Fluorescent stilbazolium dyes as probes of the norepinephrine transporter: Structural insights into substrate binding

Article abstract of DOI:10.1039/c2ob26633d

We report the synthesis, binding kinetics, optical spectroscopy and predicted binding modes of a series of sterically demanding, fluorescent norepinephrine transporter (NET) ligands. A series of bulky stilbazolium dyes, including six newly synthesized com

Full text of DOI:10.1039/c2ob26633d

View Article Online / Journal Homepage / Table of Contents for this issue
Dynamic Article Links
Organic &
Biomolecular
Chemistry
Cite this: Org. Biomol. Chem., 2012, 10, 8710
www.rsc.org/obc
PAPER
Fluorescent stilbazolium dyes as probes of the norepinephrine transporter:
structural insights into substrate binding†
James N. Wilson,* Adrienne S. Brown, W. Michael Babinchak, Clark D. Ridge‡ and Jamie D. Walls
Received 16th August 2012, Accepted 13th September 2012
DOI: 10.1039/c2ob26633d
We report the synthesis, binding kinetics, optical spectroscopy and predicted binding modes of a series of
sterically demanding, fluorescent norepinephrine transporter (NET) ligands. A series of bulky
stilbazolium dyes, including six newly synthesized compounds, were evaluated to determine the effect of
extending the molecular probes’ ‘heads’ or ‘tails’. Taking advantage of the dyes’ characteristic ‘turn-on’
emission, the kinetic binding parameters, kon and koff were determined revealing that extension of the
molecules’ tails is well tolerated while expansion of the head is not. Additionally, a ‘headfirst’ orientation
appears to be preferred over a ‘tail-first’ binding pose. Further details of the possible binding modes were
obtained from the emission spectra of the bound probes. A small range of interplanar twist angles,
approximately 35° to 60°, is predicted to produce the observed emission. Docking experiments and
molecular modelling support the kinetic and spectroscopic data providing structural insights into substrate
binding.
homologous monoamine transporters (MATs) such as the seroto-
Introduction
nin transporter (SERT).6 The transport of ASP+ and a related
The norepinephrine transporter (NET) is a member of the solute
carrier 6 (SLC6) family of sodium : neurotransmitter symporters;
it is responsible for reuptake of norepinephrine from the synaptic
cleft, but may bind or transport other substrates as well. NET is
the target of several exogenous ligands, including recreational
drugs such as cocaine as well as reuptake inhibitors such as desi-
pramine or venlafaxine.1 Several fluorescent probes have also
been demonstrated as NET ligands and/or transportons:2 Hadrich
et al. reported that NET is capable of binding bulky fluorescent
derivatives of guanidine, cocaine and nisoxetine3 while DeFelice
and coworkers demonstrated that NET can not only bind bulky
substrates, but is capable of transporting them, as well.4 Utilizing
4-[4-(dimethyl amino)styryl]-N-methylpyridinium (ASP+ or A1,
Fig. 1), a fluorescent stilbazolium dye, binding stoichiometry
and kinetics of uptake were also investigated.5 These studies
demonstrate that NET can accommodate expanded transportons,
however, it is not clear what the limits of substrate binding are.
Furthermore, little is known about the underlying mechanism by
which ASP+ emission is enhanced upon binding to NET and
why such an enhancement is not observed when binding to
stilbazolium dye, 4-[2-(6-hydroxy-2-naphthalenyl)-ethenyl]-1-
methylpyridinium (HNEP+ or H1),7 indicates that NET may
accommodate substrates at least 15 in length in the putative
binding pocket (Fig. 1), which is significantly greater than the
8.8 of the native substrate, norepinephrine. We recently reported
that stilbazolium dimers, based on A1 and H1 (e.g. D1, Fig. 2),
may bind to, but are not transported via NET.8 These probes
extended to approximately 35 Å in length, demonstrating that
inclusion of a bulky ‘tail’ appears to be a key feature that limits
transport activity while still permitting probe binding to NET.
The difference in length between A1 and D1 is approximately
20 Å and it is likely that substrates with dimensions intermediate
to A1 and D1 may exhibit both binding to and transport via NET.
In the present contribution, we investigate three aspects of
substrate binding to NET with the overall goal of understanding
the observed ‘turn-on’ emission of NET-active stilbazolium
dyes. First, we examine the effect of extending the aliphatic tail
to increase the overall length, l, of the fluorescent probes
(Fig. 1). Second, we explore the impact of expanding the
aromatic core perpendicular to the long axis of the dyes, i.e.
d, through benzo-fusion to the electron-donating head group.
Finally, we attempt to identify if there is a preference for sub-
strate orientation in the NET binding pocket, i.e. do stilbazolium
dyes insert headfirst or tail-first? In addressing these three para-
meters, we aim to understand the functional limits of the NET
transporter (and by extension, the related sodium : neurotransmit-
ter symporters). Taking advantage of stilbazolium dyes’ excellent
sensitivity to their chemical microenvironment, we are able to
Department of Chemistry, University of Miami, 1301 Memorial Drive,
Coral Gables, FL 33124, USA. E-mail: jnwilson@miami.edu;
Fax: +1 305 284 4571; Tel: +1 305 284 2619
†Electronic supplementary information (ESI) available: 1H-NMR and
13C-NMR of newly synthesized compounds, HSQC of 11, table of
TD-DFT calculated photophysical data. See DOI: 10.1039/c2ob26633d
‡Present address: Office of Regulatory Science, CFSAN, FDA, 5100
Paint Branch Parkway, College Park, MD 20740, USA
8710 | Org. Biomol. Chem., 2012, 10, 8710–8719
This journal is © The Royal Society of Chemistry 2012

View Article Online
Fig. 1 (A) The structure of the transmembrane norepinephrine trans-
porter (NET) adapted from reference 10 with the position of the binding
pocket indicated. (B) Expanded view of the binding pocket showing the
two aromatic ‘gatekeeper’ residues, TYR152 and PHE317 (orange, at
middle) and two additional aromatic residues, PHE72 and PHE323 that
construct the pocket ‘floor’ (orange, at bottom). ASP75 and ASP473, in
pink, may stabilize bound cationic stilbazolium dyes. The binding
pocket is approximately 15 in height between VAL325 and GLY383,
with a minimum cross section of 4.2 between TYR152 and PHE317.
(C) Derivatives of stilbazolium dyes A1 and H1 were constructed with
expanded aryl ‘heads’ and extended ‘tails’ to probe the limits of sub-
strate binding in this pocket.
