Processive rotaxane systems. Studies on the mechanism and control of the threading process

Article abstract of DOI:10.1021/ja068714i

The threading behavior of a zinc analogue of a previously reported processive manganese porphyrin catalyst onto a series of polymers of different lengths is reported. It is demonstrated that the speed of the threading process is determined by the opening of the cavity of the toroidal porphyrin host, which can be tuned with the help of axial ligands that coordinate to the metal center in the porphyrin.

Full text of DOI:10.1021/ja068714i

Published on Web 04/07/2007
Processive Rotaxane Systems. Studies on the Mechanism
and Control of the Threading Process
Pilar Hidalgo Ramos, Ruud G. E. Coumans, Alexander B. C. Deutman,
Jan M. M. Smits, Rene de Gelder, Johannes A. A. W. Elemans,
Roeland J. M. Nolte,* and Alan E. Rowan*
Contribution from the Institute for Molecules and Materials, Radboud UniVersiteit Nijmegen,
ToernooiVeld 1, 6525 ED Nijmegen, The Netherlands
Received January 8, 2007; E-mail: r.nolte@science.ru.nl; a.rowan@science.nl
Abstract: The threading behavior of a zinc analogue of a previously reported processive manganese
porphyrin catalyst onto a series of polymers of different lengths is reported. It is demonstrated that the
speed of the threading process is determined by the opening of the cavity of the toroidal porphyrin host,
which can be tuned with the help of axial ligands that coordinate to the metal center in the porphyrin.
Introduction
Processive enzymes such as DNA polymerase III and
λ-exonuclease are beautiful examples of molecular nanotech-
nology in nature. These enzymes are highly successful in
performing their tasks because they work in a pseudorotaxane
topology, in which the enzyme threads on the DNA strand and
slides along the chain while performing numerous rounds of
catalysis before the complex dissociates.1 These biocatalysts
have served as a source of inspiration for the development of
the first synthetic processive catalytic rotaxane.2 By constructing
a cavity-containing porphyrin macrocycle H2-1 (Figure 1a),3
which after insertion of a manganese center (Mn-1) was
threaded onto a polybutadiene polymer, it was possible to mimic
the catalytic action of the natural systems. It was shown that
the macrocyclic catalyst can move along this polymer thread
while, in the presence of an oxygen donor, catalyzing the
conversion of the polymer double bonds into the corresponding
epoxide functions (Figure 1c). The “outside” of the porphyrin
catalyst is blocked by a bulky axial ligand, allowing the
conversion of the substrate to take place preferentially (∼80%)
within the cavity.2,4 In more recent work, the threading and
dethreading process of H2-1 on polymers of different lengths
was investigated.5,6 Kinetic and thermodynamic studies revealed
that the barrier that has to be overcome is of entropic origin
and most likely related to the stretching and unfolding of the
polymer chain. After finding the end of the polymer chain, the
macrocycle has to move along the chain over a certain critical
length before the threading process can continue. As a conse-
quence, the rate-limiting step is dependent on the polymer
Figure 1. (a) Macrocycles H2-1 and Zn-1; (b) crystal structure of H2-1
(including a CHCl3 solvent molecule); (c) schematic representation of the
catalytic action of the processive catalyst.
length. This proposed mechanism is similar to the nucleation
model described by Muthukumar for the translocation of a
polymer through a hole.7 Further conclusions from the kinetic
data suggested that the aforementioned processive enzyme
mimic Mn-1 most likely operates by randomly sliding along
the polymer thread while performing its catalytic function. Here
(6) Although not related to processive catalysis, several authors have studied
the threading and dethreading, and the interaction between macrocycles
and polymers: (a) Mason, P. E.; Bryant, W. S.; Gibson, H. W. Macro-
molecules, 1999, 37, 1559-1569. (b) Hodge, P.; Monvisade, P.; Owen, G.
J.; Heatley, F.; Pang, Y. New J. Chem. 2000, 24, 703-709. (c) Inoue, Y.;
Miyauchi, M.; Nakajima, H.; Takashima, Y.; Yamaguchi, H.; Harada, A.
J. Am. Chem. Soc. 2006, 128, 8994-8995. (d) Mason, P. E.; Parsons, I.
W.; Tolley, M. S. Angew. Chem., Int. Ed. Engl. 1996, 35, 2238-2241. (e)
Mason, P. E.; Parsons, I. W.; Tolley, M. S. Polymer 1998, 39, 3981-
3991. (f) Venturi, M.; Dumas, S.; Balzani, V.; Cao, J.; Stoddart, J. F. New
J. Chem. 2004, 28, 1032-1037. (g) Wenz, G.; Gruber, C.; Keller, B.;
Schilli, C.; Albuzat, T.; Muller, A. Macromolecules 2006, 39, 8021-8026.
