition of nonapeptides by cyclodextrin trimers evaluated
in this study was unselective, yet better binding affinities
were obtained for the cyclodextrin trimers than for the un-
modified cyclodextrins. Trimer recognition of nonapeptides
was experimentally determined to be superior to the re-
cognition of a heptapeptide. The experimental results, how-
ever, suggested poor selectivity of nonapeptides by the
cyclodextrin trimers and the recognition of nonapeptides
was considered to be unselective. The experimental pro-
cedure involved immobilization of peptides, which could
have contributed to incomplete overlap of the peptides due
to steric hindrance at the surface.
Experimental Section
General: All reactions were carried out in flame-dried glassware
under a nitrogen atmosphere unless otherwise stated. Solvents were
distilled and/or dried according to standard procedures and the
reactions were monitored either by TLC analysis or by MALDI-
TOF MS. Flash chromatography was carried out with Merck silica
gel 60 (230–400 mesh) as the stationary phase. TLC (Merck silica
gel 60, F254) was visualized by UV light or the use of a Ce-Mol
solution [cerium(IV) sulfate (10 g) and ammoniummolybdate (15 g)
dissolved in 10% aqueous sulfuric acid (1000 mL)] with subsequent
heating. 1H and 13C NMR spectra were recorded on a Varian Gem-
Figure 2. Different possible binding modes of cyclodextrin trimers
17–20 (green) with surface-attached peptides (backbone yellow,
aromatic side chains red) in the BIAcore. The measured binding
constant may be based on contributions for all three binding
modes.
This is also why all cyclodextrin trimers bind GGGF
with essentially the same affinity (observation 3). The
phenyl group is exposed and the cyclodextrin trimer can
only bind with one cyclodextrin, which means that binding
is similar to that of a simple cyclodextrin.
1
ini 400 spectrometer (400 MHz for H and 100 MHz for 13C). The
chemical shifts are reported in ppm downfield to TMS (δ = 0 ppm)
1
for H NMR spectroscopy and relative to the solvent peak for 13C
The higher binding of trimers 18 and 20, and to some
extent 17 (observation 4), to hepta- and nonapeptides (KA
≈ 6ϫ103 m–1) shows that these compounds can uncoil the
peptide and bind to two or more aryl groups (Figure 2).
There appears to be little preference for hepta- versus nona-
peptide or between different nonapeptides, showing very
poor selectivity both for the distance between aryl groups
or for the identity. This suggests that this type of binding
predominantly involves two cyclodextrin···aryl interactions
NMR spectroscopy. Mass spectra were obtained by using a Micro-
mass LC-TOF mass spectrometer and MALDI-TOF spectra were
recorded on a Bruker Daltonics mass spectrometer using a 2,5-
dihydroxybenzoic acid (DHBA)-based matrix. Reverse-phase (RP)
HPLC was performed on an Agilent 1100 series instrument using
a Vydac preparative RP column (C8, 208TP-510), unless otherwise
stated.
Hexadeca-O-benzyl-6A,6D-dideoxy-6A,6D-diiodo-α-cyclodextrin (1α):
2A–F,3A–F,6B,C,E,F-Hexadeca-O-benzyl-α-cyclodextrin[27]
(2.00 g,
(Figure 2, middle). Compound 19 is unusual because it has 0.82 mmol), PPh3 (1.35 g, 5.14 mmol), and imidazole (0.69) were
5284
www.eurjoc.org
© 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Org. Chem. 2011, 5279–5290

Recognition of Peptides by Cyclodextrin Trimers
added to freshly distilled toluene (75 mL) and heated to 75°C.
Then I2 (1.30 g, 5.14 mmol) was added and the resulting mixture
was left stirring overnight at 75°C under N2. An equal volume of a
saturated aqueous solution of NaHCO3 was added to the reaction
mixture and it was stirred for 5 min. The organic layer was sepa-
rated, washed with a saturated aqueous solution of Na2S2O3, di-
luted with EtOAc (150 mL), and washed with water. The organic
layer was dried with Na2SO4, filtered, concentrated in vacuo, and
the resulting oil was purified by flash column chromatography
(EtOAc/pentane; 1:5 Ǟ 1:3), resulting in a white foam (1.69 g,
0.64 mmol, 78%) ; m.p. 59.1–66.8°C. 1H NMR (CDCl3): δ = 7.27–
7.22 (m, 40 H, Ph), 7.15–7.11 (m, 40 H, Ph), 5.20 (d, J = 10.8 Hz,
2 H), 5.15 (d, J = 14.4 Hz, 2 H), 5.03 (d, J = 11.2 Hz, 2 H), 4.98
(d, J = 3.2 Hz, 2 H), 4.92 (d, J = 3.2 Hz, 2 H), 4.86 (d, J = 10.8 Hz,
2 H), 4.80 (dd, J = 11.2, 10.8 Hz, 4 H), 4.52 (d, J = 3.6 Hz, 2 H),
2A–F,3A–F,6B–F-heptadeca-O-benzyl-α-cyclodextrin (1.00 g,
0.40 mmol) in DMF (10 mL) at 0°C and stirred for 15 min. Propar-
gyl bromide (0.16 mL, 0.88 mmol) was added dropwise, the reac-
tion mixture was allowed to reach room temperature, and the pro-
gress of the reaction was monitored by TLC. The reaction was
quenched by adding MeOH (5 mL), diluted with water, and ex-
tracted with Et2O. The combined organic phases were washed with
brine, dried with Na2SO4, and concentrated in vacuo to give an
oil, which was purified by flash column chromatography (EtOAc/
pentane; 1:4 Ǟ 1:3), resulting in a white foam (0.94 g, 0.37 mmol,
1
93%); m.p. 54.5–62.9°C. H NMR (CDCl3): δ = 7.27–7.14 (m, 85
H, Ph), 5.20 (dt, J = 3.2, 11.2 Hz, 6 H), 5.14–5.11 (m, 4 H), 5.09
(d, J = 3.2 Hz, 1 H), 5.06 (d, J = 3.2 Hz, 1 H), 4.89 (d, J = 11.2 Hz,
6 H), 4.54 (t, J = 12.8 Hz, 2 H), 4.51–4.44 (m, 15 H), 4.43 (d, J =
4.0 Hz, 2 H), 4.39 (dd, J = 1.6, 2.8 Hz, 2 H), 4.36 (dd, J = 2.4,
4.49 (d, J = 3.2 Hz, 2 H), 4.42–4.39 (m, 8 H), 4.37 (d, J = 4.4 Hz, 3.2 Hz, 1 H), 4.19 (d, J = 1.6 Hz, 1 H), 4.17–4.12 (m, 6 H), 4.10
2 H), 4.13–4.06 (m, 6 H), 3.97 (d, J = 4.8 Hz, 4 H), 3.89 (d, J = (d, J = 4.0 Hz, 1 H), 4.07 (d, J = 3.6 Hz, 2 H), 4.06–4.02 (m, 7 H),
7.2 Hz, 4 H), 3.73 (d, J = 10.0 Hz, 2 H), 3.68–3.64 (m, 2 H), 3.58
(d, J = 10.8 Hz, 2 H), 3.54 (d, J = 8.8 Hz, 2 H), 3.50–3.48 (m, 4
H), 3.46 (t, J = 3.6 Hz, 2 H), 3.43 (d, J = 3.6 Hz, 2 H), 3.36 (dd,
J = 3.2, 10.0 Hz, 2 H) ppm. 13C NMR (CDCl3): δ = 139.5, 139.4,
139.3, 138.5, 138.3, 138.2, 138.19, 138.1, 128.5, 128.4, 128.3,
4.00 (d, J = 2.4 Hz, 2 H), 3.96–3.94 (m, 8 H), 3.65 (dd, J = 10.4,
11.6 Hz, 2 H), 3.57 (d, J = 10.8 Hz, 2 H), 3.55 (d, J = 7.6 Hz, 1
H), 3.52 (dd, J = 3.2, 6.4 Hz, 4 H), 3.50 (d, J = 3.2 Hz, 2 H), 3.46
(dd, J = 3.2, 9.6 Hz, 2 H), 2.23 (t, J = 2.4 Hz, 1 H, CϵC-H) ppm.
13C NMR (CDCl3): δ = 139.3, 138.3, 138.2, 138.19, 138.15, 138.12,
128.28, 128.26, 128.23, 128.1, 127.96, 127.9, 127.88, 127.82, 127.7, 128.4, 128.39, 128.3, 128.2, 128.1, 127.9, 127.8, 127.74, 127.70,
127.6, 127.5, 127.3, 127.1, 127.0, 99.5, 98.5, 84.5, 81.0, 80.8, 80.7,
80.3, 80.2, 79.4, 78.9, 78.7, 75.9, 75.6, 75.4, 73.7, 73.6, 73.0, 72.9,
127.66, 127.61, 127.59, 127.54, 127.3, 127.2, 127.1, 126.9, 98.7,
98.62, 98.59, 98.53, 80.9, 80.8, 79.8, 79.4, 79.3, 79.2, 79.1, 79.06,
72.7, 72.0, 71.4, 70.4, 69.6, 69.3 ppm. HRMS (MALDI-TOF): m/z 79.00, 78.9, 75.6, 75.5, 75.4, 74.7, 73.4, 73.3, 72.8, 72.7, 72.6, 71.5,
calcd. for C148H154I2O28 [M + Na]+ 2655.8613; found 2654.6694;
calcd. for [M + K]+ 2671.8353; found 2670.6228. Rf 0.77 (EtOAc/
pentane; 1:3.5).