Fig. 2 Chemical structures of the stilbazolium dyes utilized to probe
the NET binding pocket; norepinephrine (NE) is shown for comparison.
functional limits of NET. H2–6 were synthesized by Knoevena-
gel condensation11 of alkyl-picolinium bromides with 6-
hydroxy-2-naphthaldehyde (Scheme A, Fig. 2) adapting our pre-
viously described synthesis;7,8 A2–10 have previously been
described.12–16 The synthesis of D2 (Scheme B, Fig. 2) first
required the generation of dialdehyde 11, via nucleophilic substi-
tution of 4-fluorobenzaldehyde with N,N′-dimethyl-1,6-hexane-
diamine; D2 was then synthesized via condensation of 11 with
1,4-dimethylpyridinium iodide. Characterization details of the
six novel compounds, D2 and H2–6, are in the Experimental
section below; the identities of the remaining compounds were
monitor binding to NET with spatial, spectral and temporal resol-
ution. Most probes were found to bind to NET, with several com-
pounds exhibiting NET-dependent transport. The bound substrates
exhibit emission spectra that are hypsochromically shifted relative
to spectra obtained in solvents of moderate to high polarity. From
this data, we conclude that probes with extended tails, but not
expanded heads may access a solvent-excluding and sterically
constrained substrate binding pocket. We evaluate the observed
binding behavior against predicted binding modes to a recently
reported ‘open-to-out’ NET model.9,10 The predicted binding
modes do not appear to be entirely congruent with the observed
binding behavior, suggesting that the stilbazolium dyes bind to a
transporter conformation intermediate to the current open-to-out
model and the predicted ‘open-to-in’ conformation.
1
confirmed by H NMR and HRMS. Fig. 2 shows the chemical
structures of the probes; the series A2–4 and H2–4 incorporate
additional methylene units building on the N-methyl tails of A1
and H1, to produce ethyl, n-propyl and n-butyl functionalized
probes. Fig. 3 depicts the surface electrostatic potential of the
probes; calculations were performed using DFT at the 6-31*
level with the B3LYP basis set. The corresponding probe dimen-
sions are listed in Table 1. In the extended conformation, the
length of the probes increases by approximately 1.2 Å per
methylene unit. The isoamyl and benzyl groups of A5,6 and
H5,6 place additional steric bulk on the probes’ tails
Results and discussion
Probe design and synthesis
Using the known NET substrates A1,4,5 D18 and H17 as tem-
plates, we designed a family of related dyes to probe the
This journal is © The Royal Society of Chemistry 2012
Org. Biomol. Chem., 2012, 10, 8710–8719 | 8711

View Article Online
Fig. 4 TD-DFT optimized geometry of H1 showing the twisted intra-
molecular charge transfer (TICT) excited state. Panels A and B show
alternate in-plane views.
Fig. 3 Electrostatic potential maps of the stilbazolium dyes utilized to
probe the NET binding pocket.
Table 1 Physical dimensions of probes from optimized geometriesa
l (Å)
d (Å)
A (Å2)
V (Å3)
Fig. 5 Optical spectra of representative stilbazolium probes containing
the N,N-dimethylaniline (D2) and hydroxynaphthalene (H6) moieties in
PBS and octanol. The lowest energy absorption bands (solid lines) are
moderately bathochromically and hyperchromically shifted in octanol.
The relative emission intensities are shown with dashed lines; probes are
emissive in octanol or PEG (data not shown), but quenched in PBS.
A1
A2
A3
A4
A5
A6
A7
A8
A9
A10
D1
D2
H1
H2
H3
H4
H5
H6
14.7
15.8
16.7
18
18.1
17.1
15.9
16.7
14.4
16
33.6
30.9
15.2
16.4
17.4
18.7
18.6
17.5
4.3
4.3
4.3
4.3
4.3
5.9
4.3
4.3
6.8
7.2
4.3
5.0
5.6
5.6
5.6
5.6
5.6
5.9
300.2
320.6
340.8
360.9
379.0
381.2
323.0
338.3
340.7
367.8
641.9
647.9
304.6
325.0
345.2
365.4
383.3
386.0
280.5
299.1
317.5
335.9
354.0
364.5
306.0
322.9
330.7
353.8
617.8
621.4
289.7
308.3
326.7
345.1
363.2
373.8
exhibit broad amorphous absorption peaks that are characteristic
of charge transfer (CT) transitions and, in the excited state, may
relax to twisted intramolecular charge transfer (TICT) excited
states.17–19 Quantum chemical calculations suggest TICT is
the primary mechanism of thermal relaxation for A113,17 and
time dependent density functional (TD-DFT calculations at the
6-31G* level using the B3LYP basis set predict a TICT state for
H1 as well (Fig. 4). Thus, in polar solvents such as phosphate
buffered saline (PBS), it was expected that the A and H series
would be weakly emissive or quenched. In less polar solvents
such as octanol or viscous solvents such as polyethylene glycol
(PEG), they should display good to moderate quantum yields
enabling their use as so-called ‘turn-on’ reporters in the context
of NET binding.
Probes A2–8 as well as D1 and D2 possess essentially iden-
tical conjugated cores as A1 and were anticipated to exhibit
similar optical properties. UV-vis and fluorescence spectroscopy
confirm that the absorption and emission of the parent chromo-
phore is not significantly changed upon introduction of alternate
tails (Fig. 5, Table 2). A2–8 were found to have absorption
maxima centered near 480 nm in octanol with molar absorptiv-
ities of approximately 35 000 M−1, cm−1; emission maxima
were 598 nm in octanol with quantum yields of 0.06. The H2–6
series exhibited nearly identical optical properties as that of
the parent chromophore, H1. One exception is H6, for which the
absorption and emission maxima w ere slightly red shifted. The
absorption of D2 is slightly hypsochromically shifted relative to
A1, in contrast to the bathochromic shift previously observed for
D1. The emission maximum of both dimers is 481 nm, which is
a 17 nm blue shift relative to A1. While structurally related to
a Spartan 08, DFT 6-31G* B3LYP.
perpendicular to the long axis of the molecule. Probes A7–A10
extend the head of A1, with pyrrolidine and piperidine units
replacing the dimethylamino group on A7 and A8, respectively.
A9 is a benzofused analog of A1, while A10 incorporates the
most bulky head with a N-ethyl carbazole replacing the N,N-
dimethylaniline of A1. Finally, D2 is the head-to-head isomer of
D1 with virtually identical dimensions, however, the pyridinium
groups are located on either end of D2, while placed internally
in D1. While the monomer probes may be able to bind headfirst
or tail-first, these two dimers allow us to explicitly test how the
orientation of the charge influences binding.
Optical properties
As a result of the donor–acceptor architecture, stilbazolium dyes
are sensitive to their chemical microenvironment and respond
with changes in their absorption and emission parameters. They
8712 | Org. Biomol. Chem., 2012, 10, 8710–8719
This journal is © The Royal Society of Chemistry 2012

View Article Online
Table 2 Summary of optical data for representative probesa
Octanol
PBS
λmax, abs (nm)
log ε
λmax, em (nm)
Φ
λmax, abs (nm)
log ε
λmax, em (nm)
Φ
I
oct/IPBS
A1
A2
A6
A9
A10
D1
D2
H1
H2
H6
481
481
481
458
446
484
475
428
428
434
4.59
4.59
4.49
4.36
4.62
4.60
4.65
4.54
4.54
4.56
598
598
598
628
566
581
581
571
570
579
0.06
0.06
0.06
0.02
0.45
0.16
0.06
0.16
0.16
0.16
449
449
462
413
419
458
460
391
391
399
4.43
4.43
4.36
4.34
4.56
4.45
4.53
4.43
4.43
4.52
590
590
612
658
581
614
610
559
559
561
0.003
0.003
0.004
10−4
0.01
15
15
17
>30
43
>160
>100
16
10−4
10−4
0.01
0.01
0.008
16
20
a A3–5 and H3–5 values are identical to A2 and H2, respectively.
