Spectral Properties and Orientation of Voltage-Sensitive Dyes in Lipid

Article pubs.acs.org/Langmuir

Spectral Properties and Orientation of Voltage-Sensitive Dyes in Lipid Membranes Maria Matson, Nils Carlsson, Tamás Beke-Somfai, and Bengt Nordén* Department of Chemical and Biological Engineering, Chalmers University of Technology, SE − 412 96 Gothenburg, Sweden S Supporting Information *

ABSTRACT: Voltage-sensitive dyes are frequently used for probing variations in the electric potential across cell membranes. The dyes respond by changing their spectral properties: measured as shifts of wavelength of absorption or emission maxima or as changes of absorption or fluorescence intensity. Although such probes have been studied and used for decades, the mechanism behind their voltage sensitivity is still obscure. We ask whether the voltage response is due to electrochromism as a result of direct field interaction on the chromophore or to solvatochromism, which is the focus of this study, as result of changed environment or molecular alignment in the membrane. The spectral properties of three styryl dyes, di-4ANEPPS, di-8-ANEPPS, and RH421, were investigated in solvents of varying polarity and in model membranes using spectroscopy. Using quantum mechanical calculations, the spectral dependence of monomer and dimer ANEPPS on solvent properties was modeled. Also, the kinetics of binding to lipid membranes and the binding geometry of the probe molecules were found relevant to address. The spectral properties of all three probes were found to be highly sensitive to the local environment, and the probes are oriented nearly parallel with the membrane normal. Slow binding kinetics and scattering in absorption spectra indicate, especially for di-8-ANEPPS, involvement of aggregation. On the basis of the experimental spectra and time-dependent density functional theory calculations, we find that aggregate formation may contribute to the blue-shifts seen for the dyes in decanol and when bound to membrane models. In conclusion, solvatochromic and other intermolecular interactions effects also need to be included when considering electrochromic response voltage-sensitive dyes.

■

INTRODUCTION Voltage-sensitive styryl dyes have been widely exploited to visualize variations of electric potential across cell membranes,1−4 in order to image the change of membrane potential in neurons5−8 or to study reaction mechanisms of ion pumps like the Na+/K+-ATPase.9−13 These generally amphiphilic dyes insert themselves into the lipid membrane and there change their optical properties as a response to the variation of the local electric potential, generally either seen as a shift in the wavelength or variation of the intensity for the absorption and fluorescence spectra. Different mechanisms behind the voltage sensitivity of the probes have been suggested: charge-shift electrochromic mechanisms14−16 or solvatochromic mechanisms17−19 where in the latter the electrical field is thought to affect position and orientation of the chromophore leading to optical changes due to changes in the microenvironment. The voltage sensing has also been suggested to depend on a combination of several mechanisms.20−23 Irrespective of the exact mechanism, the orientation of the dye molecule in the lipid bilayer is an important parameter for the voltage-sensing. A poor orientation with a wide orientational distribution can be anticipated to make each molecule be affected differently by the field, which would provide a less distinct response to the electric field. The aim of some studies has been to determine the orientation angle of the membranesolubilized voltage-sensitive dyes. Initially, the probes were © 2012 American Chemical Society

assumed to align themselves parallel with the membrane normal,15,16,24 but more recent studies suggest that they are not perfectly aligned. The manner of orientation has been in some cases subject to considerable controversy. For example, Lambacher and Fromherz25 reported that di-8-ANEPPS sits at an angle of 37.8 ± 1.6° relative to the membrane normal in bilayers of the synthetic lipid POPC on a silicon support, as measured by fluorescence interferometry. The same probe has also been studied in outer hair cells with fluorescence polarization microscopy where the probe was concluded to be oriented at 63.3° to the membrane normal.26 Both di-4ANEPPS and di-8-ANEPPS have been studied in black lipid membranes where they were found to orient at 36 ± 3° with respect to the membrane normal using polarization-resolved second harmonic generation imaging technique.27 The absorbance as well as fluorescence spectra of voltagesensitive probes have been shown to be generally sensitive to their local environment.14,17,19,28−37 Parameters such as solvent polarity,32−35 pH,28,37 and ionic strength28,37 have been shown to be important not only for the absorption and emission wavelengths and for the emission intensity of the dyes, but also for the fluorescence lifetime.19,29−31,34,36 For example, the Received: April 27, 2012 Revised: June 25, 2012 Published: June 27, 2012 10808

dx.doi.org/10.1021/la301726w | Langmuir 2012, 28, 10808−10817

Langmuir

Article

octanoate, purchased from Sigma, and n-decanol, from L. Light & Co., was used for liquid crystal formation. Pluronic F127 was from BASF. The buffer used was potassium phosphate (10 mM, pH 7.4) if not mentioned. Preparation of Large Unilamellar Lipid Vesicles (LUVs). DOPC and DOPG, dissolved in chloroform, were mixed in a ratio of 4:1 to yield LUVs with 20% negatively charged lipids and dried by rotary evaporation to form a lipid film. The film was dried further under vacuum for at least 2 h and then dispersed in buffer while vortexing. Five freeze−thaw cycles (liquid nitrogen/50 °C) were performed and the vesicles were extruded 21 times through two polycarbonate filters with pore size 100 nm using a hand-held syringe LiposoFast-Pneumatic extruder (Avestin, Canada) to obtain LUVs of uniform size. Preparation of Lamellar Liquid Crystals. To form liquid crystals, a 7.2% sodium octanoate, 20.4% n-decanol, and 72.4% MilliQ-water by weight was mixed and the samples were left at room temperature in the dark to equilibrate for at least one week. The dyes, dissolved in ethanol, were added in the preparation step in exchange for water. The resulting phase is composed of parallel 2.5-nm-thick double layers separated by a 6.0-nm-thick water layers.38,39 For spectroscopic measurements, the samples were placed between two parallel quartz plates with a 0.1 mm spacer chiseled in one of them. The samples were then heated to 100 °C and cooled to room temperature in order to increase the macroscopic orientation of the bilayers parallel to the quartz plates.40 Absorbance and Emission in Different Solvents. Absorption and emission of the probes (2 μM) were investigated in Milli-Q-water and different alcohols (methanol, ethanol, iso-propanol, n-hexanol, and n-decanol) as well as in samples of different glycerol/buffer content (10%, 30%, 50%, 70%, and 90% glycerol (w/w)). Absorption spectra 200−800 nm were recorded on a Cary 5000 or a Cary 50 Bio (Agilent Technologies, US). In addition, samples of each probe in 90% glycerol (w/w) were heated from 20 to 80 °C, and absorbance and emission were recorded at every 10 °C. For the heated samples, absorption spectra were recorded in the 350−700 nm range with a Cary 4 (Agilent Technologies, US). The equilibrium time was set to 5 min for each temperature, for both absorption and emission measurements. The absorption maximum was determined and used as excitation wavelength in fluorescence measurements, which were made on a SPEX τ-3 Fluorolog spectrometer (Jobin Yvon Horiba, France). For glycerol/buffer samples, the excitation wavelength was 490 nm. Emission was recorded at 510−850 nm or 520−850 nm and the excitation and emission slits were 2 nm. Integration time was set to 0.2 s (alcohols) or 0.1 s (glycerol, temperature), respectively. Additionally, titration of 1 μL LUV (10 mM) was performed for samples of di-4ANEPPS and RH421 (4 μM) and emission for each aliquot was recorded. The Stokes’ shift is the difference between the fluorescence maximum wavelength and the absorption maximum wavelength. The shift was calculated for water and the alcohols and compared to the wavelength shift dependence of the “solvent orientation polarizability” (Δf), defined as41

absorption of RH421 red-shifts when the solvent polarity decreases, whereas the fluorescence displays a blue-shift under the same conditions.35 The red-shift of absorption is also seen for di-8-ANEPPS in solutions of decreased DMSO/chloroform ratio (and hence decreased polarity).33 When salt concentration is increased, a blue-shift of the absorbance of RH421 occurs, whereas the fluorescence has been observed to decrease and red-shift.28 In this study, we examine three commonly used and wellstudied styryl-dyes, di-4-ANEPPS, di-8-ANEPPS, and RH421 (see Figure 1) and characterize their binding geometries in two