71.4, 69.1, 69.0, 58.5 ppm. HRMS (MALDI-TOF): m/z calcd. for
C
158H164O30 [M + Na]+ 2564.1205; found 2564.0170; calcd. for
[M + K]+ 2580.0995; found 2580.3392. Rf 0.66 (EtOAc/pentane;
1:3.5).
6A,6D-Diazido-hexadeca-O-benzyl-6A,6D-dideoxy-α-cyclodextrin
(2α):[12] Compound 1α (1.00 g, 0.38 mmol) was stirred with NaN3
(0.124 g, 1.90 mmol) in DMF (20 mL) at 75°C overnight under N2.
Then a saturated aqueous solution of NaCl (100 mL) and EtOAc
(150 mL) was added to the solution and the resulting mixture was
stirred for 5 min. The organic layer was separated and the water
phase was extracted with EtOAc. The combined organic phases
were washed with water, dried with Na2SO4, filtered, and concen-
trated in vacuo to give an oil, which subsequently was purified by
flash column chromatography (EtOAc/pentane; 1:4), resulting in
2α as a white foam (0.84 g, 0.34 mmol, 89%); m.p. 60.5–64.9°C.
1H NMR (CDCl3): δ = 7.27–7.22 (m, 40 H, Ph), 7.17–7.11 (m, 40
H, Ph), 5.24 (d, J = 10.8 Hz, 2 H), 5.16 (d, J = 3.2 Hz, 2 H), 5.12
(d, J = 10.8 Hz, 2 H), 4.98 (d, J = 11.2 Hz, 2 H), 4.93 (dd, J = 2.8,
3.2 Hz, 4 H), 4.85 (d, J = 10.8 Hz, 2 H), 4.82 (d, J = 7.2 Hz, 2 H),
4.79 (d, J = 7.2 Hz, 2 H), 4.58 (d, J = 12.4 Hz, 2 H), 4.49 (d, J =
12.0 Hz, 2 H), 4.44 (d, J = 3.2 Hz, 2 H), 4.40 (d, J = 15.6 Hz, 6
H), 4.36 (d, J = 12.0 Hz, 2 H) 4.12 (d, J = 8.4 Hz, 2 H), 4.09 (d, J
= 11.6 Hz, 2 H), 4.07 (d, J = 8.0 Hz, 2 H), 4.01 (d, J = 6.8 Hz, 2
H), 3.99 (d, J = 10.4 Hz, 2 H), 3.95 (d, J = 8.8 Hz, 6 H), 3.88 (d,
J = 9.6 Hz, 2 H), 3.82 (dd, J = 2.8, 9.6 Hz, 2 H), 3.68 (dd, J = 8.4,
9.6 Hz, 4 H), 3.58 (d, J = 10.4 Hz, 2 H), 3.48 (d, J = 3.6 Hz, 2 H),
3.44 (dd, J = 3.2, 10.0 Hz, 2 H), 3.38 (dd, J = 3.2, 9.6 Hz, 2 H)
ppm. 13C NMR (CDCl3): δ = 139.41, 139.39, 139.36, 138.6, 138.4,
138.3, 138.18, 138.15, 128.49, 128.48, 128.43, 128.36, 128.3, 128.2,
128.1, 128.0, 127.83, 127.77, 127.69, 127.67, 127.65, 127.6, 127.52,
127.48, 127.15, 127.11, 127.05, 126.98, 99.1, 98.9, 98.4, 80.97,
80.92, 80.8, 80.5, 80.2, 79.7, 79.4, 79.2, 78.5, 75.98, 75.85, 75.2,
73.6, 73.56, 73.2, 73.0, 72.7, 72.0, 71.8, 70.9, 69.6, 69.0, 52.4 ppm.
HRMS (MALDI-TOF): m/z calcd. for C148H154N6O28 [M + Na]+
2486.0709; found 2486.7663; calcd. for [M + K]+ 2502.0448; found
2502.9878. Rf 0.77 (EtOAc/pentane; 1:3.5).
6A,6D-Diazido-nonadeca-O-benzyl-6A,6D-dideoxy-β-cyclodextrin
(2β): Compound 1β (2.00 g, 0.65 mmol) was stirred with NaN3
(0.21 g, 3.26 mmol) in DMF (40 mL) at 75°C overnight under N2.
A saturated aqueous solution of NaCl (200 mL) and EtOAc
(250 mL) was added to the solution and the resulting mixture was
stirred for 5 min. The organic layer was separated and the water
phase was extracted with EtOAc. The combined organic phases
were washed with water, dried with Na2SO4, filtered, and concen-
trated in vacuo to give an oil, which subsequently was purified by
flash column chromatography (EtOAc/pentane; 1:4), resulting in a
1
white foam (1.67 g, 0.58 mmol, 89%); m.p. 54.3–60.5°C. H NMR
(CDCl3): δ = 7.25–7.10 (m, 95 H), 5.27 (d, J = 4.0 Hz, 1 H), 5.16
(d, J = 3.6 Hz, 1 H), 5.13 (d, J = 3.6 Hz, 2 H), 5.10 (t, J = 4.0 Hz,
2 H), 5.07–5.05 (m, 2 H), 5.04 (d, J = 3.6 Hz, 1 H), 4.99 (d, J =
3.6 Hz, 1 H), 4.96 (d, J = 11.2 Hz, 1 H), 4.82 (d, J = 11.2 Hz, 1
H), 4.77 (d, J = 11.6 Hz, 1 H), 4.73–4.69 (m, 7 H), 4.57 (dd, J =
9.6, 11.6 Hz, 2 H), 4.50–4.38 (m, 19 H), 4.10–3.88 (m, 25 H), 3.67
(d, J = 8.4 Hz, 2 H), 3.62 (d, J = 10.4 Hz, 2 H), 3.59 (d, J =
10.4 Hz, 4 H), 3.54 (dd, J = 2.8, 10.4 Hz, 2 H), 3.50 (dd, J = 3.6,
8.8 Hz, 2 H), 3.47–3.43 (m, 5 H), 3.39 (dt, J = 3.2, 8.8 Hz, 4 H)
ppm. 13C NMR (CDCl3): δ = 139.43, 139.35, 139.22, 139.18,
139.15, 138.6, 138.5, 138.38, 138.35, 138.29, 138.25, 138.2, 138.1,
128.7, 128.64, 128.60, 128.48, 128.46, 128.4, 128.34, 128.32, 128.29,
128.18, 128.16, 128.09, 128.05, 128.01, 127.95, 127.9, 127.8, 127.7,
127.62, 127.59, 127.55, 127.3, 127.2, 127.12, 127.08, 127.0, 98.9,
98.74, 98.68, 98.66, 98.4, 98.2, 98.0, 81.0, 80.90, 80.85, 80.4, 80.0,
79.8, 79.5, 79.41, 79.39, 79.36, 79.0, 78.90, 78.88, 78.8, 78.2, 77.8,
76.0, 75.9, 75.8, 75.3, 75.08, 75.05, 73.5, 73.4, 73.02, 72.99, 72.97,
72.8, 72.7, 71.9, 71.8, 71.7, 71.6, 70.92, 70.86, 69.62, 69.57, 69.4,
69.10, 69.06, 52.4 ppm. HRMS (MALDI-TOF): m/z calcd. for
C175H182N6O33 [M + Na]+ 2918.2646; found 2918.4578; calcd. for
[M + K]+ 2934.2385, found 2934.7091. Rf 0.66 (EtOAc/pentane;
1:3).