A1, the extended conjugation of A9 and A10 alters the photo-
physical parameters of these two probes. The absorption
maximum of both of these compounds is hypsochromically
shifted relative to A1; the emission of A9 is bathochromic shift
of 30 nm, while the emission of A10 is hypsochromically shifted
by 22 nm with respect to A1.
In aqueous solutions, such as PBS, emission of the stilbazo-
lium dyes is quenched (Table 2), likely due to the formation of
an excited state with strong charge transfer (CT) character. This
results in good to excellent ‘on–off’ ratios as determined by
comparing the emission intensity in octanol ET(30) = 48.3 kcal
mol−1, η0 = 7.24)20,21 versus PBS. While octanol cannot pre-
cisely mimic the binding pocket of NET, this ON–OFF ratio pro-
vides a reasonable approximation of the emission enhancement
that might be observed upon entering a relatively nonpolar and
constricted binding pocket.22
NET binding studies
The switchable emission of the stilbazolium dyes enables their
use as ‘turn-on’ reporters without need for rinsing steps to
remove unbound probes. In cellular assays, the polar environ-
ment of the aqueous media limits probe emission through stabi-
lization of the TICT excited state. As previously reported for A1,
D1, and H1, introduction of the probes results in a rapid increase
in emission intensity as a result of binding to NET.4,5,7,8 In the
case of A1 and H1 this initial rapid rise is followed by a more
gradual increase due to probe transport and intracellular accumu-
lation in mitochondria.4,5 Thus, NET binding and uptake can be
temporally and spatially resolved by confocal microscopy.
We performed time course experiments to evaluate probe
binding and displacement using both desipramine and norepi-
nephrine as competitive inhibitors. HEK-293 cells stably expres-
sing hNET (provided by R. Blakely, Vanderbilt)4,23 were
exposed to 1 μM solutions of the stilbazolium dyes, A1–10, D1,
D2, and H1–6 at t = 0 s followed by desipramine at t = 60 s. For
A and D series probes, excitation was achieved with either a 476
or 488 nm laser lines and emission was collected from 500 to
575 nm; for the H series excitation was at 405 nm and emission
was collected between 475 and 575 nm. Representative images
from a 120 s experiment are shown in Fig. 6A; typical binding
and displacement profiles captured from these experiments are
Fig. 6 (A) Images from time-dependent NET binding and displace-
ment studies of A2. Panels B–E show binding and displacement curves
for A2, A6, H2 and H6, respectively; scale bar is 10 μm.
shown in Fig. 6B–E (A2, A6, H2, H6). Table 3 shows kon, koff
and KD values calculated from these curves. Characteristic ‘turn-
on’ emission was observed for most of the probes investigated,
however, no binding was observed for A9, A10 and D2. The
values for kon are quite similar for most probes at approximately
2.0 × 105 M−1, s−1; A6, D1 and H5 are slightly lower although
This journal is © The Royal Society of Chemistry 2012
Org. Biomol. Chem., 2012, 10, 8710–8719 | 8713

View Article Online
Table 3 Summary of binding and displacement kinetics
kon (M−1, s−1 a
)
koff (s−1 a
)
KD (nM)a
R2
A1
A2
A3
A4
A5
A6
A7
A8
A9
A10
D1
D2
H1
H2
H3
H4
H5
H6
2.1 × 105 0.4 × 105
2.3 × 105 0.9 × 105
2.1 × 105 0.6 × 105
2.0 × 105 8.5 × 104
1.9 × 105 4.6 × 104
8.8 × 104 2.3 × 104
2.3 × 105 6.0 × 104
1.8 × 105 1.4 × 105
9.4 × 10−2 1.4 × 10−3
7.7 × 10−2 2.2 × 10−3
4.7 × 10−2 8.4 × 10−3
6.1 × 10−2 2.0 × 10−3
2.9 × 10−2 4.6 × 10−3
3.5 × 10−2 5.4 × 10−3
3.4 × 10−2 6.1 × 10−3
2.4 × 10−1 9.0 × 10−2
456 129
340 206
222 86
300 194
147 51
397 157
147 58
0.99
0.91
0.96
0.93
0.95
0.96
0.95
0.95
—
—
0.95
—
0.91
0.95
0.99
0.94
0.94
0.91
1354 1448
—
—
b
b
b
b
—
—
b
b
b
b
—
1.1 × 105 2.1 × 105
—
1.6 × 10−1 2.1 × 10−2
141 39
b
b
b
b
—
—
—
2.8 × 105 7.6 × 105
2.5 × 105 6.6 × 105
1.8 × 105 2.3 × 1045
2.2 × 105 5.6 × 105
1.4 × 105 6.0 × 105
2.2 × 105 5.5 × 105
2.5 × 10−1 7.9 × 10−3
5.8 × 10−1 1.2 × 10−2
4.4 × 10−1 5.6 × 10−3
2.9 × 10−1 6.0 × 10−3
6.0 × 10−1 1.8 × 10−2
2.3 × 10−1 4.5 × 10−3
92 45
234 98
240 54
132 52
414 266
101 40
a Calculated from four independent experiments. b No binding observed.
the highest value (H1) and lowest value (A7) differ by a factor
of only 3; off rates ranged from 0.60 s−1 to 0.03 s−1. With the
exception of A8, the compounds possess very similar KD values,
ranging from 92 nM to 456 nM with an average value of 240 nM.
From these time-dependent experiments we can make three
conclusions regarding the binding of the stilbazolium dyes to
NET. First, turn-on emission is the result of NET binding events
as this behavior was only observed for cells stably expressing
hNET and not in control HEK-293 cells. This notion is sup-
ported by the desipramine and norepinephrine induced displace-
ment. Second, extension of the probe tails is well tolerated based
on the binding behaviour of A1–6 and H1–6. In contrast, the
expansion of the aryl heads is not well tolerated as evident by
the lack of binding observed for A9 and A10, which possess
additional aromatic rings on their electron donating heads. Third,
the orientation of the cationic picolinium group may determine
whether probes bind or not. For the monomeric dyes, both head
first and tail-first orientations are possible. However, for the two
isomeric dimers, D1 can only bind in a head first orientation,
while D2 can only bind in a tail-first fashion. As only D1 was
found to bind to NET, it appears that the head first orientation is
preferred. The lack of binding dependent emission for A9 and
A10 also supports this final point. The bulky head groups of
these dyes likely prevents their binding in a headfirst orientation,
and while we cannot exclude the existence of a spectroscopically
silent tail-first binding mode, there is no data to suggest such
binding occurs.
Fig. 7 Linear regression analysis reveals a significant (R2 > 0.8)
relationship between substrate volume and kon (H1–5) or koff (A1–5).