Figure 1. Molecular structures of the three investigated styryl dyes.

different model membranes: in the lipid bilayer of large unilamellar lipid vesicles (LUVs) and in a lamellar liquid crystal system. The results show that all probes are preferentially aligned close to the membrane normal, but di-4-ANEPPS is somewhat better aligned than the other two probes. The optical properties of the dyes in different environments, of varying polarity and viscosity, are investigated with the purpose of comparing the properties of the three probes in order to assess mechanisms of interaction with environment and potentials for probing membrane voltage. The sensitivity to the environment for the ANEPPS dyes was also investigated by quantum mechanical (QM) computations employing time-dependent density functional theory (TD-DFT). Both experimental and computational results confirm that the solvent has a significant impact on the optical properties of the dyes. The comparison of experimental and theoretical results indicate that the optical behavior of the ANEPPS dyes is likely to be due to formation of dimers or aggregates in water, decanol, or lipid bilayers, while the dyes are in monomeric form in alcohols with shorter chain length. This could explain the blue-shift seen in the absorption of dyes in membranes compared to in ethanol. Experimentally, di-8-ANEPPS shows spectral properties with a departing trend compared to the other two probes, also seen in the binding kinetics to lipid vesicles where di-8-ANEPPS is shown to have a slower binding rate compared to the other two probes.

■

Δf =

ε−1 n2 − 1 − 2 2ε + 1 2n + 1

(1)

where ε is the dielectric constant and n the refractive index of the solvent. Computational Details. To support the experimental observations and estimate the differences in excitation energies between the monomer and dimer forms of the ANEPPS dyes in different environments, quantum chemical calculations were employed. The size of the ANEPPS models was reduced, the optically inactive hydrocarbon part was not considered in our theoretical models, however, to force charge neutrality, the monovalent SO3− group was included. These models, with a total of 52 and 104 atoms for the monomer and dimer, respectively, were submitted to geometry optimization using a recently developed density functional theory with dispersion correction, ωB97-XD,42 with the 6-31G(d) basis set.43

EXPERIMENTAL METHODS

Materials. Di-4-ANEPPS (4-(2-[6-(dibutylamino)-2naphthalenyl]ethenyl)-1-(3-sulfopropyl)pyridinium), di-8-ANEPPS (4-(2-[6-(dioctylamino)-2-naphthalenyl]ethenyl)-1-(3-sulfopropyl)pyridinium), RH421 (4-[4-[4-(dipentylamino)phenyl]-1,3-butandienyl]-1-(4-sulfobutyl)pyridinium hydroxide), and retinoic acid were purchased from Sigma Aldrich. The lipids DOPC (1,2-dioleoyl-snglycero-3-phosphocholine) and DOPG (1,2-dioleoyl-sn-glycero-3phosphoglycerol) were purchased from Avanti Polar Lipids Inc. and used for preparation of large unilamellar lipid vesicles. Sodium 10809


Langmuir

Article

LUVs described by the membrane surface: β is the angle between membrane normal and the LUV long-axis. For the case of an infinitely long (tube-shaped), perfectly aligned LUV, we have Sm = 1. As the normal to the membrane plane is then obviously perpendicular to the tube axis, β = 90°, and the third factor becomes equal to −1/2. In a partially aligned (realistic) system of LUVs, Sm ≪ 1 and 0° ≪ β ≪ 90°. However, we may redefine Sm to include the third, normal orientation factor, and any deviation of it from −1/2, getting the simpler equation

The latter functional also includes long-range corrections or Coulombic attenuation, which was shown to be important for ANEPPS dyes to reproduce charge-transfer excitations within accepted accuracy.44 Excited-state energies were calculated using time-dependent density functional theory (TD-DFT), which was shown to determine spectral characteristics of larger organic dye molecules with relatively good accuracy.45 In a related case, medium-sized basis sets, e.g., 6-31+G(d), have been reported to give satisfactorily accurate results on such molecules with low computational cost.46,47 Unfortunately, the DFT and ab initio methods available for such large systems show a systematic error by overestimating excitation energies for several compounds, which results in markedly shorter absorption wavelengths than obtained by experimental measurements46,48,49 Note that these deviations from experiments are systematic. Thus, several functionals with different basis sets were employed here for the calculation of excited-state energies: ωB97-XD/6-31+G(d)//ωB97-XD/6-31G(d), ωB97-XD/6-311+G(d,p)//ωB97-XD/6-31G(d), CAM-B3LYP/6311++G(2d,2p)//ωB97-XD/6-31G(d). Excitation energies were calculated using the ground-state structure of ANEPPS models. Note that between the monomer and dimer models oscillatory strengths of the calculated excited states considered can be different. Thus, to compare results with experiments, only the states with highest oscillator strength were considered here. Solvent effects of water, methanol, ethanol, 2-propanol, 1-hexanol, heptane, and 1-decanol, the latter two mimicking the hydrophobic environment of the liquid crystals and the lipid bilayers, were considered by using the integral equation formalism polarizable continuum solvent model (IEFPCM).50 The calculations were performed using the Gaussian 09 software package.51 Linear Dichroism (LD) Spectroscopy. Linear dichroism (LD) is the difference in absorption of linearly polarized light parallel and perpendicular to a macroscopic orientation axis. To be able to achieve an LD signal, macroscopically oriented molecules are required. Starting with LUVs, the orientation was gained by exposing the LUVs to a shear flow. Molecules which interact with the LUVs in a nonrandom mode will gain orientation. The LD spectroscopic application of shear-deformed vesicles is a well-established methodology.52−55 The deformation of the spherical shape of the liposome is small compared to the natural undulations of the lipid bilayer, so the effective membrane structure, as well as the distribution of membrane solutes, can be considered the same as in a planar membrane. Orientations of symmetric probes such as retinal and pyrene in liposome membranes thus agree well with those observed in planar (lamellar) bilayer.56 A Couette flow cell, consisting of two cylinders where the sample is applied in the gap between the cylinders was used to align the LUVs. The total optical path length of the cell was 1 mm (the gap 0.50 mm), and a shear flow was created by rotating the outer cylinder with 1000 rpm. Baseline was collected without rotation and subtracted for each spectrum. The lipid concentration was 1.25 mM and the concentration of di-4-ANEPPS, di-8-ANEPPS, and RH421 was 3 μM and for retinoic acid 4 μM. The buffer used was 5 mM potassium phosphate (pH 7.4) containing 50% (w/w) sucrose. There are two reasons for adding sucrose: to reduce light scattering by matching the refractive index of the buffer with that of the LUVs and to increase the viscous drag of the shear flow resulting in enhanced LUV deformation and increased macroscopic orientation of the sample.54 The LD normalized with respect to the isotropic absorption gives the reduced LD (LDr), which provides a quantitative assessment of orientation from which the apparent angle, α, between the transition dipole moment of the dyes and the lipid membrane normal, can be determined56 LDr =

