Light-driven transport of a molecular walker in either direction along a molecular track

Article abstract of DOI:10.1002/anie.201004779

Walk this way! A walker unit is able to walk in either direction along a four-foothold molecular track, depending on the sequence of application of four external stimuli: acid, base, UV light, and visible light in the presence of iodine (see picture). The

Full text of DOI:10.1002/anie.201004779

DOI: 10.1002/anie.201004779
Molecular Devices
Light-Driven Transport of a Molecular Walker in Either Direction
along a Molecular Track **
Michael J. Barrell, Araceli G. Campaꢀa, Max von Delius, Edzard M. Geertsema, and
David A. Leigh*
Nature uses bipedal motor proteins that “walk” down intra-
cellular tracks to perform essential tasks in a variety of key
biological processes.[1] Although the molecular mechanisms
through which these fascinating linear motors operate are
beginning to be understood,[2] there are still few synthetic
mimics that exhibit the most important characteristics of
natural motors, namely repetitive, progressive, processive,
and directional walker transport along a molecular track.
Several walker–track systems based on DNA have been
described,[3] and recently our research group reported a small-
molecule system,[4] in which the migration of a walker unit
along a four-foothold track could be biased in one direction
through an information ratchet type of Brownian ratchet
mechanism.[5,6]
Herein we report the design, synthesis, and operation of a
small-molecule walker–track conjugate, in which the walker
can be transported in either direction along a four-foothold
molecular track (roughly 1.5 times more likely to take a step
in one direction than the other), depending on the sequence
of four applied stimuli: acid or base for mutually exclusive
“foot” dissociation and UV light or visible light (plus iodine)
to induce or release ring strain between the walker and the
track.[7] The design (Figure 1) is closely related to the
previously reported small-molecule walker–track system,
Figure 1. a) Operating mechanism of a light-driven walker–track
system based on selectively labile “feet” and adjustable ring strain
between the walker (red) and the track in one positional isomer. The
reaction sequence shown results in transport of the walker from left to
which features a walker with one hydrazone foot (labile in
acid; locked in base) and one disulfide foot (labile in base;
locked in acid).[4] The crucial difference is that a stilbene unit
has been added between the internal aldehyde and the
disulfide footholds of the track (Scheme 1). The key to
achieving directionality lies in the isomerization of the
stilbene moiety, through which significant ring strain can be
induced in the positional (constitutional) isomer in which the
walker unit bridges the stilbene linkage (Figure 1 and
Scheme 2b).[8] E!Z isomerization provides a driving force
right; switching steps (ii) and (iv) would cause the walker to be
transported from right to left. b) Potential energy profile experienced
by the walker unit (an energy ratchet mechanism[6,10]). c) Repeat unit
of a hypothetical polymeric system along which the walker could be
transported in either direction depending on the order of the stimuli
applied. Stimuli: (i) UV light, (ii) base, (iii) visible light and iodine,
(iv) acid.[11]
[*] Dr. M. J. Barrell, Dr. A. G. Campaꢀa, M. von Delius,
Dr. E. M. Geertsema, Prof. D. A. Leigh
for the walker to “step” onto the central stilbene unit, while
subsequent Z!E isomerization results in a majority of the
walkers being transported away from the stilbene group in a
direction determined by which foot–track interaction is
labilized next. Such a manipulation of the thermodynamic
minima (here by strain induction through stilbene isomer-
ization) and kinetic barriers (here by addition of either base
or acid) experienced by a substrate corresponds to an energy
ratchet type of Brownian ratchet mechanism (Fig-
ure 1b).[6,9,10]
School of Chemistry, University of Edinburgh
The King’s Buildings, West Mains Road, Edinburgh EH9 3JJ (UK)
Fax: (+44)131-650-6453
E-mail: david.leigh@ed.ac.uk
Homepage: http://www.catenane.net
[**] We thank the EPSRC National Mass Spectrometry Service Centre
(Swansea, UK) for high-resolution mass spectrometry. This research
was funded through the European Research Council Advanced
Grant WalkingMols. A.G.C. thanks the Fundaciꢁn Ramꢁn Areces for
a postdoctoral fellowship. D.A.L. is an EPSRC Senior Research
Fellow and holds a Royal Society Wolfson Research Merit Award.
Molecular walker–track conjugate E-1,2-1[12] was synthe-
sized according to Scheme 1 (see the Supporting Information
for experimental procedures and characterization data). The
Supporting information for this article is available on the WWW
under http://dx.doi.org/10.1002/anie.201004779.