The emission spectra of the bound probes confirm that they
experience a similar hydrophobic and/or restricted environment.
The NET-bound probes (in green, Fig. 8) are spatially segregated
from the intracellularly accumulated population (in red, Fig. 8)
while probes in the bulk aqueous media are non-emissive. Thus,
the spectra of the bound probes can be captured by imaging in
xyλ mode and compared with solution spectra and calculated
spectra. Emission of the bound probes is hypsochromically
shifted to most solvents studied, which possess ET30 values
ranging from 33.9 to 63.1 kcal mol .
−1 20 For the A series of dyes
(excluding A9 and A10, which were not observed to bind), the
emission maximum of the bound probes is 565 nm; for H1–5,
the emission maximum is 545 nm. One explanation this effect is
that the bound probes may experience a less polar environment
than in bulk solution. This is consistent with the presence of
multiple hydrophobic and aromatic residues that line the
putative NET binding pocket (Fig. 1). A second factor may be
the restrictive environment of the binding pocket that limits
access to the non-emissive TICT excited state. In order to
examine the possibility of this second effect, we performed
TD-DFT calculations at different interplanar rotations in range of
solvent polarities then evaluated several possible conformations
of bound probes.
A general trend was not found for the effect of tail extension
on binding or displacement as poor or no correlation was found
by regression analysis (R2 < 0.35) when relating physical para-
meters (l, d, A or V, see values in Table 1) to kon or koff
.
However, when considering only the series of extended alkyl
tails, i.e. A1–5 and H1–5, a significant relationship was found
between volume of the probe and either kon (H1–5) or koff
(A1–5). Additional methylenes result in lower koff in the case of
A1–5 (Fig. 7, R2 = 0.84) and lower kon for the series of H1–5
(Fig. 7, R2 = 0.80).
8714 | Org. Biomol. Chem., 2012, 10, 8710–8719
This journal is © The Royal Society of Chemistry 2012

View Article Online
Fig. 9 Transition energies of A1 (left) and H1 (right) as a function of
interplane twist angle. The oscillator strength for each transition is pro-
portional to the data point area; weaker transitions were observed at
larger twist angles, reflecting poor HOMO–LUMO overlap.
(35° to 60°) is predicted for the H series of dyes that emit at
550 nm (2.25 eV). Furthermore, the calculated oscillator
strengths provide additional limits on the polarity of the NET
binding pocket. In more polar environments such as H2O
(63.1 kcal mol−1) or THF (37.4 kcal mol−1), the dyes are pre-
dicted to have weak or no oscillator strength (ƒ ≤ 0.15) at tran-
sition energies corresponding to the observed 550 nm emission.
This suggests that not only do the A and H stilbazolium dyes
adopt similar twist angles in the NET binding pocket, but that
the environment in which they bind has an equivalent polarity to
Fig. 8 (A) The emission of the NET bound probes such as A4 (in
green) is spatially distinct from the intracellular pool (in red). (B) Phase
contrast image; scale bar is 10 μm. (C) Spectral resolution of the NET
bound probes (e.g. A4) allows comparison to different solvent
environments.
solvents falling within 33.9 kcal mol−1 and 37.4 kcal mol−1
.
To relate the torsional dependence of emission to possible
binding modes, we next examined conformations of the dyes in
both rigid and flexible docking experiments. AutoDock
Vina25–27 was utilized to identify the putative poses of the A, D
and H series of dyes in the ‘open-to-out’9 homology model10 of
NET in rigid docking experiments (Fig. 10). The dyes are pre-
dicted to align with their long axis parallel to the substrate trans-
port channel; dyes with longer tails (e.g. butyl, isoamyl or
benzyl) as well as the dimers, D1 and D2, were found to have
moderate (<30°) displacements from dyes with shorter tails.
These results are consistent with our previous report on related
stilbazolium dimers.8 Using the representative poses obtained
from AutoDock Vina, we next performed geometry optimiz-
ations (MMFF, Spartan 08) on A1 and H1 in the NET binding
pocket with the following flexible residues: ASP75, ASP473,
PHE72, PHE317, PHE329, and TYR152 (Fig. 11). For A1,
three key differences are predicted between a headfirst and tail-
first orientation in the binding pocket: First, the tail-first orien-
tation results in A1 sitting approximately 2.0 Å higher in the
binding pocket. This results in slightly greater exposure to the
polar solvent environment, as well as a less rotational restriction
in the excited state. The headfirst orientation sits lower in the
binding pocket and is better shielded from solvent. The differ-
ence in ligand depth in the binding pocket is related to the
second difference between these two binding poses. For the
headfirst orientation, the cationic pyridinium is in close proxi-
mity and likely stabilized by ASP473. In contrast, the tail-first
orientation is predicted to interact with ASP75, on the opposite
side of the binding pocket (Fig. 11B, D). The third difference
between headfirst and tail-first orientations is the interplane
torsion angles. The tail-first orientation exhibits a large inter-
plane twist of 70.3° while the headfirst orientation shows a much
smaller twist of 30.0°. Comparing these two geometries with the
Quantum chemical calculations and molecular modelling
The energy of the electronic transitions in dyes exhibiting TICT
excited states can be correlated to interplanar torsions, thereby
providing information about the fluorophore conformation
during photoemission. In order to relate the observed fluor-
escence with possible binding modes, we performed TD-DFT
calculations at the 6-31G(d) on the chromophore cores, A1 and
H1, using the IEFPCM solvent model.24 Ten geometries were
examined beginning from the ground state optimized geometries
and progressing through 10° interplane torsions to 90°. Plots of
the S0 → S1 transition energies as a function of torsion angle are
shown in Fig. 9 with the area of the data points proportional to
the oscillator strength. The vertical excitation energies for the
ground state optimized geometry match very well with the exper-
imentally determined values (less than 0.04 eV or 1.6% differ-
ence). Increasing the twist angle from 0° to 90° decreases the S0
→ S1 transition energy, while simultaneously decreasing the
probability of the transition. This result is consistent with pre-
vious reports of cyanine dyes18 as well as the measured emission
maxima and quantum yields.
While the NET binding pocket cannot necessarily be com-
pared to a single solvent environment, the range of polarities
covered in the TD-DFT calculations do provide a benchmark
against which we may evaluate the observed fluorescence. The
emission maximum observed for A1 and related dyes is centered
at 565 nm, corresponding to 2.2 eV. The intercept of the line cor-
responding to 2.19 eV (gray area, Fig. 9A) shows that a narrow
range of twist angles (45° to 60°) is predicted to yield the
observed emission at 565 nm. A similar range of twist angles
This journal is © The Royal Society of Chemistry 2012
Org. Biomol. Chem., 2012, 10, 8710–8719 | 8715

View Article Online
Fig. 10 (A) Representative poses of the A series (green) and H series
(cyan) determined by AutoDock Vina.25–27 (B) One subunit of D1 (red)
and D2 (yellow) is predicted to adopt a similar pose with the long-axis
parallel to the substrate channel.
predicted allowability of the optical transitions (Fig. 9A)
suggests that the headfirst orientation should produce an emis-
sive excited state, while the tail-first orientation is more represen-
tative of the thermally relaxed TICT excited state. This notion is
supported by the finding that D2, which can only bind in a tail-
first orientation, does not demonstrate the characteristic ‘turn-on’
emission upon binding, while D1, which can only bind in a
headfirst geometry, does. For H1, the relationship between orien-
tation and ligand depth is consistent with that predicted for A1:
the headfirst orientation sits lower in the binding pocket than the
tail-first orientation. Similarly, the pyridinium group of the tail-
first orientation is in close proximity to ASP75, while the pyrdi-
nium tail of the headfirst pose is placed close to ASP473. The
predicted twist angles of the two orientations of H1 are exactly
opposite those predicted for A1 with the head-first orientation
exhibiting an interplane torsion of 66.0° while the twist angle for
the tail-first pose is 36.6°. From the relationship of torsion angle,
transition energy and oscillator strength (Fig. 9B), it is possible
that both of these poses yield emissive states.