Heptadeca-O-benzyl-6-O-propargyl-α-cyclodextrin (4α): NaH
(38 mg, 0.96 mmol) in a solid portion was added to a solution of
Eur. J. Org. Chem. 2011, 5279–5290
© 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjoc.org
5285

M. Bols et al.
FULL PAPER
Icosa-O-benzyl-6-O-propargyl-β-cyclodextrin (4β): NaH (0.016 g, (m, 20 H, 3-H, 4-H, 6-H), 3.58–3.37 (m, 14 H, 5-H, 5ϫ 2-H), 3.19–
0.41 mmol) in a solid portion was added to a solution of 3β (0.50 g, 3.13 (dd, 2 H, 2ϫ 2-H), 1.29 (k, 6 H, 6ϫ SiCHMe), 1.06 (d, 36
0.17 mmol) in DMF (4.0 mL) at 0 °C and stirred for 15 min. Pro- H, 12ϫ CH3CH) ppm. 13C NMR (75 MHz, CDCl3): δ = 139.6,
pargyl bromide (0.02 mL, 0.19 mmol) was added dropwise, the re-
action mixture was allowed to reach room temperature, and the
progress of the reaction was monitored by TLC. The reaction was
quenched by adding MeOH (5 mL), diluted with water, extracted
with Et2O, the combined organic phases were washed with brine,
dried with Na2SO4, concentrated in vacuo to give an oil, which was
purified by flash column chromatography (EtOAc/pentane; 1:4 Ǟ
1:3) resulting in a white foam (0.44 g, 0.15 mmol, 88%); m.p. 55.1–
62.2°C. 1H NMR (CDCl3): δ = 7.24–7.15 (m, 10 0 H, Ph), 5.25 (d,
J = 3.6 Hz, 1 H), 5.21 (d, J = 3.6 Hz, 1 H), 5.17 (d, J = 3.6 Hz, 1
H), 5.15 (d, J = 3.6 Hz, 1 H), 5.14 (d, J = 4.0 Hz, 1 H), 5.11 (t, J
= 3.6 Hz, 3 H), 5.07 (d, J = 10.8 Hz, 4 H), 5.03 (d, J = 8.4 Hz, 1
H), 4.98 (d, J = 6.0 Hz, 1 H), 4.76 (d, J = 11.2 Hz, 7 H), 4.55 (d,
J = 12.4 Hz, 1 H), 4.49–4.46 (m, 12 H), 4.43 (d, J = 6.4 Hz, 5 H),
4.41 (d, J = 2.4 Hz, 2 H), 4.38 (d, J = 8.0 Hz, 7 H), 4.05–3.93 (m,
32 H), 3.86 (d, J = 8.4 Hz, 1 H), 3.81 (d, J = 8.8 Hz, 1 H), 3.68 (d,
J = 10.0 Hz, 1 H), 3.62 (d, J = 10.4 Hz, 1 H), 3.55 (d, J = 10.4 Hz,
5 H), 3.50 (d, J = 9.2 Hz, 2 H), 3.48 (d, J = 4.0 Hz, 3 H), 3.45 (dd,
J = 2.0, 3.6 Hz, 3 H), 3.42 (d, J = 3.6 Hz, 1 H), 2.19 (t, J = 2.0 Hz,
1 H) ppm. 13C NMR (CDCl3): δ = 139.53, 139.50, 138.61, 138.55,
138.46, 129.10, 129.04, 128.99, 128.97, 128.94, 128.90, 128.79,
128.72, 128.53, 128.38, 128.205, 128.13, 128.09, 128.05, 127.99,
127.80, 127.77, 127.66, 127.56, 127.15, 98.87, 98.82, 98.67, 98.62,
98.56, 98.52, 81.23, 81.12, 81.12, 81.06, 81.00, 79.94, 79.51, 79.30,
79.19, 79.14, 79.02, 78.98, 78.89, 78.82, 78.13, 75.77, 75.68, 75.65,
75.48, 75.39, 75.03, 73.49, 72.98, 72.91, 72.87, 72.77, 71.80, 71.73,
71.59, 71.23, 71.21, 69.50, 69.42, 69.33, 69.09, 69.07, 58.64 ppm.
HRMS (MALDI-TOF): m/z calcd. for C185H192O35 [M + Na]+
2996.3142; found 2995.9890; calcd. for [M + K]+ 3012.2881; found
3012.2943. Rf 0.68 (EtOAc/pentane; 1:3).
2A–F,3A–F,6B,C,E,F-Hexadecakis-O-benzyl-6A,D-di-C-(4-phenyl-1H-
1,2,3-triazol-1-yl)-β-cyclodextrin (5): Derivative 2β (150 mg,
0.052 mmol) was dissolved in MeCN (5 mL) and EtOAc (2.5 mL),
together with CuI (2 equiv.), DIPEA (3 equiv.), and phenylacetyl-
ene (21.2 mg, 0.207 mmol, 4 equiv.) were added and the resulting
mixture was stirred overnight at 40 °C. The reaction mixture was
poured into a saturated aqueous solution of NH4Cl (10 mL) and
water (20 mL). The organic phase was separated and the aqueous
phase was extracted three times with EtOAc. The combined or-
ganic phases were washed with brine, dried with MgSO4, filtered,
and the solvents were evaporated. The resulting oil was purified by
flash chromatography (EtOAc/pentane; 1:4 Ǟ 1:2), which gave 5
as a clear solid residue (129.4 mg, 81 %). 1H NMR (300 MHz,
CDCl3): δ = 7.49 (s, 2 H, triazole H), 7.25–6.93 (m, 105 H, Ph),
5.83 (d, J = 3.2 Hz, 1 H), 5.55 (d, J = 3.15 Hz, 1 H), 5.32–5.12 (m,
J = 11.2 Hz, 5 H), 5.08–4.92 (m, 5 H), 4.88–4.64 (m, 7 H, CHPh),
4.50–4.32 (m, 26 H, CHPh), 4.24–3.82 (m, 28 H, 7ϫ3-H, 7ϫ4-H,
7ϫ5-H, 7ϫ6-H), 3.70–3.25 (m, 12 H, 7ϫ6-H, 5ϫ2-H), 2.93–2.89
(d, 2 H, 2 ϫ 2-H) ppm. HRMS (MALDI-TOF): m/z calcd. for
C191H194N6O33 [M + Na]+ 2690.1648; found 2690.0348; calcd. for
[M + K]+ 2706.1387; found 2706.2119.
139.2, 138.6, 138.5, 138.4, 138.2 (Cquat. triazole, Cipso), 128.7, 128.5,
128.4, 128.2, 128.1, 127.8, 127.7 (Ar), 99.6, 98.6, 98.5, 98.3, 98.1,
97.6 (1-C), 80.8, 80.6, 77.5, 73.6, 73.5, 72.8, 72.7, 69.5, 18.9 (CH3),
11.3 [C(CH3)2] ppm. HRMS (MALDI-TOF): m/z calcd. for
C
197H226N6O33Si2 [M + 2H]+ 3264.108; found 3264.41.
Nonadecakis-O-benzyl-6A,6D-di-C-[4-(methyl 2,3,4-tri-O-benzyl-α-
-glucosidyl)-1H-1,2,3-triazol-1-yl]-β-cyclodextrin (8): Diazide 2β
D
(202 mg, 70 μmol) and methyl 2,3,4-tri-O-benzyl-6-O-propargyl α-
d-glucoside (326 mg, 10 equiv., 0.70 mmol) were dissolved in
MeCN/EtOAc (2:1, 2 mL). CuI (27 mg, 2 equiv., 140 μmol) and di-
isopropylethylamine (5 equiv., 60 μL) were added and the reaction
was mixed in a closed flask for 2 weeks. The yellow crude reaction
mixture was diluted with EtOAc, washed with a saturated aqueous
solution of NH4Cl, HCl (1 m), a saturated aqueous solution of
NaHCO3, and brine. After concentration in vacuo, the crude prod-
uct was purified by flash chromatography (EtOAc/pentane 1:5 to
1:2) to give the desired product in quantitative yield. The Glaser
coupled byproduct (alkyne–alkyne coupling) was isolated as a side
product. 1H NMR (500 MHz, CDCl3): δ = 7.51–6.98 (m, 127 H,
Ar, triazole-H), 5.61 (d, J = 3.2 Hz, 1 H, 1-H), 5.53 (d, J = 3.2 Hz,
1 H, 1-H), 5.36–5.30 (m, 3 H, CHPh, 1-H), 5.25 (d, J = 3.0 Hz, 1
H, 1-H), 5.19 (d, J = 3.0 Hz, 1 H, 1-H), 5.17–5.10 (m, 4 H), 5.07
(d, J = 10.3 Hz, 1 H, CHPh), 5.02–4.98 (m, 3 H, 7 H, CHPh, 6-
H), 4.88 (t, J = 10.4 Hz, 8 H, 6-H), 4.84–4.76 (m, 8 H, CHPh, 3-
H), 4.75–4.71 (m, 4 H, CHPh), 4.67 (d, J = 12.2 Hz, 6 H, CHPh),
4.64–4.35 (m, 42 H, 1-H, 6-H), 4.27–4.13 (m, 11 H, CHPh, 2-H),
4.09–3.87 (m, 24 H, CHPh, 3-H, 6-H), 3.87–3.70 (m, 10 H, 4-H),
3.65 –3.41 (m, 18 H, 5-H, 4-H), 3.37 (s, 6 H), 3.34–3.28 (m, 2 H, 2-
H), 3.26 (dd, J = 9.7, 3.4 Hz, 1 H, 2-H) ppm. 13C NMR (126 MHz,
CDCl3): δ = 144.1 (Cquat triazole), 138.9, 138.30, 138.26, 138.22,
138.17, 138.1 (Cipso, Ar), 128.50, 128.48, 128.46, 128.43, 128.38,
128.33, 128.28, 128.25, 128.20, 128.16, 128.12, 128.09, 128.07,
128.02, 127.98, 127.94, 127.91, 127.87, 127.79, 127.75, 127.71,
127.63, 127.58, 127.52, 127.4, 127.3, 127.2, 127.1, 127.0 (Ar), 125.2
(CH triazole), 99.1, 98.4, 98.2, 97.7 (C-1), 82.1 (C-3), 80.4 (C-4),
79.9 (C-5), 78.9, 77.3, 77.1, 76.9, 75.8, 75.1 (CH2Ph), 73.6, 73.42,
73.38, 73.1, 72.8, 72.54, 72.47, 72.3, 71.9, 71.5 (C-3, C-2), 70.1 (C-
4, C-5), 69.2 (C-6), 65.0 (C-6), 55.3 (CH3), 50.7 (C-6) ppm. MS
(MALDI-TOF): m/z (%) = 3901.82 (100.0) [M]+.