Angew. Chem. Int. Ed. 2011, 50, 285 –290
ꢀ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
285

Communications
Scheme 1. Synthesis of molecular walker–track conjugate E-1,2-1.
a) NaH, DMF, RT, 16 h; b) methanesulfonyl chloride (MsCl), NEt3,
CH2Cl2, 08C!RT, 1.5 h; c) potassium thioacetate (KSAc), DMF, RT,
3 h, 59% (three steps); d) AcOH (cat.), MeOH, RT, 2 h, 83%;
e) NaOMe, MeOH, RT, 2 h; f) I2, KI, CH2Cl2, RT, 5 min, 48% (two
steps); g) NaH, DMF, RT, 16 h; h) MsCl, NEt3, CH2Cl2, 08C!RT, 2 h;
i) KSAc, DMF, RT, 3 h, 35% (three steps); j) HC(OMe)3, p-toluenesul-
fonic acid (p-TsOH), MeOH, RT, 1 h; k) NaOMe, MeOH, RT, 3 h;
l) 3-mercaptopropionic acid, I2, KI, CH2Cl2, RT, 5 min; m) HCl (1m),
CH2Cl2, RT, 15 min, 69% (four steps); n) AcCl, MeOH, 08C!RT, 5 h,
76%; o) WA-SIMes-CF3 catalyst,[15] CH2Cl2, microwave (300 W), 1008C,
3 h, 23%. See the Supporting Information for details.
Scheme 2. Z!E and E!Z stilbene isomerization of individual posi-
tional isomers of 1 and ring opening of E-2,3-1 in either direction
under dynamic covalent conditions (acid or base) to release ring strain
between the walker and track. Conditions: E!Z: 0.1–10 mm, hn
(365 nm; bandwidth: 10 nm), CD2Cl2, RT, 5 min; Z!E: 0.1 mm,[20] I2
(ca. 10 equiv), hn (500 nm; bandwidth: 10 nm[20]), CD2Cl2, RT, 4–8 h.
Dynamic disulfide exchange (basic conditions): 0.1 mm, d,l-dithio-
threitol (DTT, 10 equiv), 1,8-diazabicyclo[5.4.0]undec-7-ene (DBU,
40 equiv), (MeO2CCH2CH2S)2 (20 equiv), CHCl3, RT, 24 h. Dynamic
hydrazone exchange (acidic conditions): 0.1 mm, excess CF3CO2H
(TFA), CHCl3, RT, 48 h.
initial position of the walker unit at footholds 1 and 2 of the
track was established by the synthesis of macrocycle 7,
starting from the simple aromatic building blocks 2 and 3 and
the bipedal walker unit 6. Compound 8, which contains thiol
foothold 3 (masked as a disulfide with methyl 3-mercapto-
propionate, a “placeholder” thiol[13]) and aldehyde foothold 4,
was prepared from precursors 4 and 5. The synthesis was
completed by a ruthenium-catalyzed cross-metathesis reac-
tion,[14] which afforded E-1,2-1 in 23% yield.[15,16] The
positional isomer where the walker unit is located on the
other end of the molecular track, E-3,4-1, was prepared
unambiguously through an analogous synthetic route (see the
Supporting Information).
of the track, depending on the use of either base or acid
(Scheme 2b). Irradiation of pristine Z-2,3-1 (obtained from
Z-1,2-1 by base-induced disulfide exchange) at 365 nm
generated a 75:25 (E/Z) photostationary state, which was
subjected to conditions for disulfide (base) or hydrazone
exchange (acid) in separate experiments. Under basic con-
ditions, approximately 80% of the walker units from this E/Z
mixture were transported to the 1,2-positional isomer (> 95%
of E-2,3-1 converted into the 1,2 isomer; 25% of residual
Z-2,3-1); under acidic conditions approximately 80% of the
3,4-positional isomer was formed (> 95% of E-2,3-1 con-
verted into the 3,4 isomer; 30% of residual Z-2,3-1).
Having established that the key principles of the energy
ratchet mechanism would operate by using individual posi-
tional isomers, we needed to establish a procedure for
analyzing the mixture of isomers expected to be generated
during the full sequence of operations of a statistical
molecular-motor mechanism. Pleasingly, the composition of
the walker–track system could be accurately determined after
each step by 500 MHz 1H NMR spectroscopy, even for
complex mixtures that contain all eight isomers (Z/E-1,2-1,
Z/E-2,3-1, Z/E-3,4-1, Z/E-1,4-1[11]). The partial 1H NMR
spectra of the six isomers involved in the major “passing-leg
gait”[10a] mechanism (Figure 1a) are shown in Figure 2.