Fig. 11 (A) Side view of A1 in the NET binding pocket with a ‘head
first’ orientation showing the MMFF calculated geometry with a 30°
twist interplanar twist angle. (B) The ‘tail first’ orientation of A1 sits
2.0 higher leaving the dye slightly more exposed to the solvent environ-
ment. A larger interplanar twist angle (70.3°) is predicted as a result of
the less restrictive upper vestibule of the binding pocket. (C) Top view
of the head first pose showing the shallow twist and close contact (2.5)
of the pyridinium 2H with ASP473. (D) Top view of the tail first pose
showing the larger rotation and the close contact (2.2) of the pyridinium
2H with ASP75.
characteristic fluorescence observed for NET-bound probes and
suggest that emission from the bound dyes requires an inter-
planar twist angle between 35° and 60°. Molecular modelling of
probes in the NET pocket shows that two factors, interplanar
twisting and probe depth in the pocket, may control the emission
response. The MMFF predicted geometries correlate well with
the quantum chemical calculations and observed emission
energies.
These results shed some light on the molecular interactions
and conformations necessary for inducing the emission enhance-
ment upon probe binding to NET. It is worth noting that not all
MATs possess the appropriate binding pocket environments
necessary to induce such a response. Indeed, Solis et al. recently
reported that ASP+ and structurally related 4-(4-(dimethylamino)-
phenyl)-1-methyl-pyridinium, bind, but do not exhibit the
characteristic turn-on emission observed for stilbazolium dyes
binding to NET.6 The generation of fluorogenic probes targeting
specific MATs may be with an improved understanding of the
binding geometries and related photophysical responses.
Conclusions
We have examined a series of sterically-demanding stilbazolium
dyes with the aim of understanding substrate binding at the NET
transporter and the observed ‘turn-on’ emission. All of the syn-
thesized probes exhibit excellent environmental sensitivity
mimicking the optical properties of the parent chromophores, A1
and H1. The binding and displacement kinetics demonstrate that
NET may accommodate significantly expanded stilbazolium dyes,
provided the expansion is along the long axis of the probe.
Specifically, the extension of the N-alkyl pyridinium tails is well-
tolerated, as evidenced by the binding behaviour of A2–6 and
H2–6, while expansion of the electron-donating aryl head, such as
A8–10, results in reduced affinity for the substrate-binding pocket.
The contrasting behavior of dimers, D1 and D2 suggests that, at
least for ASP+ analogues, a headfirst orientation is necessary to
observe the turn-on emission characteristic of stilbazolium dyes.
Quantum chemical calculations provide some insight into
the possible excited state geometries responsible for the
Experimental
Synthesis
The reagents and materials for synthesis were used as obtained
from Sigma Aldrich and TCI America and used without further
8716 | Org. Biomol. Chem., 2012, 10, 8710–8719
This journal is © The Royal Society of Chemistry 2012

View Article Online
purification. ACS reagent grade solvents were obtained from
EMD Chemicals. H NMR spectra were recorded on a Bruker
50 mL of EtOAc. The crude precipitate was crystallized from a
mixture of MeOH–EtOAc to produce a yellow powder that was
collected and dried under vacuum. IR νmax: 2690.55, 2969.16,
1608.05, 1478.68, 1391.20, 1240.68, 1170.37, 1033.20, 866.45,
1
500 MHz spectrometer using tetramethylsilane (TMS) as the
internal standard and DMSO-d6 as solvent. HRMS was per-
formed on a Bruker micrOTOF-Q II and IR spectra were
obtained on a Perkin-Elmer Spectrum One FT-IR spectrometer.
A1–10 and H1–6 were synthesized via Knoevenagel conden-
sation11 of the appropriate aldehyde and N-alkyl picolinium salt
using conditions recently reported for H1 (Scheme 1A).8 The
identities of A1–10 and D1, which have been previously
reported,12–16 were confirmed by 1H-NMR, 13C-NMR and
HRMS. H2–6, D2 and 11 are novel compounds that were fully
808.12 cm−1 1H NMR (500 MHz, DMSO/MeOH): δ 10.16
.
(1H, s), 8.97 (2H, d, J = 6.5 Hz), 8.26 (2H, d, J = 6.7 Hz), 8.16
(1H, d, J = 16.5 Hz), 8.08 (1H, s), 7.86–7.84 (2H, m), 7.80 (1H,
d, J = 8.5 Hz), 7.58 (1H, d, J = 16 Hz), 7.19–7.15 (2H, m), 4.55
(2H, q, J = 7.2 Hz), 1.55 (3H, t, J = 7.3 Hz). 13C NMR
(125.74 MHz, DMSO): δ 157.63, 153.46, 144.34, 136.15,
130.81, 130.62, 130.18, 127.82, 127.44, 124.34, 123.97, 122.28,
119.98, 109.64, 55.49, 16.71. HRMS (ESI+) m/z calcd for
C19H18NO+: 276.1422; Found: 276.1391. MP: 253 °C. Yield:
1.7 g (48%).
1
characterized by H-NMR, 13C-NMR, HRMS, IR, MP, UV-vis
and fluorescence spectroscopy.
4,4′-[1,6-Hexanediylbis(methylimino)]bisbenzaldehyde (11).