⎛ 3 cos2 α − 1 ⎞⎛ 3 cos2 β − 1 ⎞ LD = 3Sm⎜ ⎟ ⎟⎜ A iso 2 2 ⎝ ⎠⎝ ⎠

LDr =

3 S(1 − 3 cos2 α) 4

(3)

The macroscopic orientation factor, S, was calibrated from the LD measured on retinoic acid, which has been shown to bind effectively parallel with the lipid normal (α = 0°),57 and the apparent binding angles were calculated for the three dyes. While for the flow-aligned lipid vesicles the macroscopic orientation is stationary, turning to the liquid crystal samples the sample orientation is static. In order to measure the linear dichroism, the plates were mounted in a goniometer cell holder56 with the grazing angle ω adjusted to 60°. A spectrum recorded at the same grazing angle on a sample without any dye was used as a baseline. Absorption spectra were recorded on the same anisotropic samples on a Cary 50 Bio (Agilent Technologies, US) at a 90° grazing angle (normal incidence) and baseline corrected in the same manner. The isotropic absorption used in eq 2 can be obtained by correcting the recorded absorption for the lack of sample isotropy with the LD extrapolated to a 0° grazing angle so the LDr is now determined by40

LDr =

LD LD = A iso A + (LDω= 0 /3)

(4) r

For the lamellar liquid crystal system, the expression for LD in terms of S and α becomes LDr =

3 S(3 cos2 α − 1) 2

(5)

S is the membrane orientation factor: it is unity if the membranes are perfectly parallel with the quartz surfaces of the cell holder. LD was measured using a Chirascan CD spectrometer with an LD detector (Applied Photophysics, U.K.). Spectra were recorded between 200 and 700 nm with 1 nm increments and the time per point was set to 0.5 s. Binding Kinetics to LUV. Binding kinetics experiments were performed on a Cary Eclipse Fluorolog (Agilent Technologies, US). Samples with the probes (1 μM) were prepared and the LUVs were added (ratio probe/lipid was 1:600) and the fluorescence intensity was monitored over time. The kinetics of di-8-ANNEPS dye was also examined in the presence of 0.05% Pluronic F127 copolymer. Excitation was at 490 nm and emission was collected at 625 nm, both slits were 5 nm, and the PMT was 600 V.

■

RESULTS Spectral Properties of Probes in Solution. In order to compare the spectral properties of the three probes in different solvents, two series of solvent mixtures differing in polarity and/or viscosity were used. The spectral characteristics observed were used to provide information about the potential environmental effects in the model membranes (see below). We chose to focus on the absorption and fluorescence emission wavelengths, as well as emission intensity in the comparison of the three probes. The solvents used to investigate the effect of polarity were water and a series of alcohols. The wavelengths of the absorption peaks and the wavelengths of the fluorescence emission peaks for the three probes dissolved in the different solvents are presented in Table 1. The excitation spectra were found to have maxima at the same wavelengths as the absorption peaks for all probes (results not shown) and will

(2)

where Sm is a macroscopic membrane orientation factor considering both deformation of LUV and the orientation of the long-axis of the LUVs in the flow parallel to the flow direction. The third factor accounts for the macroscopic orientation of the lipid chains in the 10810


Langmuir

Article

further investigated in ethanol/Milli-Q-water mixtures at different proportions. Besides the wavelength shift of absorption, the spectra in low ethanol content showed pronounced light-scattering which, however, decreased with increasing ethanol content, as expected with aggregates. Also, the emission spectra changed, showing an intensity increase, especially seen when going from 20% to 40% ethanol content (absorption and emission spectra are shown in Supporting Information, Figure S3). To study the viscosity dependence of the spectral properties, the probes were dissolved in a set of solvent mixtures of buffer and glycerol. In Figure 3 (left panels), the absorption wavelength, emission wavelength, and emission intensity are all seen to vary with the glycerol content in the solution. The absorption peaks become red-shifted with increasing glycerol content for all probes (see Figure 3A). We note that the absorption of the two ANEPPS probes is different, which might be due to the fact that the increase in glycerol content does not break the dye aggregates. For di-4-ANEPPS and RH421, the emission wavelength blue-shifts with increasing glycerol concentration, while di-8 ANEPPS does not follow this pattern (Figure 3C). All the differences between the di-4-ANEPPS and di-8-ANEPPS probes are suggested to arise from di-8-ANEPPS forming aggregates and not reflecting solution behavior. The emission intensity is seen to increase for all three probes with increasing glycerol concentration (Figure 3E). While solution mixtures of glycerol and buffer vary in viscosity, the polarity of these solutions also changes with the composition. In order to separate these effects from each other, a 90% glycerol (by weight) solution was studied at different temperatures, since the same solution will sense a decreased viscosity with increasing temperature, while the polarity will change only slightly with the change in temperature. In Figure 3 (right panels), the effect of temperature on the spectral properties is presented. The absorption peak slightly blue-shifts with increasing temperature for di-4-ANEPPS and RH421, while there is a clear red-shift for di-8-ANEPPS (see Figure 3B). The emission wavelength is seen to red-shift with increasing temperature for all three probes (Figure 3D), while a decrease in the fluorescence intensity is observed (Figure 3F). Spectral Properties of Probes in Model Membranes. As can be seen in Table 1, the wavelengths of absorption and fluorescence maxima are all blue-shifted when the dyes are bound to either membrane model compared with the wavelengths in the alcohols, and the fluorescence intensity was also observed to increase. Titrations of LUVs into samples containing di-4-ANEPPS or RH421 were performed and the emission was recorded for each aliquot (Figure 4). In the spectra, it can be seen that RH421 emission blue-shifts from the wavelength of the free dye to that of the bound form, and when titrating LUVs, the wavelength only shifts slightly for the first titrations, which is due to emission from dye free in solution. For di-4-ANEPPS, the picture is more complex since the emission peak is shifting with the dye lipid ratio (cannot be explained by free dye emission), where the emission maximum peak at high dye to lipid ratio is found at 635 nm, whereas at lower dye to lipid ratio it is shifted to 678 nm. Theoretical Calculations. To estimate the environment sensitivity of the ANEPPS dyes, TD-DFT calculations were carried out in various solvents representing water, alcohols, as well as nonpolar matrices, such as heptane and decanol, corresponding to the inner parts of the membrane; see Table 2.

Table 1. Absorption and Emission Maximum Wavelength of the Probes in Different Solvents di-4-ANEPPS

Water MeOH EtOH iPrOH HexOH DeOH DOPC/ DOPG Liquid crystals

di-8-ANEPPS

RH421

λmax, abs (nm)

λmax. fluo (nm)

λmax. abs (nm)

λmax. fluo (nm)

λmax. abs (nm)

λmax. fluo (nm)

459 496 500 500 504 489 468

744 730 722 709 700 672 635

455 498 500 500 506 490 464

680 729 724 710 699 668 622

475 518 519 519 524 508 481

718 713 709 702 698 673 677

480

649

476

648

495

671

not be considered further. We found that the fluorescence emission intensities increased (about 20-fold for the ANEPPS dyes and about 2.5-fold for RH421 when comparing emission in methanol and decanol), and the emission peaks were generally blue-shifted with decreasing solvent polarity for the probes, while by contrast, the absorption peak wavelengths were red-shifted (absorption and emission spectra are found in Supporting Information, Figures S1 and S2). However, with two interesting exceptions, di-8-ANEPPS in water has its emission peak blue-shifted compared to the one in alcohols, contrary to the behavior of the other dyes, a deviation we shall suggest is due to aggregate formation as discussed below. Dilution did not decrease the blue-shift but rather reinforced it (result not shown) virtually opposing the idea of aggregation. The second exception from the trend is seen in decanol where for all three probes both absorption and emission appear blueshifted compared to the other alcohols. The difference between the emission and excitation wavelength maxima, referred to as the Stokes’ shift, is shown in Figure 2. For all three probes, the

Figure 2. The Stokes’ shift (difference between wavelengths for emission and absorption maximum) for di-4-ANEPPS (black), di-8ANEPPS (red), and RH421 (green) in water, methanol, ethanol, isopropanol, hexanol, and decanol. The lines are added to guide the eye.

