Unraveling unidirectional threading of α-cyclodextrin in a [2]rotaxane through spin labeling approach

Article abstract of DOI:10.1021/ja3073484

We present here the results of a CW-ESR investigation of a double spin labeled α-cyclodextrin-based [2]rotaxane that is characterized by the presence of nitroxide labels both at the wheel and at the dumbbell. This was accomplished by synthesizing a spin l

Full text of DOI:10.1021/ja3073484

Article
pubs.acs.org/JACS
Unraveling Unidirectional Threading of α‑Cyclodextrin in a
2]Rotaxane through Spin Labeling Approach
Costanza Casati, Paola Franchi, Roberta Pievo, Elisabetta Mezzina,* and Marco Lucarini*
[
†
†
‡
,†
,†
†
Department of Organic Chemistry “A. Mangini”, University of Bologna, Via San Giacomo 11, I-40126 Bologna, Italy
‡
Max Planck Institute for Biophysical Chemistry am Faßberg 11, 37077 Go
̈
ttingen, Germany
*
S Supporting Information
ABSTRACT: We present here the results of a CW-ESR investigation of
a double spin labeled α-cyclodextrin-based [2]rotaxane that is
characterized by the presence of nitroxide labels both at the wheel and
at the dumbbell. This was accomplished by synthesizing a spin labeled α-
CD (the wheel) that was mechanically blocked on a thread containing a
nitroxide unit by a Cu(I) catalyzed azide−alkyne cycloaddition
(
CuAAC). Both ESI-MS analysis and NMR spectroscopy were used to
provide evidence of the threading processes. Because of the unsym-
metrical structure of both the wheel and the axle, two different
geometrical isomers could be predicted on the basis of the orientation of the CD along the thread. By combining molecular
dynamic calculations and information extracted from the CW-ESR spectra, we were able to determine the geometrical nature of
the isomer that was isolated as the only species. The ESR spectra showed J-coupling between the two mechanically connected
nitroxide units and were analyzed by a model assuming three main molecular states. The intramolecular noncovalent nature of
spin exchange was confirmed by reversibly switching the magnetic interaction on−off by changing the pH of the solution in the
presence of a competing macrocyclic host.
INTRODUCTION
Chart 1
■
Cyclodextrins (CDs) continue to be attractive wheel
components in constructing mechanically interlocked assem-
blies, such as rotaxanes.
1
,2
A rotaxane is described as a
molecular system in which a macrocycle (wheel) threads a
linear subunit (dumbbell) with two bulky stoppers. The
number of components in a rotaxane is indicated in brackets
as a prefix. Therefore, a rotaxane consisting of a single dumbbell
1,2
and a single macrocycle is indicated as [2]rotaxane.
The basis for the construction of a CD-based [2]rotaxane is
the interaction between the hydrophobic cavity of the CD and
the special hydrophobic unit in the linear component. Since the
early 1980s, CD-based [2]rotaxanes have been prepared by
encapsulation of a symmetrical linear molecule with CD and
subsequent “locking” by coordination or by covalent linking to
1
,2
bulky stoppers. At the same time, many unsymmetrical
2]rotaxanes, in which a CD ring encircles an unsymmetrical
[
3
−6
dumbbell, were also set up. In this case, when an asymmetric
CD was threaded by a nonsymmetrical dumbbell, it would give
two isomers differing in orientation of CD with respect to the
different ends of the dumbbell. As a result, two isomeric
ring by a linear subunit resulting in only one rotaxane isomer
[2]rotaxanes were always obtained in these systems. The
7,8
was observed. In principle, unidirectional threading of a CD
ring can be improved by using a functionalized CD at the
primary or at the secondary face with appropriate function-
alities. Fort and co-workers reported an example of [2]rotaxane
orientations of the CD ring in the isomeric [2]rotaxanes were
deduced from NOE measurements or by recording circular
dichroism spectra.
It was expected that unidirectional threading of a CD ring by
a linear subunit during the formation of a [2]rotaxane could be
achieved in order to construct more complicated rotaxanes.
However, only in a few cases unidirectional threading of a CD
Received: July 26, 2012
Published: October 29, 2012
©
2012 American Chemical Society
19108
dx.doi.org/10.1021/ja3073484 | J. Am. Chem. Soc. 2012, 134, 19108−19117

Journal of the American Chemical Society
Article
Scheme 1
having a lactosyl-α-CD conjugate at the primary face and a
having a paramagnetic label on the wheel. By combining
molecular dynamic calculations and information extracted from
the CW-ESR spectra, we were able to determine the
geometrical nature of the isolated isomer. Because of great
interest in controlling through space spin exchange between
two nearby radical centers, we also examined the possibility of
reversibly triggering the spin interaction in the reported
[2]rotaxane.
symmetric decane-based axle carrying lactosyl stoppers at the
9
extremities. Because of the symmetric nature of the axle,
however, only one isomer was possible. Thus, the preparation
and separation of unidirectional rotaxane in which the CD is
monofunctionalized remain totally unexplored.
In past decades, double spin labeling of macromolecules by
stable nitroxide radicals has provided remarkable information:
in particular, continuous wave (CW) and pulse (e.g., double
quantum coherence (DQC) and PELDOR) ESR spectra of
RESULTS AND DISCUSSION
10,11
■
double spin labeled systems have been studied.
sensitivity of DQC and PELDOR
The
spectra allows the
reliable determination of distances between labels in the range
.6−6.0 nm in frozen solution, whereas shorter distances are
1
2,13
Synthesis and NMR Characterization of Rotaxanes. It
has already been shown that the TEMPO unit is large enough
to be used as the end-cap group in paramagnetic rotaxanes
1
19a
18
having cucurbit[6]urils
or α-cyclodextrin as the wheel.
generally not accessible because the electron dipolar interaction
becomes too large and the presence of relevant scalar electron
exchange interactions prevents the irradiation of a single
electron spin, which is the prerequisite for their application
While the development of straightforward procedures to
functionalize selectively cucurbiturils is still a challenging
21
task, reliable synthetic procedures are available for the
functionalization of the primary or the secondary rims of
1
4
(
vide infra).
On the other hand, when the spin labels are at relatively
22
cyclodextrins. For this reason we decided to employ as a
paramagnetic wheel an α-CD functionalized with a nitroxide
unit.
short distances, the liquid solution CW-ESR spectrum could be
very informative because its shape depends on several structural
and dynamic parameters that characterize the double labeled
molecule. Actually, to probe supramolecular complexation by
Several examples of β-CD containing one or more nitroxide
17,23,24
spin labels have been reported in the past.
However, to
the best of our knowledge, no paramagnetic α-CD has been
reported in the literature. To this purpose, the new synthesis of
two TEMPO-monofunctionalized α-CDs were undertaken,
bringing the radical arm in the larger secondary rim at the 3
position or at the 6 position of the primary rim (Scheme 1).
The reactions for the functionalization of the primary rim
1
5
16
macrocyclic hosts, resorcinarenes, calixarenes, and cyclo-
1
7
dextrin containing two nitroxide labels were reported and
investigated by CW-ESR.
We have already shown that 2,2,6,6-tetramethylpiperidine-N-
oxyl (TEMPO) ring cannot pass through both the cavity of an
1
8
9
,25a
α-cyclodextrin and that of cucurbit[6]uril (CB[6]), a
were realized by selective tosylation of α-CD,
affording
9
synthetic macrocyle largely employed in the preparation of
pure 6-O-Ts-α-CD in 20% yield. The azido derivative was
obtained in N,N-dimethylformamide (DMF) by tosyl displace-
ment with sodium azide (90% yield). Coupling of the nitroxide
1
9
rotaxanes. This feature allowed us to prepare α-CD- and
CB[6]-based rotaxanes having the TEMPO radical as the end-
cap group and the above-mentioned macrocycles as the wheel.