4-Fluorobenzaldehyde (1.24 g, 10.0 mmol), N,N′-dimethyl-1,6-
hexanediamine (202 mg, 1.40 mmol) and K2CO3 (1.0,
7.2 mmol) were combined in 20 mL of DMSO and heated at
90 °C for 24 h. DMSO was removed under reduced pressure and
the product was isolated as a pale yellow solid by column
(E)-4-[2-(6-Hydroxy-2-naphthalenyl)ethenyl]-1-propyl-pyridinium
bromide (H3). 6-Hydroxy-2-naphthaldehyde (1.7 g, 10 mmol),
4-methyl-1-propylpyridinium bromide (2.6 g, 12 mmol) and
100 μL piperidine were combined in 15 mL of MeOH and
heated to 50 °C for 24 h. After cooling, the product was precipi-
tated as a yellow powder with addition of 250 μL of AcOH
and 50 mL of EtOAc. The crude precipitate was crystallized
from a mixture of MeOH–EtOAc to produce fine yellow crystals
that were collected and dried under vacuum. IR νmax: 3010.80,
1642.81, 1516.60, 1602.71, 1516.60, 1473.14, 1363.85,
1278.15, 1216.66, 1162.09, 963.63, 870.70, 803.64,
chromatography using a CH2Cl2/EtOAc gradient. IR νmax
:
2921.4, 2857.1, 2804.8, 2731.2, 1660.7, 1587.9, 1526.7, 1468.4,
1431.5, 1386.0, 1365.7, 1222.8, 1161.0, 10947, 996.9, 945.2,
1
809.3, 736.6, 727.5 cm−1. H NMR (500 MHz, CDCl3): δ 9.7
(2H, s), 7.73 (4H, d, J = 8.9 Hz), 6.68 (4H, d, J = 8.8 Hz),
3.4 (4H, t, J = 7.2 Hz), 3.04 (6H, s), 1.63 (4H, s), 1.40 (4H, d,
J = 7.2 Hz). 13C NMR (125.74 MHz, CDCl3): δ 190.04, 153.44,
132.08, 125.04, 110.84, 52.32, 38.52, 26.93 (The signal at
26.93 ppm corresponds to two carbons, β and γ to the amine, of
the hexane linker. See HSQC spectrum in ESI.†) HRMS (ES+)
m/z calcd for C22H28N2O2Na+: 375.2043; Found: 375.2033.
MP: 133 °C. Yield: 0.23 g (47%).
657.09 cm−1 1H NMR (500 MHz, DMSO/MeOH): δ 10.14
.
(1H, s), 8.97 (2H, d, J = 7.0 Hz), 8.27 (2H, d, J = 7.0 Hz), 8.17
(1H, d, J = 16.0 Hz), 8.08 (1H, s), 7.86 (2H, d, J = 8.5 Hz), 7.79
(1H, d, J = 8.5 Hz), 7.59 (1H, d, J = 16.0 Hz), 7.18–7.15 (2H,
m), 4.49 (2H, t, J = 16.0 Hz), 1.98–1.90 (2H, m, J = 7.0 Hz),
0.91 (3H, t, J = 7.5). 13C NMR (125.74 MHz, DMSO): δ
157.57, 153.55, 144.54, 141.85, 136.17, 130.88, 130.64, 130.22,
127.84, 127.46, 124.38, 123.94, 122.32, 119.95, 109.63, 61.24,
24.48, 10.73. HRMS (ES+) m/z calcd for C20H20NO+:
290.1539; Found: 290.1546. MP: 291 °C. Yield: 1.5 g (41%).
4,4′-[1,6-Hexanediylbis[(methylimino)-4,1-phenyl-ene-(E)-2,1-
ethenediyl]]bis[1-methylpyridinium]
diiodide
(D2). 1,4-
Dimethylpyridinium iodide (0.32 g, 1.4 mmol), 11 (0.22,
0.63 mmol) and 100 μL of piperidine were dissolved in 6 mL of
MeOH and heated at 50 °C for 12 h. After cooling, the product
was precipitated as a red powder with addition of EtOAc. The
crude precipitate was crystallized from a mixture of MeOH/
EtOAc to produce a dark red powder that was collected and
dried under vacuum. IR νmax: 3444.23, 2934.76, 1644.69,
1574.99, 1518.16, 1387.63, 1313.31, 1162.37, 1046.71, 979.45,
(E)-1-Butyl-4-[2-(6-hydroxy-2-naphthalenyl)eth-enyl]pyridinium
bromide (H4). 6-Hydroxy-2-naphthaldehyde (1.7 g, 10 mmol),
1-butyl-4-methylpyridinium bromide (2.8 g, 12 mmol) and
100 μL piperidine were combined in 15 mL of MeOH and
heated to 50 °C for 24 h. After cooling, the product was precipi-
tated as a yellow powder with addition of 250 μL of AcOH and
50 mL of EtOAc. The crude precipitate was crystallized from a
mixture of MeOH–EtOAc to produce fine yellow crystals that
1
823.09 cm−1. H NMR (500 MHz, DMSO/MeOH): δ 8.68 (4H,
d, J = 6.9 Hz), 8.03 (4H, d, J = 6.7 Hz), 7.91 (2H, d, J =
16.0 Hz), 7.6 (4H, d, J = 9.0 Hz), 7.15 (2H, d, J = 16.0 Hz),
6.77 (4H, d, J = 9.0 Hz), 4.17 (6H, s), 3.43 (4H, t, J = 7.5 Hz),
2.99 (6H, s), 1.54 (4H, s), 1.34 (4H, s). 13C NMR (125.74 MHz,
DMSO): δ 153.82, 151.26, 144.75, 142.37, 130.80, 122.53,
122.48, 117.24, 112.14, 51.79, 46.81, 38.54, 26.74, 26.69.
HRMS (ES+) m/z calcd for C36H44N4 266.1777; Found:
266.1781. MP: 259 °C (dec). Yield: 98 g (20%).
were collected and dried under vacuum. (5–23b) IR νmax
2965.81, 1640.50, 1602.03, 1515.22, 1482.41, 1389.36,
1236.90, 1163.87, 968.03, 867.13, 812.12, 663.09 cm−1 1H
:
.
NMR (500 MHz, DMSO/MeOH): δ 10.58 (1H, s), 8.96 (2H, d,
J = 6.6 Hz), 8.25 (2H, d, J = 6.8 Hz), 8.16 (1H, d, J = 16.0 Hz),
8.07 (1H, s), 7.85 (2H, d, J = 8.8 Hz), 7.77 (1H, d, J = 8.5 Hz),
7.57 (1H, d, J = 16.0 Hz), 7.16–7.14 (2H, m), 4.52 (2H, t,
J = 7.5 Hz), 1.92–1.86 (2H, m), 1.35–1.27 (2H, m), 0.94 (3H,
s). 13C NMR (125.74 MHz, DMSO): δ 158.08, 153.52,
144.50, 141.89, 136.26, 130.83, 130.68, 130.04, 127.69, 127.38,
124.31, 123.92, 122.16, 120.18, 109.65, 59.68, 33.00, 19.26,
13.84. MP: 255 (dec) °C. HRMS (ES+) m/z calcd for
C21H22NO+: 304.1696; Found: 304.1701. MP: 255 °C. (dec.)
Yield: 1.2 g (31%).
