Molecular switch based on a biologically important redox reaction

Article abstract of DOI:10.1021/jp045793g

Building on our earlier report of a single-molecule probe,¹ we show how biologically important redox centers, nicotinamide and quinone, incorporated into a fluorophore-spacer-receptor molecular structure, form redox active molecular switches, w

Full text of DOI:10.1021/jp045793g

130
J. Phys. Chem. B 2005, 109, 130-137
Molecular Switch Based on a Biologically Important Redox Reaction
Ping Yan, Michael W. Holman, Paul Robustelli, Arindam Chowdhury, Fady I. Ishak, and
David M. Adams*
Department of Chemistry, Columbia UniVersity, 3000 Broadway, New York, New York 10027
ReceiVed: September 16, 2004; In Final Form: October 11, 2004
Building on our earlier report of a single-molecule probe,1 we show how biologically important redox centers,
nicotinamide and quinone, incorporated into a fluorophore-spacer-receptor molecular structure, form redox
active molecular switches, with the photoinduced electron-transfer behavior of each depending on the oxidation
state of the receptor subunit. The switch based on nicotinamide (3/6) is strongly fluorescent in its oxidized
state (ΦF ≈ 1.0) but nonfluorescent in the reduced state (ΦF < 0.001) due to electron transfer from the
reduced nicotinamide to the photoexcited fluorophore. The fluorescence can be reversibly switched off and
on chemically by successive reduction with NaBH3CN and oxidation with tetrachlorobenzoquinone and
switched electrochemically over 10 cycles without significant degradation. A similar switch based on
quinonimine turned out to be nonfluorescent in both reduced and oxidized states: in addition to a similar
quenching mechanism in the reduced state, quenching also occurs in the oxidized state, due to electron transfer
from the fluorophore to the receptor. Ab initio quantum chemical calculations of orbital energy levels were
used to corroborate these quenching mechanisms. Calculations predicted photoinduced electron transfer to be
energetically favorable in all cases where quenching was observed and unfavorable in all cases where it was
not. Application of the perylene analogue as a biosensor has also been demonstrated by coupling the switch
to the catalytic pathway of yeast alcohol dehydrogenase, a common NADH/NAD+-utilizing enzyme.
Introduction
Molecular switches2smolecules whose optical or electronic
properties change dramatically in response to an environmental
perturbationshave been extensively studied and developed due
to interest in their potential applications as molecular memo-
ries,3,4 molecular valves,5 and biological or chemical sensors.1,6
Fluorescent switches are of special interest since they allow for
sensitive detection and monitoring, even down to the level of
single molecules in solution and on surfaces under ambient
conditions.1,7-9 A common motif in fluorescent switches is
photoinduced electron transfer (PET)10sin the off state of the
switch, PET either to or from the fluorophore quenches the
fluorescence. Since these switches are based on the modulation
of electron transfer, they are also of interest in the realm of
molecular electronics.11-13 Sensors based on PET6,14-20 have
been successfully designed for a wide variety of chemical
species, including cations such as Ca2+ 21
,
Zn2+ 22,23
,
and
Figure 1. Fluorescence switch design based on photoinduced electron
transfer: (a) aromatic amine as proton receptor, R ) -CH(n-C9H19)2;
(b) nicotinamide as redox receptor; (c) quinonimine as redox receptor;
and (d) general molecular orbital diagramsthe position of the receptor
HOMO relative to the HOMO of the photoexcited fluorophore
determines whether PET is thermodynamically favorable.
Hg2+ 24,25
;
anions such as pyrophosphate26 and azide ion;27 and
neutral molecules such as amino acids28 and saccharides.29 While
many fluorescent sensors for cations are now commercially
available,30 switches based on reversible redox reactions are
fewer and less well-studied.17,31-35
so photoexcitation of the fluorophore leads to rapid PET from
the amine to the fluorophore, quenching the fluorescence. After
protonation, on the other hand, the HOMO of the amine is lower
than the HOMO of the fluorophore, so PET quenching is
thermodynamically unfavorable, and the excited fluorophore can
relax radiatively.
Here, we report a rational design of a redox switch analogous
to the NDAPP proton sensor. Since part of the motivation for
redox switches of fluorescence is for real-time visualization of
cellular events,36 two biologically important redox centers were
chosen here as the redox receptors: nicotinamide, the redox
We have previously reported a molecule (ΦF < 0.01),
N-(1-nonyldecyl)-N′-(p-aminophenyl)-perylene-3,4:9,10-tetracarbox-
ylbisimide (NDAPP, Figure 1a) suitable for single molecule
level sensing and imaging that becomes strongly fluorescent
(ΦF ≈ 1.0) upon protonation or coordination to a metal ion or
surface.1 The general mechanism in Figure 1d is typical for a
PET-based fluorescence switch: the highest-occupied molecular
orbital (HOMO) of the receptor subunit (here the free amine)
is higher in energy than the HOMO of the fluorophore subunit,
* Corresponding author. E-mail: dadams@chem.columbia.edu.
10.1021/jp045793g CCC: $30.25 © 2005 American Chemical Society
Published on Web 12/13/2004

Molecular Switch Based on Bio-Redox Reaction
J. Phys. Chem. B, Vol. 109, No. 1, 2005 131
SCHEME 1a
Figure 2. Comparison of normalized absorption and emission spectra
of the NAD+ analogue and NADH analogue in 2:1 MeOH/CH2Cl2.
The inset shows the difference in absorption between NAD+ and NADH
analogues.
a R) -CH(n-C6H13)2, conditions: (a) imidazole, 100 °C, 78%; (b)
imidazole, 130 °C, 90%; (c) MeOH, 65 °C, 41%; (d) Pd/C, H2, RT,
76%; (e) imidazole, 130 °C, 37%; and (f) CHCl3, NaClO, RT, 38%.
center in the ubiquitous nicotinamide adenine dinucleotide
(NADH/NAD+), and nicotinamide adenine dinucleotide phos-
phate (NADPH/NADP+) redox couples; and quinone, which
also plays an important role in cellular electron-transfer reac-
tions.37 Redox switches incorporating nicotinamide and quinon-
imine redox receptors are shown in Figure 1b and 1c, respec-
tively. As in the case of NDAPP, the HOMO of an aromatic
amine is higher than the HOMO of the fluorophore, and the
reduced forms are nonfluorescent due to PET quenching (Figure
1d); after oxidation, the HOMO of the receptor subunit is lower
than HOMO of the fluorophore, and this PET quenching no
longer occurs. This paper represents a general strategy of
converting amine-based cation sensors to redox active switches
and sensors that can be easily coupled to enzymatic reactions.