By using the PELDOR technique, we were able to determine
the distances among the paramagnetic end units of the CB[6]-
26
alkyne 4-propargyloxy-TEMPO (1a, Scheme 1) onto the
cyclodextrin, in the presence of a catalytic amount of copper
sulfate and sodium ascorbate as reducing agent (click
conditions), afforded the single 6-triazolyl-TEMPO-α-CD (2a)
in 48% yield after purification by gel filtration over a Sephadex
G-15 column.
2
0
based rotaxane.
As a part of our continuing studies on the characterization of
new supramolecular architectures containing open-shell mole-
cules, we present here the results of a CW-ESR investigation of
the double spin labeled α-CD-based [2]rotaxane that is
characterized by the presence of two nitroxide free radicals at
the wheel and at the dumbbell at several temperatures. In this
work we succeed to synthesize for the first time a spin labeled
α-CD (the wheel) and to prepare an unidirectional rotaxane
The functionalization of the secondary edge of α-CD was
realized according the reaction sequence reported in Scheme 1
starting from α-cyclodextrin in DMF in the presence of p-
toluenesulfonylimidazole and freshly activated 4 Å powder
molecular sieves, which afforded prevalently mono-2-p-
27
25b
toluenesulfonate.
Tosyl displacement with sodium azide
1
9109
dx.doi.org/10.1021/ja3073484 | J. Am. Chem. Soc. 2012, 134, 19108−19117

Journal of the American Chemical Society
Article
28
afforded the mono-3-azidodeoxy-α-cyclodextrin. Cu(I) cata-
lyzed 1,3-dipolar cycloaddition of this azide with alkyne 1a gave
3
a in 36% yield.
In order to prepare the bis-labeled rotaxane, we decided to
9
follow the approach reported by Fort and co-workers based on
the formation and assembly by click reaction of a
pseudorotaxane composed by α-CD and a C-10 alkyl chain,
blocked at one end by a saccharidic stopper and carrying an
azido group at the other end. To check the feasibility of this
approach, we initially treated under click conditions the
lactosyldecane azide 4 with nitroxide 1a in the presence of
the unmodified α-CD (Scheme 2). The reaction afforded a
Scheme 2
1
Figure 1. H NMR partial spectra (600 MHz, DMSO-d , 1.5 mM) of a
6
mixture of rotaxane 6b−7b (a) and bis-labeled rotaxane (b) showing
H triazole signal pattern at 353 K. In the latter case, the presence of
three signals corresponding to the different triazole moieties in the
molecule is indicative of the formation of only one isomer.
two signals for the proton in the triazole bound to the lactosyl
end with the amidic linkage (see Supporting Information). The
two signals arise from the restricted rotation of the amidic
bond. Actually, tertiary amides are known to undergo a Z−E
29
isomerization in solution. At room temperature, two sets of
signals corresponding to endo and exo rotamers are commonly
observed. Peak coalescence is generally achieved by increasing
the temperature, thus making the NMR spectra simpler. Also in
our case, at 80 °C, peak coalescence was observed accompanied
by a substantial signal broadening for the protons close to the
amide group. In the mixture 6b−7b, all the signals due to the
aromatic proton of triazoles are doubled at room temperature,
two singlets and two couples of signals that at high temperature
coalesce into one, this being an indication of the presence of
the two isomeric forms of the rotaxane (see Figure 1a).
The success of the synthesis of this α-CD-based para-
magnetic rotaxane containing lactosyl and TEMPO moieties as
stoppers is the basis for the preparation of a new interlocked
molecule, in which the bead component consisted of a radical
TEMPO-α-CD conjugate. We used both monofunctionalized
α-CDs, bringing the radical arm in the larger secondary rim (3a,
Scheme 1) and at the 6 position of the primary rim (2a,
Scheme 1), respectively.
The “click” rotaxanation of lactosyl azide 4 with alkyne 1a in
the presence of the monofuntionalized CDs in the condition of
the reaction reported in the Scheme 3 provided the desired
product only in the case of the 6-armed macrocycle 2a (vide
infra). In the presence of the secondary-armed macrocyle 3a,
no formation of the rotaxane was observed.
The bis-labeled rotaxane was obtained in 29% isolated yield
after purification by gel filtration over a Sephadex G-15 column.
Its structural determination was provided by ESI-MS and NMR
data (Figure 2b). Cross-peaks were detected in the 2D ROESY
spectrum involving the alkyl protons of the axle with the inner
protons H3 and H5 of α-CD resonating in the range 3.7−4.0
ppm, which are absent in the free thread.
mixture of the isomeric rotaxanes 6a and 7a (see discussion
reported below), differing in the orientation of CDs with
respect to the rod’s ends, in 22% isolated yield.
Both ESI-MS analysis and NMR spectroscopy were used to
provide evidence of the threading processes. In all cases H
NMR spectra of the investigated products were recorded after
in situ reduction of the sample by Na S O or phenylhydrazine
2
2
4
to obtain the corresponding diamagnetic N-hydroxy forms
1b−3b, 5b−9b). Although the N-hydroxy form of the free
(
thread (5b) shows small chemical shift changes in the
corresponding H NMR spectra when mechanically trapped
inside the macrocycle to form rotaxanes 6b and 7b, the 2D
ROESY spectrum (Supporting Information) gave clear
evidence of the interlocked structure. In particular, we detected
strong cross-peaks between the alkyl chain protons of the
thread (1−1.4 ppm) and the inner proton of α-CD, falling in
the range 3.7−4.0 ppm. These interactions, which are totally
absent in the 2D ROESY spectrum of the free thread 5b,
indicated that the cyclodextrin is located over the rotaxane
decamethylene chain, between the lactosyl and TEMPO units.
Because of the unsymmetrical structure of both CD and the
axle, a mixture of the two isomers differing in the orientation of
CDs with respect to the rod’s ends was obtained. The presence
of the two isomers with a relative ratio of 70:30 was revealed in
the H NMR spectrum of 6b−7b at 80 °C, by looking at the
region around 8 ppm (see Figure 1a). This zone is
characterized by the signal due to the aromatic proton of the
triazole. The free thread 5b shows one singlet due to the
aromatic proton of triazole next to the nitroxide rod’s end and
Because of the unsymmetrical structure of both 2a and the
axle, a mixture of isomers should be predicted on the base of
the orientation of the CD along the thread. However, the
inspection of the H NMR region around 8 ppm revealed the
presence of only one isomer. The N-acetate group displays
restricted rotation, giving rise to multiplicity of signals in proton
NMR spectra. Heating of the material to 60 °C in DMSO
shows peak coalescence. At 353 K, the presence of three signals
1
9110
dx.doi.org/10.1021/ja3073484 | J. Am. Chem. Soc. 2012, 134, 19108−19117

Journal of the American Chemical Society
Article
Scheme 3
the azide 4a to occur only from the secondary edge of CD. If
this is true, we can ascribe the unique isolated isomer to the one
in which the two spin labels are directed to the same part (8a,
“
close” isomer). However, the NMR data do not permit making
any definitive conclusion and CW-EPR spectra were recorded
to determine the correct assignment.
ESR Results. ESR spectrum of the mixture of monolabeled
rotaxanes 6a−7a (0.05 mM) in water at 328 K is reported in
Figure 3a. The expected three-line spectrum is characterized by
Figure 3. ESR spectra of the investigated nitroxides (0.05 mM)
recorded in water at 328 K.
the coupling of the unpaired electron with a nitrogen nucleus
with a hyperfine splitting constant, a , of 16.97 G (g = 2.0056).