(E)-1-Ethyl-4-[2-(6-hydroxy-2-naphthalenyl)ethenyl]-pyridinium
bromide (H2). 6-Hydroxy-2-naphthaldehyde (1.7 g, 10 mmol),
1-ethyl-4-methylpyridinium bromide (2.4 g, 12 mmol) and
100 μL piperidine were combined in 15 mL of MeOH and
heated to 50 °C for 24 h. After cooling, the product was precipi-
tated as a yellow powder with addition of 250 μL of AcOH and
This journal is © The Royal Society of Chemistry 2012
Org. Biomol. Chem., 2012, 10, 8710–8719 | 8717

View Article Online
(E)-4-[2-(6-Hydroxy-2-naphthalenyl)ethenyl]-1-(3-methyl-butyl)-
pyridinium bromide (H5). 6-Hydroxy-2-naphthaldehyde (1.7 g,
10 mmol), 4-methyl-1-(3-methylbutyl)pyridinium bromide
(2.9 g, 12 mmol) and 100 μL piperidine were combined in
15 mL of MeOH and heated to 50 °C for 24 h. After cooling,
the product was precipitated as a yellow powder with addition of
250 μL of AcOH and 50 mL of EtOAc. The crude precipitate
was crystallized from a mixture of MeOH–EtOAc to produce
fine yellow crystals that were collected and dried under vacuum.
IR νmax: 3695.6, 2961.21, 2868.68, 1611.59, 1472.67, 1279.81,
in DMEM (Dulbecco’s Modification of Eagle’s Medium with
4.5 g L−1 glucose, without L-glutamine and sodium pyruvate)
containing 10% dialyzed FBS, 2 mM glutamine, 100 units per
ml penicillin, 100 μg ml−1 streptomycin, and 250 μg ml−1
.
Geneticin at 37 °C and 5% CO2. Cells were seeded at a density
of 105 cell per cm2 in 96 microwell plates and incubated for at
least 48 h until a visible monolayer was established. Prior to
imaging, DMEM was removed and L-15 (Leibovitz’s
L-15 modified, pH 7.3) was added to each well. All probe and
inhibitor solutions were prepared in L15 media with 5% FBS.
Probe stock solutions were prepared to either 25 μM (D1, D2) or
100 μM (all other probes) and diluted to 10 μM in the micro-
wells; desipramine stock solutions were prepared to 500 μM and
diluted to 10 μM in the microwells. Cells maintained a normal
morphology during the course of the experiments (maximum
of 1) and remained adhered to the imaging dish.
Imaging was performed on a Leica SP5 confocal microscope
housed within the UM Biology Imaging Core Facility. Excitation
was achieved using 405 nm, 476 nm or 488 nm lasers. For time
dependent experiments, emission was collected between 475 and
575 nm for H1–6 and between 525 and 600 nm for A1–10,
D1,2; images were obtained in xyt mode at 2 s intervals. Emis-
sion spectra of membrane bound and internalized probes were
obtained in xyλ mode with data points at 5 nm intervals (10 nm
bandwidth). Images were analyzed using Fiji/ImageJA software
(NIH, USA).
1159.63, 866.90, 809.27, 711.01 cm−1 1H NMR (500 MHz,
.
DMSO/MeOH): δ 10.16 (1H, s), 8.99 (2H, d, J = 12.0 Hz), 8.26
(2H, d, J = 7.2 Hz), 8.17 (1H, d, J = 16.5 Hz), 8.08 (1H, s),
7.86–7.84 (2H, m), 7.79 (1H, d, J = 8.5 Hz), 7.58 (1H, d, J =
16.5 Hz), 7.19–7.16 (2H, m), 4.54 (2H, t, J = 7.5 Hz), 1.84 (2H,
q, J = 7.3 Hz), 1.61 (1H, t, J = 7.2 Hz), 0.96 (6H, d, J = 6.5 Hz).
13C NMR (125.74 MHz, DMSO): δ 157.62, 153.48, 144.53,
141.84, 136.15, 130.82, 130.65, 130.20, 127.82, 127.43, 124.36,
123.98, 122.27, 119.97, 109.65, 58.48, 25.63, 22.59. HRMS
(ES+) m/z calcd for C22H24NO+: 318.1852; Found: 318.1858.
MP: 280 °C (dec). Yield: 1.3 g (33%).
(E)-1-Benzyl-4-[2-(6-hydroxy-2-naphthalenyl)ethenyl]-pyridinium
bromide (H6). 6-Hydroxy-2-naphthaldehyde (1.7 g, 10 mmol),
4-methyl-1-(phenylmethyl)pyridinium
bromide
(3.2
g,
12 mmol) and 100 μL piperidine were combined in 15 mL of
MeOH and heated to 50 °C for 24 h. After cooling, the product
was precipitated as a yellow powder with addition of 250 μL of
AcOH and 50 mL of EtOAc. The crude precipitate was crystal-
lized from a mixture of MeOH–EtOAc to produce fine yellow
Computational methods
The excited geometries of A1 and H1 were calculated by
TD-DFT using the B3LYP functional at the 6-31G(d) level using
Guassian’09.29 Ground state geometries used for AutoDock
Vina25–27 and flexible docking experiments were calculated by
DFT using the B3LYP functional at the 6-31G(d) level using
Spartan’08 (Wavefunction, Inc.). The optimized geometry of A1
was used to construct the dimers, D1 and D2; to determine
maximum length, the tetramethylene linker was inserted in an
extended, ‘all-anti,’ conformation. Molecular docking was per-
formed using AutoDock Vina on the DFT and TD-DFT opti-
mized geometries; torsions were enabled for aliphatic carbons,
all other bonds were restricted. The ‘open-to-out’ NET homo-
logy model10 was prepared for docking in AutoDockTools
(Molecular Graphics Laboratory, Scripps Research Institute).
The grid box was centered in the reported binding pocket
(−25.3, −6.1, −20.5) with dimensions were x = 20 Å, y = 30 Å
and z = 16 Å. Flexible docking experiments were performed in
Spartan using MMFF; residues were limited to within 5 Å of the
AutoDock Vina predicted binding sites. Flexible residues were
PHE72, ASP75, TYR152, PHE317, PHE323 and ASP473.
crystals that were collected and dried under vacuum. IR νmax
3689.61, 2957.62, 2844.66, 1604.27, 1156.93, 1055.19,
1033.17 cm−1 1H NMR (500 MHz, DMSO/MeOH): δ 9.06
:
.
(2H, d, J = 6.5 Hz), 8.27 (2H, d, J = 6.5 Hz), 8.16 (1H, d, J =
16.0 Hz), 8.08 (1H, s), 7.87 (2H, t, J = 8.5 Hz), 7.80 (1H, d, J =
8.7 Hz), 7.58–7.53 (3H, m), 7.48–7.42 (3H, m), 7.17–7.14 (2H,
m). 13C NMR (125.74 MHz, DMSO): δ 157.58, 153.99, 144.61,
142.29, 136.23, 135.19, 130.94, 130.80, 130.20, 129.67, 129.15,
127.86, 127.50, 124.40, 124.32, 122.30, 119.94, 109.64, 62.41.
