Yan et al.
2
observed for the R series (1d, 1g, and 1h, see Table 1) since
these alkyl chains are quite remote from the chromophore.
Although the effects on spectra are not substantial, side chains
do play critical roles in optimizing solubility, membrane binding
kinetics and avidity, and even voltage sensitivities (vide infra).
Applicability Assessment A: Sensitivity to Transmem-
brane Potential. Historically hemicyanines are best known for
their voltage sensitivities. A typical hemicyanine dye has a
hydrophobic tail and a hydrophilic head that promote binding
and alignment within cell membranes. Since optical excitation
of a hemicyanine dye involves charge redistribution, the
transmembrane potential has a measurable effect on the excita-
tion energy, and indeed voltage sensitivities can be conveniently
1
8
measured by using a hemispherical lipid bilayer apparatus.
The voltage sensitivities of these dyes, expressed in fluorescence
change per 100 mV, are included in Table 1. A few points are
clear from a glance at the table: first, the average sensitivity of
monothiophene series (1a,b), 1.5%, is much lower than that of
the bithiophene series (1c-h), 10%; second, the side chains
have significant effects on the sensitivity, i.e., 1d and 1e differ
only in side chains but have very different sensitivities.
Extensive optimization of side chains results in 1e with a
sensitivity of 18%, which is among the highest reported for fast
FIGURE 2. Normalized absorption and emission spectra of a series
of dyes with different numbers of thienylene units, measured in vesicle
suspensions containing 1 mg/mL soybean phosphatidylcholine in PBS
buffer.
dyes (see Figure 2 and Table 1). The fluorescence quantum
yields of 1a and 1d are around 10%, but that of 1i is only 2%,
discouraging further work on higher homologues. This low
fluorescence quantum yield is most likely caused by torsional
26,42
40
voltage sensitive dyes.
On the other hand, no attempts have
relaxation by rotation around intervening single bonds, which
are most abundant in 1i.
been made to optimize other series (1i, 2, 3, and 4) due to their
relatively low fluorescence quantum yields (e2%).
C. Effect of Acceptor. Using a larger heterocyclic acceptor
has proven effective in red-shifting the optical spectra of a
Applicability Assessment B: Sensitivity to Lipid Composi-
tions. Even in the absence of proteins, the cell membrane is a
highly heterogeneous 2-dimensional liquid phase. “Rafts”,
domains enriched in cholesterol and saturated lipids, floating
in a fluidic lipid bilayer have been proposed as a feature of cell
4
hemicyanine dye. Shown in Figure 3 are the absorption and
emission spectra of a series of dyes with different acceptors:
pyridinium for 1d, quinolinium for 2, and acridinium for 3. For
each additional fused benzo ring, there is an ∼75 nm red-shift
in both absorption and emission spectra, but there is also a
successive drop in fluorescence quantum yield, with acridinium
salt 3 being not fluorescent at all. The mechanism might be
due to steric hindrance induced by the fused benzo rings, which
effectively prevents the chromophore from attaining a fluores-
cent planar conformation.
4
3
membrane structure. Rafts play important roles in cellular
functions such as signal transduction and membrane traffick-
44
ing. Fluorescence imaging of domains in model biomembranes
can be realized by using two dyes having complementary
4
5
staining preferences or one dye with domain-sensitive fluo-
rescence characteristics (polarization, wavelength, lifetime,
2
1,46-49
etc.).
The large Stoke shifts and solvatochromic effects
D. Effect of Rigidification. Incorporating methine groups
in a ring has been very successful in red-shifting the absorption
of these new hemicyanines prompted us to assess the sensitivity
to membrane composition. Model membranes for liquid disor-
dered phase and liquid ordered phase were prepared from 1,2-
dioleoyl-sn-glycero-3-phosphocholine (DOPC) and 7:3 1,2-
dipalmitoyl-sn-glycero-3-phosphocholine (DPPC)/cholesterol,
respectively. Figure 4 shows the emission spectra of two
representative hemicyanines, 1b and 1d, in two types of vesicle
suspensions. 1d shows increased and blue-shifted (30 nm)
fluorescence in 7:3 DPPC/cholesterol vesicles relative to DOPC
vesicles; in contrast, 1b shows decreased and red-shifted (20
nm) fluorescence in 7:3 DPPC/cholesterol vesicles. All mono-
thiophene and bithiophene derivatives tested show similarly
contrasting lipid composition dependent emission spectra.
41
peaks and improving the photostabilities of polymethine dyes.
To test the effect of rigidification in our amino(oligo)thiophene-
based dyes we synthesized a CMe2-bridged dye, 4 (see Scheme
2
). Compared with the nonbridged 1d, 4 shows a large red-
shift (∆λ ) 65 nm) in the absorption spectrum and a small red-
shift (∆λ ) 18 nm) in the fluorescence spectrum, both of which
were measured in vesicle suspensions.
E. Effect of Side Chains. Two series of dyes with varying
1
2
alkyl substitutes, R on the hydrophobic side and R on the
1
hydrophilic side, have been synthesized. The R series (1c-f,
see Table 1) show progressive red-shifts in both absorption and
emission spectra when the alkyl chain increases from methyl
to n-butyl. This trend may be understood in terms of the
HOMO-LUMO gap: for hemicyanines the HOMO resides
mostly on the amine side and the LUMO on the pyridinium
(42) Kuhn, B.; Fromherz, P. J. Phys. Chem. 2003, 107, 7903.
(
43) Simons, K.; Ikonen, E. Nature 1997, 387, 569.
(
44) Simons, K.; Toomre, D. Nat. ReV. Mol. Cell Biol. 2000, 1, 31.
15
side. The larger the size of the alkyl chain the more the amino
nitrogen is shielded from hydrogen-bonded solvation, destabiliz-
ing the ground state relative to the excited state, thus red-shifting
the absorption and emission maxima. It is not surprising that
only minimal variations in absorption and emission spectra are
(45) Baumgart, T.; Hess, S. T.; Webb, W. W. Nature 2003, 425, 821.
(
46) Parasassi, T.; Gratton, E.; Yu, W. M.; Wilson, P.; Levi, M. Biophys. J.
997, 72, 2413.
47) Owen, D. M.; Lanigan, P. M. P.; Dunsby, C.; Munro, I.; Grant, D.;
Neil, M. A. A.; French, P. M. W.; Magee, A. I. Biophys. J. 2006, 90, L80.
48) Margineanu, A.; Hotta, J. I.; Van der Auweraer, M.; Ameloot, M.; Stefan,
1
(
(
A.; Beljonne, D.; Engelborghs, Y.; Herrmann, A.; Muellen, K.; De Schryver,
F. C.; Hofkens, J. Biophys. J. 2007, 93, 2877.
(49) Kim, H. M.; Jeong, B. H.; Hyon, J.-Y.; An, M. J.; Seo, M. S.; Hong,
J. H.; Lee, K. J.; Kim, C. H.; Joo, T.; Hong, S.-C.; Cho, B. R. J. Am. Chem.
Soc. 2008, 130, 4246.
(
40) Silva, G. L.; Ediz, V.; Yaron, D.; Armitage, B. A. J. Am. Chem. Soc.
007, 129, 5710.
41) Reynolds, G. A.; Drexhage, K. H. J. Org. Chem. 1976, 42, 885.
2
(
6
590 J. Org. Chem. Vol. 73, No. 17, 2008