(1) Wang, J.; Sattar, A.; Wang, C. C.; Karam, J. D.; Konigsberg, W. H.; Steitz,
T. A. Cell 1997, 89, 1087-1099.
(2) Thordarson, P.; Bijsterveld, E. J. A.; Rowan, A. E.; Nolte, R. J. M. Nature
2003, 424, 915-918.
(3) Rowan, A. E.; Aarts, P. P. M.; Koutstaal, K. W. M. Chem. Commun. 1998,
611-612.
(4) Elemans, J. A. A. W.; Bijsterveld, E. J. A.; Rowan, A. E.; Nolte, R. J. M.
Chem. Commun. 2000, 2443-2444.
(5) Coumans, R. G. E.; Elemans, J. A. A. W.; Rowan, A. E.; Nolte, R. J. M.
Proc. Natl. Acad. Sci. U.S.A. 2006, 103, 19647-19651.
9
10.1021/ja068714i CCC: $37.00 © 2007 American Chemical Society
J. AM. CHEM. SOC. 2007, 129, 5699-5702
5699

A R T I C L E S
Hidalgo Ramos et al.
introduction of a zinc ion into the macrocyclic host leads to a
much slower threading rate for each of the three polytetrahy-
drofuran polymers studied (Table 1). Our first impression was
that this remarkable difference was caused by coordination of
the zinc ion to the oxygen atoms in the polymer chains, since
it is well-known that this metal center, when present in a
porphyrin, prefers to bind a fifth ligand. In this way, the
movement of the porphyrin macrocycle will slow down, and
the average time to reach the viologen trap will increase because
of the “stickiness” of the metal center to the multiple oxygen
atoms of the polymer chain. However, when the ∆Gqon -values
for the threading of H2-1 and Zn-1 where plotted versus the
polymer length (Figure 3b), similar slopes were obtained (76
( 21 J mol-1 nm-1 for H2-1 and 64 ( 17 J mol-1 nm-1 for
Zn-1), which indicates that the energy penalty for the macro-
cycle to move along the polymer thread is apparently the same
for both macrocycles. The above-mentioned zinc-oxygen bind-
ing can therefore not be the major reason for the slower
threading of Zn-1.
Figure 2. (a) Structure of polymers 2-4; (b) schematic representation of
the threading of host 1 on polymers 2-4.
we describe studies on the threading behavior of Zn-1,8 an
analogue of Mn-1, which has been designed to give more insight
in the initial steps of the threading process and the role of the
cavity in the motion along the thread. More specifically, we
show that the initial binding of the catalyst to the polymer chain
can be controlled by the addition of ligands that axially
coordinate to the porphyrin metal. Coordination on the outside
of Zn-1 facilitates threading, whereas coordination on the inside
of the cavity inhibits the threading process.
To obtain more insight in the threading behavior we
determined the thermodynamic parameters of the process by
carrying out a series of threading experiments of Zn-1 on
polymer 2 at different temperatures. From the Eyring plots,
values of ∆Hq ) 24 ( 4 kJ mol-1 and ∆Sq ) -93 ( 10 J
on
on
K-1 mol-1, were calculated. When these numbers are compared
to the values measured for H2-1 and 2 (∆Hq ) 20 kJ mol-1
on
and ∆Sqon ) -88 J K-1 mol-1),5 it can be concluded that, within
experimental error, the enthalpy and the entropy values of the
movement along the chain are identical. We may conclude
therefore that the mechanism of this motion is the same for
both compounds and that Zn-1 has no additional interactions
with the thread. The difference of 6 kJ mol-1 in ∆Gqon between
H2-1 and Zn-1 (Figure 3b) is apparently related to the initial
threading process, which is most likely the blocking of the cavity
of Zn-1 by a molecule that has to be removed before the
threading can start. Since all the threading experiments were
carried out in a mixture of chloroform and acetonitrile (1:1,
v/v), the latter solvent may bind to the zinc ion,9 preferentially
on the inside of the macrocycle, acting as a competitive species
for the accommodation of other guests in the cavity. If this is
the case, the dimensions of the cavity are reduced to such an
extent that it becomes difficult for the polymer chain to thread
through the hole of Zn-1. To investigate if such a ligand
coordination is indeed a factor that influences the threading rate,