Figure 3. Photograph showing different fluorescence properties for 5
µM oxidized form (6), left, and reduced form (3), right, in 2:1 MeOH/
CHCl3.
carboxylbisimide (4), which in turn was synthesized by con-
densation of phenylenediamine with 1.1
Optical Properties. The absorption and fluorescence spectra
of NAD+ analogue 6 and NADH analogue 3 are shown in
Figure 2. The absorption bands above ∼450 nm are identical
and characteristic of perylene bisimide absorption; the absorption
difference spectrum between 300 and 400 nm (Figure 2, inset)
shows a strong peak at 350 nm, which is characteristic of 1,4-
dihydronicotinamide and has been widely used in assays for
NAD(P)H-utilizing enzymes.39 Despite the similarity in absorp-
tion spectra, the fluorescence properties are very different:
NAD+ analogue is strongly fluorescent (Φ ≈ 1) while the
NADH analogue has almost no fluorescence (Φ < 0.001).
Figure 3 illustrates this difference, showing the reduced and
oxidized forms in solution under UV excitation.
The other redox pair 9/10, also has similar absorption spectra,
but while the reduced form (9) is nonfluorescent as expected
from the model in Figure 1, here the oxidized form (10) was
also found to be nonfluorescent. We believe this is also due to
PET, but now from the photoexcited fluorophore to the
quinonimine, as described in detail later.
Results and Discussion
Synthesis. Shown in Scheme 1 are the synthetic pathways
for both reduced and oxidized forms of the redox switches:
NADH analogue 3 and NAD+ analogue 6 and hydroquinon-
imine 9 and quinonimine 10. Nonsymmetric perylene bisimide
dyes are generally synthesized by condensation of a perylene-
3,4:9,10-tetracarboxyl-3,4-anhydride-9,10-imide 1 with appro-
priate amines.38 The long-chain secondary alkyl group in 1
improves the solubility of the products in organic solvents; other
side chains could be chosen to make the products water soluble.
The amines for the reduced forms, 2 and 8, are very reactive
and can condense with 1 straightforwardly. The oxidized forms
were synthesized indirectly: quinonimine 10 was prepared by
oxidizing the reduced form 9 with NaClO; NAD+ analogue 3
was synthesized by a pyridinium exchange reaction between
3-(carbamoyl)-1-(2,4-dinitrophenyl)-pyridinium chloride (5) and
N-(4-aminophenyl)-N′-(1-hexylheptyl)-perylene-3,4:9,10-
tetra-
Chemical Redox Switching of Fluorescence. Reversible
chemical switching requires a careful selection of reductant and
oxidant: (a) the redox reaction should be selective so that the

132 J. Phys. Chem. B, Vol. 109, No. 1, 2005
Yan et al.
Figure 4. Variation in the fluorescence intensity for 3/6 upon repeated
reduction with NaBH3CN and oxidation with TCBQ, over three cycles.
Figure 6. Cyclic voltammograms at a scan rate of 100 mV/s: (a) 0.3
mM N,N′-bis(1-hexylheptyl)perylene-3,4:9,10-tetracarboxylbisimide and
(b) 0.1 mM 6. Reduction/oxidation potentials used in the spectroelec-
trochemistry are indicated by the vertical arrows.
Figure 5. 1H NMR of 6 in 1:1 CD2Cl2/CD3OD: (a) before reduction;
(b) after reduction with NaBH3CN (2, 1,4-dihydro isomer; 4, 1,6-
dihydro isomer); and (c) after reoxidation with TCBQ.
fluorophore is not destroyed, (b) the reaction should be clean
and the byproduct should not affect further switching or quench
the fluorescence, and (c) the oxidation and reduction should be
carried out under similar conditions (solvent, pH, temperature).
As a result, reversible chemical switching is rare in the literature,
and investigators typically resort to photochemical and elec-
trochemical methods for reversible switching. The redox inter-
conversion between NAD+ analogue and NADH analogue
involves two electrons and one proton, which is equivalent to
a hydride (H-) transfer, so a hydride donor NaBH3CN and
hydride acceptor, tetrachlorobenzoquinone (TCBQ), were se-
lected for reversible reduction and oxidation, respectively.
NaBH3CN is also known to be a selective reducing reagent for
pyridinium salt in the presence of carbonyl groups.40 TCBQ
has been used successfully for the oxidation of dihydronicoti-
namides at room temperature, and it has been shown that this
oxidation can be further accelerated by UV irradiation since
the photoexcited TCBQ is a more powerful oxidant.41 The
changes in fluorescence intensity for 3/6 upon repeated reduction
with NaBH3CN and oxidation with TCBQ over three cycles
are shown in Figure 4.
conversion back to the original NAD+ analogue 6 (Figure 5c)
in ∼80% yield.