N
The high field ESR line is characterized by a lower height with
respect to that obtained with free thread, this being due to the
slower motion in solution of the rotaxane biradical, resulting in
incomplete averaging of the anisotropic components of the
hyperfine and g-tensors. The spectral resolution is not sufficient
to allow the detection of the two geometrical isomers, this
1
Figure 2. H NMR spectra (600 MHz, 298 K, D O, 1.5 mM) of (a)
2
macrocyle 2b and (b) bis-labeled rotaxane: (A) H triazole; (B) H1
CD, CH OTEMPOH, H1 Glu; (C) H1 Gal, N-CH -triazolyl,
2
2
CH OTEMPOH and H decane; (D) H-2,3,4,5,6-lactosyl moiety and
2
being an indication that the difference in the a values for the
N
CD; (E) CH −CO; (F) TEMPOH; (G) H decane. Signal assignment
3
two paramagnetic isomers is smaller than the spectral line
width.
was achieved by measuring the 2D ROESY spectrum of diamagnetic
bis-labeled rotaxane. Square signals give NOE cross-peaks indicating
the encapsulation of the decyl thread into the CD.
ESR spectrum of the bis-labeled rotaxane (0.05 mM) in
30
water at 328 K is reported in Figure 3c. In nitroxide biradicals,
the shape of the ESR spectrum is dictated by the degree of J-
coupling between the two electron spins and isotropic
(
Figure 1b) corresponding to the different triazole moieties in
the molecule is indicative of the formation of only one species.
It is obvious that the presence of unidirectional bis-labeled
hyperfine coupling, a , to the nitroxide nitrogen nucleus. J-
N
Coupling increases with decreasing distance between labels and
31−33
[2]rotaxane arises from the formation of only one precursor
therefore reports on the relative interlabel distances.
When
isomer, 4@2a [2]pseudorotaxane in aqueous media. Such
orientation specificity was not observed in the 4@α-CD
J is zero, the ESR spectrum is characterized by three hyperfine
lines as found for a monoradical nitroxide species. As J
increases, the hyperfine lines broaden and begin to form
complex multiplets. When J becomes large relative to aN
(strong exchange), the spectrum is characterized by a five-line
[
2]pseudorotaxane, as two isomeric [2]rotaxanes 6a and 7a
were obtained in a 70:30 ratio.
The lack of rotaxane formation in the case of the branched
CD functionalized in the secondary rim (3a) suggests that the
substitution of CD in the secondary rim may hamper the
inclusion and the threading process of the decamethylene part
into the cavity of the host, allowing the inclusion direction of
pattern with an intensity ratio of 1:2:3:2:1 separated by a /2.
N
As observed in Figure 3c, at 328 K, biradical coupling is clearly
demonstrated by the presence of five-line pattern, although the
second and fourth lines are broader (discussed further below).
1
9111
dx.doi.org/10.1021/ja3073484 | J. Am. Chem. Soc. 2012, 134, 19108−19117

Journal of the American Chemical Society
Article
Possible mechanisms for J-coupling can be operative by
through-bond or through-space or both. Because of the
mechanical linkage between the two nitroxide units, the
through bond mechanism can be obviously excluded and the
interaction must be related exclusively to through space
interaction.
To discard intermolecular J-coupling between nitroxide
labels of two different rotaxane molecules, we recorded under
the same condition the ESR spectrum of a mixture containing
both the thread, 5a, and the spin-labeled CD 2a. As expected,
this spectrum is characterized by J = 0 (see Figure 3b).
Because of the through-space nature of J, we checked by ESR
the possibility of reversibly turning on/off the spin interaction
by means of an external trigger. This was achieved by simply
changing the pH of the solution containing the bis-labeled
nitroxide in the presence of anilinium chloride and a
macrocyclic host like cucurbit[7]uril (CB[7], Chart 1). Because
of the complexation of one or both nitroxide units, addition of
CB[7] resulted in the disappearance of the exchange peaks in
the ESR spectrum (see Figure 4a). Actually, complexation by
In addition to J-coupling, nitroxides weakly associated by
36
noncovalent forces can also exhibit electron dipolar coupling.
3
The strength of this interaction is determined by 1/D , where
D is the distance between the two nitroxide units. For rapidly
tumbling peptides, the dipole−dipole tensor averages to zero
and does not contribute to the ESR spectrum. However, for
slowly tumbling or immobile systems, the strength of the
dipolar interaction serves as a measure of the distance between
the coupled spins. At 203 K in water the bis-labeled rotaxane
gave the expected spectrum for an immobile nitroxide showing
a substantial broadening of the spectral lines as expected for a
biradical exhibiting dipolar coupling (see Supporting Informa-
tion). Because of large distribution of distances, however, a
quantitative analysis was not possible.
The ESR spectra of the bis-labeled rotaxane were registered
in temperature range 299−360 K (see Figure 5). Considerable
Figure 5. ESR spectra of the bis-labeled rotaxane recorded at different
temperatures in water.
narrowing of second and fourth lines when the temperature
increases indicates unambiguously the fast modulation of the
exchange interaction by an intramolecular motion. However, at
the temperatures (>350 K) when the widths of second and
fourth lines are close to those of outermost lines, the ratio of
line intensities differs essentially from 1:2:3:2:1, which is
predicted by the model of the diradical with two conformations
Figure 4. EPR spectra of the bis-labeled rotaxane (0.05 mM) in the
presence of CB[7] 10 mM: (a) initial solution and in the presence
anilium chloride 25 mM at (b) pH 3 and (c) pH 8.
and with the strong fast exchange (J ≫ a ). Thus, the observed
N
distribution of line intensities should be explained by assuming
a model with more than two conformations. In this model,
37
proposed by Luckhurst and further developed by Parmon et
38
al., three main molecular states are considered. In the present
case, these states correspond to different coconformations
adopted by the α-CD on the dumbbell component and/or
different conformations of the covalent bonds connecting the
thread and the spin label. In the first group of conformations
(conformation A), the two labels are too far and the radical
centers have no opportunity to interact (J = 0). In two others
(B and C), the TEMPO radical centers get some possibilities to
give exchange coupling. In this case, the appearance of the
exchange depends on the distance between the radical centers
and their mutual orientation. If the average lifetimes of the
3
4
19a,35
macrocyles like CD or CB
is known to significantly
reduce the probability of collisions of the nitroxide termini,
giving rise to a complete suppression of spin exchange.
Addition of anilinium chloride (25 mm) at pH 3 (Figure 4b)
restored the original five-line pattern, this being an indication
that the radicals are completely displaced from the macrocyclic
cavity by the aromatic ammonium cation. When increasing the
basicity of the solution to pH 8 with NaOH, the second and
fourth lines were suppressed again and the monoradical-like
spectrum is observed (see Figure 4c). This result strongly
suggests that aniline (pK = 4.6), which is present in alkaline
radical fragments in conformations A, τ , and in conformations
a
A
solution, is released from the cavity being replaced by the
nitroxide guests. The process is fully reversible: addition of HCl
to give back pH 3 results in the formation of a five-line
spectrum.
B and C, τB+C, are sufficiently long (i.e., a τ > 1 and a τ
1), a superposition of two ESR spectra will be observed (with
>
N A
N B+C
statistical weight τ /(τ +τB+C) and τB+C/(τ + τB+C)). If the
A
A
A
movement of the radical fragments in conformations B and C is
1
9112
dx.doi.org/10.1021/ja3073484 | J. Am. Chem. Soc. 2012, 134, 19108−19117

Journal of the American Chemical Society
Article
fast, the corresponding ESR spectrum will be described by the
model of fast exchange.