Differences in chemical-shift values that are indicative of
the position of the walker unit on the track arise for the
protons of the four methylene groups in the walker unit
(shown in red, HB–HE). Protons Hk and Hl serve as distinctive
markers for the configuration of the stilbene olefinic bond
An investigation of the photochemistry of E-1,2-1 and
E-3,4-1, the positional isomers in which the walker unit is not
located over the track stilbene unit, showed that it was
possible to efficiently carry out direct (i.e., unsensitized)
E!Z stilbene photoisomerization[17] at 365 nm in CD2Cl2
(Scheme 2a). The excellent diastereomeric ratios at the
photostationary states (ca. 9:1 Z/E) are a result of the high
molar absorptivities of the E isomers compared to the
corresponding Z isomers.[18] The reverse process, that is,
Z!E stilbene isomerization, also proceeded well in CD2Cl2
(> 9:1 E/Z) by using iodine and narrow-bandwidth green light
(500 nm; 10 nm bandwidth).[19,20]
We next confirmed that the strained[21] E-2,3-1 isomer
could be formed from Z-2,3-1 and that, in a subsequent
dynamic covalent[22] exchange reaction, the walker unit could
be efficiently transported to either the 1,2 or the 3,4 position
286
www.angewandte.org
ꢀ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2011, 50, 285 –290

to dissociate from the track and rebind at footholds 1 or 3.
This process leads to an equilibrium in which the positional
isomer Z-2,3-1 is favored. Light-induced Z!E isomerization
(iii) of the stilbene in this mixture results in 75% conversion
of Z-2,3-1 into the strained E-2,3-1 isomer. In the following
acid-catalyzed step (iv), the hydrazone foot is able to
dissociate from the track and rebind at either foothold 2 or
4, while the disulfide foot acts as a fixed pivot.[23] This process
results in a large majority (> 95%) of walker units being
transported away from the E-stilbene unit (2,3 position)
towards the now energetically favorable 3,4 position (see
Figure 1b for the schematic energy diagram that corresponds
to these processes). Note that all the energy required to fuel
directional transport in Scheme 3 is supplied through the
E!Z photoisomerization reaction (i), which creates config-
urational strain in the track. The other three reactions (ii–iv),
which are all under thermodynamic control, each dissipate
some of that energy (even reaction (iii), which uses the energy
to induce conformational strain between the walker and the
track in the 2,3 position) in a way designed to achieve the
desired directional migration of the walker.
The behavior of the walker–track system 1 under these
operating conditions (Figure 3a), and when the stimuli are
applied in a different order (iii–ii–i–iv; Figure 3b) or starting
from a different positional isomer (E-3,4-1; Figure 3c,d), are
shown in Figure 3. The amount of each E/Z isomer pair (1,2-1,
2,3-1, 3,4-1, and 1,4-1[11]) is shown after each step of the
applied sequence of stimuli. The graphs show both the
behavior of the system (solid symbols and lines) extrapolated
from the experimentally determined equilibrium and steady-
state ratios between each pair of exchanging isomers
(Scheme 2a for example), and the experimentally determined
composition during full-system operation (hollow sym-
bols).[24]
Figure 3a shows the change in composition of the system
during the reaction sequence intended to transport the walker
from left to right, (i.e., that shown in Scheme 3), starting from
E-1,2-1. After the full cycle of reactions, which corresponds to
up to two steps being taken by the walker, 48% of the walker
units are on the right-hand side of the track (3,4-1), 30% of
the walker units are on the left-hand side (1,2-1), and the
remainder are in the middle (16% 2,3-1) or have taken a
“double step” from left to right (6% 1,4-1). In other words,
50% more walkers have taken two steps to the right under the
operating sequence than have taken one step to the right and
one step back (or remained on the left-hand side throughout).
Figure 3b shows the results of a change in the sequence of
reactions; following the E!Z ring-strain-inducing reaction,
the hydrazone foot–track interaction is labilized (under acidic
conditions) rather than that of the disulfide. This reaction
sequence (iii–ii–i–iv) should bias transport of the walker from
right to left (see caption to Figure 1) and indeed, only 4% of
3,4-1 is present in the mixture after these reactions are applied
to 1,2-1 (interestingly, a significant amount (33%) of the 1,4
“double step” isomer is formed through folding of the
track[11]).