1,6-Di-O-(methyl 2,3,4-tri-O-benzyl-6-yl-α-D-glucopyranosidyl)-2,4-
1
hexadiyne (9): H NMR (500 MHz, CDCl3): δ = 7.39–7.29 (m, 30
H, Ar), 5.02 (d, J = 10.8 Hz, 2 H, Bn) 4.86 (m, 4 H, Bn) 4.84 (d,
J = 12.0 Hz, 2 H, Bn), 4.69 (d, J = 12.1 Hz, 1 H, Bn), 4.65 (d, J =
10.8 Hz, 2 H, Bn), 4.63 (d, J = 3.3 Hz, 2 H, 1-H, 1 HЈ), 4.25 (d, J
= 16.6 Hz, 2 H, CH2-ϵ), 4.18 (d, J = 16.6 Hz, 2 H, CH2-ϵ), 4.01
(br. t, J = 9.3 Hz, 2 H, 3-H, 3Ј-H), 3.79 (m, 4 H, 6-H, 6Ј-H, 5-H,
5Ј-H), 3.63 (m, 4 H, 6-H, 6Ј-H, 4-H, 4Ј-H), 3.56 (m, 2 H, 2-H, 2Ј-
H), 3.40 (s, 6 H, Me) ppm. 13C NMR (126 MHz, CDCl3): δ =
139.0, 138.4, 138.3 (6 Cipso), 128.6–127.8 (30 C, Ar), 98.5 (2 C, C1,
C1Ј), 82.2 (2 C, C3, C3Ј), 79.9 (2 C, C2, C2Ј), 77.4 (2 C, C4, C4Ј),
75.9 (2 C, Bn), 75.4 (2 C, Bn), 75.3 (2 C, Bn), 73.6 (alkyn), 71.0
(alkyn), 69.9 (2 C, C5, C5Ј), 68.5 (2 C, C6, C6Ј), 59.2 (2 C, 2ϫ
CH2-ϵ), 55.4 (2 C, 2ϫ Me) ppm.
Nonadecakis-O-benzyl-6A,6D-di-C-[4-(triisopropylsilyl)-1H-1,2,3-tri-
azol-1-yl]-β-cyclodextrin (6): Following the same procedure as that
used for 5, derivative 2β (600 mg, 0.208 mmol) was coupled with
triisopropylsilyl acetylene (151.2 mg, 0.829 mmol) to obtain 6 as a
2A–G,3A–G,6B,C,E,F,G-Nonadecakis-O-benzyl-6A,D-di-O-propargyl-β-
cyclodextrin (10): tBuOK (0.122 g, 2.63 mmol) and propargyl bro-
1
clear solid residue (391 mg, 58%). H NMR (500 MHz, CDCl3): δ
= 7.48, 7.46 (s, 2 H, triazol H), 7.24–6.93 (m, 105 H, Ph), 5.73 (d, mide (0.44 mL, 4 mmol; dropwise) were added to a solution of β-
J = 3.4 Hz, 1 H, 1-H), 5.52 (d, J = 3.2 Hz, 1 H, 1-H), 5.49 (d, J =
3.5 Hz, 1 H, 1-H), 5.40 (d, J = 3.3 Hz, 1 H, 1-H), 5.21–4.97 (m, 12
cyclodextrin diol (0.751 g, 0.263 mmol) in DMF (5 mL) at –78 °C.
The reaction mixture was allowed to reach room temp. and the
H, CHPh, 3ϫ1-H), 4.84–4.19 (m, 35 H, CHPh, 6-H), 4.12–3.72 progress of the reaction was monitored by TLC. The reaction was
5286 Eur. J. Org. Chem. 2011, 5279–5290
www.eurjoc.org
© 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Recognition of Peptides by Cyclodextrin Trimers
quenched by adding water and was extracted with EtOAc. The
combined organic phases were washed with brine, dried with
Na2SO4, concentrated in vacuo, and purified by flash chromatog-
raphy (EtOAc/pentane; 1:4 Ǟ 1:3), resulting in 10 as a clear solid
residue (0.497 g, 0.17 mmol, 64%). 1H NMR (500 MHz, CDCl3):
δ = 7.33–7.13 (m, 95 H, Ph), 5.34 (d, J = 3.6 Hz, 1 H, 1-H), 5.28
(d, J = 3.2 Hz, 1 H, 1-H), 5.26 (d, J = 3.6 Hz, 1 H, 1-H), 5.20–5.10
(m, 7 H, CHPh), 5.04 (d, J = 11.6 Hz, 2 H, CHPh), 5.0 (d, J =
water (20 mL) and a saturated aqueous solution of NH4Cl (10 mL).
The aqueous layer was extracted three times with EtOAc, the com-
bined organic phases were washed with brine, dried with MgSO4,
and the solvent removed in vacuo. The resulting oil was purified
by flash column chromatography (EtOAc/pentane; 1:4 Ǟ 1:2) to
give 13 as a white foam (0.63 g, 0.08 mmol, 83 %); m.p. 81.6–
85.9°C. 1H NMR (CDCl3): δ = 7.30 (s, 2 H, triazole-H), 7.24–7.02
(m, 250 H), 5.76 (d, J = 3.2 Hz, 1 H), 5.55 (d, J = 3.2 Hz, 1 H),
11.2 Hz, 1 H, CHPh), 4.85–4.80 (m, 7 H), 4.57–4.46 (m, 28 H), 5.37 (d, J = 10.0 Hz, 1 H), 5.27 (dd, J = 2.0, 10.8 Hz, 2 H), 5.02
4.09–3.94 (m, 28 H), 3.77–3.68 (m, 7 H), 3.63 (d, J = 10.8 Hz, 2
H), 3.58 (d, J = 3.2 Hz, 2 H), 3.55–3.48 (m, 4 H), 2.30 (t, J =
(t, J = 2.8 Hz, 2 H), 5.16–5.14 (m, 6 H), 5.11 (d, J = 3.2 Hz, 2 H),
5.09–5.04 (m, 6 H), 4.94 (d, J = 2.8 Hz, 1 H), 4.89 (s, 2 H), 4.87
2.4 Hz, 1 H, CH), 2.28 (t, J = 2.4 Hz, 1 H, CH) ppm. 13C NMR (d, J = 2.0 Hz, 8 H), 4.85–4.79 (m, 18 H), 4.75 (d, J = 10.0 Hz, 2
(126 MHz, CDCl3): δ = 139.4–138.2 (Cipso, Ar), 128.40, 128.37,
H), 4.70 (d, J = 12.0 Hz, 2 H), 4.60 (dd, J = 4.8, 12.0 Hz, 2 H),
128.23, 128.22, 128.07, 128.04, 127.99, 127.94, 127.89, 127.86, 4.54 (d, J = 6.0 Hz, 2 H), 4.49 (d, J = 2.0 Hz, 6 H), 4.46–4.40 (m,
127.70, 127.66, 127.62, 127.60, 127.57, 127.50, 127.46, 127.41, 30 H), 4.38 (d, J = 2.8 Hz, 2 H), 4.36 (d, J = 3.6 Hz, 6 H), 4.32
127.23, 127.20, 127.10, 127.01 (Ar), 98.6, 98.5, 98.4, 98.3 (C-1),
80.9, 79.77 (Cquat), 79.75, 79.1, 78.9, 78.8, 77.4, 77.1 (CH), 76.9,
(dd, J = 3.2, 4.4 Hz, 6 H), 4.29 (d, J = 3.2 Hz, 2 H), 4.20 (d, J =
11.6 Hz, 2 H), 4.15–4.05 (m, 50 H), 3.95 (t, J = 11.6 Hz, 20 H),
75.8, 75.7, 75.6, 75.3, 75.1, 75.0, 74.9 (CH2Ph), 73.3, 73.3, 73.0, 3.88 (d, J = 7.6 Hz, 6 H), 3.69 (d, J = 10.8 Hz, 2 H), 3.60–3.56 (m,
72.91, 72.85, 72.78, 72.75, 72.72, 72.6, 71.5, 71.0, 69.3, 69.2, 68.9, 6 H), 3.52–3.44 (m, 30 H), 3.20 (dd, J = 3.6, 9.6 Hz, 4 H) ppm.
58.53 (C-6), 58.48 (C-6) ppm. HRMS (MALDI-TOF): m/z calcd. 13C NMR (CDCl3): δ = 139.6, 139.3, 139.2, 139.0, 138.7, 138.6,
for C181H188O35 [M + Na+] 2944.2829; found 2943.9462; calcd. for
[M + K+] 2960.2568; found 2960.5916.