Analysis of the dependence of τB+C/τ on temperature afforded
A
changes of entropy and enthalpy for the equilibrium as ΔH =
+1.45 kcal mol− and ΔS = +3.85 cal mol K .
1
−1 −1
In the model of three conformations the variation in the
width of the second and fourth lines with the temperature can
be ascribed to the variation of the modulation conditions for
the exchange interaction, while the observed distribution of the
line intensities can be explained as a result of the superposition
of the ESR spectrum of A and that one due to conformations B
and C. The relative intensities of the ESR lines are proportional
to the relative lifetimes in the state A and B + C. Thus,
theoretical simulation of the experimental spectra afforded the
We can tentatively assign the three different ESR species to
three different groups of the mechanically interlocked
biradicals. The temperature dependence of τB+C/τ indicates
A
that A is the more stable group of conformations. In these
conformations the radical centers are not able to interact and J
=
0 (Scheme 4A). Any displacement of the macrocyle from this
equilibrium position results in a new group of conformations
where the TEMPO radical centers get some possibilities to give
exchange coupling. We attributed these displacements to
movements of the macrocycle along and/or around the thread
which are expected to be slow in the ESR time scale as
indicated by the superposition of the ESR signals due to group
conformations A and B + C. This is in line with previous
reports showing that threading of paramagnetic guest molecules
in the cyclodextrin cavity is generally slow in the ESR time
39
ratio I /I between the integrated intensity of the spectral
2
1
41
scale. The small enthalpy and entropy variations observed for
the transition between A and B + C conformations suggest that
the interaction of α-CD with the thread is very similar in the
two group of conformations. This is in agreement with a
theoretical investigation performed on cyclodextrin-based
rotaxanes showing that the interaction between α-CD and an
aliphatic thread does not change significantly by varying the
Figure 6. ESR spectrum of the bis-labeled rotaxane recorded at 328 K
in water (black) and the corresponding theoretical simulation (red).
42
position of the macrocyle on the linear chain.
On the other hand, the exchange between conformations B
and C is fast in the ESR time scale as indicated by the time-
dependent modulation of J-coupling. We attributed this
exchange to conformational freedom about both the two
chains bearing the spin labels (Scheme 4B,C).
lines 2 and 1 (see Figure 6) from which the ratio τB+C/τ can be
easily calculated by means of the following equation:
A
τB+C
τA
3(I /I )
2 1
=
2 − (I /I )
(1)
2
1
Molecular Dynamics. The main question about the nature
of the geometrical isomer remains open. This problem can be
translated into a question of distance distribution of the two
labels that are attached to the thread and the macrocyle. This
distance is determined both by coconformation adopted by the
α-CD on the paramagnetic dumbbell component and by the
conformation adopted by the covalent bonds on the chain and
on the label. Thus, stochastic dynamics (SD) simulations were
performed in water at 328 K on the two possible isomers 8a
and 9a by using the AMBER* force field (see Figure 8). The
N−O bond was modeled by the CO bond owing to their
similar geometries and the higher reliability of the parameters
As an example, in Figure 6 is reported the excellent agreement
between the experimental ESR spectrum of the bis-labeled
rotaxane recorded at 328 K and the corresponding theoretical
simulation obtained by using the three conformations model
and τB+C/τ = 0.76.
A
The τB+C/τ ratio increases as the temperature rises. This
A
dependence can be approximated by a straight line (Figure 7)
3
8,40
according to eq 2:
τB+C
τA
ΔS
R
ΔH
RT
ln
=
−
(2)
43
of the latter group in the AMBER* force field.
A Monte Carlo conformational search was carried out by
3
3
3
3
rotating all C(sp )−C(sp ), Ar−C(sp ), O−C(sp ), and N−
3
C(sp ) bonds in the thread and in the CD radical arm. The
most stable conformation found by this procedure was that one
used in the dynamic simulation.
The computational approach taken in this study is to dock
the thread inside the host cavity, energy-minimize the
mechanical complex, and carry out standard equilibrations
and production runs to derive distance distributions, p(r),
•
between the midpoint of N−O bonds. To accomplish this, we
began with a substituted cyclodextrin having a 6-fold symmetric
cavity with the thread docked in its interior. We selected a
symmetric CD cavity only as a guide for the placement of the
guest and with the awareness that the presence of the
paramagnetic arm will result in a deformation of the macrocyle
cavity upon geometry optimization. The origin of a Cartesian
reference frame was placed at the center of mass of the CD with
the z axis aligned with the C6 symmetry axis of CD. A second
reference frame was placed on the thread molecule with the z
Figure 7. Lifetime τB+C/τ ratio as a function of temperature.
A
1
9113
dx.doi.org/10.1021/ja3073484 | J. Am. Chem. Soc. 2012, 134, 19108−19117

Journal of the American Chemical Society
Article
Scheme 4. Possible Conformations of the Bis-Labeled Rotaxane
•
Figure 9. Distance distributions, p(r), between the midpoint of N−O
bonds determined by MD simulations at 328 K in water for carbonyl
analogues of 8a (“close” isomer, blue) and 9a (“far” isomer, red).
is negligible. Some years ago we reported the CW-ESR
spectrum of a paramagnetic rotaxane (10) having an octa-
methylene chain flanked by TEMPO groups as the thread and
18
α-CD as the wheel (Scheme 5).
Scheme 5
Figure 8. Clustered molecular display. Molecular dynamics of the bis-
labeled rotaxanes 8a (a) and 9a (b). Drawings include 20 structures
that refer to the 50 ns simulation. Hydrogens have been omitted for
clarity.
axis passing through the C(5)−C(6) bond of the alkyl chain.
The thread was translated along the z axis in order to have the
C(6) located in the plane passing through the six acetal oxygens
of the CD. At the end of this procedure, minimization of the
rotaxane led to the starting geometry for stochastic dynamics
This ESR spectrum recorded in water consists of only three
lines, indicating that the biradical thread is constricted in an
extended conformation in which the TEMPO fragments behave
as two single nitroxide radicals and J = 0. We repeated MD
simulation also for this structure (see Supporting Information),
and we obtained rather narrow distance distributions, p(r), in
which all distances are larger than 15 Å. This result clearly led
to the conclusion that the presence of exchange peaks in the
ESR spectrum of the bis-labeled rotaxane (Figure 3c) is not
compatible with the p(r) profile found by MD for the “far”
isomer.
(
SD) simulations. The simulations were run at 328 K for 50 ns
with time steps of 1 fs and an equilibrium time of 5 ns before
each dynamic run.
In Figure 9, estimates of p(r) from MD calculations are
reported. As expected for the two isomers, very different
distance profiles were obtained with ⟨p(r)⟩ corresponding to
1
1.35 Å and 21.96 Å for the “close” (blue) and “far” (red)
isomers, respectively. It is important to note that the
occurrence of distances smaller than 15 Å in the “far” isomer
1
9114
dx.doi.org/10.1021/ja3073484 | J. Am. Chem. Soc. 2012, 134, 19108−19117

Journal of the American Chemical Society
Article
Further evidence to this hypothesis was obtained by
were recorded on a Varian Inova operating at 125 MHz in D O
2
44−47
solutions using DSS (3-(trimethylsilyl)-1-propanesulfonic acid, sodium
salt) as an external standard. Chemical shifts are reported in parts per
million (δ scale). ROESY data were collected using a 90° pulse width
of 6.7 μs and a spectral width of 6000 Hz in each dimension,
respectively. The data were recorded in the phase sensitive mode using
a CW spin-lock field of 2 kHz, without spinning the sample.
