Visible light sensitized isomerization of rhodopsin-based molecular switches

Article abstract of DOI:10.1016/j.tetlet.2015.02.115

Abstract Neutral rhodopsin-based molecular photoswitches can be activated by using low-energy visible light through sensitized irradiation. The triplet state of the neutral and protonated forms are studied and compared by the use of different experimental

Full text of DOI:10.1016/j.tetlet.2015.02.115

Tetrahedron Letters 56 (2015) 1991–1993
Contents lists available at ScienceDirect
Tetrahedron Letters
journal homepage: www.elsevier.com/locate/tetlet
Visible light sensitized isomerization of rhodopsin-based molecular
switches
⇑
David Martínez-López, Marina Blanco-Lomas, Pedro J. Campos, Diego Sampedro
Departamento de Química, Centro de Investigación en Síntesis Química (CISQ), Universidad de La Rioja, Madre de Dios 51, 26006 Logroño, La Rioja, Spain
a r t i c l e i n f o
a b s t r a c t
Article history:
Neutral rhodopsin-based molecular photoswitches can be activated by using low-energy visible light
through sensitized irradiation. The triplet state of the neutral and protonated forms are studied and com-
pared by the use of different experimental conditions, laser flash photolysis, and theoretical calculations.
The use of common, visible light absorbing sensitizers allows for the photoisomerization without the
requirement of acidic water solutions.
Received 8 January 2015
Revised 17 February 2015
Accepted 20 February 2015
Available online 2 March 2015
Ó 2015 Elsevier Ltd. All rights reserved.
Keywords:
Photoswitches
Molecular devices
Energy transfer
Photochemistry
Photophysics
Light driven molecular devices1,2 have attracted considerable
attention in recent years due to several practical advantages over
chemical or electrochemical inputs, such as high temporal and spa-
tial resolution and lack of waste products.3 Among the different
devices, switches4 and motors5 have become very useful as a con-
sequence of their ability to convert light energy into mechanical
work. As a drawback, most of these systems require high-energy
UV light to operate. Therefore, practical applications of these
devices are restrained to their use in complex systems that do
not absorb in that region or, at least, are very photostable. For
instance, the use of photoswitches in a biological matrix requires
the use of visible light to avoid potential irradiation-induced dam-
age. Several strategies have been used to shift the absorption maxi-
mum of photoswitches to allow their triggering with low-energy,
long-wavelength light. Modification of the chromophore by chang-
ing substituents or increasing the conjugation may lead to a shift in
the absorption band maximum.6 Alternatively, complexation with
Lewis acids has also been reported to shift the absorption from UV
to visible light in azobenzenes.7 We have recently shown8 that
acids). For the rhodopsin-based switches,8 these conditions include
water as solvent at pH = 5. This prevents the use of these switches
with acid-labile substrates and, more importantly, in organic sol-
vents. With the aim of trying to circumvent these drawbacks, we
explored the possibility of using energy transfer to activate our
photoswitches with visible light while using common organic
solvents. This could confer a wider scope of applicability to the
rhodopsin-based molecular photoswitches. This strategy has been
proven useful previously to allow the use of low-energy light to
activate molecular devices. A rotaxane-based linear molecular
motor has been reported9 to work via an energy transfer. Also, a
catenane rotor showed a unidirectional movement.10 Similarly,
several types of molecular switches have been activated by
using sensitized photochemistry including azobenzenes,11
diarylethenes,12 and overcrowded alkenes.13 Herein, we report
our efforts of trying to use sensitized photochemistry to permit
the activation of rhodopsin-based molecular switches with the
use of visible light.
We explored the triplet excited states of a neutral14 and a
protonated15 model species 1 and 2 (Fig. 1). These compounds
are analogues of the protonated Schiff base present in rhodopsin,
may be prepared in the pure E form and have shown excellent
properties as photoswitches.14,15 Additionally, they have been used
to control the conformation of a peptide.8
First, we tested the sensitized photochemistry of 1. To do this,
we used thioxanthone as a triplet sensitizer (ET = 63.2 kcal/mol).