Stokes’ shift was found to vary with the solvent orientation polarizability (see eq 1 for definition). It can be noted that the Stokes’ shifts for the two ANEPPS probes are very similar for all solvents except water. The chromophore behaves the same way in these two probes and in alcohol solvents, whereas in water, di-8-ANEPPS differs from the trend, likely due to aggregation affecting its spectral properties. Di-8-ANEPPS was 10811


Langmuir

Article

Figure 3. Wavelength of absorption (A,B) and emission (C,D) maxima and emission intensity (normalized by the maximum intensity) (E,F) at different glycerol contents (left panel) and at different temperatures (for a sample of 90% glycerol w/w) (right panel) for di-4-ANEPPS (black), di-8ANEPPS (red), and RH421 (green).

water, the decanol bilayer and the lipid bilayer, but the inconsistency could be explained by formation of π-stacked dimer di-8-ANEPPS. Note that to have an optimized dimer structure standard DFT functionals are not suitable, as they lack description of dispersive interactions. Thus, we chose for our investigations ωB97-XD, which is a recently developed functional including empirical dispersion correction, as well as long-range corrections, making it an optimal approach to describe electronic properties of π-stacked compounds. As seen in Table 2, these dimer calculations indeed show blue-shifted absorption peaks compared to those of the monomer, for both parallel and antiparallel orientations. Very similar observations can be made when considering calculations using higher basis sets, or other density functionals which employ long-range corrections using the Coulomb attenuating method.59 These test calculations confirmed that various unfavorable effects, such as basis set superposition error (BSSE), commonly observed for lower basis sets, do not significantly alter the obtained results for optimizations (Table S1 in Supporting Information), and they indicated that dispersion correction is necessary also for the TD-DFT calculations (Table S1 in Supporting Information). Therefore, we chose the same ωB97-XD functional, which was shown recently to produce satisfactory results when calculating spectral properties,60 for our TD-DFT calculations. Probe Orientation. To investigate the orientation angles of the probes in LUVs and lamellar liquid crystals, LD experiments were performed. The probes were compared with retinoic acid, which can be used as a reference dye, since it is known to align itself parallel with the membrane normal in LUVs of DOPC.57 As seen in Figure 5A, the ANEPPS probes show small negative signals at around 310/325 and 251 nm and a large LDpeak at 468/464 nm when bound to LUVs, while RH421 has a positive peak at 300 nm and a large negative signal at 481 nm. The large negative LD peaks correspond almost exactly to their absorbance maxima in LUVs (as can be seen when comparing A and B in Figure 5). In LUVs, negative signals are due to

Figure 4. Emission of di-4-ANEPPS and RH421 (4 μM) is increasing when titrated with LUVs (aliquots of 1−2 μL from a 10 mM solution to a volume of 1 mL). The red curves are emission of dye in buffer, and the black curves represent the LUV titration aliquots (for di-4ANEPPS aliquots of 2 μL are shown, while for RH421 aliquots of 1 μL up to 10 μL and then every 2 μL are shown). The inset shows the emission in buffer (red) and the final titration step (black) normalized with respect to the maximum emission value.

A recent study investigating the emission properties of di-8ANEPPS in membrane models confirmed that TD-DFT calculations with long-range correction have comparable accuracy to multireference CASSCF calculations.44 Here, our calculations were performed on ANEPPS models without the hydrocarbon tail, but keeping the remaining groups (Figure S4 in Supporting Information). For the monomeric form, the first excited state, having the highest oscillator strength and corresponding to HOMO−LUMO transition, show a consistent red shift with the decrease of polarity, which is in agreement with properties of similar dye molecules,47,58 and with our experimental investigations on alcohols (Tables 1 and 2). However, the values for the monomeric form do not correlate with the relative shifts observed in experiments for 10812


Langmuir

Article

Table 2. Calculated Absorption Peaks of ANEPPS Monomer and Dimer Models as a Function of the Environment calculated monomer ANEPPS solventa

λmax. absa (nm)

Δλb (nm)

MeOH EtOH iPrOH HexOH DeOH Heptane Water

401 403 404 406 410 421 400

−2 0 +1 +3 +10 +21 −3

experimental

ap dimer ANEPPS λmax. absa (nm)

p dimer ANEPPS

Δλb (nm)

384 392 378

−19 −11 −25

λmax. absa (nm)

Δλb (nm)

−37 −44 −38

366 359 365

di-8-ANEPPS solvent

Δλb (nm)

MeOH EtOH iPrOH HexOH DeOH

−2 0 0 +6 −11

Water Liquid crystalc DOPC/DOPG

−41 −20 −32

Results obtained at the ωB97-XD/6-31+G(d)//ωB97-XD/6-31G(d) level of theory. Environmental effects of the various solvents were considered using the IEFPCM continuum model. bAbsorption peak wavelength shift relative to ethanol. cHaving a bilayer with 74% decanol content. a

Table 3. Apparent Binding Angles of the Transition Dipole Moment of the Probes in LUVs probe

αapparent in LUV

di-4-ANEPPS di-8-ANEPPS RH421

14° (±4°) 18° (±4°) 21° (±4°)

The orientation of the probes was also investigated in lamellar liquid crystals. The retinoic acid was found not to be appropriate as a reference for the membrane normal in this system, since all the probes were better oriented than retinoic acid, but it could still be used as a common reference for the efficiency of orientation of the LUVs. Di-4-ANEPPS was the best oriented dye, tightly followed by di-8-ANEPPS and then RH421, when the relative orientation was assigned. The orientation thus only slightly differed between the probes. Binding to LUVs. The probes were observed to bind to lipid membranes with different rates; therefore, an examination of the binding kinetics was performed. Since the dyes in buffer are only weakly fluorescent but show high increase in fluorescence quantum yield upon binding to the membranes, the fluorescence intensity could be used as a quite sensitive measure of the amount of dye bound to the membrane. The fluorescence of the three probes in buffer was measured as a function of time upon addition of LUVs. The result clearly shows that the probes have different binding kinetics at the same concentration. di-4-ANEPPS and RH421 both show a rapid increase in their fluorescence, later followed by a slow decrease, while di-8-ANEPPS shows a slow monotonic increase over a much longer time (Figure 6). This experiment demonstrates that di-8-ANEPPS binds more slowly to lipid membranes compared to di-4-ANEPPS and RH421. The latter probes also seem to have some other kind of process occurring after their initial association to the membranes, leading to decreased fluorescence intensity. A two-exponential curve was fitted to the signal of di-8ANEPPS, from which the rate constants of the probe were obtained to 6.6 (±1.6) × 10−4 s−1 and 5.9 (±1.9) × 10−5 s−1. The rate constants of the other two probes cannot be determined because they bind so rapidly that the first binding process could not be resolved accurately with this method. The concentration of di-8-ANEPPS was also reduced (up to 100 times lower) to see if faster kinetics would be achieved because the aggregation is less extensive; however, the result showed no significant variation from the higher concentration (not

Figure 5. LD (A) and absorbance (B) spectra for di-4-ANEPPS (black), di-8-ANEPPS (red), RH421 (green) (3 μM), and retinoic acid (blue) (4 μM) when bound to LUV (1.25 mM). (C) Absorbance of the probes in pure buffer (10 mM KPi/sucrose, 50:50 w/w).