Figure 2. Partial 1H NMR spectra (500 MHz, CD2Cl2, 298 K) of the six
isomers of 1 involved in the passing-leg gait mechanism[10a] (Figure 1a
and Scheme 3). The dashed lines connect some of the key signals that
are indicative of the position of the walker unit or the configuration of
the track olefin. The signals corresponding to minor isomers (e.g. the
“E-2,3-1” spectrum contains 25% of Z-2,3-1), solvent, and other
impurities are shown in gray. The lettering corresponds to the proton
labeling shown in Scheme 1.
(d < 7 ppm for Z and d > 7 ppm for E isomers). The signal at
around d = 10.0 ppm (green, Hv or Hj), which corresponds to
the free aldehyde group, and the signal at around d = 3.8 ppm
(blue, Ha and Hp), which corresponds to the methylene
disulfide protons, are also particularly useful probes because
their chemical shifts vary with both the walker position and
the stilbene configuration (see the Supporting Information
1
for full H NMR spectra of individual positional and config-
urational isomers).
Directionally biased transport of the walker from left to
right through the sequential application of the four external
stimuli in the order i–ii–iii–iv is shown in Scheme 3. Starting
from pristine E-1,2-1, photochemical E!Z isomerization (i)
gives a photostationary state that consists of 88% of the
Z-1,2-1 isomer. Subjecting this mixture to the basic conditions
(ii), in which the hydrazone linkage between the walker unit
and the track is kinetically stable,[23] allows the disulfide foot
Figure 3c shows that the walker unit can be effectively
transported from right to left by this sequence. Starting from
E-3,4-1, 48% of the walkers are found on the left-hand side of
Angew. Chem. Int. Ed. 2011, 50, 285 –290
ꢀ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.angewandte.org
287

Communications
Scheme 3. Directionally biased (left to right) walker migration over one full cycle of operation starting from E-1,2-1, following the layout used in
Figure 1. For clarity, only the major isomer is shown after each reaction (see Figure 3a for the composition of the entire system at each stage).
1
Isomer ratios were obtained from the integration of the H NMR spectra (see the Supporting Information). Conditions: (i) 0.1–10 mm, hn
(365 nm; bandwidth: 10 nm), CD2Cl2, RT, 5 min (0.1 mm) to 1 h (10 mm); (ii) 0.1 mm, DTT (10 equiv), DBU (40 equiv), (MeO2CCH2CH2S)2
(20 equiv), CHCl3, RT, 24 h; (iii) 0.1 mm,[20] I2 (ca. 10 equiv), hn (500 nm; bandwidth: 10 nm[20]), CD2Cl2, RT, 6 h; (iv) 0.1 mm, TFA (excess), CHCl3,
RT, 48 h.
Figure 3. Dynamic behavior of walker–track system 1, varying the starting position of the walker and the stimuli sequence (denoted through colored bands;
purple: E!Z isomerization at 365 nm; blue: base-catalyzed disulfide exchange; green: Z!E isomerization at 500 nm (I2 catalyst); red: acid-catalyzed
hydrazone exchange). Experimental (expt.) and calculated (calcd.) product distribution[24] over one full cycle of operation: a) starting compound E-1,2-1,
stimuli sequence i–ii–iii–iv; b) starting compound E-1,2-1, stimuli sequence iii–ii–i–iv (a mismatch as this sequence tends to transport the walker from right
to left); c) starting compound E-3,4-1, stimuli sequence i–iv–iii–ii; d) starting compound E-3,4-1, stimuli sequence iii–iv–i–ii (a mismatch as this sequence
tends to transport the walker from left to right). Calculated product distribution over seven base/acid cycles (no stilbene isomerization), starting from
e) pristine E-1,2-1 and f) pristine Z-1,2-1. The calculated data are based on simple mathematical extrapolation of the observed steady-state ratios between
each Z/E isomer pair under conditions (i) and (iii) and the observed equilibrium ratios between individual positional isomers under conditions (ii) and (iv)
1
(see Section 5 in the Supporting Information). Experimental data are based on the integrals of the H NMR spectra of mixtures obtained after each step of
the reaction sequence, margin of error ꢀ3% (see Section 4 in the Supporting Information). A small amount of oligomers (ca. 5%) was also obtained
during each experiment. Conditions: (i) 0.1–10 mm, hn (365 nm; bandwidth: 10 nm), CD2Cl2, RT, 5 min–1 h; (ii) 0.1 mm, DTT (10 equiv), DBU (40 equiv),
(MeO2CCH2CH2S)2 (20 equiv), CHCl3, RT, 12–48 h; (iii) 0.1 mm,[20] I2 (ca. 10 equiv), hn (500 nm; bandwidth: 10 nm[20]), CD2Cl2, RT, 4–8 h; (iv) 0.1 mm, TFA
(excess), CHCl3, RT, 6–96 h.