6A,D-Di-O-{methyl-4-[N-(methyl 6-deoxy-2,3,4-tri-O-benzyl-α-
D-
138.52, 138.48, 138.46, 130.0, 129.2, 128.61, 128.56, 128.49, 128.44,
128.38, 128.35, 128.33, 128.29, 128.2, 128.1, 127.91, 127.89, 127.85,
127.5, 127.4, 127.3, 98.7, 81.4, 81.19, 81.16, 81.07, 81.06, 81.02,
76.1, 74.9, 74.7, 74.3, 74.2, 73.9, 73.66, 73.63, 73.59, 73.65, 73.5,
73.2, 72.9, 72.79, 72.74, 72.72, 72.69, 72.64, 72.58, 72.56, 72.51,
71.90, 71.87, 71.85, 71.65, 71.61, 71.51, 71.48, 69.3, 69.2 ppm.
HRMS (MALDI-TOF): m/z calcd. for C414H482N6O88 [M + Na]+
7568.3324; found 7568.6194; calcd. for [M + K]+ 7584.3063; found
7583.9533. Rf 0.29 (EtOAc/pentane; 1:2.5).
glucopyranosidyl)]triazolyl}-2A–G,3A–G,6B,C,E,F,G-nonadecakis-O-
benzyl-β-cyclodextrin (12): Compound 10 (101 mg, 34.6 μmol) and
the methyl 6-azido-2,3,4-tri-O-benzyl-6-deoxy-α-d-glucopyrano-
side (169 mg, 10 equiv., 0.35 mmol) were dissolved in MeCN/
EtOAc (2:1, 4 mL) and CuI (13 mg, 2 equiv., 69 μmol) was added
together with diisopropylethylamine (5 equiv., 30 μL). The reaction
was left at room temp. for 2 weeks to give a dark solution, which
was diluted (EtOAc) and washed with NH4Cl (satd. solution) and
brine, followed by drying of the organic layer, and concentration
in vacuo to give a crude product, which was purified by flash
chromatography (EtOAc/pentane; 1:4 Ǟ 1:2) to give 12 as a clear
solid residue (79 mg, 58%). 1H NMR (500 MHz, CDCl3): δ = 7.51–
6.98 (m, 127 H, Ar, triazole-H), 5.33 (d, J = 3.4 Hz, 1 H, 1-H),
5.26 (d, J = 3.5 Hz, 2 H, 1-H), 5.24 (d, J = 3.6 Hz, 1 H, 1-H), 5.21
(d, J = 10.4 Hz, 1 H, CHPh), 5.19 (d, J = 10.6 Hz, 1 H, CHPh),
5.17 (d, J = 3.6 Hz, 2 H, 1-H), 5.10–5.00 (m, 8 H, 2ϫ 1-H, CHPh),
4.91 (dd, J = 10.8, 1.1 Hz, 4 H, CHPh), 4.87- 4.66 (m, 14 H,
CHPh), 4.66–4.58 (m, 4 H, CHPh), 4.58–4.33 (m, 30 H, 2ϫ 1-H,
6-H), 4.25–3.76 (m, 30 H, 3-H, 5-H), 3.57 (dt, J = 13.7, 6.7 Hz, 6
H), 3.52–3.44 (m, 8 H, 2-H), 3.41 (dd, J = 9.5, 3.4 Hz, 2 H, 2-
H), 3.16 (2s, 6 H, CH3), 3.14–3.10 (m, 3 H, 4-H) ppm. 13C NMR
(126 MHz, CDCl3): δ = 144.8 (Cquat triazole), 139.32, 139.30,
139.26, 139.24, 139.19, 138.55, 138.51, 138.48, 138.41, 138.34,
6A,6D-Dideoxy-6A,6D-bis{N-[4Ј-(2A–F 3A–F,6B–F-heptadeca-O-benzyl-
,
α-cyclodextrin-6A-oxy)-1,2,3-triazolyl-4-methyl]}nonadeca-O-
benzyl-β-cyclodextrin (14): To a solution of alkyn 4α (1.35 g,
0.53 mmol) and diazide 2β (0.73 g, 0.25 mmol) in 4 mL of MeCN
and 4 mL of EtOAc, CuI (97 mg, 0.51 mmol) and DIPEA
(0.132 mL, 0.76 mmol) was added and the resulting mixture was
left stirring until TLC showed no diazide. The reaction was then
quenched by adding 20 mL of water and 10 mL of satd. aq. NH4Cl.
The aqueous layer was extracted thrice with EtOAc, the combined
organic phases were washed with brine, dried with MgSO4 and the
solvent removed in vacuo. The resulting oil was purified by flash
column chromatography (EtOAc/pentane; 1:4 Ǟ 1:2) giving 1.61 g
of 14 (0.20 mmol, 81%) as a white foam. The compound was a
10:9 mixture of isomers. Mp 83.0–87.7 °C. 1H NMR (CDCl3): δ =
7.31 (s, 4 H, triazole-H), 7.42–6.98 (m, 504 H, -Ph), 5.67 (d, J =
3.2 Hz, 1 H), 5.57 (d, J = 2.4 Hz, 1 H), 5.43 (d, J = 2.8 Hz, 0.90
H), 5.27 (dd, J = 6.4, 10.8 Hz, 7 H), 5.22 (d, J = 2.4 Hz, 0.90 H),
5.20–5.18 (m, 11 H), 5.14–5.05 (m, 30 H), 5.02 (d, J = 3.2 Hz, 2
H), 4.92–4.89 (m, 6 H), 4.85 (t, J = 9.2 Hz, 21 H), 4.75 (d, J =
10.0 Hz, 4 H), 4.70 (d, J = 11.2 Hz, 4 H), 4.66 (s, 3 H), 4.61 (dd,
J = 1.2, 10.4 Hz, 2 H), 4.57–4.55 (m, 2 H), 4.52 (t, J = 4.4 Hz, 6
H), 4.48–4.44 (m, 40 H), 4.42 (d, J = 5.2 Hz, 25 H), 4.39–4.33 (m,
39 H), 4.28 (d, J = 12.0 Hz, 11 H), 4.22 (d, J = 12.0 Hz, 7 H),
4.15–4.10 (m, 24 H), 4.07 (d, J = 3.6 Hz, 10 H), 4.06–3.98 (m, 60
H), 3.92 (d, J = 10.8 Hz, 27 H), 3.86 (d, J = 10.0 Hz, 16 H), 3.78
(d, J = 7.6 Hz, 8 H), 3.72 (d, J = 11.6 Hz, 7 H), 3.58 (d, J =
10.0 Hz, 7 H), 3.55–3.53 (m, 6 H), 3.49–3.41 (m, 60 H), 3.22 (dd,
J = 3.2, 10.0 Hz, 3.6 H), 3.17 (dd, J = 2.8, 10.0 Hz, 4 H) ppm. 13C
NMR (CDCl3): δ = 139.53, 139.47, 138.55, 138.50, 138.47, 138.37,
138.32, 129.11, 129.06, 129.04, 128.91, 128.74, 128.55, 128.41,
138.32, 138.28, 138.25, 138.23, 138.20, 137.98, 137.95, 134.7 (Cipso
,
Ar), 129.8, 129.0, 128.57, 128.52, 128.47, 128.38, 128.36, 128.34,
128.16, 128.13, 128.02, 127.98, 127.94, 127.81, 127.79, 127.75,
127.66, 127.58, 127.55, 127.49, 127.44, 127.3, 127.2, 127.1 (Ar),
126.96, 126.94, 123.85 (CH triazole), 98.63, 98.58, 98.5, 98.4, 98.3
(7ϫ C-1), 97.93, 97.91 (2ϫ C-1), 81.9, 80.9 (C-3), 79.9, 79.2 (C-2),
78.1, 77.3, 77.1, 76.8 (C-2, C-4), 75.8, 75.0, 73.4, 73.3, 73.2, 72.6
(CH2Ph,C-6), 71.5 (C-5), 69.3 (CH2), 69.1 (C-2), 65.0 (C-6), 55.3
(CH3), 50.5 (C-6-triazole) ppm. MS (MALDI-TOF): m/z (%) =
3924.43 (100.0) [M + Na]+.
6A,6D-Dideoxy-6A,6D-bis{N-[4Ј-(2A–F 3A–F,6B–F-heptadeca-O-benzyl-
,
α-cyclodextrin-6A-oxy)-1,2,3-triazolyl-4-methyl]}hexadeca-O-
benzyl-α-cyclodextrin (13): CuI (0.039 g, 0.20 mmol) and DIPEA
(0.05 mL, 0.30 mmol) were added to a solution of 4α (0.54 g, 128.28, 128.23, 128.06, 127.92, 127.75, 127.67, 127.58, 127.45,
0.21 mmol) and 2α (0.25 g, 0.10 mmol) in MeCN (2 mL) and 127.37, 127.27, 127.25, 127.15, 127.10, 98.67, 98.60, 98.58, 98.55,
EtOAc (2 mL) and the resulting mixture was stirred until TLC
showed no diazide. The reaction was then quenched by adding
78.90, 73.50, 73.46, 73.34, 73.30, 73.04, 72.73, 72.65, 72.62, 72.57,
72.51, 72.48, 71.78, 71.74, 71.71, 71.65, 71.51, 71.48, 71.45, 71.39,
Eur. J. Org. Chem. 2011, 5279–5290
© 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjoc.org
5287

M. Bols et al.
FULL PAPER
71.35, 69.12, 69.02 ppm. HRMS (MALDI-TOF): m/z calcd. for
C491H510N6O93 [M + Na]+ 8000.5260; found 8000.1963; calcd. for
[M + K]+ 8016.5000; found 8016.0023. Rf 0.26 (EtOAc/pentane;
1:2.5).