Electrochemical Redox Switching of Fluorescence. To
increase the understanding of biological electron-transfer reac-
tions, intensive research efforts have been made on the
electrochemistry of NAD+/NADH and their analogues.42 NAD+/
NADH and their analogues are infamous for their electrochemi-
cal irreversibility: the two forms can be interconverted but only
with large over-potentials,43-45 which reflects the bistability of
the reduced and oxidized forms, a necessary property of a
molecular switch.3 It has also often been found that NAD+ (or
its analogues) forms dimeric productssin which a C-C bond
between two nicotinamide units is formed in lieu of the new
C-H bondswhen reduced on a naked electrode (rather than a
mediator or enzyme-modified electrode) but that the dimer can
be still reoxidized back to NAD+ (or its analogues) at an anode
wave with high over-potential.43-45 Figure 6 shows the cyclic
voltammograms (CV) both of N,N′-bis(1-hexylheptyl)perylene-
3,4:9,10-tetracarboxylbisimide (Figure 6a)sthe fluorophore
alonesand of NAD+ analogue 6 (Figure 6b)sboth fluorophore
and receptor. The CV of the ordinary perylene bisimide (Figure
6a) shows clearly one reversible oxidation and two reversible
reduction peaks, in agreement with previous reports.46 For the
NAD+ analogue (Figure 6b), the peak for the irreversible
reduction of the nicotinamide is close to the reduction peaks
for the fluorophore and is not fully resolved, but the irreversible
oxidation (back to the NAD+ analogue) is clearly distinguished
at +0.6 V versus Fc+/Fc. The reduction product may be a dimer
NMR experiments confirm that the switching is between the
expected products 3 and 6 (Figure 5). After reduction of the
NAD+ analogue 6 (Figure 5a) with NaBH3CN, the major
product is clearly the NADH analogue 3 (Figure 5b), from
comparison of this spectrum to the spectrum of independently
prepared 3, and the minor product is a 1,6-dihydro isomer. Both
products are apparently nonfluorescent, and reoxidation gives

Molecular Switch Based on Bio-Redox Reaction
J. Phys. Chem. B, Vol. 109, No. 1, 2005 133
resemble the orbitals of the isolated species. Here also, in the
reduced species the HOMO of the hydroquinonimine was found
to be higher in energy than the HOMO of the fluorophore,
predicting quenching by electron transfer, while the HOMO of
the oxidized quinonimine was lower in energy than the HOMO
of the fluorophore, predicting no quenching, as in the NADH/
NAD+ analogues. Recall, however, that while the hydroquinon-
imine 9 was nonfluorescent as expected, the oxidized form 10
was also found to be nonfluorescent. The frontier molecular
orbital energies plotted in Figure 9 illustrate why: in the
oxidized species (Figure 9b), the LUMO of the receptor is now
lower in energy than the LUMO of the perylene, so the
fluorescence can now be quenched by electron transfer from
the fluorophore to the receptor. Instead of switching electron
transfer on and off, as with the NADH/NAD+ analogues, here
the redox reaction changes the direction of electron transfer.
While potentially also of interest, this process is not so amenable
to study via fluorescence spectroscopy. Note (Figure 8b) that
the calculations correctly predict that this reversed electron-
transfer quenching cannot occur for the NAD+ analogue because
there the LUMO of the receptor is still higher in energy than
the LUMO of the fluorophore.
Figure 7. Reversible switching of fluorescence upon repeated elec-
trochemical reduction/oxidation of 6 in acetonitrile. The control is N,N′-
bis(1-hexylheptyl)perylene-3,4:9,10-tertracarboxylbisimide. The reduc-
tion and oxidation were carried out at -0.81 and +0.76 V vs Fc+/Fc,
respectively, for 4 min each.
or mixture of dimers rather than the NADH analogue 3, but it
is in any case nonfluorescent and is reoxidized back to the
NAD+ analogue 6.
Spectroelectrochemistry was used to follow the fluorescence
changes as a function of applied voltage. At a carefully selected
voltage, -0.81 V versus Fc+/Fc, the nicotinamide receptor can
be reduced selectively without appreciable reduction of the
fluorophore, and at +0.76 V versus Fc+/Fc, the reduction
product can be reoxidized cleanly. Figure 7 shows the reversible
switching of fluorescence upon repeated electrochemical reduc-
tion/oxidation on the NAD+ analogue in acetonitrile. There is
about 13% degradation of the system after 10 redox cycles. N,N′-
Bis(1-hexylheptyl)perylene-3,4:9,10-tertracarboxylbisimide, as
the control, only shows small variations in fluorescence under
the same conditions, probably due to a small amount of
reduction of the fluorophore.
Molecular Orbital Calculations. Molecular orbital calcula-
tions have been found useful in rational design of PET probes
based on benzofurazan47 and fluorescein.48,49 Here, we used ab
initio quantum chemical calculations to determine the relative
energies of the molecular orbitals, to corroborate the quenching
mechanism shown in Figure 1. We found that calculations using
DFT at the B3LYP level, with a dielectric continuum solvent
model (vide infra), accurately explained the observed electron-
transfer behavior of the molecules studied here. Figure 8 shows
the frontier molecular orbitals for the NADH and NAD+
analogues 3 and 6. Note that the fluorophore and the redox
receptor are sufficiently decoupled that molecular orbitals are
localized on either one or the other, due to the large (ca. 70°)
interplanar angle between the bisimide and the phenylene
moieties and to the node in the HOMO and LUMO of the
perylene bisimide on the linking nitrogen atom.50 The molecular
orbitals in general are localized on either the fluorophore or
the receptor and have similar contours to those calculated for
an isolated perylene bisimide or isolated phenylnicotinamide.
These molecular orbital calculations support the proposed
quenching mechanism: in the NADH analogue (Figure 8a), the
HOMO of the receptor is higher in energy than the HOMO of
the fluorophore and can donate an electron to the excited state
of the fluorophore to quench its fluorescence; while in the NAD+
analogue (Figure 8b), the HOMO of the receptor is lower than
the HOMO of the fluorophore and quenching via electron
transfer is not energetically favorable.
Coupling the Redox Switch to an Enzymatic Pathway. As
a first step toward combinatorial screening and cellular imaging,
we coupled the redox switch to a catalytic enzymatic reaction
utilizing NADH/NAD+ cofactors, as shown in Figure 10a. The
redox potential of 3/6, which is difficult to ascertain precisely
from the electrochemical measurements, is only slightly per-
turbed from that of natural NADH/NAD+. We studied the
equilibrium reaction (eq 1) of 3/6 with NADH/NAD+ in two
different limiting environments: organic solution involving
acetonitrile/H2O (97:3) and aqueous solution involving micelles
NAD+ analogue 6 + NADH h
NADH analogue 3 + NAD+ (1)
Addition of 6 equiv of NADH to 6 in acetonitrile (1.6 µM)
resulted in an 85% decrease in fluorescence. Similarly, in an
aqueous environment, where 6 was solubilized by n-dodecyl-
â-D-maltoside (DDM) and the reaction was facilitated by a
phase transfer catalyst, cetyltrimethylammonium bromide (CTAB),
6 could be easily reduced by NADH. The reverse reaction,
involving addition of NAD+ to 3, gave an insignificant increase
in fluorescence, suggesting that the oxidation of 3 with NAD+
is not energetically favorable and that the equilibrium in eq 1
lies to the right.