we performed several experiments in the presence of axial
ligands that bind with a higher affinity to the zinc porphyrin
than the acetonitrile molecule. From previous work, it was
known that a 4-tert-butylpyridine (tbpy) ligand binds to Zn-1
on the outside of the cavity (Figure 4a). Since this binding to
the outside will release the coordinated acetonitrile molecule
from the inside of the macrocycle, the cavity is opened and
Results and Discussion
We recently succeeded in solving the X-ray structure of H2-1
(Figure 1b), which allowed us to determine the diameter of its
cavity, being 8 Å. A hole of these dimensions is perfectly suited
to accommodate a polymer such as polytetrahydrofuran as a
thread (which has a diameter of 4.7 Å). In the current study we
have employed three polytetrahydrofuran derivatives of different
lengths (2-4),5 which are selectively blocked at one end by a
bulky stopper group (Figure 2a). Directly linked to the blocking
group, the polymers contain a viologen function (1,1′-dialkyl-
4,4′-bipyridinium ion), which is known to have a high binding
affinity for the cavity of 1 (Ka ) 106 - 107 M-1).3
Consequently, the porphyrin-containing macrocycle has to
completely traverse the polymer chain to reach the viologen
binding site, which traps the macrocycle (Figure 2b). The
complexation of 1 to the viologen trap is monitored by
fluorometric techniques, that is, the porphyrin emission is
quenched by the viologen once it is accommodated inside the
cavity of 1. In the case of H2-1, the threading processes as
followed by fluorescence quenching initially followed second-
order kinetics, as plots of 1/[H2-1] versus time gave straight
lines. From the slope of these lines, kon (rate constant for the
threading process) and ∆Gq (the difference in free energy
between the ground state of the uncomplexed components and
the transition state) were obtained.5
on
(9) A solution of Zn-1 in CHCl3 was titrated with increasing amounts of CH3-
CN and the process was followed by UV-vis spectroscopy. A similar
titration was carried out with zinc(II) tetrakis (2-methoxyphenyl) porphyrin
and CH3CN. From a comparison of the data it was concluded that the
acetonitrile molecule binds significantly stronger to Zn-1 than to the
reference compound, suggesting strong complexation of acetonitrile inside
the cavity.
(10) Using larger amounts of tbpy might significantly affect the polarity of the
system, causing a change in the affinity of the viologen moiety for the
cavity of 1. A blanc experiment using H2-1 under the applied conditions,
however, did not reveal any notable effect on the threading rate constant
(kon).
The fluorescence curves obtained for the threading processes
of Zn-1 onto polymers 2-4 are shown in Figure 3a. Analysis
of the data showed that these threading processes also follow
second-order kinetics, and when the calculated kon-values are
compared to those found for H2-1, it is evident that the
(7) Muthukumar, M. Phys. ReV. Lett. 2001, 86, 3188-3191.
(8) Elemans, J.; Claase, M. B.; Aarts, P. P. M.; Rowan, A. E.; Schenning, A.;
Nolte, R. J. M. J. Org. Chem. 1999, 64, 7009-7016.
9
5700 J. AM. CHEM. SOC. VOL. 129, NO. 17, 2007

Processive Rotaxane Systems
A R T I C L E S
Figure 3. (a) Normalized fluorescence intensity of Zn-1 vs time measured after the addition of 1 equiv of polymers 2-4 to the solution of this host ([Zn-1]
) [2-4] ≈ 0.7 µM, CHCl3/MeCN (1/1, v/v)). (b) ∆Gqon-values for H2-1 ([) and for Zn-1 (2) plotted vs the length of 2-4.
Table 1. Measured kon (M-1 s-1) and ∆Gq (kJ mol-1) Values for
on
the Threading Processes Involving Macrocycles H2-1 and Zn-1 and
Polymers 2-4 at 296Ka
Zn
−
1
+
tbpy
Zn
−
1
+
py
Zn−1 + py
H2
−1
Zn−1
(50% bound)
(50% bound) (99% bound)
polymer
kon
∆
Gq
kon
∆
Gq
kon
Gq
∆
∆
Gq Gq
kon kon
on
∆
on
on
on
on
2
3
4
36000 47 2300 53 4690 52 700 56 34 64
19000 48 1400 54 2560 53 560 57 21 65
12000 49
980 55 1990 54 325 58 12 66
a Estimated error on kon ) 30%.