transition dipole moments that are preferentially aligned parallel with the membrane normal. By comparing the absorbance spectra of the dyes in LUV with the corresponding spectra in pure buffer, the probes are concluded to bind to the LUVs, since the wavelengths of the absorption maxima are shifted and the peaks sharpened (Figure 5B,C). If a significant amount of dye had remained free in solution, this would have been seen as a corresponding broadening of the absorption peak, blurring the spectrum. By normalizing the LD spectra with respect to the absorbance spectra, the reduced linear dichroism (LDr) is obtained and can be used for quantitative assessment of the binding angles. LDr can also be used to provide information on the environment of the dye molecules: should all dye molecules reside in the same molecular environment, then LDr will be nearly constant, whereas for a heterogeneous environment, the slight wavelength shifts would vary in the wavelength range of the absorption band. For these three dyes, LDr is constant, which clearly indicates that the vast majority of the dye molecules have inserted into the lipid bilayer. By assigning a binding angle of 0° for retinoic acid, the orientation factor S of the flow-aligned LUV system was determined to be 0.05. Using this orientation factor, the binding angles of the light-absorbing transition dipole moment were obtained for each probe; see Table 3. The transition dipole moments of the first two excited states are almost aligned with the long axis of the chromophore as determined from QM calculations for ANEPPS dyes.44 10813


Langmuir

Article

but this effect could not explain the shift we observed in the isotropic decanol solvent compared to the other alcohols. Another solvent property that could affect the spectral properties of a dye molecule is viscosity. The absorption wavelength was concluded to not be dependent on the viscosity, while the emission wavelength, on the other hand, was found to be affected by the viscosity when varying both the glycerol content and the temperature. The wavelength shifts of di-8-ANEPPS, however, do not follow the same trends as the other two probes, indicating that the probe aggregation is also found in the glycerol samples. Glycerol is probably too polar a solvent to dissolve the di-8-ANEPPS aggregates, although at the highest temperature, in 90% glycerol the aggregates seem to dissolve as seen by comparing di-4-ANEPPS and di-8-ANEPPS data (Figure 3D). Viscosity in terms of membrane fluidity has been shown to be important for the emission of di-8-ANEPPS and RH421, but a way to eliminate the influence of this effect in the spectral response could be to detect the fluorescence at the red edge of the emission spectra in a temperature above the phase transition of the lipids.62 It is hence important to have the viscosity effects and phase transition temperatures of lipids in mind when studying spectra of dye molecules of this kind. To further investigate the wavelength shifts seen in the experimental results, quantum mechanical calculations were performed for the ANEPPS chromophore. In search for an explanation of the different spectral responses seen in the experiments, two dimer configurations were considered besides the monomeric one. On the basis of a comparison of the wavelength in different environments, we propose that the ANEPPS dyes are mainly in monomeric form in ethanol, 2propanol, methanol, and hexanol. In pure decanol, the magnitude of the blue-shift from the absorption peak compared to ethanol is ∼11 nm, which is in good agreement with the blue-shifts calculated for the antiparallel dimers in nonpolar solvents with long hydrocarbon chains, i.e., 19 nm in 1-decanol and 11 nm in heptane (Table 1 and Table 2). Note that these values suggest that a similar antiparallel orientation should be present in the liquid crystals with 74% decanol, where the experimental blue-shift is ∼24 nm compared to ethanol. The optimized antiparallel dimer has an overall head-to-tail length of ∼20 Å, which is reasonably close to the proposed thickness of the bilayer in liquid crystals (∼25 Å) (see Supporting Information, Figure S5A). However, in the case of the lipid bilayer formed by the DOPC and DOPG mixture the thickness of the bilayer is larger (∼40−50 Å), and such an antiparallel alignment would not be preferred, as the charged sulfonate groups would be located in the middle of the hydrophobic core. Therefore, most likely the presence of the parallel dimers or higher aggregates is going to be preferred, which is supported by the close agreement between the magnitude of the experimental blue shift (32 nm) and theoretical ones of the parallel dimer in more hydrophobic solvents (heptane, 44 nm; 1-decanol, 37 nm) (Table 2 and Supporting Information Figure S5B). The smaller deviations in wavelength shift magnitudes between experimental values and the excited-state calculations may partly arise not only from limitations in the accuracy of TD-DFT methods and implicit solvent models, but also from the lack of sampling several dye monomer and dimer conformations surrounded by explicit solvent or lipid molecules,20,44 an investigation beyond the scope of this study. In fact, the applied models and methods show excellent agreement with experimental results for most of the solvents, and also close agreement for the more ordered hydrophobic

Figure 6. Emission intensity variation over time for di-4-ANEPPS (black), di-8-ANEPPS (red), and RH421 (green) (1 μM) after addition of LUV (600 μM). The binding kinetics is much slower for di-8-ANEPPS than for the other two probes. The inset compares the binding kinetics of di-8-ANEPPS with (blue) and without (red) pluronic F127.

shown). The ANEPPS probes are recommended by the supplier to be used together with the block copolymer surfactant pluronic F127 (to increase the solubilization of the probes). Therefore, the kinetics of di-8-ANEPPS was examined in the presence of this agent and compared to the surfactantfree kinetics. The results showed (Figure 6, inset) that Pluronic F127 indeed speeds up the binding toward LUVs, but the time required for saturation of the probe in the lipid membrane is still quite long (at least 6 h incubation is needed when using Pluronic F127 in the medium compared to 14 h for an untreated di-8-ANEPPS sample).

■

DISCUSSION The spectral properties of all three styryl dyes are clearly very sensitive to the local environment. Generally, the ground state of polar molecules is destabilized by nonpolar solvents, and excitation of the molecule is easier and a red-shift is commonly observed for the absorbance with decreasing polarity. Polar solvents, on the other hand, are usually able to stabilize the excited state of a molecule by reorientation of the solvent molecules, which leads to a decreased energy of the excited state, resulting in an emission at longer wavelength compared to the wavelength in a less polar solvent (i.e., up to hexanol). These solvatochromic effects were seen for all three probes, and the results are consistent with previous studies, where separate probes have been studied with respect to effects of polarity on the spectral properties.32−35 When comparing the three probes, they all show the same trend in absorbance and emission wavelength. The trend is, however, broken for di-8-ANEPPS, which shows a decreased Stokes’ shift in water compared to methanol due to a blue-shift of the emission wavelength. This could be explained by aggregation of the dye, which would lead to a more hydrophobic environment, compared to the chromophore being surrounded by water. It is also notable that the trend of red-shifting absorbance with decreased polarity is not seen for decanol and the membrane models (LUV and liquid crystals), which for all probes show a blueshift compared with the more polar alcohols. There have been reports of blue-shifts in lipid bilayers for several dyes compared to the wavelength in alcohols, and for some dyes, this is also seen with respect to water.14,17,24,61 This blue-shift has been concluded to be due to the anisotropic nature of membranes,24 10814