Acquisitions were recorded at mixing times of 300 ms. Other
instrumental settings were 64 increments of 2K data points, 8 scans
performing PELDOR
measurements on rotaxane 8a.
The X-band PELDOR time traces of the bis-labeled rotaxane
were performed in H O at 40 K and did not display any
2
pronounced dipolar oscillation (see Supporting Information).
Attempts to deduce a distance distribution profile from the
analysis of the data were unsuccessful, this being an indication
of the absence of spin pairs with distances of ≥15 Å.
Many experimental and theoretical studies have been
devoted to the dependence of the exchange coupling upon
per t , 1.5 s delay time for each scan.
1
ESI-MS spectra were recorded with Micromass ZMD spectrometer
by using the following instrumental settings: positive ions; desolvation
3
1,32
structural parameters.
Nevertheless, the dependence of the
gas (N ), 230 L/h; cone gas (skimmer), 50 L/h; desolvation temp,
electron spin−spin exchange interaction between unpaired
electrons (J) on three-dimensional structural parameters
remains challenging because the modulation of the exchange
interaction, due to the large conformational freedom of the
investigated systems, limits the utility of their applications. It
has been shown that the exchange coupling constant J
2
1
20 °C; capillary voltage, 3.2 kV; cone voltage, 40 and 100 V; hexapole
extractor, 3 V.
All reagents were commercially available and were used without
9
26
further purification. Compounds 4, 1, 6-O-p-toluenesulfonyl-α-
9,25a
9
cyclodextrin,
6-azido-6-deoxy-α-cyclodextrin, 2-O-p-toluenesul-
25b 27
fonyl-α-cyclodextrin,
synthesized according to literature procedure.
and 3-azido-deoxy-α−cyclodextrin were
48
decreases exponentially with the distance and that it depends
on the orientation of the orbitals containing the unpaired
Compound 2a. Ascorbic acid (0.032 g, 0.18 mmol) and copper
sulfate (0.017 g, 0.07 mmol) were added to a solution of 6-azido-6-
deoxy-α-cyclodextrin (0.35 g, 0.35 mmol) and 1a (0.074 g, 0.35
49
electron with respect to each other. It is also generally
accepted that J is negligible when two N−O bonds are
separated by a distance larger than 8−9 Å and interact only
mmol) in 2:1 H O/i-PrOH (10 mL). The reaction mixture was stirred
2
48b,50
for 72 h at room temperature and evaporated under reduced pressure.
The crude residue was purified by column chromatography (silica gel,
eluent CH CN/H O, 85:15) and gel exclusion over Sephadex column
through space.
By combination of our ESR and MD results, it is possible to
support the last supposition. Simulation of the ESR spectrum of
3
2
G15 (0.2 g, 48.5%). δH 2.80−2.90 (m, 1 H), 3.15−3.25 (m, 1 H),
8
a recorded at 328 K afforded a ratio τB+C/τ = 0.76 (see
A
3.50−4.10 (m, 34 H), 4.19 (m, 1 H), 4.60−4.70 (m, 2 H), 4.95−5.19
+
Figure 6). This ratio corresponds to a 57% of structures in
which J is 0. In the p(r) profile obtained by MD calculations,
this corresponds to distances larger than 10.1 Å for the “close”
isomer and 21.0 Å for the “far” isomer (see Figure 9). While the
above distance limit is unrealistic for the “far” isomer, the value
found with the “close” isomer is in excellent agreement with the
hypothesis considering J = 0 when distances are larger than 8−
(m, 8 H), 8.00−8.10 (bs, 1 H). ESI-MS m/z 1208.3 [M + H] , 1230.3
+
1
[M + Na] . The H NMR spectrum of 2b was measured after Na S O
2
2
4
in situ reduction of the sample containing the nitroxide 2a. δ
2 H), 1.76 (t, J = 12 Hz, 2 H), 2.42 (t, J = 12 Hz, 2 H), 2.86 (d, J = 12
1.46 (s,
H
1
Hz, 1 H), 3.15 (d, J = 12 Hz, 1 H), 3.50−4.10 (m, 32 H), 4.19 (m, 1
H), 4.60−4.70 (m, 2 H), 4.96 (s, 1 H), 4.99 (d, J = 2.4 Hz, 1 H),
5
.03−5.10 (m, 5 H), 5.16 (s, J = 1.8 Hz, 1 H), 8.07 (s, 1 H).
Compound 3a. Ascorbic acid (0.023 g, 0.13 mmol) and copper
5
1
9
Å.
On the basis of the above consideration we can definitively
conclude that the rotaxane was isolated in the 8a isomer form
sulfate (0.032 g, 0.13 mmol) were added to a solution of 3-azidodeoxy-
α-cyclodextrin (0.13 g, 0.13 mmol) and 1a (0.027 g, 0.13 mmol) in
H O (5 mL). The reaction mixture was stirred at room temperature
2
(
“close” isomer).
overnight and evaporated under reduced pressure. The crude residue
was purified by gel exclusion over Sephadex column G15 (0.1 g,
CONCLUSIONS
63.8%). δ
H
3.50−4.15 (m, 34 H), 4.46 (m, 2 H), 5.00−5.20 (m, 6 H),
■
+
+
8
.30−8.40 (bs). ESI-MS m/z 1208.3 [M + H] , 1230.3 [M + Na] .
We reported the synthetic procedure for the preparation of a
bis-nitroxide rotaxane by click chemistry. The findings
presented here demonstrate that ESR of doubly TEMPO
labeled rotaxanes is a useful technique to unravel unidirectional
threading of α-cyclodextrin in a [2]rotaxane and to study
quantitatively the dynamics of intramolecular collisions of the
molecular fragments mechanically connected. While it is
reassuring that crystallography and solution magnetic resonance
are primary techniques to characterize mechanically interlocked
molecules, it is also clear that important new information is
available from solution studies of spin-labeled rotaxanes by
ESR.
1
The H NMR spectrum of 3b was measured after Na S O in situ
2
2
4
reduction of the sample containing the nitroxide 3a. δ 1.44 (s, 12 H),
H
1
4
.75 (m, 2 H), 2.37 (m, 2 H), 3.50−4.10 (m, 35 H), 4.13 (m, 1 H),
.46 (m, 2 H), 5.02−5.16 (m, 6 H), 8.34 (s, 1 H).
Rotaxanes 6a/7a. To an aqueous solution (5 mL) of α-CD (0.06
g, 0.062 mmol) and lactosyl azide (4) (0.04 g, 0.062 mmol) stirred for
15 min, alkyne 1a (0.015 g, 0.073 mmol), ascorbic acid (0.005 g, 0.025
mmol), and copper sulfate (0.003 g, 0.012 mmol) were added. The
mixture was stirred for 72 h at room temperature. After evaporation
under reduced pressure the crude was purified by column
chromatography (silica gel, eluent CH CN/H O, 9:1, to CH CN/
3
2
3
H O/30% NH , 8:2:1) and gel exclusion over Sephadex column G15
2
3
(
1
0.024 g, 22%). δ 1.00−1.55 (m, 12 H), 1.58−1.66 (m, 2 H) 1.72−
H
.90 (m, 4 H), 1.94−2.04 (m, 2 H), 2.20−2.28 (m, 3 H), 3.50−4.00
EXPERIMENTAL SECTION
■
(m, 47 H), 4.32−4.54 (m, 7 H), 4.60−4.66 (m, 2 H), 5.04−5.15 (m, 7
+
1
ESR spectra has been recorded by using the following instrument
settings: microwave power 0.79 mW, modulation amplitude 0.04 mT,
modulation frequency 100 kHz, scan time 180 s, 2K data points.