288
www.angewandte.org
ꢀ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2011, 50, 285 –290

the track after the full reaction sequence and only 36%
remain on the right-hand side.
Figure 3d shows the effect of applying the right-to-left
directionally biased reaction sequence starting with the
isomer for which the walker is already on the left.
would actually “ratchet” the walker unit energetically uphill.
Work aimed at overcoming these deficiencies is currently
underway.
Received: August 2, 2010
Revised: September 3, 2010
Published online: October 15, 2010
These results collectively illustrate the characteristics
typical of energy ratchet mechanisms.[6,10a] Irrespective of
the initial walker position on the track (e.g., starting from 1,2-
1 (Figure 3a), 3,4-1 (Figure 3d), or any combination of walker
positions) after the sequence i–ii–iii–iv (or iii–iv–i–ii) the
majority of walkers are found on the right-hand side (3,4-1).
Conversely, irrespective of the initial walker position on the
track (Figure 3b and c), after the sequence iii–ii–i–iv (or i–iv–
iii–ii) the majority of walkers are on the left-hand side (1,2-1).
The position of the walker does not influence the sequence of
reactions applied, nor their basic effect on the track,[25] and
statistical “errors” (e.g., the walker does not take a step every
time a foot–track interaction is labilized) are an intrinsic
feature of the mechanism.
Figures 3e and 3 f (E-stilbene track and Z-stilbene track,
respectively) show the behavior of the system when no
stilbene isomerization reactions are carried out. Repeatedly
switching between treatment with acid and base (labilizing
first one foot–track interaction and then the other) allows the
distribution of the walkers on the track to approach the
minimum-energy distribution.[4] In the case of the E-stilbene
track (Figure 3e), the gap between footholds 2 and 3 is too
wide for the walkers to bridge easily (note the very small
amounts of the 2,3 isomer formed) and it would take seven
full operational cycles to approach the steady state of the
system, which features roughly equal amounts of walkers at
each end of the track. With the Z-stilbene track (Figure 3 f),
the 2,3 position is, of course, very accessible to the walker
units and so the steady-state is reached after only two cycles.
However, the 2,3 position is actually energetically more
favored than the ends of the tracks, so the majority of walkers
remain in the center (49% Z-2,3-1). The differences in
behavior and walker distribution produced by switching
between acid and base alone (Figure 3e,f) and when addi-
tional stilbene isomerization is carried out (e.g., Figure 3a,c)
show the dramatic effect that the energy ratchet mechanism
has on walker transport.
The operations of walker–track conjugate 1 shown in
Scheme 3 and Figure 3a,c correspond to a light-driven linear
molecular motor system.[26] Significant directional bias, which
stems from an energy ratchet mechanism, can transport the
walker unit in either direction along the molecular track; the
control over the sense of direction is a feature not found in
biological linear motor proteins. While both the photochem-
ical stilbene isomerization and the acid- and base-induced
reversible foot-migration processes proceed with good effi-
ciencies, a weakness of the present system lies in the flexibility
of the track and the resulting folding products, which reduce
the net directionality of transport and prevent improvement
of the bias over multiple cycles. Furthermore, a hypothetical
polymeric track based on the current design would utilize the
light energy used to power directional transport very ineffi-
ciently by switching the configuration of many of the double
bonds in the track each time, whereas only one double bond
Keywords: Brownian ratchets · dynamic covalent chemistry ·
.
molecular devices · molecular motors · photochemistry
[1] a) Molecular Motors (Ed.: M. Schliwa), Wiley-VCH, Weinheim,
2003; b) R. D. Vale, Cell 2003, 112, 467 – 480; c) M. Schliwa, G.
Woehlke, Nature 2003, 422, 759 – 765.
[2] For concise reviews on motor proteins, see: kinesin family:
a) S. M. Block, Biophys. J. 2007, 92, 2986 – 2995; b) N. Hirokawa,
Y. Noda, Y. Tanaka, S. Niwa, Nat. Rev. Mol. Cell Biol. 2009, 10,
682 – 696; myosin family: c) J. R. Sellers, C. Veigel, Curr. Opin.
Cell Biol. 2006, 18, 68 – 73; d) H. L. Sweeney, A. Houdusse,
Annu. Rev. Biophys. 2010, 39, 539 – 557; dynein family: e) M. P.