While 3/6 is too sterically hindered to be bound by the
enzyme directly, this NADH cross-reaction allows us to couple
any enzymatic reaction that produces NADH to our fluorescent
switch. Figure 10a shows this process for a typical NADH/
NAD+-utilizing enzyme, yeast alcohol dehydrogenase (ADH).
ADH catalyzes the oxidation of ethanol to acetaldehyde as
NAD+ is reduced to NADH. Coupling the reaction in eq 1 into
this biochemical pathway allows for detection of the enzymatic
turnover. Figure 10b shows the change in fluorescence versus
time for a micelluar aqueous solution of probe (6), enzyme
(ADH), substrate (ethanol), and cofactor (NAD+). Fluorescence
intensity (9) decreases as the probe monitors the production of
NADH due to the enzymatic reaction. In control experiments,
where the enzymatic cycle is halted by leaving out either
NAD+(b), ADH (2), or ethanol (f), little fluorescence change
is observed.
Similar calculations on the quinonimine-based molecules 9
and 10 produced similar results: the molecular orbitals were
found to be localized on either fluorophore or receptor and to

134 J. Phys. Chem. B, Vol. 109, No. 1, 2005
Yan et al.
Figure 8. Frontier molecular orbitals for (a) NADH analogue 3 and (b) NAD+ analogue 6.
Figure 9. Calculated molecular orbital energies for (a) reduced and
(b) oxidized quinonimine-based molecules 9 and 10, showing the
quenching mechanism for each.
Experimental Procedures
Material and Methods. NAD+, NADH, and yeast alcohol
dehydrogenase were purchased from Roche Applied Science.
All other chemicals were purchased from Aldrich and used
without further purification unless otherwise specified. The
acetonitrile used for electrochemical studies was freshly distilled
from CaH2. N-(1-Hexylheptyl)perylene-3,4:9,10-tetracarboxyl-
3,4-anhydride-9,10-imide (1),51,52 1-(4-aminophenyl)-1,4-dihy-
dronicotinamide (2),53-55 3-(carbamoyl)-1-(2,4-dinitrophenyl)-
pyridinium chloride (5),56 and 4-hydroxy-4′-nitro-diphenylamine
(7)57 were synthesized according to the literature methods.
Column chromatography was performed on SiliCycle ultrapure
silica gel (70-230 mesh). Thin-layer chromatography (TLC)
was carried out on Selecto Scientific silica gel 60 F-254 flexible
TLC plates. NMR spectra were recorded on a Bruker 300, 400,
or 500 MHz instrument as noted, and the chemical shifts were
reported relative to TMS at 0 ppm (for 1H NMR) and CD3OD
at 49.00 ppm (for 13C NMR). Atmospheric pressure chemical
ionization (APCI) mass spectra were taken on a JEOL JMS-
LCmate mass spectrometer (JEOL Ltd., Tokyo, Japan); high-
resolution FAB mass spectra were obtained on a JMS HX-110/
110A tandem mass spectrometer (JEOL); and matrix assisted
laser desorption ionization (MALDI) mass spectra were obtained
on a Voyager DE mass spectrometer (AB Biosystems, Framing-
ham, MA). UV/vis absorption spectra were recorded on a
Perkin-Elmer Lambda 25 UV/VIS spectrometer and fluores-
cence emission spectra on a Perkin-Elmer LS 55 fluorometer
with excitation at 488 nm. Fluorescence quantum yields were
calculated from the integrated emission from 500 to 800 nm
over the absorption at 488 nm, relative to N,N′-bis(1-hexylhep-
Figure 10. Coupling of the fluorescence switch to the catalytic pathway
of alcohol dehydrogenase (ADH): (a) the enzymatic catalytic cycle
and (b) experimental realization: the fluorescence is quenched in
response to enzymatic reaction (9) while the fluorescence is stable in
the absence of either NAD+ (b), ADH (2), or ethanol (f).
tyl)perylene-3,4:9,10-tetracarboxylbisimide in chloroform as a
reference (ΦF ) 1.0).58
Chemical Redox Switching. Fluorescence was measured in
a quartz fluorometer cell with a screw cap (Starna, path length:
10 mm). The fluorescence experiment began with 3 mL of 0.4
µM NAD+ analogue (6) in 2:1 (v/v) CH2Cl2/ MeOH. For
reductions, 4 µL of 1 mM NaBH3CN in 2:1 (v/v) CH2Cl2/MeOH
was added and after 6.5 h at room temperature, and the
fluorescence dropped to ∼6% of the initial value. For reoxi-
dation, 6 µL of 4 mM tetrachlorobenzoquinone in 2:1 (v/v) CH2-
Cl2/MeOH was added, and the mixture was exposed to a UV

Molecular Switch Based on Bio-Redox Reaction
J. Phys. Chem. B, Vol. 109, No. 1, 2005 135
as for the other cyclic voltammetry and fluorescence experi-
ments.
Biochemical Study. The reaction in micellar system was
carried out in a Starna cell (4 mm path length) containing 50
mM Tris-HCl buffer (pH ) 7.6), 150 mM NaCl, 2.0%
n-dodecyl-â-D-maltoside (DDM) surfactant, 0.25% cetyltri-
methylammonium bromide (CTAB) phase transfer catalyst, 10%
(v/v) ethanol, 1.5 mM NAD+, and 1.2 µM NAD+ analogue (6).
Fifty µL of yeast alcohol dehydrogenase (1 mg/mL) was injected
to initiate the reaction, and then the progress was monitored by
UV/vis absorption and fluorescence spectra. The reduced
product 3 was confirmed by thin-layer chromatography after
the solution was evaporated and extracted with CHCl3.