127.0, 98.7, 98.5, 81.1, 81.0, 79.04, 78.97, 78.93, 78.91, 78.86, 78.78,
78.73, 78.1, 75.54, 75.46, 75.1, 73.5, 73.43, 73.39, 72.90, 72.85, 72.7,
72.6, 71.7, 71.7, 69.5, 69.44, 69.41 ppm. HRMS (MALDI-TOF):
m/z calcd. for C545H566N6O103 [M + Na]+ 8864.9134; found
8865.4036; calcd. for [M + K]+ 8880.8873; found 8881.3359. Rf
0.23 (EtOAc/pentane; 1:2.5).
6A,6D-Dideoxy-6A,6D-bis{N-[4Ј-(2A–G 3A–G,6B–G-icosa-O-benzyl-β-
,
cyclodextrin-6A-oxy)-1,2,3-triazolyl-4-methyl]}hexadeca-O-benzyl-
α-cyclodextrin (15): CuI (19.1 mg, 0.10 mmol) and DIPEA
(0.025 mL, 0.15 mmol) were added to a solution of 4β (0.32 g,
0.11 mmol) and 2α (0.12 g, 0.05 mmol) in MeCN (2 mL) and
EtOAc (2 mL) and the resulting mixture was stirred until TLC
showed no diazide. The reaction was then quenched by adding
water (20 mL) and a saturated aqueous solution of NH4Cl (10 mL).
The aqueous layer was extracted three times with EtOAc, the com-
bined organic phases were washed with brine, dried with MgSO4,
and the solvent removed in vacuo. The resulting oil was purified
by flash column chromatography (EtOAc/pentane; 1:4 Ǟ 1:2) to
give 15 as a white foam (0.36 g, 0.043 mmol, 85 %); m.p. 81.7–
88.5°C. 1H NMR (CDCl3): δ = 7.25 (s, 2 H), 7.23–7.03 (m, 280 H,
Ph), 5.77 (d, J = 2.0 Hz, 2 H), 5.53 (d, J = 3.2 Hz, 2 H), 5.37 (d,
J = 10.0 Hz, 2 H), 5.29 (d, J = 3.6 Hz, 4 H), 5.24 (d, J = 2.8 Hz,
6 H), 5.21 (t, J = 4.0 Hz, 6 H), 5.16 (d, J = 3.6 Hz, 2 H), 5.13–5.10
(m, 7 H), 5.07 (d, J = 7.6 Hz, 4 H), 4.98 (t, J = 10.8 Hz, 10 H),
4.82–4.69 (m, 30 H), 4.60 (d, J = 12.0 Hz, 6 H), 4.54–4.46 (m, 40
H), 4.42 (d, J = 7.6 Hz, 12 H), 4.38 (d, J = 3.2 Hz, 12 H), 4.36–
4.32 (m, 12 H), 4.27 (d, J = 9.6 Hz, 4 H), 4.19 (d, J = 11.6 Hz, 4
H), 4.06–3.95 (m, 80 H), 3.67 (d, J = 12.0 Hz, 2 H), 3.61 (d, J =
11.6 Hz, 5 H), 3.58–3.53 (m, 12 H), 3.48–3.44 (m, 18 H), 3.20 (dd,
J = 3.2, 10.0 Hz, 4 H) ppm. 13C NMR (CDCl3): δ = 139.5, 139.40,
139.37, 139.34, 139.2, 139.1, 138.9, 138.54, 138.51, 138.46, 138.43,
138.37, 138.34, 138.12, 138.05, 138.02, 137.96, 134.6, 129.9, 129.1,
128.7, 128.5, 128.4, 128.34, 128.27, 128.20, 128.10, 128.05, 127.99,
127.94, 127.90, 127.76, 127.72, 127.68, 127.55, 127.50, 127.4,
127.33, 127.28, 127.2, 127.1, 127.0, 126.9, 98.7, 98.5, 81.8, 81.1,
81.0, 80.7, 80.4, 79.8, 79.7, 79.5, 79.1, 79.0, 78.91, 78.89, 78.84,
78.81, 78.7, 78.21, 78.19, 76.0, 75.5, 75.43, 75.42, 75.3, 74.2, 73.9,
73.5, 73.42, 73.38, 73.0, 72.9, 72.83, 72.79, 72.7, 72.6, 72.5, 71.71,
71.68, 71.6, 71.5, 69.4 ppm. HRMS (MALDI-TOF): m/z calcd. for
Hydrogenolysis of Trimers. General Procedure: Trimer was dis-
solved in CH2Cl2/EtOH (1:1), Raney nickel (3 mL) was added and
the resulting mixture was stirred for 1 h before being filtering
through a cotton plug. The filtrate was evaporated in vacuo and
dissolved in 2-methoxyethanol, then Pd(OH)2 was added and 1 atm
of hydrogen gas applied, and finally a few drops of TFA were
added. The progress of the reaction was monitored by MALDI-
TOF MS and took from 7 to 30 d to go to completion. The reac-
tion mixture was filtered through a silica plug and the solvents
were evaporated. The white solid residue was freeze-dried to give a
powder.
6A,6D-Dideoxy-6A,6D-bis{N-[4Ј-(α-cyclodextrin-6-oxy)-1,2,3-triaz-
1
olyl-4-methyl]}-α-cyclodextrin (17): H NMR (D2O): δ = 8.36 (s, 2
H), 5.17 (d, J = 3.2 Hz, 4 H), 5.05 (s, 18 H), 4.09 (d, J = 9.2 Hz, 4
H), 3.98–3.96 (m, 22 H), 3.89–3.87 (m, 30 H), 3.83 (s, 32 H), 3.66–
3.60 (m, 40 H), 3.58–3.55 (m, 22 H), 3.26 (d, J = 11.2 Hz, 2 H),
2.80 (d, J = 11.2 Hz, 2 H) ppm. 13C NMR (D2O): δ = 110.0, 95.1,
94.4, 81.7, 73.5, 72.2, 71.9, 65.5, 62.7, 60.6, 60.5, 59.6, 58.2, 57.2,
51.7 ppm. HRMS (MALDI-TOF): m/z calcd. for C114H182N6O88
[M + Na+ 3065.9849; found 3064.8098; calcd. for [M + K]+
3081.9588; found 3080.7580. Elution time: 16.97 min [1 mLminϪ1
60 min, 5–90% MeCN in H2O (0.1% TFA)].
,
6A,6D-Dideoxy-6A,6D-bis{N-[4Ј-(α-cyclodextrin-6-oxy)-1,2,3-triaz-
1
olyl-4-methyl]}-β-cyclodextrin (18): H NMR (D2O): δ = 8.33 (s, 1
H), 7.99 (s, 3 H), 5.07 (s, 5 H), 4.95 (d, J = 3.2 Hz, 46 H), 4.90–
4.89 (m, 5 H), 3.82 (t, J = 3.6 Hz, 53 H), 3.75–3.72 (m, 88 H), 3.61
(d, J = 4.4 Hz, 8 H), 3.60–3.59 (m, 7 H), 3.58 (d, J = 0.8 Hz, 4 H),
3.575 (d, J = 0.8 Hz, 2 H), 3.57 (d, J = 2.4 Hz, 4 H), 3.56 (d, J =
1.2 Hz, 4 H), 3.54 (s, 23 H), 3.52–3.47 (m, 74 H), 2.82 (d, J =
10.4 Hz, 5 H) ppm. 13C NMR (D2O): δ = 101.6, 96.0, 95.7, 81.5,
73.5, 73.3, 72.2, 71.9, 69.8, 61.7, 60.6 ppm. HRMS (MALDI-
TOF): m/z calcd. for C120H192N6O93 [M + Na]+ 3228.0377; found
3228.7444; calcd. for [M + K]+ 3244.0116; found 3244.1412. Elu-
tion time: 17.41, 18.03 min [1 mLminϪ1, 60 min, 5–90% MeCN in
H2O (0.1% TFA)].
C
518H588N6O58 [M + Na]+ 8432.7197; found 8433.8854; calcd. for
[M + K]+ 8448.6937; found 8449.4727. Rf 0.23 (EtOAc/pentane;
1:2.5).
6A,6D-Dideoxy-6A,6D-bis{N-[4Ј-(2A–G 3A–G,6B–G-heptadeca-O-
,
benzyl-β-cyclodextrin-6A-oxy)-1,2,3-triazolyl-4-methyl]}hexadeca-
O-benzyl-α-cyclodextrin (16): CuI (32.9 mg, 0.17 mmol) and
DIPEA (0.05 mL, 0.26 mmol) were added to a solution of 4β
(0.54 g, 0.18 mmol) and 2β (0.25 g, 0.09 mmol) in MeCN (2 mL)
and EtOAc (2 mL) and the resulting mixture was stirred until TLC
showed no diazide. The reaction was then quenched by adding
water (20 mL) and a saturated aqueous solution of NH4Cl (10 mL).