Koonce, M. Samsꢀ, Trends Cell Biol. 2004, 14, 612 – 619; f) K.
Oiwa, H. Sakakibara, Curr. Opin. Cell Biol. 2005, 17, 98 – 103.
[3] a) W. B. Sherman, N. C. Seeman, Nano Lett. 2004, 4, 1203 – 1207;
b) J.-S. Shin, N. A. Pierce, J. Am. Chem. Soc. 2004, 126, 10834 –
10835; c) P. Yin, H. Yan, X. G. Daniell, A. J. Turberfield, J. H.
Reif, Angew. Chem. 2004, 116, 5014 – 5019; Angew. Chem. Int.
Ed. 2004, 43, 4906 – 4911; d) Y. Tian, Y. He, Y. Chen, P. Yin, C.
Mao, Angew. Chem. 2005, 117, 4429 – 4432; Angew. Chem. Int.
Ed. 2005, 44, 4355 – 4358; e) P. Yin, H. M. T. Choi, C. R. Calvert,
N. A. Pierce, Nature 2008, 451, 318 – 322; f) S. J. Green, J. Bath,
A. J. Turberfield, Phys. Rev. Lett. 2008, 101, 238101; g) T.
Omabegho, R. Sha, N. C. Seeman, Science 2009, 324, 67 – 71;
h) H. Gu, J. Chao, S.-J. Xiao, N. C. Seeman, Nature 2010, 465,
202 – 205; i) K. Lund, A. J. Manzo, N. Dabby, N. Michelotti, A.
Johnson-Buck, J. Nangreave, S. Taylor, R. Pei, M. N. Stojanovic,
N. G. Walter, E. Winfree, H. Yan, Nature 2010, 465, 206 – 210.
[4] a) M. von Delius, E. M. Geertsema, D. A. Leigh, Nat. Chem.
2010, 2, 96 – 101; b) S. Otto, Nat. Chem. 2010, 2, 75 – 76.
[5] a) V. Serreli, C.-F. Lee, E. R. Kay, D. A. Leigh, Nature 2007, 445,
523 – 527; b) M. Alvarez-Pꢁrez, S. M. Goldup, D. A. Leigh,
A. M. Z. Slawin, J. Am. Chem. Soc. 2008, 130, 1836 – 1838.
[6] We use the term “ratcheting” in a manner that is consistent with
its use in biology and nonequilibrium statistical physics: namely,
it is the capturing of a positional displacement of a substrate
through the imposition of a kinetic energy barrier that prevents
the displacement being reversed when the thermodynamic
driving force is removed; see M. N. Chatterjee, E. R. Kay,
D. A. Leigh, J. Am. Chem. Soc. 2006, 128, 4058 – 4073. An
information ratchet is a Brownian ratchet mechanism in which
the position of the particle on a potential energy surface causes
the potential energy surface to change (at an energetic cost),
thus leading to directional transport of the particle. In an energy
ratchet mechanism the potential energy surface is periodically or
stochastically varied irrespective of the position of the particle in
order to cause directional transport (for example, the relative
depths of two pairs of minima and the relative heights of the
maxima that connect them could be repeatedly switched, as in
Figure 1b). See ref. [10a].
[7] For examples of light-driven synthetic molecular machines, see:
a) S. Shinkai, T. Nakaji, T. Ogawa, K. Shigematsu, O. Manabe,
J. Am. Chem. Soc. 1981, 103, 111 – 115; b) H. Murakami, A.
Kawabuchi, K. Kotoo, M. Kunitake, N. Nakashima, J. Am.
Chem. Soc. 1997, 119, 7605 – 7606; c) N. Koumura, R. W. J.
Zijlstra, R. A. van Delden, N. Harada, B. L. Feringa, Nature
1999, 401, 152 – 155; d) T. Hugel, N. B. Holland, A. Cattani, L.
Angew. Chem. Int. Ed. 2011, 50, 285 –290
ꢀ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.angewandte.org
289

Communications
Moroder, M. Seitz, H. E. Gaub, Science 2002, 296, 1103 – 1106;
[16] The styrene (Type II) olefins in 7 and 8 have similar reactivities
and so the maximum yield of E-1,2-1 expected from their
statistical cross-metathesis is 50% (see ref. [14a]). This synthetic
disconnection was chosen to prevent scrambling of the sensitive
(different) disulfide and aldehyde/hydrazone functionalities in
the final molecule.