HRMS (ES+) m/z calcd for C24H20NO+: 338.1539; Found:
338.1545. MP: 272 °C. Yield: 0.91 g (22%).
Steady-state absorption and emission spectroscopy
Spectroscopy and HPLC grade solvents were utilized for all
spectroscopic measurements; all path lengths were 1 cm. UV-vis
absorption spectra were obtained on a Perkin-Elmer Lambda 35
UV-vis spectrometer using chromophore solutions of 10 μM to
20 μM. Fluorescence studies were performed on a Perkin-Elmer
LS55 Fluorometer. For determination of Φem, solutions were pre-
pared to an optical density of less than 0.05 in order to minimize
inner filter effects. Perylene in cyclohexane was used as a refer-
ence for quantum yields.28
Acknowledgements
This work was supported by a grant from the James & Esther
King Biomedical Research Program (1KF08). C.D.R. and
J.D.W. acknowledge the American Chemical Society Petroleum
Research Funds (ACS PRF# 51521-DNI6) for partial support of
this research. The National Science Foundation is acknowledged
for providing funds towards the purchase of an LC-ESI-MS
Cell culture and confocal microscopy
HEK-293 cells stably expressing hNET,
a
gift of
Prof. R. Blakely, Vanderbilt University, were cultured as pre-
viously described in sterile T-75 flasks.7,8 Cells were maintained
8718 | Org. Biomol. Chem., 2012, 10, 8710–8719
This journal is © The Royal Society of Chemistry 2012

View Article Online
16 K. Lunkenheimer and A. Laschewsky, Prog. Colloid Polym. Sci., 1992,
89, 239.
17 Y. Huang, T. Cheng, F. Li, C.-H. Huang, S. Wang, W. Huang and
Q. Gong, J. Phys. Chem. B, 2002, 106, 10041.
(CHE-0946858). We thank Prof. Randy Blakely for providing
the hNET-HEK293 cells and Dr James Baker (UM, Biology)
helpful discussions and assistance with imaging experiments.
18 G. L. Silva, V. Ediz, D. Yaron and B. A. Armitage, J. Am. Chem. Soc.,
2007, 129, 5710.
19 X. Cao, R. W. Tolbert, J. L. McHale and W. D. Edwards, J. Phys. Chem.
A, 1998, 102, 2739.
Notes and references
1 A. Schlessinger, E. Geier, H. Fan, J. J. Irwin, B. K. Shoichet,
K. M. Giacomini and A. Sali, Proc. Natl. Acad. Sci. U. S. A., 2011, 108,
15810.
2 O. Phanstiel and J. J. Archer, RSC Drug Discovery Ser., 2012, 17, 162.
3 D. Hadrich, F. Berthold, E. Steckhan and H. Bonisch, J. Med. Chem.,
1999, 42, 3101.
20 C. Reichardt, Angew. Chem., Int. Ed. Engl., 1979, 18, 98.
21 S. Matsuo and T. Makita, Int. J. Thermophys., 1989, 10, 833.
22 R. Sicard, J. Dhuguru, W. Liu, N. Patel, R. Landgraf and J. N. Wilson,
Bioorg. Med. Chem. Lett., 2012, 22, 5532.
23 A. Galli, L. J. DeFelice, B. J. Duke, K. R. Moore and R. D. Blakely,
J. Exp. Biol., 1995, 198, 2197.
4 J. W. Schwartz, R. D. Blakely and L. J. DeFelice, J. Biol. Chem., 2003,
278, 9768.
5 J. W. Schwartz, G. Novarino, D. W. Piston and L. J. DeFelice, J. Biol.
Chem., 2005, 280, 19177.
6 E. Solis, I. Zdravkovic, I. D. Tomlinson, S. Y. Noskov, S. J. Rosenthal
and L. J. De Felice, J. Biol. Chem., 2012, 287, 8852.
7 A. S. Brown, L. M. Bernal, T. L. Micotto, E. L. Smith and J. N. Wilson,
Org. Biomol. Chem., 2011, 9, 2142.
24 G. Scalmani and M. J. Frisch, J. Chem. Phys., 2010, 132, 114110.
25 O. Trott and A. J. Olson, J. Comput. Chem., 2010, 31, 455.
26 G. M. Morris, R. Huey, W. Lindstrom, M. F. Sanner, R. K. Belew,
D. S. Goodsell and A. J. Olson, J. Comput. Chem., 2009, 30,
2785.
27 M. F. Sanner, J. Mol. Graphics Modell., 1999, 17, 57.
28 I. Berlman, Handbook of Fluorescence Spectra of Aromatic Molecules,
Academic Press, New York, 2nd edn, 1971.
8 E. L. Smith, A. S. Brown, E. Adjaye-Mensah and J. N. Wilson, Org.
Biomol. Chem., 2012, 10, 1493.
9 S. K. Singh, C. L. Piscitelli, A. Yamashita and E. Gouaux, Science, 2008,
322, 1655.
10 M. Gabrielsen, I. Sylte, S. G. Dahl and A. W. Ravna, BMC Res. Notes,
2011, 4, 559.
11 E. Knoevenagel, Ber. Dtsch. Chem. Ges., 1898, 31, 2596.
12 A. F. Pozharskii, N. V. Vistorobskii, A. A. Bardin and E. A. Filatova,
Russ. J. Org. Chem., 2006, 42, 131.
13 G. S. He, T.-C. Lin, S.-J. Chung, Q. Zheng, C. Lu, Y. Cui and
P. N. Prasad, J. Opt. Soc. Am. B, 2005, 22, 2219.
14 J. K. Mishra, P. K. Behera, S. K. Parida and B. K. Mishra, Indian
J. Chem., Sect. B, 1992, 31B, 118.
29 M. J. T. Frisch, G. W. H. B. Schlegel, G. E. Scuseria, M. A. Robb,
J. R. Cheeseman, G. Scalmani, V. Barone, B. Mennucci, G. A. Petersson,
H. Nakatsuji, M. Caricato, X. Li, H. P. Hratchian, A. F. Izmaylov,
J. Bloino, G. Zheng, J. L. Sonnenberg, M. Hada, M. Ehara, K. Toyota,
R. Fukuda, J. Hasegawa, M. Ishida, T. Nakajima, Y. Honda, O. Kitao,
H. Nakai, T. Vreven, J. A. Montgomery, Jr., J. E. Peralta, F. Ogliaro,
M. Bearpark, J. J. Heyd, E. Brothers, K. N. Kudin, V. N. Staroverov,
R. Kobayashi, J. Normand, K. Raghavachari, A. Rendell, J. C. Burant,
S. S. Iyengar, J. Tomasi, M. Cossi, N. Rega, J. M. Millam, M. Klene,
J. E. Knox, J. B. Cross, V. Bakken, C. Adamo, J. Jaramillo, R. Gomperts,
R. E. Stratmann, O. Yazyev, A. J. Austin, R. Cammi, C. Pomelli,
J. W. Ochterski, R. L. Martin, K. Morokuma, V. G. Zakrzewski,
G. A. Voth, P. Salvador, J. J. Dannenberg, S. Dapprich, A. D. Daniels,
Ö. Farkas, J. B. Foresman, J. V. Ortiz, J. Cioslowski and D. J. Fox,
Gaussian 09 (Revision A.1), Gaussian, Inc., Wallingford CT., 2009.
15 B. Jędrzejewska, J. Kabatc, B. Ośmiałowski and J. Pączkowski, Spectro-
chim. Acta, Part A, 2007, 67, 306.
This journal is © The Royal Society of Chemistry 2012
Org. Biomol. Chem., 2012, 10, 8710–8719 | 8719