Langmuir

Article

fact that the dye molecules are self-quenched due to dimerization or interaction between adjacent membranebound dye molecules above a certain concentration in the membrane. All dyes are amphiphilic and show spectral properties that could possibly arise or partly arise from the formation of aggregates, but di-8-ANEPPS, which due to its hydrophobic character can be expected to be most prone to form aggregates, exhibits a much slower binding kinetics to membranes compared to the other dyes. A slow disassembly of the aggregate can explain the longer time it will take for the dye to reach into the membrane. Furthermore, we notice the presence of considerable light scattering in the absorption spectra for di8-ANEPPS in aqueous solution, which seems to not be found in the di-4-ANEPPS and RH421 at the same concentration. If the latter two dyes aggregate, the constructs of di-8-ANEPPS are likely to be both more stable and larger in comparison. Binding kinetics when using the dyes for probing the voltage variation could be of importance. The probe di-8-ANEPPS, in the concentrations used in this work, needs an incubation time of several hours to reach saturation, even in the presence of the solubilizing agent Pluronic F127, which might not be feasible in all applications. When using the dye in experiments over longer times, it appears necessary to wash away the excess dye to avoid a continuously increasing amount of membrane-bound dye. As to the mechanism of voltage-sensing, we consider solvatochromism to be a very important part influencing the spectral properties as an effect of both changed solvent polarity and viscosity. If the voltage, in addition to perturbing the electronic states of the chromophore, will also change the average probe position with respect to the membrane surface, the probe may sense a different local environment with respect to the bulk water, the polar head-groups, and the lipid interior of the membrane, which can all significantly affect its spectral properties.

bilayer phases when the dimer models are considered. These observations indicate that the blue-shift occurring upon lipid bilayer insertion of the ANEPPS could also be due to formation of dimers or higher aggregates in addition to sensing the more anisotropic environment. Considering the quite different polarities of the solvents and membrane, it is hard to draw general conclusions as to the relation between dimerization/ aggregation and solvent polarity: for example, a dimerization or higher-order π-stacking is anticipated in polar solvent and but not in nonpolar solvent. Thus, while in a polar solvent like water aggregation is promoted by hydrophobic interactions, in the lipid bilayer one has to search for a different explanation of the dye aggregation evidenced from the blue-shift. The orientation of the probe in the membrane can also be crucial when using it for indication of voltage variation. Poor orientation could be due to a wide orientation distribution as a result of sparse packing in the sample, in which case the differently oriented dye molecules would be differently affected by the field, which in turn would give a less distinct response to the variation of the voltage. The results show that the probes are preferentially more aligned parallel with the membrane normal than parallel with the membrane surface. Comparing with retinoic acid, which is known to bind parallel to the membrane normal in LUVs composed of DOPC,57 the binding angles of the transition dipole moment of the molecules bound to LUVs were determined. The dye di-4-ANEPPS was found to be closely aligned with the normal compared to the other two probes (Table 3). The results in the lamellar liquid crystals are similar to the best orientation for di-4-ANEPPS, but the binding angle could not be determined, since the probes gained an orientation even better than retinoic acid (note that the bilayer thickness is smaller compared to LUVs which could affect the molecular orientation). The investigated probes have the structural features of molecular rotors with internal flexibility allowing mechanisms for radiationless return to the ground state63 (Figure 1). For molecular rotors, an increased intensity of emission is usually found when increasing the viscosity of the solvent, as a result from restricted internal rotation of the molecule.63 Indeed, the emission intensity was increased with the viscosity, which was seen both when the temperature was varied and when samples of different glycerol content were examined (Figure 3E,F). In addition, the emission intensity was generally also increased when decreasing the solvent polarity (see Figure S2 in Supporting Information), since less energy can be transferred to the solvent as heat. The environment in a lipid membrane is hydrophobic and more viscous compared to water solution, and consequently, upon binding to membranes the fluorescence quantum yield of the dyes is increased (Figure 4). The difference in quantum yield, low in water but high in membrane, was used to study the membrane binding kinetics of the dyes. Interestingly, the probes show different binding kinetics for LUVs where di-8-ANEPPS showed a very slow binding process compared to the two other probes at the same concentration (1 μM and dye to lipid ratio 1:600) (Figure 6). The probe RH421 has been shown to have a concentrationdependent binding kinetics to lipid membranes where the binding was observed to be very slow at high concentrations (13 μM and dye to lipid ratio 1:12). At low concentrations (0.064 μM and dye to lipid ratio 1:240), however, the fluorescence rapidly increases in the presence of lipids, followed by a decreased intensity,64 similar to di-4-ANEPPS and RH421 in this study. The decreased fluorescence is probably due to the

■

CONCLUSIONS The three styryl dyes have been shown to be sensitive to both polarity and viscosity changes in the local environment. Interestingly, di-8-ANEPPS shows different trends compared to the other two probes, which likely is due to aggregation in more polar solution. The binding to lipid membranes was also found to be slower for this probe compared to the other probes, which is in agreement with the aggregation hypothesis requiring time for disassembly of aggregates in water solution phase and potential reassembly to other aggregates in membrane. The combination of the experimental and theoretical calculations suggests that, in ethanol as well as in higher alcohols, up to hexanol, the ANEPPS dyes are preferentially monomeric, while in decanol and in a lipid bilayer, they exhibit antiparallel and parallel dimers or higher aggregates, respectively. The three probes were also found to orient themselves near-perfectly parallel with the membrane normal. The results of this study suggest that the field response of these voltage-sensitive dyes is likely to strongly involve solvatochromism, including changes of orientation and location of chromophores in membrane. Although we cannot judge here the relative importance of electrochromism as a contributor to the measurable field response (the subject of an ongoing separate investigation), we note that the sensitivity to environment makes field-induced solvatochromism a very strong mechanistic candidate. While it is thus a possible scenario that both solvatochromism and electrochromism are 10815