[17] For an excellent review on the photochemistry of stilbene
derivatives, see: H. Meier, Angew. Chem. 1992, 104, 1425 – 1446;
Angew. Chem. Int. Ed. Engl. 1992, 31, 1399 – 1420.
[18] The photostationary equilibrium ratio for direct Z!E isomer-
ization can be calculated as a function of the quantum yields (F)
and molar absorptivities (e) of the E and Z species (at a given
wavelength): [Z]/[E] = (FE!Z eE)/(FZ!E eZ), the quantum yield
ratio is usually approximately 1 (see ref. [17]).
[19] The mechanism for iodine-mediated stilbene Z!E isomeriza-
tion involves the reversible addition of IC radicals (here photo-
generated with green light) to the double bond. See, for
example: S. Yamashita, Bull. Chem. Soc. Jpn. 1961, 34, 972 – 976.
[20] The iodine-mediated stilbene Z!E isomerization reactions
were conducted at relatively low concentrations (0.1 mm) of the
walker–track conjugate and with a narrow (10 nm) bandwidth of
green light (500 nm) to avoid side-reactions of the disulfide
groups.
[21] The length of the methylene spacer in the walker unit is crucial in
order to generate ring strain in E-2,3-1, but not in Z-2,3-1. The
choice of a four-carbon atom spacer was based on model studies
and molecular modeling (semi-empirical, PM3).
[22] a) S. J. Rowan, S. J. Cantrill, G. R. L. Cousins, J. K. M. Sanders,
J. F. Stoddart, Angew. Chem. 2002, 114, 938 – 993; Angew. Chem.
Int. Ed. 2002, 41, 898 – 952; b) P. T. Corbett, J. Leclaire, L. Vial,
K. R. West, J.-L. Wietor, J. K. M. Sanders, S. Otto, Chem. Rev.
2006, 106, 3652 – 3711; c) J.-M. Lehn, Chem. Soc. Rev. 2007, 36,
151 – 160.
e) T. Muraoka, K. Kinbara, Y. Kobayashi, T. Aida, J. Am. Chem.
Soc. 2003, 125, 5612 – 5613; f) E. M. Pꢁrez, D. T. F. Dryden, D. A.
Leigh, G. Teobaldi, F. Zerbetto, J. Am. Chem. Soc. 2004, 126,
12210 – 12211; g) Q.-C. Wang, D.-H. Qu, J. Ren, K. Chen, H.
Tian, Angew. Chem. 2004, 116, 2715 – 2719; Angew. Chem. Int.
Ed. 2004, 43, 2661 – 2665; h) J. Bernꢂ, D. A. Leigh, M. Lubom-
ska, S. M. Mendoza, E. M. Pꢁrez, P. Rudolf, G. Teobaldi, F.
Zerbetto, Nat. Mater. 2005, 4, 704 – 710; i) R. Eelkema, M. M.
Pollard, J. Vicario, N. Katsonis, B. S. Ramon, C. W. M. Bastiaan-
sen, D. J. Broer, B. L. Feringa, Nature 2006, 440, 163; j) V.
Balzani, M. Clemente-Leꢀn, A. Credi, B. Ferrer, M. Venturi,
A. H. Flood, J. F. Stoddart, Proc. Natl. Acad. Sci. USA 2006, 103,
1178 – 1183; k) T. Muraoka, K. Kinbara, T. Aida, Nature 2006,
440, 512 – 515; l) M. Yamada, M. Kondo, J.-i. Mamiya, Y. Yu, M.
Kinoshita, C. J. Barrett, T. Ikeda, Angew. Chem. 2008, 120,
5064 – 5066; Angew. Chem. Int. Ed. 2008, 47, 4986 – 4988; m) M.
Klok, N. Boyle, M. T. Pryce, A. Meetsma, W. R. Browne, B. L.
Feringa, J. Am. Chem. Soc. 2008, 130, 10484 – 10485; n) M. R.
Panman, P. Bodis, D. J. Shaw, B. H. Bakker, A. C. Newton, E. R.
Kay, A. M. Brouwer, W. J. Buma, D. A. Leigh, S. Woutersen,
Science 2010, 328, 1255 – 1258.
[8] For elegant mechanistic studies based on ring strain in macro-
cycles that contain a “stiff stilbene” linkage, see: a) Q. Z. Yang,
Z. Huang, T. J. Kucharski, D. Khvostichenko, J. Chen, R.
Boulatov, Nat. Nanotechnol. 2009, 4, 302 – 306; b) T. J. Kuchar-
ski, Z. Huang, Q. Z. Yang, Y. C. Tian, N. C. Rubin, C. D.