Theoretical Methods. Quantum chemical calculations were
performed using density functional theory (DFT) in Jaguar 4.1,60
at the B3LYP level, with a 6-31G** basis set. Orbitals were
visualized in Molden to determine for each orbital whether it
was localized primarily on the fluorophore or primarily on the
receptor. Solvation was implemented using Jaguar’s standard
Poisson-Boltzmann dielectric continuum boundary method;
settings used were for chloroform (dielectric ) 4.806; molecular
weight ) 119.38; density ) 1.4460) for the quinonimine
molecules 9 and 10, and a weighted average of the solvent
properties for 2:1 (v/v) CH2Cl2/MeOH (dielectric ) 25.0;
molecular weight ) 49.6; density ) 0.9822) for the NADH/
NAD+ analogues 3 and 6. The solution-phase optimized
geometries were used for 3 and 6, as well as for the phenyl-
nicotinamide and phenyl-quinonimine model systems, and found
to differ only very slightly from the gas-phase optimized
geometries, leading to changes of generally <1.5% in the orbital
energy levels.
Figure 11. Cell for spectroelectrochemistry.
lamp (UVGL-25, UVP Inc., 366 nm, 18.4 W) for 30 min. For
the subsequent switching cycles, the amount of redox reagents
used increased gradually: 8 µL of 1 mM NaBH3CN, 6 µL of
4 mM tetrachlorobenzoquinone, 28 µL of 1 mM NaBH3CN,
Synthesis of NADH Analogue (3). A total of 6.5 mg (0.03
mmol) of 1-(4-aminophenyl)-1,4-dihydronicotinamide (2) and
22.9 mg (0.04 mmol) of N-(1-hexylheptyl)perylene-3,4:9,10-
tetracarboxyl-3,4-anhydride-9,10-imide (1) were mixed in 5 mL
of CH2Cl2, and 2 g of imidazole was added. The mixture was
heated to 40 °C under a stream of argon to remove the CH2Cl2,
and the solid mixture was heated to 100 °C. The reaction was
followed by TLC and was complete ∼10 min after the imidazole
melted. After being cooled, the mixture was dispersed in 5 mL
of MeOH, and 90 mL of H2O was added to precipitate out the
product, which was collected by filtration and washed with 30
mL of H2O. Column chromatography (SiO2, 2:98 MeOH/
CH2Cl2) gave 3 as a red powder (18 mg, 78%). Rf (silica gel,
1
and 51 µL of 4 mM tetrachlorobenzoquinone. The H NMR
experiment started with 0.55 mL of 1 mM NAD+ analogue in
1:1 (v/v) CD2Cl2/MeOD; 10 µL 55 mM NaBH3CN in 1:1 (v/v)
CD2Cl2/CD3OD was added, and the reaction was complete
within 5 min. For reoxidation, 1.5 mg tetrachlorobenzoquinone
was added, and the mixture was exposed to UV lamp for 30
min. The tetrachlorobenzoquinone was not completely dissolved
but did not affect the spectrum.
Cyclic Voltammetry. Cyclic voltammetry (CV) was carried
out using a Model AFRDE5 bi-potentiostat from Pine Instrument
Company with data acquisition controlled by in-house written
software (Labview, National Instruments). The working com-
partment contained the analyte in 1:2 CH2Cl2/MeCN with 0.1
M NBu4PF6 (TBAHFP), a Pt working electrode, and an SCE
reference electrode. A silver wire immersed in the same solution
was employed as the counter electrode, separated from the
working compartment by a glass frit. All potentials were
referenced to an internal Fc/Fc+ couple. The working compart-
ment was purged with argon for 6 min and left under a slight
positive pressure of argon during the measurements.
1
5:95 MeOH/CH2Cl2) ) 0.25; H NMR (500 MHz, CDCl3): δ
0.83 (t, 6 H, 2 CH3), 1.16-1.40 (m, 16 H, 8 CH2), 1.87 (m, 2
H, R-CH2), 2.25 (m, 2 H, R-CH2), 3.26 (br s, 2 H, 4-H
nicotinamide), 5.04 (dt, J ) 8.1 Hz, 3.3 Hz,1 H, 5-H nicotina-
mide), 5.19 (m, 1 H, CH 1-hexylheptyl), 5.24 (s, 2 H, NH2),
6.45 (d, J ) 8.1 Hz,1 H, 6-H nicotinamide), 7.31 (d, J ) 9.2
Hz, 2H, phenylene), 7.34 (d, J ) 9.2 Hz, 2 H, phenylene), 7.64
(br s, 1 H, 2-H nicotinamide), 8.65-8.79 (m, 8 H, perylene);
MS (MALDI, negative): m/z ) 770.96 [M]- (calc. for
C49H46N4O5: 770.91).
N-(4-Aminophenyl)-N′-(1-hexylheptyl)-perylene-3,4:9,10-
tetracarboxylbisimide (4). A total of 100 mg (0.174 mmol) of
N-(1-hexylheptyl)perylene-3,4:9,10-tetracarboxyl-3,4-anhydride-
9,10-imide (1), 142 mg of (1.31 mmol) 1,4-phenylenediamine,
and 2 g of imidazole was heated at 130 °C under argon for 5 h.
After being cooled, the mixture was dissolved in 100 mL of
1:1 MeOH/CH2Cl2 and then concentrated to about 15 mL to
precipitate out the product. The red precipitate was collected
by membrane filtration (0.45 µm, Osmonics), washed with 5
mL of MeOH, redissolved in 100 mL of 2:98 CH3COOH/CH2-
Spectroelectrochemistry. The spectroelectrochemistry was
carried out in a 2 × 10 mm path-length quartz fluorometer cell
fused to the bottom of a three-neck flask, with platinum mesh
as the pseudo-transparent working electrode (see Figure 11), in
a modification of a previously reported setup.59 Analyte
concentration was 5-10 µM. A thin fluorometer cell (2 mm ×
10 mm) is used so that most of the analyte is confined near the
working electrode, and polar solvent (electrolyte in acetonitrile)
ensures a fast switching rate. Other conditions were the same

136 J. Phys. Chem. B, Vol. 109, No. 1, 2005
Yan et al.
Cl2, and stirred at room temperature for 2 h. The solvent was
evaporated, and the residue was purified by column chroma-
tography (SiO2, CHCl3). Yield 104 mg (90%) Rf (silica gel,
perylene); MS (MALDI, negative): m/z ) 755.54 [M]- (calcd
for C49H45N3O5: 755.34).