Langmuir

Article

(11) Pratap, P. R.; Robinson, J. D. Rapid Kinetic Analyses of the Na +/K+-ATPase Distinguish among Different Criteria for Conformational Change. Biochim. Biophys. Acta 1993, 1151, 89−98. (12) Heyse, S.; Wuddel, I.; Apell, H. J.; Sturmer, W. Partial Reactions of the Na,K-ATPase - Determination of Rate Constants. J. Gen. Physiol. 1994, 104, 197−240. (13) Clarke, R. J.; Apell, H. J.; Kong, B. Y. Allosteric Effect of ATP on Na(+),K(+)-ATPase Conformational Kinetics. Biochemistry 2007, 46, 7034−7044. (14) Fluhler, E.; Burnham, V. G.; Loew, L. M. Spectra, Membrane Binding, and Potentiometric Responses of New Charge Shift Probes. Biochemistry 1985, 24, 5749−5755. (15) Loew, L. M.; Scully, S.; Simpson, L.; Waggoner, A. S. Evidence for a Charge-Shift Electrochromic Mechanism in a Probe of Membrane Potential. Nature 1979, 281, 497−499. (16) Loew, L. M.; Simpson, L. L. Charge-Shift Probes of Membrane Potential. A Probable Electrochromic Mechanism for P-Aminostyrylpyridinium Probes on a Hemispherical Lipid Bilayer. Biophys. J. 1981, 34, 353−365. (17) Clarke, R. J.; Zouni, A.; Holzwarth, J. F. Voltage Sensitivity of the Fluorescent Probe RH421 in a Model Membrane System. Biophys. J. 1995, 68, 1406−1415. (18) Osakai, T.; Sawada, J.; Nagatani, H. Potential-Modulated Fluorescence Spectroscopy of the Membrane Potential-Sensitive Dye di-4-ANEPPS at the 1,2-Dichloroethane/Water Interface. Anal. Bioanal. Chem. 2009, 395, 1055−1061. (19) Visser, N. V.; Vanhoek, A.; Visser, A.; Frank, J.; Apell, H. J.; Clarke, R. J. Time-Resolved Fluorescence Investigations of the Interaction of the Voltage-Sensitive Probe RH421 with LipidMembranes and Proteins. Biochemistry 1995, 34, 11777−11784. (20) Callis, P. R. Electrochromism and Solvatochromism in Fluorescence Response of Organic Dyes: A Nanoscopic View. In Advanced Fluorescence Reporters in Chemistry and Biology I: Fundamentals and Molecular Design, Demchenko, A. P., Ed.; Springer: Berlin, 2010; Vol. 8, pp 309−330. (21) Fromherz, P.; Lambacher, A. Spectra of Voltage-Sensitive Fluorescence of Styryl-Dye in Neuron Membrane. Biochim. Biophys. Acta 1991, 1068, 149−156. (22) Hübener, G.; Lambacher, A.; Fromherz, P. Anellated Hemicyanine Dyes with Large Symmetrical Solvatochromism of Absorption and Fluorescence. J. Phys. Chem. B 2003, 107, 7896−7902. (23) Kuhn, B.; Fromherz, P. Anellated Hemicyanine Dyes in a Neuron Membrane: Molecular Stark Effect and Optical Voltage Recording. J. Phys. Chem. B 2003, 107, 7903−7913. (24) Loew, L. M.; Simpson, L.; Hassner, A.; Alexanian, V. An Unexpected Blue Shift Caused by Differential Solvation of a Chromophore Oriented in a Lipid Bilayer. J. Am. Chem. Soc. 1979, 101, 5439−5440. (25) Lambacher, A.; Fromherz, P. Orientation of Hemicyanine Dye in Lipid Membrane Measured by Fluorescence Interferometry on a Silicon Chip. J. Phys. Chem. B 2001, 105, 345−346. (26) Greeson, J. N.; Raphael, R. M., Application of Fluorescence Polarization Microscopy to Measure Fluorophore Orientation in the Outer Hair Cell Plasma Membrane. J. Biomed. Opt. 2007, 12. (27) Ries, R. S.; Choi, H.; Blunck, R.; Bezanilla, F.; Heath, J. R. Black Lipid Membranes: Visualizing the Structure, Dynamics, and Substrate Dependence of Membranes. J. Phys. Chem. B 2004, 108, 16040− 16049. (28) Clarke, R. J.; Schrimpf, P.; Schoneich, M. Spectroscopic Investigations of the Potential-Sensitive Membrane Probe RH421. Biochim. Biophys. Acta 1992, 1112, 142−152. (29) Ephardt, H.; Fromherz, P. Anillnopyridinium - SolventDependent Fluorescence by Intramolecular Charge-Transfer. J. Phys. Chem. 1991, 95, 6792−6797. (30) Ephardt, H.; Fromherz, P. Fluorescence of Amphiphilic Hemicyanine Dyes without Free Double-Bonds. J. Phys. Chem. 1993, 97, 4540−4547.

important, the disentanglement of these effects, as demonstrated here by a systematic change of bulk solvent properties, is needed for any quantitative understanding of the effect. One should also recall that the polar−hydrophobic−polar structure of the lipid bilayers induces a particular orientation of the inserted probe chromophores that may change their spectral properties compared to when in a randomly oriented state. Such an alignment might also promote dimerization or oligomerization in accordance with observed spectral shift, scattering, and computational data. Intermolecular interactions of the probe molecules may clearly influence their spectroscopic behavior in the membrane. As several other molecules have shown similar behavior upon membrane insertion, dimerization may be a general phenomenon for membranebound aromatic dyes.

■

ASSOCIATED CONTENT

S Supporting Information *

Absorption and emission spectra of the dyes in various solvents, structures of the ANEPPS models used in the calculations, the calculated absorption peaks (obtained by different levels of theory), and schematic representation of di-4-ANEPPS dimers in a decanol bilayer and lipid bilayer. This material is available free of charge via the Internet at http://pubs.acs.org.

■

AUTHOR INFORMATION

Corresponding Author

*E-mail: [email protected]. Notes

The authors declare no competing financial interest.

■

ACKNOWLEDGMENTS This work was funded by the European Research Council (ERC Senior Advanced Grant to B.N.) and by a grant to B.N. from the King Abdullah University of Science and Technology.

■

REFERENCES

(1) Chemla, S.; Chavane, F. Voltage-Sensitive Dye Imaging: Technique Review and Models. J. Physiol. Paris 2010, 104, 40−50. (2) Grinvald, A.; Hildesheim, R. Vsdi: A New Era in Functional Imaging of Cortical Dynamics. Nat. Rev. Neurosci. 2004, 5, 874−885. (3) Zochowski, M.; Wachowiak, M.; Falk, C. X.; Cohen, L. B.; Lam, Y. W.; Antic, S.; Zecevic, D. Imaging Membrane Potential with Voltage-Sensitive Dyes. Biol. Bull. 2000, 198, 1−21. (4) Loew, L. M. Potentiometric Dyes: Imaging Electrical Activity of Cell Membranes. Pure Appl. Chem. 1996, 68, 1405−1409. (5) Grinvald, A.; Frostig, R. D.; Lieke, E.; Hildesheim, R. Optical Imaging of Neuronal-Activity. Physiol. Rev. 1988, 68, 1285−1366. (6) Fromherz, P.; Muller, C. A. Cable Properties of a Straight Neurite of a Leech Neuron Probed by a Voltage-Sensitive Dye. Proc. Natl. Acad. Sci. U. S. A. 1994, 91, 4604−4608. (7) Gogan, P.; Schmiedeljakob, I.; Chitti, Y.; Tycdumont, S. Fluorescence Imaging of Local Membrane Electric-Fields During the Excitation of Single Neurons in Culture. Biophys. J. 1995, 69, 299− 310. (8) Canepari, M.; Campani, M.; Spadavecchia, L.; Torre, V. CCD Imaging of the Electrical Activity in the Leech Nervous System. Eur. Biophys. J. Biophy. 1996, 24, 359−370. (9) Bartolommei, G.; Devaux, N.; Tadini-Buoninsegni, F.; Moncelli, M.; Apell, H. J. Effect of Clotrimazole on the Pump Cycle of the Na,KATPase. Biophys. J. 2008, 95, 1813−1825. (10) Habeck, M.; Cirri, E.; Katz, A.; Karlish, S. J.; Apell, H. J. Investigation of Electrogenic Partial Reactions in Detergent-Solubilized Na,K-ATPase. Biochemistry 2009, 48, 9147−9155. 10816