Table 2. Calculated Ka (M-1) for the Binding of H2-1 and Zn-1 to
Polymers 2-4 at 296Ka
Zn
−
1
+
tbpy
Zn
−
1
+
py
Zn
−1 + py
Figure 4. (a) Schematic representation of the complexation of tbpy to the
“outside” of Zn-1; (b) schematic representation of the cavity of Zn-1 filled
with py.
polymer
H2−1
Zn−1
(50% bound)
(50% bound)
(99% bound)
2
3
4
2.8 × 107 2.7 × 106 1.49 × 107 9.72 × 105 1.72 × 105
2.9 × 107 2.8 × 106 7.10 × 106 1.23 × 106 1.72 × 105
3.0 × 107 2.8 × 106 9.50 × 106 1.15 × 106 2.60 × 105
or absence of axial ligands are compiled, illustrating the clear
difference in time required for the toroidal host to reach the
viologen trap at the blocked end of the polymer. From the plots
of the ∆Gqon-values versus the polymer lengths, the differences
in the energy barriers for the various threading processes are
visible (Figure 5b). The lines that can be drawn through the
data points have similar slopes, indicating similar sliding
energies for the threading processes. Once equilibrium is
achieved, it is possible to estimate the binding constants of
1 with polymers 2-4 under the different conditions studied.
Table 2 shows that the affinity of macrocycles H2-1 and Zn-1
for the viologen moiety changes depending on the experimental
conditions. After zinc insertion, there is a decrease in the
association constant owing to the abovementioned solvent
competition. A similar effect is observed when py is present,
although it should be noted that after removal of the axial ligands
upon threading on the polymer, they will probably bind to
the zinc ion on the outside of the cavity. An increased value of
Ka is obtained in the presence of tbpy, since this ligand
eliminates the solvent competition and in addition causes a
positive allosteric effect on Zn-1, enhancing the binding for
viologens as we showed previously.11 The measured values are
in close accordance to the ones previously obtained for the
complexation of 1,1′-dimethyl-4,4′-bipyridinium ion in the
cavity of Zn-1.11
a Estimated error on Ka ) 50%.
ready to thread. Because of the rather low association constant
between Zn-1 and tbpy (Ka ) 250 M-1), ∼6300 equiv of the
ligand were used in the threading experiments to ensure an
approximate 50% binding to the porphyrin.10 Upon the addition
of each of the polymers 2-4 to Zn-1 in the presence of tbpy in
a CH3CN/CHCl3 solution (1:1, v/v), a faster quenching of
porphyrin fluorescence was observed (Table 1), indicating that
the rate of the threading of Zn-1 is indeed related to the release
of an axially binding solvent molecule (acetonitrile) from the
cavity. To obtain additional evidence for this hypothesis, the
cavity of Zn-1 was intentionally blocked by the addition of
pyridine (py) to the host (Figure 4b). As a result of stabilizing
π-π stacking interactions and a cavity effect, this smaller axial
ligand binds very strongly within the cavity of Zn-1 (Ka ) 1.1
× 105 M-1 in CDCl3),8 and at low concentrations (∼10-6 M)
only a few equivalents are needed to achieve a nearly quantita-
tive binding. In a first experiment, a calculated amount of
pyridine was added so that on average 50% of the cavities of
Zn-1 were occupied, and upon the addition of each of the
polymers this indeed resulted into a much slower threading rate
(Figure 5a). Upon the addition of more equivalents of py to the
system (so that ∼99% of the cavities were occupied), the kon-
values dropped dramatically, and a nearly complete blocking
of the threading process was achieved (Table 1). In Figure 5a,
all threading experiments of Zn-1 on polymer 2 in the presence
(11) Thordarson, P.; Coumans, R. G. E.; Elemans, J.; Thomassen, P. J.; Visser,
J.; Rowan, A. E.; Nolte, R. J. M. Angew. Chem., Int. Ed. 2004, 43, 4755-
4759.
9
J. AM. CHEM. SOC. VOL. 129, NO. 17, 2007 5701

A R T I C L E S
Hidalgo Ramos et al.
Figure 5. (a) Normalized fluorescence intensity of H2-1 and of Zn-1 in the absence and presence of different ligands measured vs time after the addition
of 1 equiv of polymer 2 ([M-1] ) [2] ≈ 0.7 µM-1, CHCl3/MeCN (1/1, v/v)). (b) ∆Gqon-values for the threading of H2-1 ([) and Zn-1 in the absence of
ligand (2) and in the presence of tbpy (/) or py (O for 50% bound; b for 99% bound) plotted vs the length of 2-4.