The aqueous layer was extracted three times with EtOAc, the com-
bined organic phases were washed with brine, dried with MgSO4,
and the solvent was removed in vacuo. The resulting oil was puri-
fied by flash column chromatography (EtOAc/pentane; 1:4 Ǟ 1:2)
to give 16 as a white foam in a 1:1 mixture of isomers (0.58 g,
6A,6D-Dideoxy-6A,6D-bis{N-[4Ј-(β-cyclodextrin-6-oxy)-1,2,3-triaz-
olyl-4-methyl]}-α-cyclodextrin (19): H NMR (D2O): δ = 8.13 (s, 2
1
H), 5.23 (s, 2 H), 5.17 (d, J = 14.0 Hz, 2 H), 5.10 (s, 9 H), 5.07 (s,
4 H), 5.01 (d, J = 7.6 Hz, 3 H), 4.09–4.03 (m, 7 H), 3.97 (d, J =
9.2 Hz, 23 H), 3.90 (s, 20 H), 3.87 (s, 16 H), 3.84–3.73 (m, 11 H),
3.68 (d, J = 10.8 Hz, 18 H), 3.62 (d, J = 9.2 Hz, 18 H), 3.60–3.54
(m, 7 H), 3.20 (d, J = 13.2 Hz, 2 H), 2.98 (d, J = 12.0 Hz, 2 H)
ppm. 13C NMR (D2O): δ = 110.0, 95.3, 95.2, 82.1, 73.4, 72.4, 72.3,
71.9, 71.24, 71.20, 69.5, 69.3, 66.8, 64.1, 62.8, 60.7, 60.6, 58.2 ppm.
HRMS (MALDI-TOF): m/z calcd. for C126H202N6O98 [M + Na]+
3390.0905; found 3386.0380; calcd. for [M + K]+ 3406.0644; found
3403.4028. Elution time: 16.38 min [1 mLminϪ1, 60 min, 5–90%
MeCN in H2O (0.1% TFA)].
1
0.07 mmol, 73%); m.p. 81.8–88.3 °C. H NMR (CDCl3): δ = 7.29
(s, 4 H), 7.21–7.05 (m, 590 H, Ph), 5.65 (s, 1 H), 5.56 (s, 1 H), 5.43
(s, 1 H), 5.31 (s, 1 H), 5.26 (d, J = 15.6 Hz, 7 H), 5.19–4.92 (m, 63
6A,6D-Dideoxy-6A,6D-bis{N-[4Ј-(β-cyclodextrin-6-oxy)-1,2,3-triaz-
1
H), 4.76–4.72 (m, 28 H), 4.64 (d, J = 9.6 Hz, 6 H), 4.58 (d, J = olyl-4-methyl]}-β-cyclodextrin (20): H NMR (D2O): δ = 8.33 (s, 2
12.4 Hz, 3 H), 4.45–4.31 (m, 154 H), 4.22 (t, J = 11.2 Hz, 14 H), H), 8.00 (s, 2 H), 5.08 (s, 3 H), 4.95 (s, 27 H), 4.89 (s, 6 H), 4.85
4.03–3.96 (m, 175 H), 3.53 (t, J = 9.6 Hz, 28 H), 3.46–3.43 (m, 49 (s, 2 H), 3.84 (t, J = 9.2 Hz, 37 H), 3.76 (s, 50 H), 3.72 (s, 35 H),
H), 3.20 (dd, J = 10.4, 19.2 Hz, 8 H) ppm. 13C NMR (CDCl3): δ
= 139.48, 139.45, 139.41, 138.56, 138.52, 138.48, 138.44, 138.37,
3.63–3.62 (m, 3 H), 3.61 (s, 4 H), 3.60 (d, J = 1.2 Hz, 5 H), 3.59
(t, J = 1.2 Hz, 4 H), 3.57 (d, J = 2.0 Hz, 5 H), 3.56 (t, J = 1.2 Hz,
128.8, 128.6, 128.4, 128.3, 128.2, 128.12, 128.10, 128.06, 127.99, 6 H), 3.54 (d, J = 1.6 Hz, 18 H), 3.53 (t, J = 1.2 Hz, 14 H), 3.52–
127.94, 127.90, 127.88, 127.7, 127.6, 127.5, 127.32, 127.25, 127.19, 3.51 (m, 22 H), 3.49–3.45 (m, 43 H), 3.27 (d, J = 1.2 Hz, 12 H),
5288
www.eurjoc.org
© 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Org. Chem. 2011, 5279–5290

Recognition of Peptides by Cyclodextrin Trimers
3.25 (d, J = 1.2 Hz, 2 H), 3.14 (s, 4 H) ppm. 13C NMR (D2O): δ 8.4 Hz, 1 H), 3.06 (dd, J = 8.0, 14.0 Hz, 1 H), 3.01 (dd, J = 5.6,
= 102.1, 102.0, 81.3, 81.2, 73.23, 73.15, 72.21, 72.15, 72.0, 71.9,
71.6, 71.3, 69.5, 62.7, 60.6, 60.4, 58.2 ppm. HRMS (MALDI-
TOF): m/z calcd. for C132H212N6O103 [M + Na]+ 3552.1434; found
3557.6064; [M + K]+ 3568.1173; found 3573.4937. Elution time:
12.86, 13.69 min [1 mLminϪ1, 60 min, 5–90% MeCN in H2O
(0.1% TFA)].
8.8 Hz, 1 H), 2.98 (dd, J = 6.4, 9.6 Hz, 1 H) ppm. HRMS: m/z
calcd. for C35H42N8O7 [M + Na]+ 709.3074; found 709.2978. Elu-
tion time: 16.05 min.
GGGF: 1H NMR (D2O/CD3OD): δ = 7.47–7.33 (m, 5 H, Ph), 4.68
(dd, J = 6.0, 6.4 Hz, 1 H), 3.97 (d, J = 16.4 Hz, 1 H), 3.90 (d, J =
17.2 Hz, 1 H), 3.75 (dd, J = 16.4, 17.2 Hz, 2 H), 3.29 (s, 2 H), 3.25
(dd, J = 6.4, 14.0 Hz, 1 H), 3.16 (dd, J = 5.6, 14.8 Hz, 1 H) ppm.
HRMS: m/z calcd. for C15H21N5O4 [M + Na]+ 358.1491; found
358.1479. Elution time: 14.52 min.
Peptide Synthesis. General Procedure: The peptides were automati-
cally synthesized by utilizing Fmoc solid-phase peptide synthesis
on a Rink-amide resin. Couplings were performed on a Liberty
automated microwave peptide synthesizer with an in situ Fmoc
amino acid activation protocol with HBTU/DIPEA. Fmoc depro-
tection was achieved by treatment with 20% piperidine in DMF.
After final deprotection the peptidyl resin was transferred to a fil-
tration tube, washed several times with CH2Cl2, and then shrunk
in MeOH for 30 min, and left overnight in a desiccator containing
KOH (s). Cleavage was achieved by adding TFA/TIS/water (5 mL;
95:2.5:2.5 v/v) and swirling for 1 h, then filtering. The peptide was
isolated by evaporation and subsequent freeze-drying. The peptides
were dissolved in 20% MeCN in water and preparative purification
achieved by RP-HPLC on an Agilent 1100 system with a Vydac
column (C8, 208TP-510) 5–40% MeCN in milliQ H2O (0.1%
TFA).
Determination of Association Constants by Plasmon Surface Reso-
nance (SPR): A manually operated BIAcore X® instrument and
standard CM5® sensor chips were used. The peptides were cova-
lently attached to the chip surface by standard amine coupling re-
actions in 10 mm acetate buffer adjusted to pH 4.6 by addition of
HCl.[25] Experiments were carried out in 10 mm phosphate buffer at
pH 4.6 with various concentrations of cyclodextrins ranging from 5
to 1000 μm.
Acknowledgments
We thank the Lundbeck Foundation for financial support.
FGGGFGGGF: 1H NMR (D2O): δ = 7.65–7.49 (m, 15 H, Ph), 4.50
(t, J = 7.2 Hz, 1 H), 4.24 (d, J = 16.8 Hz, 1 H), 4.17 (d, J = 16.8 Hz,
1 H), 4.14 (d, J = 2.8 Hz, 2 H), 4.11 (d, J = 2.8 Hz, 2 H), 4.09 (d,
J = 2.8 Hz, 2 H), 4.06 (dd, J = 2.8, 6.4 Hz, 2 H), 3.45–3.38 (m, 4
H), 3.24 (dd, J = 8.8, 14.8 Hz, 1 H), 3.20 (dd, J = 9.2, 14.0 Hz, 1
H), 2.28 (qv, J = 2.4 Hz, 4 H) ppm. HRMS: m/z calcd. for
C39H48N10O9 [M + Na]+ 823.3503; found 823.3509. Elution time:
13.89 min.
[1] H. Chen, W. S. Weiner, A. D. Hamilton, Curr. Opin. Chem.
Biol. 1997, 1, 458–466.
[2] H. J. Schneider, A. K. Yatsimirsky, Chem. Soc. Rev. 2008, 37,
263–277.
[3] G. Arena, A. Casnati, A. Contino, A. Magrì, F. Sansone, D.