Quinonimine (10). To a solution of 15 mg (0.02 mmol) of
9 in 10 mL of CHCl3, 150 µL of NaClO solution (available Cl2
10-13%) was added, and the mixture was stirred at room
temperature for 10 min. The solvent was evaporated, and the
residue was purified by column chromatography (SiO2, 0.5:99.5
MeOH/CH2Cl2). Yield 5.7 mg (38%). Rf (silica gel, 5:95 MeOH/
CH2Cl2) ) 0.51; 1H NMR (400 MHz, CD2Cl2): δ 0.83 (t, 6 H,
2 CH3), 1.16-1.39 (m, 16 H, 8 CH2), 1.86 (m, 2 H, R-CH2),
2.24 (m, 2 H, R-CH2), 5.17 (m, 1 H, 1 CH), 6.58 (dd, J ) 10.2
Hz, 2.2 Hz, 1 H, quinonimine), 6.71 (dd, J ) 10.1 Hz, 2.2 Hz,
1 H, quinonimine), 7.09 (d, J ) 8.4 Hz, 2 H, phenylene), 7.24
(dd, J ) 10.2 Hz, 2.7 Hz, 1 H, quinonimine), 7.35 (dd, J )
10.1 Hz, 2.7 Hz, 1 H, quinonimine), 7.42 (d, J ) 8.4 Hz, 2 H,
phenylene), 8.59-8.79 (m, 8 H, perylene); MS (MALDI,
negative): m/z ) 753.43 [M]- (calcd for C49H43N3O5: 753.32).
1
5:95 MeOH/CH2Cl2) ) 0.24; H NMR (500 MHz, CDCl3): δ
0.83 (t, 6 H, 2 CH3), 1.16-1.40 (m, 16 H, 8 CH2), 1.87 (m, 2
H, R-CH2), 2.25 (m, 2 H, R-CH2), 3.84 (s, 2 H, NH2), 5.19 (m,
1 H, 1 CH), 6.84 (d, J ) 8.4 Hz, 2 H, 2 CH phenylene), 7.11
(d, J ) 8.4 Hz, 2 H, 2 CH phenylene), 8.63-8.79 (m, 8 H,
perylene); HRMS (FAB+): m/z ) 664.3193 [M+H]+ (calc.
for C43H42O4N3: 664.3175).
NAD+ Analogue (6). A solution of 40 mg (0.06 mmol) of
the amine 4 in 60 mL of 2:1 CH2Cl2/MeOH was added slowly
to a solution of 130 mg (0.4 mmol) 3-(carbamoyl)-1-(2,4-
dinitrophenyl)-pyridinium chloride (5) in 50 mL of dry MeOH
in a pressure vessel. After two drops of pyridine (catalyst) was
added, the mixture was sealed and stirred at 65 °C for 17 h.
After being cooled, the solvent was evaporated, and the residue
was purified by column chromatography (SiO2, 1:10 MeOH/
CH2Cl2 to elute byproduct and then 2:5 MeOH/CH2Cl2 to elute
the product). To get rid of small amount of impurity 5, the solid
was dissolved in 40 mL of 1:1 MeOH/ CH2Cl2, and the solution
was concentrated to about 5 mL, at which point the product
precipitated out and was collected by membrane filtration (0.45
µm, Osmonics) and washed with 1 mL of MeOH. Yield: 22
mg (41%). Rf (silica gel, 1:5 MeOH/CH2Cl2) ) 0.52; 1H NMR
(500 MHz, 1:1 CD2Cl2/CD3OD): δ 0.85 (t, 6 H, 2 CH3), 1.19-
1.43 (m, 16 H, 8 CH2), 1.89 (m, 2 H, R-CH2), 2.28 (m, 2 H,
R-CH2), 5.19 (m, 1 H, 1 CH), 7.85 (d, J ) 8.4 Hz, 2 H, 2 CH
phenylene), 8.07 (d, J ) 8.4 Hz, 2 H, 2 CH phenylene), 8.45
(t, J ) 6.8 Hz, 1 H, 5-H nicotinamide), 8.54-8.85 (m, 8 H,
perylene), 9.24 (d, J ) 7.7 Hz, 1 H, nicotinamide), 9.43 (d, J
) 5.9 Hz, 1 H, nicotinamide), 9.74 (s, 1 H, 2-H nicotin-
amide); HRMS (FAB+): m/z ) 769.3439 [M-Cl]+ (calcd for
C49H45O5N4: 769.3390).
Conclusions
We have extended previously developed fluorescent probes
based on photoinduced electron transfer1 to rationally design a
new redox switch of fluorescence, featuring a biologically
important redox center, by incorporating nicotinamide (the redox
center for NAD+/NADH) into a fluorophore-spacer-receptor
structure. The NAD+ analogue was found to be strongly
fluorescent (ΦF ≈ 1.0) while the NADH analogue was non-
fluorescent (ΦF < 0.001). Repeated off/on switching of the
fluorescence was demonstrated chemically, by reduction with
NaBH3CN and oxidation with tetrachlorobenzoquinone, and
electrochemically, over at least 10 cycles with about 13%
degradation. Since many proton and metal ion sensors are based
on coordination by an amine group, our strategy of replacing a
free amine with a nicotinamide group might be a general method
to convert other cation sensors to redox switches. We have also
shown that ab initio electronic structure calculations of the
molecular orbital energy levels accurately predict the observed
PET quenching behavior. The potential application of such redox
switches as biosensors has also been demonstrated by coupling
the switch to the catalytic pathway of yeast alcohol dehydro-
genase, a common NADH/NAD+-utilizing enzyme.
4-Hydroxy-4′-aminodiphenylamine (8). To a Parr hydro-
genator was added 115 mg (0.5 mmol) of 4-hydroxy-4′-
nitrodiphenylamine (7), 3 mL of methanol, and 10 mg of 10%
Pd/C under argon. The system was charged with 40 psi H2,
and the resulting mixture was stirred at room temperature for
18 h. The Pd/C was removed by filtration through Celite, and
the solvent was evaporated under reduced pressure. The residue
was further purified by column chromatography (SiO2, 1:9
MeOH/CHCl3) twice to give a gray powder (76 mg, 76%). Rf
Acknowledgment. This work was supported primarily by
the NSF MRSEC DMR-0213574 and in part by Department of
Energy Office of Basic Energy Sciences under the Award
Agency Project DE-FG02-02ER15375. D.M.A. thanks Research
Corporation for a Cottrell Scholar award RC#CS0937.