Concepcion, R. Boulatov, Angew. Chem. 2009, 121, 7174 – 7177;
Angew. Chem. Int. Ed. 2009, 48, 7040 – 7043. For a dynamic
covalent library that consists of macrocycles with internal acyl
hydrazone and (photoswitchable) azobenzene linkages, see:
c) L. A. Ingerman, M. L. Waters, J. Org. Chem. 2009, 74, 111 –
117.
[9] a) D. A. Leigh, J. K. Y. Wong, F. Dehez, F. Zerbetto, Nature
2003, 424, 174 – 179; b) J. V. Hernandez, E. R. Kay, D. A. Leigh,
Science 2004, 306, 1532 – 1537; c) S. P. Fletcher, F. Dumur, M. M.
Pollard, B. L. Feringa, Science 2005, 310, 80 – 82.
[23] For studies on the dynamic chemistry of hydrazone–disulfide
systems under mutually exclusive (acid–base) conditions, see:
a) A. G. Orrillo, A. M. Escalante, R. L. E. Furlan, Chem.
Commun. 2008, 5298 – 5300; b) Z. Rodriguez-Docampo, S.
Otto, Chem. Commun. 2008, 5301 – 5303; c) M. von Delius,
E. M. Geertsema, D. A. Leigh, A. M. Z. Slawin, Org. Biomol.
Chem. 2010, 8, 4617 – 4624.
[10] a) E. R. Kay, F. Zerbetto, D. A. Leigh, Angew. Chem. 2007, 119,
72 – 196; Angew. Chem. Int. Ed. 2007, 46, 72 – 191; b) R. D.
Astumian, Phys. Chem. Chem. Phys. 2007, 9, 5067 – 5083;
c) R. D. Astumian, I. Derꢁnyi, Eur. Biophys. J. 1998, 27, 474 –
489; d) P. Hꢃnggi, F. Marchesoni, Rev. Mod. Phys. 2009, 81, 387 –
442; e) R. D. Astumian, Biophys. J. 2010, 98, 2401 – 2409.
[11] The 1,4 isomers of 1, which result from folding of the track (see
the Supporting Information for structures), add an additional
double-step mechanism to the major passing-leg gait mechanism
(see ref. [4]). This pathway has the opposite bias to the main
mechanism and so actually reduces the net directionality of the
walker transport.
[12] E or Z denotes the configuration of the stilbene double bond in
the track; the numerical prefixes (e.g., 1,2) specify the position of
the walker unit on the four-foothold track.
[13] The sulfur foothold of the track was protected as a disulfide with
a “placeholder” thiol in order to prevent oxidation of the free
thiol to a dimeric disulfide by atmospheric oxygen.
[24] Possible reasons for the minor differences between the data
extrapolated from the individual isomer experiments and the
experimental results from operation on the full system include
insufficient equilibration time for dynamic covalent exchange
processes, nonideal photochemical steady-state ratios in mix-
1
tures, and inaccuracies in H NMR integration.
[25] The presence of the walker at the 2,3-position lowers the E/Z
ratio at the photostationary state of the Z!E reaction
(Scheme 2b), thus reducing the net directionality of the trans-
port mechanism.
[26] In a double-labeling crossover experiment on the previously
reported small-molecule walker–track system (see ref. [4]), an
average step number (the number of steps after which 50% of
the walkers are no longer attached to their original track) of 37
was obtained for the loss of processivity during disulfide and
hydrazone exchange. The conditions for dynamic covalent bond
exchange are the same in the present study, and walker
dissociation is not observed during the photochemical experi-
ments. Therefore, the average step number during the operation
of 1 should be similar. The average step number for wild-type
kinesin is approximately 100 (R. B. Case, D. W. Pierce, N. Hom-
Booher, C. L. Hart, R. D. Vale, Cell 1997, 90, 959 – 966).
[14] a) A. K. Chatterjee, T.-L. Choi, D. P. Sanders, R. H. Grubbs,
J. Am. Chem. Soc. 2003, 125, 11360 – 11370; b) S. P. Nolan, H.
Clavier, Chem. Soc. Rev. 2010, 39, 3305 – 3316.
[15] Screening gave best results with the commercially available
catalyst WA-SIMes-CF3, compound 2h in: D. Rix, F. Caijo, I.
Laurent, F. Boeda, H. Clavier, S. P. Nolan, M. Mauduit, J. Org.
Chem. 2008, 73, 4225 – 4228.
290
www.angewandte.org
ꢀ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2011, 50, 285 –290