To ensure the triplet state population of 1, we used a sodium meta-
vanadate solution filter (cuts all wavelengths below 370 nm, see
protonation of rhodopsin-based photoswitches also causes
a
ca. 50 nm shift in the absorption maximum, which allows the
use of blue and purple light to modify a peptide conformation.
Although any of these alternatives enable the use of visible light
compatible with biological systems, all of them require the modi-
fication of the chromophore by designing a specific synthetic route
(substituent effect) or altering the reaction conditions (Lewis
⇑
Corresponding author.
http://dx.doi.org/10.1016/j.tetlet.2015.02.115
0040-4039/Ó 2015 Elsevier Ltd. All rights reserved.

1992
D. Martínez-López et al. / Tetrahedron Letters 56 (2015) 1991–1993
Ph
isomer can be obtained under these conditions. Thus, the use of
Ph
H
rhodopsin-based molecular photoswitches can be extended to dif-
ferent reaction conditions avoiding the requirements of acidic
water solutions for the use of visible light. This result greatly
increases the applicability of these systems beyond the excellent
photophysical properties as photoswitches already reported.8,14,15
The triplet state of 1 was also studied by laser flash photolysis.
The triplet state decay of a solution of xanthone in acetonitrile was
measured. Subsequently, an increasing amount of 1 was added as a
quencher, and the decay plots were recorded (see Fig. 2). All graphs
were fitted with exponential functions to find the xanthone life-
N
N
Ph
Ph
1
2
Figure 1. Rhodopsin-based photoswitches.
Supplementary material). Under these conditions, only thiox-
anthone can absorb light. Thus, the activation of the switch must
come from an energy transfer from the sensitizer. We followed
the progress of the reaction by 1H NMR until a photostationary
state (PSS) of 80:20 (E:Z) was found (see Table 1). Interestingly,
the PSS found under direct irradiation is 76:24.14 Therefore, a simi-
lar situation can be achieved by using both direct irradiation of 1
(band maximum at 289 nm) and sensitized irradiation (band maxi-
mum of thioxantone at 380 nm). This could be very useful in the
application of these switches under sensitized, low-energy irradia-
tion as the wavelength of effective irradiation can be shifted ca.
100 nm.
time, s, for each experiment (see Fig. S5). The inverse values of s
(disappearance constants, kd) were then plotted versus the amount
of photoswitch added (see Supplementary material for details). The
fit obtained indicates that 1 is a triplet quencher. The quenching
rate, kd, was measured to be 7.27 Â 109 MÀ1 sÀ1 from the linear
fit slope.
A similar study was also performed for the protonated version
of our switch, 2. We have previously shown8 the utility of this
photoswitch under visible light irradiation. However, this requires
a slightly acidic medium to operate. Therefore, we are interested in
a generalization of the use of this switch with low-energy light
under different conditions. Thus, we performed the sensitized
irradiation using the same reaction conditions and sensitizers as
for 1 (see Table 1). Surprisingly, no isomerization was found in
any case. This could be due to a high energy triplet state that
cannot be reached under these sensitization conditions or to an
inactive triplet state. A laser flash photolysis exploration was also
performed for 2 using xanthone as described before (Fig. 3).
As shown in Figure 3, 2 is also capable of inducing the triplet
state decay of xanthone. In this case, the quenching rate, kd, was
measured to be 1.14 Â 1011 MÀ1 sÀ1 from the linear fit slope (see
Fig. S6). Thus, the lack of photoisomerization of 2 should be under-
stood in different terms. In both compounds the triplet state can be
populated but a different behavior was found. The different fea-
tures of the potential energy surfaces for the triplet states may
cause that 2 does not lead to an efficient isomerization as it
happens for 1.
In order to delimit the energy of the triplet state for 1, we
performed similar experiments with other related sensitizers of
varying ET. Results are shown in Table 1.