1
(silica gel, 1:9 MeOH/CHCl3) ) 0.26; H NMR (500 MHz,
CDCl3): δ 3.47 (s, 2 H, NH2), 4.33 (s, 1 H, OH), 5.17 (s, 1 H,
NH), 6.64 (d, J ) 8.8 Hz, 2 H), 6.72 (d, J ) 8.4 Hz, 2 H), 6.83
(d, J ) 8.4 Hz, 2 H), 6.86 (d, J ) 8.8 Hz, 2 H). 13C NMR (75
MHz, CD3OD): δ 116.68, 118.16, 120.35, 120.65, 138.97,
139.49, 141.46, 152.10. MS (APCI+): m/z ) 201 [M+H]+.
Supporting Information Available: 1H NMR spectra for
3, 4, 6, 8-10. This material is available free of charge via the
Internet at http://pubs.acs.org.
Hydroquinonimine (9). A total of 40 mg (0.2 mmol) of
4-hydroxy-4′-aminophenylamine (8), 57 mg (0.1 mmol) of N-(1-
hexylheptyl)perylene-3,4:9,10-tetracarboxyl-3,4-anhydride-9,10-
imide (1), and 2 g of imidazole was stirred at 130 °C under
argon for 5 h. After being cooled, the mixture was dispersed in
3 mL of ethanol, acidified to pH ) 1 with 2 N HCl, and stirred
for 1 h. The red precipitate was collected by vacuum filtration,
thoroughly washed with H2O, redissolved in CH2Cl2, and
purified by column chromatography (SiO2, CHCl3). Yield 28
mg (37%). Rf (silica gel, 5:95 MeOH/CH2Cl2) ) 0.26; 1H NMR
(500 MHz, CDCl3): δ 0.83 (t, 6 H, 2 CH3), 1.16-1.40 (m, 16
H, 8 CH2), 1.87 (m, 2 H, R-CH2), 2.25 (m, 2 H, R-CH2), 4.62
(s, 1 H, OH), 5.19 (m, 1 H, 1 CH), 5.66 (s, 1 H, NH), 6.83 (d,
J ) 8.8 Hz, 2 H), 7.02 (d, J ) 8.4 Hz, 2 H), 7.12 (d, J ) 8.8
Hz, 2 H), 7.16 (d, J ) 8.4 Hz, 2H), 8.62-8.79 (m, 8 H,
References and Notes
(1) Zang, L.; Liu, R.; Holman, M. W.; Nguyen, K. T.; Adams, D. M.
J. Am. Chem. Soc. 2002, 124, 10640.
(2) Molecular Switches; Feringa, B. L., Ed.; Wiley-VCH: Weinheim,
Germany, 2001.
(3) Feringa, B. L.; Jager, W. F.; de Lange, B. Tetrahedron 1993, 49,
8267.
(4) Irie, M., Ed. Special Issue: Photochromism: Memories and
Switches. Chem. ReV. 2000, 100, 1683-1890.
(5) Hernandez, R.; Tseng, H.; Wong, J. W.; Stoddart, J. F.; Zink, J. I.
J. Am. Chem. Soc. 2004, 126, 3370.
(6) de Silva, A. P.; Gunaratne, H. Q.; Gunnlaugsson, T.; Huxley, A. J.
M.; McCoy, C. P.; Rademacher, J. T.; Rice, T. E. Chem. ReV. 1997, 97,
1515.
(7) Brasselet, S.; Moerner, W. E. Single Mol. 2000, 1, 17.
(8) Weiss, S. Science 1999, 283, 1676.

Molecular Switch Based on Bio-Redox Reaction
J. Phys. Chem. B, Vol. 109, No. 1, 2005 137
(9) Xie, X. S.; Trautman, J. K. Annu. ReV. Phys. Chem. 1998, 49, 441.
(10) Bissell, R. A.; de Silva, A. P.; Gunaratne, H. Q. N.; Lynch, P. L.
M.; Maguire, G. E. M.; McCoy, C. P.; Sandanayake, K. R. A. S. Top. Curr.
Chem. 1993, 168, 223.
(11) Aviram, A.; Ratner, M. A. Chem. Phys. Lett. 1974, 29, 277.
(12) Carroll, R. L.; Gorman, C. B. Angew. Chem., Int. Ed. 2002, 41,
4378.
(35) Suzuki, T.; Migita, A.; Higuchi, H.; Kawai, H.; Fujiwara, K.; Tsuji,
T. Tetrahedron Lett. 2003, 44, 6837.
(36) Yee, D. J.; Balsanek, V.; Sames, D. J. Am. Chem. Soc. 2004, 126,
2282.
(37) Underwood, A. L.; Burnett, R. W. Electrochemistry of Biological
Compounds. In Electroanalytical Chemistry; Bard, A. J., Ed.; Marcel
Dekker: New York, 1973; Vol. 6, p 1.
(13) Adams, D. M.; Brus, L.; Chidsey, C. E. D.; Creager, S.; Creutz,
C.; Kagan, C. R.; Kamat, P. V.; Lieberman, M.; Lindsay, S.; Marcus, R.
A.; Metzger, R. M.; Michel-Beyerle, M. E.; Miller, J. R.; Newton, M. D.;
Rolison, D. R.; Sankey, O.; Schanze, K. S.; Yardley, J.; Zhu, X. J. Phys.
Chem. B 2003, 107, 6668.
(14) Fluorescent Chemosensors for Ion and Molecule Recognition;
Czarnik, A. W., Ed.; ACS Symposium Series 538; American Chemical
Society: Washington, DC, 1993.
(15) Probe Design and Chemical Sensing. Topics in Fluorescence
Spectroscopy; Lakowicz, J. R.; Ed.; Plenum: New York, 1994; Vol. 4.
(16) de Silva, A. P.; Fox, D. B.; Moody, T. S.; Weir, S. M. TRENDS
Biotechnol. 2001, 19, 29.