PELDOR experiments at 40 K of the bis-labeled rotaxane (100 μM
H), 7.90−8.10 (m, 2 H). ESI-MS m/z 1851.6 [M + Na] . The H
NMR spectrum of 6b/7b was measured after Na in situ
reduction of the sample containing the nitroxides 6a/7a. δ 1.00−1.40
S O
2 2 4
H
(m, 12 H), 1.44 (brs, 12 H), 1.50−1.65 (m, 2 H), 1.75−1.85 (m, 4 H),
1.94−2.04 (m, 2 H), 2.20−2.26 (m, 3 H), 3.50−4.00 (m, 47 H), 4.15
(m, 1 H), 4.34−4.54 (m, 7 H), 4.62−4.66 (m, 2 H), 5.04−5.15 (m, 7
H), 7.86−8.09 (m, 2 H).
in H O, 10% glycerol) were performed using a Bruker ELEXSYS E-
2
5
80 X/Q-band spectrometer equipped with a ER4118X-MD5-W1
resonator at the X band (microwave frequency 9.4 GHz) (see also
Supporting Information).
Compound 5a. This product was obtained as byproduct by
1H and 2D NMR spectra were recorded at 298 K on a Varian Inova
purification of rotaxane 6a/7a or 8a. δ 1.00−1.25 (m, 12 H), 1.80−
1.92 (m, 4 H), 2.23 (brs, 3 H), 3.52−4.10 (m, 12 H), 4.30−4.60 (m, 7
H), 5.10−5.18 (m, 0.75 H), 5.60−5.64 (m, 0.25 H), 7.87−8.16 (m, 2
H
spectrometer operating at 600 MHz in D O solutions using the
solvent peak as an internal standard (4.76 ppm). C NMR spectra
2
13
1
9115
dx.doi.org/10.1021/ja3073484 | J. Am. Chem. Soc. 2012, 134, 19108−19117

Journal of the American Chemical Society
Article
+
1
H). ESI-MS m/z 878.67 [M + Na] .The H NMR spectrum of 5b was
Joiner, C. S.; Baum, L. G.; Stoddart, J. F. J. Am. Chem. Soc. 2004, 126,
11914−11922. (i) Nijhuis, C. A.; Huskens, J.; Reinhoudt, D. N. J. Am.
Chem. Soc. 2004, 126, 12266−12267. (j) Oshikiri, T.; Takashima, Y.;
Yamaguchi, H.; Harada, A. J. Am. Chem. Soc. 2005, 127, 12186−12187.
(k) Liu, Y.; Wang, H.; Chen, Y.; Ke, C. F.; Liu, M. J. Am. Chem. Soc.
2005, 127, 657−666. (l) Murakami, H.; Kawabuchi, A.; Matsumoto,
R.; Ido, T.; Nakashima, N. J. Am. Chem. Soc. 2005, 127, 15891−15899.
measured after Na S O in situ reduction of the sample containing the
2
2
4
nitroxide 5a. δH 1.12−1.28 (m, 12 H), 1.30−1.42 (m, 12 H), 1.50−
.74 (m, 2 H), 1.80−1.95 (m, 4 H), 2.10−2.26 (m, 5 H), 3.55−4.05
1
(
=
m, 14 H), 4.30−4.70 (m, 7 H), 5.12 (d, J = 7.8 Hz, 0.75 H), 5.61 (d, J
9.6 Hz, 0.25 H), 7.88 (s, 1 H), 8.01 (s, 0.25 H), and 8.05 (s, 0.75 H).
Rotaxane 8a. To an aqueous (3 mL) solution of 2a (0.05 g, 0.041
mmol) and 4 (0.026 g, 0.041 mmol) stirred for 15 min, alkyne 1a
0.01 g, 0.05 mmol), ascorbic acid (0.003 g, 0.016 mmol), and copper
sulfate (0.002 g, 0.008 mmol) were added. The mixture was stirred for
2 h at room temperature. After evaporation under reduced pressure
(
m) Wenz, G.; Han, B.-H.; Muller, A. Chem. Rev. 2006, 106, 782−817.
n) Liu, Y.; Chen, Y. Acc. Chem. Res. 2006, 39, 681−691. (o) Park, J.
(
(
S.; Wilson, J. N.; Hardcastle, K. I.; Bunz, U. H. F.; Srinivasarao, M. J.
Am. Chem. Soc. 2006, 128, 7714−7715. (p) Klotz, E. J. F.; Claridge, T.
D. W.; Anderson, H. L. J. Am. Chem. Soc. 2006, 128, 15374−15375.
(q) Cheetham, A. G.; Hutchings, M. G.; Claridge, T. D. W.; Anderson,
H. L. Angew. Chem., Int. Ed. 2006, 45, 1596−1599. (r) Wang, Q. C.;
Ma, X.; Qu, D. H.; Tian, H. Chem.Eur. J. 2006, 12, 1088−1096.
7
the crude was purified twice by gel exclusion over Sephadex column
G15 (0.025 g, 29%). δH 1.08−1.44 (m, 12 H), 1.80−1.90 (m, 4 H),
1
.93−2.03 (m, 2 H), 2.23 (m, 3 H), 2.80−3.04 (m, 1 H), 3.50−4.20
(
m, 45 H), 4.32−4.54 (m, 8 H), 4.58−4.70 (m, 2 H), 4.88−5.24 (m, 7
+ 1
H), 7.86−8.20 (m, 3 H). ESI-MS m/z 2085.98 [M + Na] . The H
(s) Ma, X.; Wang, Q. C.; Qu, D. H.; Xu, Y.; Ji, F.; Tian, H. Adv. Funct.
NMR spectrum of 8b was measured after Na S O in situ reduction of
2
2
4
Mater. 2007, 17, 829−837. (t) Frampton, M. J.; Anderson, H. L.
the sample containing the rotaxane 8a. δ 1.08−1.38 (m, 12 H), 1.40−
H
Angew. Chem., Int. Ed. 2007, 46, 1028−1064.
1
(
.48 (m, 24 H), 1.50−1.65 (m, 4 H), 1.80−1.92 (m, 4 H), 1.95−2.03
(
(
3) Isnin, R.; Kaifer, A. E. J. Am. Chem. Soc. 1991, 113, 8188−8190.
4) Buston, J. E. H.; Marken, F.; Anderson, H. L. Chem. Commun.
m, 2 H), 2.20−2.28 (m, 5 H), 2.93−2.99 (m, 1 H), 3.30−3.34 (m, 1
H), 3.50−4.20 (m, 46 H), 4.40−4.53 (m, 7 H), 4.60−4.78 (m, 4 H),
2
(
(
001, 1046−1047.
5) (a) Park, J. W.; Song, H. J. Org. Lett. 2004, 6, 4869−4872.
b) Park, J. W.; Song, H. J.; Chang, H. J. Tetrahedron Lett. 2006, 47,
4
3
6
7
1
.98−5.21 (m, 9 H), 7.86−8.15 (m, 3 H). δ 23.9, 27.5, 32.2, 32.7,
C
3.5, 33.9, 38.9, 42.3, 42.8, 42.9, 45.5, 53.3, 53.9, 60.3, 61.9, 62.3, 62.8,
3.0, 63.4, 63.8, 71.3, 72.6, 73.5, 74.2, 74.4, 74.7, 74.9, 75.2, 75.8, 76.1,
6.5, 78.1, 79.5, 83.4, 83.9, 84.0, 85.7, 89.6, 104.2, 104.3, 104.7, 105.6,
27.2, 129.4, 146.7, 147.4, 177.7, 177.9.
3
(
831−3834.
6) Zhao, Y. L.; Dichtel, W. R.; Trabolsi, A.; Saha, S.; Aprahamian, I.;
Stoddart, J. F. J. Am. Chem. Soc. 2008, 130, 11294−11296.