Table 3. Number Averaged Molecular Weight (Mn), Degree of
Polymerization (DPn), Polymer Length (Ln) and Polydispersity
Index (PDI) of Polymer Appended Viologens 2-4
Table 4. Excitation and Emission Wavelengths Used in the
Different Threading Experiments Involving Macrocycles H2-1 and
Zn-1 and Polymers 2-4 at 296K
1
polymer
Mn (g mol-
)
DPn
Ln (nm)
PDI
Zn
−1
+
tbpy
Zn
−1
+
py
Zn−1 + py
H2
−
1
Zn
−
1
(50% bound)
(50% bound)
(99% bound)
2
3
4
3140
5030
7040
34
60
88
21
37
54
1.03
1.03
1.03
λexc/nm
λem/nm
421
643
427
607
429
609
428
608
428
610
reaction becomes 2A f C. The integrated rate law for this reaction is
[A]-1 ) kont. By now plotting [1]-1 versus time, a linear plot is obtained
with the slope equal to kon which is obtained via the least-squares
method.
Determination of Ka. Since kon is known, it can be now used to
also calculate Ka by fitting the kinetic data to12
Conclusion
In conclusion, we have obtained useful insight in the threading
of a toroidal host on a polymer chain and its movement toward
a viologen trap. Although the process is clearly dependent on
the polymer length, there is now substantial evidence that the
rate of threading of Zn-1 and therefore also of our processive
catalyst2 can be controlled by coordination of axial ligands. In
contrast to the catalytic system, which interacts with the
polymer, the “cost of movement”, 70 J mol-1 nm-1, is the same
for both metal and metal-free toroids. The described approach
offers a novel method to study polymer mechanics, and current
studies are directed at larger and more functional polymer
threads.
1 - exp(kon(p - q)t)
[C] ) p
(1)
q
1 - exp(kon(p - q)t)
p
with p ) [C]eq and q ) [A]o[B]o/[C]eq, where
1
2
1
Ka
[C]eq
)
[A]o + [B]o +
-
(
2
1
[A]o + [B]o +
- 4([A]o[B]o)
)
(
x
)
Ka
Experimental Section
Acknowledgment. The Dutch National Research School for
Combination Catalysis (NRSC-C) and the Council for the
Chemical Sciences of The Netherlands Organization for Sci-
entific Research (CW-NWO) are acknowledged for financial
support to J.A.A.W.E. (Veni grant), A.E.R. (Vidi grant) and
R.J.M.N. (Top grant).
Polymer appended viologens 2-4 were synthesized according to a
previously reported procedure.5 The properties of the polymers are
summarized in Table 3.
Fluorescence experiments were performed on a Perkin-Elmer LS50B
luminescent spectrometer equipped with a thermostatted cuvette holder.
The excitation and emission slits were both set to 10 nm. All the
experiments were done in a CHCl3/MeCN (1:1, v/v) solvent mixture.
The threading kinetics were measured using a time-drive application
of the spectrometer software. Table 4 shows the wavelength at which
each sample was excited and the emission wavelength that was recorded
versus time.
Supporting Information Available: Crystallographic data of
H2-1. This material is available free of charge via the Internet
at http://pubs.acs.org.
JA068714I
Typically, to a weighed solution of ∼0.7 µM of 1, 1 equiv of 2-4
(obtained from a ∼60 µM stock solution) was added and mixed. After
the mixing time (∼10 s) the measurement was started.
(12) See (a) Koppelman, S. J.; vanHoeij, M.; Vink, T.; Lankhof, H.; Schiphorst,
M. E.; Damas, C.; Vlot, A. J.; Wise, R.; Bouma, B. N.; Sixma, J. J. Blood
1996, 87, 2292-2300. (b) Asakawa, M.; Ashton, P. R.; Ballardini, R.;
Balzani, V.; Belohradsky, M.; Gandolfi, M. T.; Kocian, O.; Prodi, L.;
Raymo, F. M.; Stoddart, J. F.; Venturi, M. J. Am. Chem. Soc. 1997, 119,
302-310.
Determination of kon. The data were analyzed by assuming that
the first part of the curve is a simple second-order reaction: A + B f
C. All experiments were carried out with [A] ) [B], therefore the
9
5702 J. AM. CHEM. SOC. VOL. 129, NO. 17, 2007