(17) Fabbrizzi, L.; Licchelli, M.; Pallavicini, P. Acc. Chem. Res. 1999,
32, 846.
(38) Langhals, H. Heterocycles 1995, 40, 477.
(39) England, P. A.; Harford-Cross, C. F.; Stevenson, J.-A.; Rouch, D.
A.; Wong, L.-L. FEBS Lett. 1998, 424, 271.
(40) Imanishi, T.; Hamano, Y.; Yoshikawa, H.; Iwata, C. J. Chem. Soc.,
Chem. Commun. 1988, 473.
(41) Rathore, R.; Le Magueres, P.; Lindeman, S. V.; Kochi, J. K. Angew.
Chem., Int. Ed. 2000, 39, 809.
(42) Gorton, L.; Dominguez, E. Electrochemistry of NAD(P)+/
NAD(P)H. In Encyclopedia of Electrochemistry; Bard, A. J., Stratmann,
M., Eds.; Wiley-VCH: New York, 2002; Vol. 9, p 67.
(43) Elving, P. J. In Topics in Bioelectrochemistry and Bioenergetics;
Milazzo, G., Ed.; Wiley: New York, 1976; p 179.
(44) Moracci, F. M.; Liberatore, F.; Carelli, V.; Arnone, A.; Carelli, I.;
Cardinali, M. E. J. Org. Chem. 1978, 43, 3420.
(45) Anne, A.; Hapiot, P.; Moiroux, J.; Save´ant, J.-M. J. Electroanal.
Chem. 1992, 331, 959.
(46) Salbeck, J.; Kunkely, H.; Langhals, H.; Saalfrank, R. W.; Daub, J.
Chimia 1989, 43, 6.
(47) Onoda, M.; Uchiyama, S.; Santa, T.; Imai, K. Anal. Chem. 2002,
74, 4089.
(18) Fabbrizzi, L.; Licchelli, M.; Rabaioli, G.; Taglietti, A. Coord. Chem.
ReV. 2000, 205, 85.
(19) Tsien, R. Y. Chem. Eng. News 1994, 72, 34.
(20) James, T. D.; Sandanayake, K. R. A. S.; Shinkai, S. Angew. Chem.,
Int. Ed. Engl. 1996, 35, 1910.
(21) Tsien, R. Y. Biochemistry 1980, 19, 2396.
(22) Burdette, S. C.; Walkup, G. K.; Spingler, B.; Tsien, R. Y.; Lippard,
S. J. J. Am. Chem. Soc. 2001, 123, 7831.
(23) Walkup, G. K.; Burdette, S. C.; Lippard, S. J.; Tsien, R. Y. J. Am.
Chem. Soc. 2000, 122, 5644.
(24) Guo, X.; Qian, X.; Jia, L. J. Am. Chem. Soc. 2004, 126, 2272.
(25) Nolan, E. M.; Lippard, S. J. J. Am. Chem. Soc. 2003, 125, 14270.
(26) Vance, D. H.; Czarnik, A. W. J. Am. Chem. Soc. 1994, 116, 9397.
(27) Fabbrizzi, L.; Pallavicini, P.; Parodi, L.; Taglietti, A. Inorg. Chim.
Acta 1995, 238, 5.
(48) Miura, T.; Urano, Y.; Tanaka, K.; Nagano, T.; Ohkubo, K.;
Fukuzumi, S. J. Am. Chem. Soc. 2003, 125, 8666.
(49) Tanaka, K.; Miura, T.; Umezawa, N.; Urano, Y.; Kikuchi, K.;
Higuchi, T.; Nagano, T. J. Am. Chem. Soc. 2001, 123, 2530.
(50) Langhals, H.; Demmig, S.; Huber, H. Spectrochim. Acta 1988, 44A,
1189.
(51) Kaiser, H.; Lindner, J.; Langhals, H. Chem. Ber. 1991, 124, 529.
(52) Holman, M. W.; Liu, R.; Adams, D. M. J. Am. Chem. Soc. 2003,
125, 12649.
(28) Fabbrizzi, L.; Francese, G.; Licchelli, M.; Perotti, A.; Taglietti, A.
Chem. Commun. 1997, 581.
(29) Nakata, E.; Nagase, T.; Shinkai, S.; Hamachi, I. J. Am. Chem. Soc.
2004, 126, 490.
(53) Brewster, M. E.; Kaminski, J. J.; Gabanyi, Z.; Czako, K.; Simay,
A.; Bodor, N. Tetrahedron 1989, 45, 4395.
(54) Brewster, M. E.; Simay, A.; Czako, K.; Winwood, D.; Farag, H.;
Bodor, N. J. Org. Chem. 1989, 54, 3721.
(30) Haugland, R. P. Handbook of Fluorescent Probes and Research
Products; 9th ed.; Molecular Probes: Eugene, OR, 2002.
(31) Fabbrizzi, L.; Licchelli, M.; Mascheroni, S.; Poggi, A.; Sacchi, D.;
Zema, M. Inorg. Chem. 2002, 41, 6129.
(32) Goulle, V.; Harriman, A.; Lehn, J.-M. J. Chem. Soc., Chem.
Commun. 1993, 1034.
(33) Daub, J.; Martin, B.; Knorr, A.; Spreitzer, H. Pure Appl. Chem.
1996, 68, 1399.
(34) Li, H.; Jeppesen, J. O.; Levillain, E.; Becher, J. Chem. Commun.
2003, 846.
(55) Park, K. K.; Han, D.; Shin, D. Bull. Korean Chem. Soc. 1986, 7,
201.
(56) Lettre, H.; Haede, W.; Ruhbaum, E. Liebigs Ann. Chem. 1953, 579,
123.
(57) Halasz, A.; Cohen, D. U.S. Patent 4 021 486 1977.
(58) Langhals, H.; Karolin, J.; Johansson, L. B.-A. J. Chem. Soc.,
Faraday Trans. 1998, 94, 2919.
(59) Giaimo, J. M.; Gusev, A. V.; Wasielewski, M. R. J. Am. Chem.
Soc. 2002, 124, 8530.
(60) Jaguar; 4.1, Schro¨dinger, Inc.: Portland, OR, 2000.