7) Craig, M. R.; Hutchings, M. G.; Claridge, T. D. W.; Anderson, H.
(
ASSOCIATED CONTENT
■
L. Angew. Chem., Int. Ed. 2001, 40, 1071−1074.
*
S
Supporting Information
(8) (a) Wang, Q. C.; Qu, D. H.; Ren, J.; Chen, K.; Tian, H. Angew.
Chem., Int. Ed. 2004, 43, 2661−2665. (b) Qu, D. H.; Wang, Q. C.;
Ren, J.; Tian, H. Org. Lett. 2004, 6, 2085−2088. (c) Wang, Q. C.; Ma,
X.; Qu, D. H.; Tian, H. Chem.Eur. J. 2006, 12, 1088−1096.
H NMR, 2D-ROESY, ESR spectra, PELDOR decay of 8a, MD
calculations details, atom coordinates for the optimized
geometries of carbonyl analogues of 8a, 9a, and 10. This
material is available free of charge via the Internet at http://
pubs.acs.org.
(d) Tomatsu, I.; Hashidzume, A.; Harada, A. Angew. Chem., Int. Ed.
2
006, 45, 4605−4608. (e) Oshikiri, T.; Takashima, Y.; Yamaguchi, H.;
Harada, A. Chem.Eur. J. 2007, 13, 7091−7098.
9) Chwalek, M.; Auzely, R.; Fort, S. Org. Biomol. Chem. 2009, 7,
(
́
AUTHOR INFORMATION
■
1
680−1688.
Corresponding Author
(
10) Steinhoff, H. J. Front. Biosci. 2002, 7, 97−110.
marco.lucarini@unibo.it
(11) Eaton, S. S.; Eaton, G. R. In Biological Magnetic Resonance;
Notes
Berliner, L. J., Eaton, S. S., Eaton, G. R., Eds.; Kluwer Academic/
Plenum Publishers: New York, 2000; Vol 19, pp 2−28.
The authors declare no competing financial interest.
(12) Tsvetkov, Yu. D. In Biological Magnetic Resonance; Berliner, L. J.,
Bender, C. J., Eds.; Kluwer Academic/Plenum Publishers: New York,
004; Vol. 21, pp 385−433.
13) Jeske, G.; Pannier, M.; Spiess, H. W. In Biological Magnetic
Resonance; Berliner, L. J., Eaton, S. S., Eaton, G. R., Eds.; Kluwer
Academic/Plenum Publishers: New York, 2000; Vol. 19, pp 493−512.
(14) Borbat, P. P.; Freed, J. F. In Biological Magnetic Resonance;
Berliner, L. J., Eaton, S. S., Eaton, G. R., Eds.; Kluwer Academic/
Plenum Publishers: New York, 2000; Vol. 19, pp 383−460.
(15) Krock, L.; Shivanyuk, A.; Goodin, D. B.; Rebek, J. Chem.
Commun. 2004, 272−273.
(16) (a) Araki, K.; Nakamura, R.; Otsuka, H.; Shinkai, S. J. Chem.
Soc., Chem. Commun. 1995, 2121−2122. (b) Rajca, A.; Mukherjee, S.;
Pink, M.; Rajca, S. J. Am. Chem. Soc. 2006, 128, 13497−13507.
(17) Chechik, V.; Ionita, G. New J. Chem. 2007, 31, 1726−1729.
(18) Mezzina, E.; Fanì, M.; Ferroni, F.; Franchi, P.; Menna, M.;
Lucarini, M. J. Org. Chem. 2006, 71, 3773−3777.
(19) (a) Mileo, E.; Casati, C.; Franchi, P.; Mezzina, E.; Lucarini, M.
Org. Biomol. Chem. 2011, 9, 2920−2924. (b) Mezzina, E.; Cruciani, F.;
Pedulli, G. F.; Lucarini, M. Chem.Eur. J. 2007, 13, 7223−7233.
(20) Pievo, R.; Casati, C.; Franchi, P.; Mezzina, E.; Bennati, M.;
Lucarini, M. ChemPhysChem 2012, 13, 2659−2661.
ACKNOWLEDGMENTS
This work was supported by MIUR (Contract 2008KRBLP5)
and Alma Mater Studiorum. University of Bologna, Italy.
■
2
(
REFERENCES
■
(
1) (a) Ogino, H. J. Am. Chem. Soc. 1981, 103, 1303−1304.
b) Manka, J. S.; Lawrence, D. S. J. Am. Chem. Soc. 1990, 112, 2440−
442. (c) Rao, T. V.; Lawrence, D. S. J. Am. Chem. Soc. 1990, 112,
614−3615. (d) Wylie, R. S.; Macartney, D. H. J. Am. Chem. Soc. 1992,
14, 3136−3138. (e) Harada, A.; Li, J.; Kamachi, M. Chem. Commun.
997, 1413−1414. (f) Anderson, S.; Claridge, T. D. W.; Anderson, H.
(
2
3
1
1
L. Angew. Chem., Int. Ed. Engl. 1997, 36, 1310−1313. (g) Murakami,
H.; Kawabuchi, A.; Kotoo, K.; Kunitake, M.; Nakashima, N. J. Am.
Chem. Soc. 1997, 119, 7605−7606. (h) Nepogodiev, S. A.; Stoddart, J.
F. Chem. Rev. 1998, 98, 1959−1976. (i) Raymo, F. M.; Stoddart, J. F.
Chem. Rev. 1999, 99, 1643−1664.
(
3
3
(
2) (a) Kawaguchi, Y.; Harada, A. J. Am. Chem. Soc. 2000, 122,
797−3798. (b) Onagi, H.; Easton, C. J.; Lincoln, S. F. Org. Lett. 2001,
, 1041−1044. (c) Harada, A. Acc. Chem. Res. 2001, 34, 456−464.
d) Stanier, C. A.; Alderman, S. J.; Claridge, T. D. W.; Anderson, H. L.
Angew. Chem., Int. Ed. 2002, 41, 1769−1772. (e) Onigi, H.; Carrozzini,
B.; Cascarano, G. L.; Easton, C. J.; Edwards, A. J.; Lincoln, S. F.; Rae,
A. D. Chem.Eur. J. 2003, 9, 5971−5977. (f) Wang, Q. C.; Qu, D. H.;
Ren, J.; Chen, K.; Tian, H. Angew. Chem., Int. Ed. 2004, 43, 2661−
(21) (a) Zhao, N.; Lloyd, G. O.; Scherman, O. A. Chem. Commun.
2012, 48, 3070−3072. (b) Lucas, D.; Minami, T.; Iannuzzi, G.; Cao,
L.; Wittenberg, J. B.; Anzenbacher, P.; Isaacs, L. J. Am. Chem. Soc.
2011, 133, 17966−17976.
(22) Bellia, F.; La Mendola, D.; Pedone, C.; Rizzarelli, E.; Saviano,
M.; Vecchio, G. Chem. Soc. Rev. 2009, 38, 2756−2781.
2
665. (g) Easton, C. J.; Lincoln, S. F.; Barr, L.; Onagi, H. Chem.Eur.
J. 2004, 10, 3120−3128. (h) Nelson, A.; Belitsky, J. M.; Vidal, S.;
1
9116
dx.doi.org/10.1021/ja3073484 | J. Am. Chem. Soc. 2012, 134, 19108−19117

Journal of the American Chemical Society
Article
(
23) (a) Paton, R. M.; Kaiser, E. T. J. Am. Chem. Soc. 1970, 92,
(45) Milov, A. D.; Ponomarev, A. B.; Tsvetkov, Y. D. Chem. Phys.