It can be deduced that the energy of the triplet state of com-
pound 1 must be between 55.2 and 45.4 kcal/mol, as 4-nitroaniline
allowed the isomerization while with acridine no photoreaction
was observed. The data on Table 1 are relevant to determine the
energy of the triplet state. However, the use of these specific
sensitizers also implies the use of relatively high-energy light
(370–390 nm). In order to improve these results, we aimed for
the use of a common photosensitizer that could be useful in the
photoisomerization of 1 by using visible light. In this sense, the
use of [Ru(bpy)3]2+ as an absorber-sensitizer in a variety of arrays
is well-known.16–18 This compound shows a strong metal-to-li-
gand charge transfer (MLCT) absorption at 452 nm and it has an
ET value of 48.0 kcal/mol. On the one hand, this complex could
be used to further delimit the triplet state energy of 1 and, on
the other hand, it is a common and easy way to use a visible light
sensitizer. When performing a similar experiment as those pre-
viously described, we obtained a PSS of 42:58 (E:Z). Several rele-
vant conclusions can be drawn from this result. First, the ET of 1
must be between 48.0 and 45.4 kcal/mol. Second, the activation
of 1 can take place by using visible light and a suitable sensitizer.
And third, a PSS with a higher proportion of the photochemical Z
To further explore this point, we performed a series of theoreti-
cal calculations on 1 and 2. A detailed computational analysis of
the potential energy surfaces and the photochemistry of these
photoswitches would require the use of multiconfigurational
approaches, such as CASPT2. This study is currently underway
but it falls well beyond the scope of this Letter. However,
Table 1
Sensitized irradiation of 1a
Entry
1
Sensitizer
Structure
kmax
ET (kcal/mol)
PSS (E:Z)
O
Thioxanthone
380
63.2
80:20
S
O
2
Acridone
390
58.3
56:44
N
H
NH2
NO2
3
4
4-Nitroaniline
Acridine
380
370
55.2
45.4
66:34
100:0
N
H
Figure 2. Triplet state decay of a 4.03 Â 10À4 M solution of xanthone in acetonitrile
in the presence of 1.
a
0.1 M solutions of 1 in CDCl3 with 1 equiv sensitizer at rt.

D. Martínez-López et al. / Tetrahedron Letters 56 (2015) 1991–1993
1993
of the central double bond was found for 2 (dihedral angle of 62°).
Consistently, a shorter C–C distance was also found (1.455 Å). Also
in the triplet state, the electronic natures of1 and 2 are different.
While for 1 a neat contribution of a p–
p⁄ excitation in the C@C cen-
tral double bond is present, for 2 the excitation is also located in
one of the phenyl rings. This is in agreement with the geometrical
features of both minima. Accordingly, recovery of the initial E
isomer will be exclusive in this case, in agreement with the experi-
mental observations.
We have shown that rhodopsin-based molecular photoswitches
can be activated by visible light using sensitized irradiation. This
implies a useful generalization of our precedent results as these
switches could also be used in organic solvents with low-energy
light sources. In turn, this could benefit the development of appli-
cations which require the use of low-energy light without the
restriction of acidic water solutions. While both the neutral and
protonated forms of the switch feature available triplet states in
terms of energy, only the neutral species is capable of an efficient
photoisomerization. Thus, the presence of a positive charge in the
molecule does not substantially affect the photophysics, but it is
crucial in determining the photochemistry of these systems.
Figure 3. Triplet state decay of a 4.03 Â 10À4 M solution of xanthone in acetonitrile
in the presence of 2.
Acknowledgements
This research was supported by the Spanish Ministerio de
Ciencia
e
Innovación (MICINN)/Fondos Europeos para el
Desarrollo Regional (FEDER) (CTQ2011-24800).
Supplementary data
Supplementary data (experimental procedures, computational
details and Cartesian coordinates for the computed geometries)
associated with this article can be found, in the online version, at
http://dx.doi.org/10.1016/j.tetlet.2015.02.115.
Figure 4. Minima in the triplet state energy surface for 1 (left) and 2 (right).
References and notes
preliminary results based on DFT methods are enough to under-
stand the different photochemical behavior of 1 and 2. We used
the B3LYP functional together with the standard 6-31G⁄ basis set
(see Supplementary material for Computational details). First, we
used the UV spectra to check our level of theory. The computed
UV spectrum of 1 (TD-DFT, 50 states) consists of a strong absorp-
tion (f = 0.53) at 295 nm, in good agreement with the experimental
band maximum at 303 nm (Fig. S2). Also, the computed spectrum
of 2 shows an absorption at 369 nm (f = 0.30). A shift of 74 nm is
observed from the neutral switch 1 to the protonated 2. The orbital
contributions for the two transitions feature a clear difference.