Sciotto, R. Ungaro, Org. Biomol. Chem. 2006, 4, 243–249.
[4] K. D. Shimizu, J. Rebek Jr., Proc. Natl. Acad. Sci. USA 1995,
92, 12403–12407.
FGGGNalGGGF: 1H NMR (D2O): δ = 8.27–8.24 (m, 3 H), 8.12
(d, J = 8.8 Hz, 1 H), 7.91–7.89 (m, 2 H), 7.81 (t, J = 8.0 Hz, 1 H),
7.76–7.62 (m, 10 H), 5.07 (dd, J = 7.2, 15.2 Hz, 1 H), 4.91 (dd, J
= 5.6, 8.0 Hz, 1 H), 4.60 (t, J = 7.2 Hz, 1 H), 4.32–4.07 (m, 12 H),
3.74–3.68 (m, 1 H), 3.58–3.51 (m, 4 H), 3.31 (dd, J = 9.6, 13.6 Hz,
1 H) ppm. HRMS: m/z calcd. for C43H50N10O9 [M + H]+ 851.3840;
found 851.3840. Elution time: 17.41 min.
[5] C. S. Wilcox, M. D. Cowart, Carbohydr. Res. 1987, 171, 141–
160.
[6] Peptoids are membrane-active polypeptides derived from fun-
gal sources, see: J. K. Chugh, B. A. Wallace, Biochem. Soc.
Trans. 2001, 29, 565–570.
[7] J. Szetli, Chem. Rev. 1998, 98, 1743–1753.
[8] M. V. Rekharsky, Y. Inoue, Chem. Rev. 1998, 98, 1875–1917.
[9] K. Sasaki, M. Nagasaka, Y. Kuroda, Chem. Commun. 2001,
24, 2630–31.
FGGGYGGGF: 1H NMR (CD3OD): δ = 7.37–7.26 (m, 10 H), 7.05
(d, J = 8.4 Hz, 2 H), 6.70 (d, J = 8.4 Hz, 2 H), 4.59 (dd, J = 5.2,
9.2 Hz, 1 H), 4.41 (dd, J = 6.0, 6.4 Hz, 1 H), 4.16 (dd, J = 6.4,
8.4 Hz, 1 H), 4.04 (d, J = 8.8 Hz, 1 H), 3.95–3.79 (m, 11 H), 3.27
(dd, J = 6.0, 14.4 Hz, 1 H), 3.19 (dd, J = 5.2,14.0 Hz, 1 H), 3.05
(dd, J = 8.4, 14.0 Hz, 2 H), 2.93 (dd, J = 9.6, 14.0 Hz, 1 H), 2.90
(dd, J = 8.4, 13.6 Hz, 1 H) ppm. HRMS: m/z calcd. for
C39H48N10O10 [M + Na]+ 839.3453; found 839.3533. Elution time:
12.94 min.
[10] a) Md. A. Hossain, H.-J. Schneider, J. Am. Chem. Soc. 1998,
120, 11208–11209; b) M. Kruppa, C. Mandl, S. Miltschitzky,
B. König, J. Am. Chem. Soc. 2005, 127, 3362–3365; c) M. E.
Bush, N. D. Bouley, A. R. Urbach, J. Am. Chem. Soc. 2005,
127, 14511–14517; d) M. Sirish, H.-J. Schneider, Chem. Com-
mun. 1999, 907–908; e) H.-J. Schneider, Angew. Chem. 2009,
121, 3982; Angew. Chem. Int. Ed. 2009, 48, 3924–3977.
[11] Recently click chemistry has been extensively used to attach
functional groups to cyclodextrins or connect cyclodextrins to
each other, see: a) K. Ohtsuka, S. Shinobu, S. Watanabe, K.
Komizo, S. Takenaka, Nucleic Acids Symp. Ser. 2008, 52, 717–
718; b) M. Mourer, F. Hapiot, E. Monflier, S. Menuel, Tetrahe-
dron 2008, 64, 7159–7163; c) M. Mourer, F. Hapiot, S. Tilloy,
E. Monflier, S. Menuel, Eur. J. Org. Chem. 2008, 5723–5730;
d) M. Munteanu, C. Soowhan, H. Ritter, J. Inclusion Phenom.
Macrocyclic Chem. 2008, 62, 197–202; e) S. Srinivasachari,
K. M. Fichter, T. Reineke, J. Am. Chem. Soc. 2008, 130, 4618–
4627; f) Y. Wang, Y. Xiao, T. Tan, S.-C. Ng, Tetrahedron Lett.
2008, 49, 5190–5191; g) J. M. Bryson, W.-J. Chu, J.-H. Lee,
T. M. Reineke, Bioconjugate Chem. 2008, 19, 1505–1509.
[12] A paper using the same assembly methodology as that used
herein appeared recently, see: G. K. Rawal, P. Zhang, C.-C.
Ling, Org. Lett. 2010, 12, 3096–3099.
FGGGWGGGF: 1H NMR (CD3OD): δ = 7.56 (d, J = 8.0 Hz, 1
H), 7.36–7.25 (m, 12 H), 7.15 (s, 1 H), 7.09 (dt, J = 0.8, 8.0 Hz, 1
H), 7.01 (dt, J = 0.8, 8.0 Hz, 1 H), 4.62–4.52 (m, 2 H), 4.15 (dd, J
= 6.0, 8.0 Hz, 1 H), 4.04 (dd, J = 6.0, 16.4 Hz, 1 H), 3.21 (dd, J =
4.0, 4.8 Hz, 1 H), 3.17 (dd, 1 H) 3.05 (dd, J = 8.4, 14.0 Hz, 1 H),
2.92 (dd, 1, 9.6, 14.0 Hz,
) ppm. HRMS: m/z calcd. for
C41H49N11O9 [M + Na]+ 862.3612; found 862.3554. Elution time:
15.46 min.
1
FGGFGGF: H NMR (CD3OD): δ = 7.37–7.20 (m, 15 H), 4.57 (t,
J = 4.8 Hz, 1 H), 4.55 (t, J = 5.2 Hz, 1 H), 4.15 (dd, J = 6.4, 8.0 Hz,
1 H), 3.97 (d, J = 16.0 Hz, 1 H), 3.90 (d, J = 3.6 Hz, 1 H), 3.87 (d,
J = 5.6 Hz, 1 H), 3.84 (d, J = 8.0 Hz, 1 H), 3.79 (s, 2 H), 3.74 (d,
J = 1.2 Hz, 2 H), 3.68 (d, J = 10.8 Hz, 1 H), 3.27 (dd, J = 8.0,
14.0 Hz, 1 H), 3.22 (dd, J = 2.8, 8.0 Hz, 1 H), 3.18 (dd, J = 6.4,
[13] J. Bjerre, T. H. Fenger, L. G. Marinescu, M. Bols, Eur. J. Org.
Chem. 2007, 704–710.
Eur. J. Org. Chem. 2011, 5279–5290
© 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjoc.org
5289

M. Bols et al.
FULL PAPER
[14] T. Lecourt, A. Herault, A. J. Pearce, M. Sollogoub, P. Sinaÿ,
Chem. Eur. J. 2004, 10, 2960–2971.
[15] F. Ortega-Caballero, J. Bjerre, L. S. Laustsen, M. Bols, J. Org.
Chem. 2005, 70, 7217–7226.
[16] D. Lesur, A. Gassama, V. Moreau, S. D. Pilard, F. Jedaïni-Pil-
ard, Carbohydr. Res. 2005, 340, 1225–1231.
[17] V. O. Rodionov, V. V. Fokin, M. G. Finn, Angew. Chem. 2005,
117, 2250; Angew. Chem. Int. Ed. 2005, 44, 2210–2215.
[18] M. Ahlquist, V. V. Fokin, Organometallics 2007, 26, 4389–4391.
[19] S. Hotha, S. Kashyap, J. Org. Chem. 2006, 71, 364–367.
[20] S. Kamijo, T. Jin, Z. Huo, Y. Yamamoto, J. Org. Chem. 2004,
69, 2386–2393.
[22]
[23]
[24]
V. D. Bock, H. Hiemstra, J. H. van Maarseveen, Eur. J. Org.
Chem. 2006, 1, 51–68.
K. D. Bodine, D. Y. Gin, M. S. Gin, Org. Lett. 2005, 7, 4479–
4482.
S. Menuel, N. Azaroual, D. Landy, N. Six, F. Hapiot, E. Monfl-
ier, Chem. Eur. J. 2011, 17, 3949–3955.
[25] F. A. Tanious, B. Nguyen, W. D. Wilson, Methods Cell Biol.
2008, 84, 53–77.
[26] M. V. Rekharsky, Y. Inoue, Chem. Rev. 1998, 98, 1875–1917.
[27] T. Lecourt, A. Herault, A. J. Pearce, M. Sollogoub, P. Sinaÿ,
Chem. Eur. J. 2004, 10, 2960–2971.
Received: May 12, 2011
Published Online: August 5, 2011
[21] R. E. Looper, D. Pizzirani, S. L. Schreiber, Org. Lett. 2006, 8,
206–2066.
5290
www.eurjoc.org
© 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Org. Chem. 2011, 5279–5290