Lett. 1984, 110, 67−72.
4
723−4725. (b) Ionita, G.; Chechik, V. Org. Biomol. Chem. 2005, 3,
3096−3098. (c) Bardelang, D.; Rockenbauer, A.; Jicsinszky, L.; Finet,
(46) Martin, R. E.; Pannier, M.; Diederich, F.; Gramlich, V.; Hubrich,
M.; Spiess, H. W. Angew. Chem., Int. Ed. 1998, 37, 2833−2837.
(47) Jeschke, G.; Koch, A.; Jonas, U.; Godt, A. J. Magn. Reson. 2002,
J.-P.; Karoui, H.; Lambert, S.; Marque, S. R. A.; Tordo, P. J. Org. Chem.
2
006, 71, 7657−7667. (d) Bardelang, D.; Finet, J.-P.; Jicsinszky, L.;
1
(
55, 72−82.
Karoui, H.; Marque, S. R. A.; Rockenbauer, A.; Rosas, R.; Charles, L.;
48) (a) Yamaguchi, K.; Okumura, M.; Maki, J.; Noro, T.;
Monnier, V.; Tordo, P. Chem.Eur. J. 2007, 13, 9344−9354.
Namimoto, H.; Nakano, M.; Fueno, T.; Nakasuji, K. Chem. Phys.
Lett. 1992, 190, 353−360. (b) Closs, G. L.; Forbes, M. D. E.;
Piotrowiak, P. J. Am. Chem. Soc. 1992, 114, 3285−3294.
(
24) Franchi, P.; Fanì, M.; Mezzina, E.; Lucarini, M. Org. Lett. 2008,
1
0, 1901−1904.
25) (a) Melton, L. D.; Slessor, K. N. Carbohydr. Res. 1971, 18, 29−
7. (b) Teranishi, K.; Watanabe, K.; Hisamatsu, M; Yamada, T. J.
Carbohydr. Chem. 1998, 17, 489−494.
26) Luo, J.; Pardin, C.; Lubell, W. D.; Zhu, X. X. Chem. Commun.
007, 2136−2138.
27) Martina, K.; Trotta, F.; Robaldo, B.; Belliardi, N.; Jicsinszkyc, L.;
(
3
(49) Barone, V.; Bencini, A.; Di Matteo, A. J. Am. Chem. Soc. 1997,
1
(
3
(
19, 10831−10837.
50) Fritscher, J.; Beyer, M.; Schiemann, O. Chem. Phys. Lett. 2002,
64, 393−401.
51) Another important example supporting this view is given by the
(
2
(
5
2
octapeptide reported by Sartori et al., characterized by three
Cravotto, G. Tetrahedr. Lett. 2007, 48, 9185−9189. The reported
procedure for the preparation of the azido derivative was slightly
modified as follows. Sodium azide (0.39 g, 6.0 mmol) was added to a
solution of mono-2-O-p-tosyl-α-CD (0.75 g, 0.61 mmol) in DMF (10
mL). The solution was stirred at 100 °C overnight and evaporated
under reduced pressure. The crude residue was purified by column
chromatography (CH CN/H O, 8:2), affording the product (0.46 g,
nitroxides residues. The octapeptide forms in acetonitrile a rigid 3 -
1
0
helix structure in which they are placed in a linear configuration one
on top of the other after one helix turn. The distance between the
midpoints of the farthest N−O bonds is 12 Å, and the corresponding
exchange constant, J is close to 0.
(52) Sartori, E.; Corvaja, C.; Oancea, S.; Formaggio, F.; Crisma, M.;
3
2
Toniolo, C. ChemPhysChem 2005, 6, 1472−1475.
7
1
5%). δ (300 MHz, DMSO-d ) 3.25−3.35 (m, 12 H), 3.45−4.00 (m,
H
6
8 H), 4.40−4.55 (m, 6 H), 4.79 (bs, 6 H), 5.40−5.70 (m, 11 H). MS-
+
ESI: m/z = 1020 [M + Na] .
28) It is known that nucleophilic substitution of 2-(p-toluenesulfo-
(
nyl)αCD afforded a monosubstituted cyclodextrin substituted at the 3
position. For a thorough discussion see ref 27.
(
29) (a) Spevak, W.; Dasgupta, F.; Hobbs, C. J.; Nagy, J. O. J. Org.
Chem. 1996, 61, 3417−3422. (b) Lockhoff, O. Angew. Chem., Int. Ed.
Engl. 1991, 30, 1611−1620. (c) Larpent, C.; Laplace, A.; Zemb, T.
Angew. Chem., Int. Ed. 2004, 43, 3163−3167.
(
30) According to NMR results, only one radical species was
observed indicating the presence of only one isomer. As already
observed with the rotaxane mixture 6a−7a, the possible presence of
another isomer could not be detected because the spectral parameters
are not sufficiently different.
(
31) (a) Miick, S. M.; Martinez, G. V.; Fiori, W. R.; Todd, A. P.;
Millhauser, G. L. Nature 1992, 359, 653−655. (b) Fiori, W. R.;
Lundberg, K. M.; Millhauser, G. L. Nat. Struct. Biol. 1994, 1, 374−377.
(
c) Fiori, W. F.; Miick, S. M.; Millhauser, G. L. Biochemistry 1993, 32,
1
(
1957−11962.
32) Hanson, P.; Millhauser, G.; Formaggio, F.; Crisma, M.; Toniolo,
C. J. Am. Chem. Soc. 1996, 118, 7618−7625.
33) Carlotto, S.; Cimino, P.; Zerbetto, M.; Franco, L.; Corvaja, C.;
(
Crisma, M.; Formaggio, F.; Toniolo, C.; Polimeno, A.; Barone, V. J.
Am. Chem. Soc. 2007, 129, 11248−11258.
(
34) Ionita, G.; Meltzer, V.; Pincuc, E.; Chechik, V. Org. Biomol.
Chem. 2007, 5, 1910−1914.
35) Yi, S.; Captain, B.; Ottaviani, M. F.; Kaifer, A. E. Langmuir 2011,
7, 5624−5632.
36) Nakabayashi, K.; Kawano, M.; Yoshizawa, M.; Ohkoshi, S.;
(
2
(
Fujita, M. J. Am. Chem. Soc. 2004, 126, 16694−16695.
(
(
37) Luckhurst, G. R. Mol. Phys. 1966, 30, 543−550.
38) Parmon, V. N.; Kokorin, A. I.; Zhidomirov, G. M.; Zamarev, K.
I. Mol. Phys. 1975, 30, 695−701.
2
2
(
39) I /I ≈ d (ΔH ) /d (ΔH ) where d1 and d2 are the line
2
1
2
2
1
1
amplitudes and ΔH and ΔH are the corresponding widths.
2
1
(
40) Szydłowska, J.; Pietrasik, K.; Głaz, L.; Kaim, A. Chem. Phys. Lett.
008, 460, 245−252.
41) Lucarini, M.; Luppi, B.; Pedulli, G. F.; Roberts, B. P. Chem.
Eur. J. 1999, 5, 2048−2054.
42) Yu, Y.; Cai, W.; Chipot, C.; Sun, T.; Shao, X. J. Phys. Chem. B
008, 112, 5268−5271.
43) Franchi, P.; Lucarini, M.; Mezzina, E.; Pedulli, G. F. J. Am.
Chem. Soc. 2004, 126, 4343−4354.
44) Milov, A. D.; Salikhov, K. M.; Shirov, M. D. Fiz. Tverd. Tela
981, 23, 975−982.
2
(
(
2
(
(
1
1
9117
dx.doi.org/10.1021/ja3073484 | J. Am. Chem. Soc. 2012, 134, 19108−19117