While for the neutral species 1 the main orbital contribution corre-
1. Balzani, V.; Credi, A.; Venturi, M. Molecular Devices and Machines: Concepts and
Perspectives for the Nanoworld; Wiley-VCH Verlag: Weinheim, Germany, 2008.
2. Molecular Switches; Feringa, B. L., Browne, W. R., Eds., second ed.; Wiley-VCH:
Weinheim, 2011.
3. Balzani, V.; Credi, A.; Venturi, M. Nano Today 2007, 2, 18–25.
4. Garcia-Iriepa, C.; Marazzi, M.; Frutos, L. M.; Sampedro, D. RSC Adv. 2013, 3,
6241–6266.
5. Feringa, B. L. J. Org. Chem. 2007, 72, 6635–6652.
6. Beharry, A. A.; Sadovski, O.; Woolley, G. A. J. Am. Chem. Soc. 2011, 133, 19684–
19687.
7. Yang, Y.; Hughes, R. P.; Aprahamian, I. J. Am. Chem. Soc. 2012, 134, 15221–
15224.
8. Blanco-Lomas, M.; Samanta, S.; Campos, P. J.; Woolley, G. A.; Sampedro, D. J.
Am. Chem. Soc. 2012, 134, 6960–6963.
9. Balzani, V.; Clemente-León, M.; Credi, A.; Ferrer, B.; Venturi, M.; Flood, A. H.;
Stoddart, J. F. Proc. Natl. Acad. Sci. U.S.A. 2006, 103, 1178–1183.
10. Leigh, D. A.; Wong, J. K. Y.; Dehez, F.; Zerbetto, F. Nature 2003, 424, 174–179.
11. Venkataramani, S.; Jana, U.; Dommaschk, M.; Sönnichsen, F. D.; Tuczek, F.;
Herges, R. Science 2011, 331, 445–448.
sponds with a
bond, in the case of the protonated compound 2, the transition is
basically
p⁄ located in the C@N bond and the conjugated phenyl
p–
p⁄ transition located in the central C@C double
p
–
group. Therefore, the excited states involved in the transitions sug-
gest than in the case of 1, photoisomerization will be possible,
while for 2 this process will be less probable. Upon sensitized
irradiation, the triplet state of both 1 and 2 will be populated.
Thus, the triplet potential energy surface shape will determine
the experimental outcome of the photoreaction. An optimization
along the T1 surfaces starting from the ground state geometries
led to two different minima for 1 and 2 with diverse features
(see Fig. 4). In the case of 1, the central C@C double bond is elon-
gated (1.464 Å) and completely twisted (dihedral angle of 82°).
Thus, upon ground state recovery, both E and Z isomers could be
formed to yield different PSSs. However, a more planar disposition
12. Brayshaw, S. K.; Schiffers, S.; Stevenson, A. J.; Teat, S. J.; Warren, M. R.; Bennett,
R. D.; Sazanovich, I. V.; Buckley, A. R.; Weinstein, J. A.; Raithby, P. R. Chem. Eur. J.
2011, 17, 4385–4395.
13. Cnossen, A.; Hou, L.; Pollard, M. M.; Wesenhagen, P. V.; Browne, W. R.; Feringa,
B. L. J. Am. Chem. Soc. 2012, 134, 17613–17619.
14. Blanco-Lomas, M.; Campos, P. J.; Sampedro, D. Eur. J. Org. Chem. 2012, 6328–
6334.
15. Blanco-Lomas, M.; Martínez-López, D.; Campos, P. J.; Sampedro, D. Tetrahedron
Lett. 2014, 55, 3361–3364.
16. Zhao, J.; Wu, W.; Sun, J.; Guo, S. Chem. Soc. Rev. 2013, 42, 5323–5351.
17. Wu, W.; Ji, S.; Wu, W.; Shao, J.; Guo, H.; James, T. D.; Zhao, J. Chem. Eur. J. 2012,
18, 4953–4964.
18. Ji, S.; Wu, W.; Wu, W.; Guo, H.; Zhao, J. Angew. Chem., Int. Ed. 2011, 50, 1626–
1629.