Langmuir

Article

(31) Krishna, M. M. G.; Periasamy, N. Fluorescence of Organic Dyes in Lipid Membranes: Site of Solubilization and Effects of Viscosity and Refractive Index on Lifetimes. J. Fluoresc. 1998, 8, 81−91. (32) Le Goff, G.; Vitha, M. F.; Clarke, R. J. Orientational Polarisability of Lipid Membrane Surfaces. Biochim. Biophys. Acta 2007, 1768, 562−570. (33) Li, H. W.; Mao, G. L.; Singer, K. D.; Lu, Z. K.; Weber, R.; Twieg, R. J. Solvent Effects on the Nonlinear Optical Response of a Potentiometric Biological Imaging Dye. J. Opt. Soc. Am. B 2007, 24, 1310−1318. (34) Mishra, A.; Behera, G. B.; Krishna, M. M. G.; Periasamy, N. Time-Resolved Fluorescence Studies of Aminostyryl Pyridinium Dyes in Organic Solvents and Surfactant Solutions. J. Lumin. 2001, 92, 175− 188. (35) Pham, T. H. N.; Clarke, R. J. Solvent Dependence of the Photochemistry of the Styrylpyridinium Dye RH421. J. Phys. Chem. B 2008, 112, 6513−6520. (36) Visser, N. V.; Vanhoek, A.; Visser, A.; Clarke, R. J.; Holzwarth, J. F. Time-Resolved Polarized Fluorescence of the Potential-Sensitive Dye RH421 in Organic-Solvents and Micelles. Chem. Phys. Lett. 1994, 231, 551−560. (37) Wolfbeis, O. S. Fluorescence-Based Ion Sensing Using Potential-Sensitive Dyes. Sens. Actuators, B: Chem. 1995, 29, 140−147. (38) Fontell, K.; Mandell, L.; Lehtinen, H.; Ekwall, P. 3-Component System Sodium Caprylate - Decanol - Water 3. Structure of Mesophases at 20 Degrees C. Acta Polytech. Scand. Chem. Technol. Metall. Ser. 1968, 1−56. (39) Friman, R.; Danielsson, I.; Stenius, P. Lamellar Mesophase with High Contents of Water - X-Ray-Investigations of the Sodium Octanoate Decanol Water System. J. Colloid Interface Sci. 1982, 86, 501−514. (40) Nordén, B.; Lindblom, G.; Jonas, I. Linear Dichroism Spectroscopy as a Tool for Studying Molecular Orientation in Model Membrane Systems. J. Phys. Chem. 1977, 81, 2086−2093. (41) Lippert, E. Dipolmoment und Elektronenstruktur von angeregten Molekulen. Z. Naturforsch., A 1955, 10, 541−545. (42) Chai, J. D.; Head-Gordon, M. Long-Range Corrected Hybrid Density Functionals with Damped Atom-Atom Dispersion Corrections. Phys. Chem. Chem. Phys. 2008, 10, 6615−6620. (43) Hehre, W. J.; Ditchfie., R; Pople, J. A. Self-Consistent Molecular-Orbital Methods .XII. Further Extensions of GaussianType Basis Sets for Use in Molecular-Orbital Studies of OrganicMolecules. J. Chem. Phys. 1972, 56, 2257−2261. (44) Robinson, D.; Besley, N. A.; O’Shea, P.; Hirst, J. D. Di-8-Anepps Emission Spectra in Phospholipid/Cholesterol Membranes: A Theoretical Study. J. Phys. Chem. B 2011, 115, 4160−4167. (45) Jacquemin, D.; Perpete, E. A.; Ciofini, I.; Adamo, C. Accurate Simulation of Optical Properties in Dyes. Acc. Chem. Res. 2009, 42, 326−334. (46) Fabian, J. TDDFT-Calculations of Vis/Nir Absorbing Compounds. Dyes Pigm. 2010, 84, 36−53. (47) Kitts, C. C.; Beke-Somfai, T.; Nordén, B. Michler’s Hydrol Blue: A Sensitive Probe for Amyloid Fibril Detection. Biochemistry 2011, 50, 3451−3461. (48) Fabian, J. Electronic Excitation of Sulfur-Organic Compounds Performance of Time-Dependent Density Functional Theory. Theor. Chem. Acc. 2001, 106, 199−217. (49) Schreiber, M.; Buss, V.; Fulscher, M. P. The Electronic Spectra of Symmetric Cyanine Dyes: A CASPT2 Study. Phys. Chem. Chem. Phys. 2001, 3, 3906−3912. (50) Cances, E.; Mennucci, B.; Tomasi, J. A New Integral Equation Formalism for the Polarizable Continuum Model: Theoretical Background and Applications to Isotropic and Anisotropic Dielectrics. J. Chem. Phys. 1997, 107, 3032−3041. (51) Frisch, M. J.; Trucks, G. W.; Schlegel, H. B.; Scuseria, G. E.; Robb, M. A.; Cheeseman, J. R.; Scalmani, G.; Barone, V.; Mennucci, B.; Petersson, G. A.; Nakatsuji, H.; Caricato, M.; Li, X.; Hratchian, H. P.; Izmaylov, A. F.; Bloino, J.; Zheng, G.; Sonnenberg, J. L.; Hada, M.; Ehara, M.; Toyota, K.; Fukuda, R.; Hasegawa, J.; Ishida, M.; Nakajima,

T.; Honda, Y.; Kitao, O.; Nakai, H.; Vreven, T.; Montgomery, J., J. A. ; Peralta, J. E.; Ogliaro, F.; Bearpark, M.; Heyd, J. J.; Brothers, E.; Kudin, K. N.; Staroverov, V. N.; Kobayashi, R.; Normand, J.; Raghavachari, K.; Rendell, A.; Burant, J. C.; Iyengar, S. S.; Tomasi, J.; Cossi, M.; Rega, N.; Millam, J. M.; Klene, M.; Knox, J. E.; Cross, J. B.; Bakken, V.; Adamo, C.; Jaramillo, J.; Gomperts, R.; Stratmann, R. E.; Yazyev, O.; Austin, A. J.; Cammi, R.; Pomelli, C.; Ochterski, J. W.; Martin, R. L.; Morokuma, K.; Zakrzewski, V. G.; Voth, G. A.; Salvador, P.; Dannenberg, J. J.; Dapprich, S.; Daniels, A. D.; Farkas, Ö .; Foresman, J. B.; Ortiz, J. V.; Cioslowski, J.; Fox, D. J. Gaussian 09, revision A.1; Gaussian, Inc; Wallingford, CT, 2009. (52) Ardhammar, M.; Nordén, B. Chromophore Orientation in Liposome membranes Probed with Flow Linear Dichroism. J. Am. Chem. Soc. 1998, 120, 9957−9958. (53) Ardhammar, M.; Lincoln, P.; Nordén, B. Ligand Substituents of Ruthenium Dipyridophenazine Complexes Sensitively Determine Orientation in Liposome Membrane. J. Phys. Chem. 2001, 105, 11363−11368. (54) Ardhammar, M.; Lincoln, P.; Nordén, B. Invisible Liposomes: Refractive Index Matching with Sucrose Enables Flow Dichroism Assessment of Peptide Orientation in Lipid Vesicle Membrane. Proc. Natl. Acad. Sci. U. S. A. 2002, 99, 15313−15317. (55) Caesar, C.; Esbjörner, E.; Lincoln, P.; Nordén, B. Assigning membrane binding geometry of cytochrome c by polarized-light spectroscopy. Biophys. J. 2009, 96, 3399−3411. (56) Nordén, B.; Rodger, A.; Dafforn, T. R. Linear Dichroism and Circular Dichroism - A Textbook on Polarized-Light Spectroscopy; RSC Publishing: Cambridge, UK, 2010. (57) Svensson, F. R.; Lincoln, P.; Nordén, B.; Esbjörner, E. K. Retinoid Chromophores as Probes of Membrane Lipid Order. J. Phys. Chem. B 2007, 111, 10839−10848. (58) Stsiapura, V. I.; Maskevich, A. A.; Kuzmitsky, V. A.; Uversky, V. N.; Kuznetsova, I. M.; Turoverov, K. K. Thioflavin T as a Molecular Rotor: Fluorescent Properties of Thioflavin T in Solvents with Different Viscosity. J. Phys. Chem. B 2008, 112, 15893−15902. (59) Yanai, T.; Tew, D. P.; Handy, N. C. A New Hybrid ExchangeCorrelation Functional Using the Coulomb-Attenuating Method (CAM-B3LYP). Chem. Phys. Lett. 2004, 393, 51−57. (60) Eriksson, E. S. E.; Eriksson, L. A. Predictive Power of LongRange Corrected Functionals on the Spectroscopic Properties of Tetrapyrrole Derivatives for Photodynamic Therapy. Phys. Chem. Chem. Phys. 2011, 13, 7176−7186. (61) Wuskell, J. P.; Boudreau, D.; Wei, M. D.; Jin, L.; Engl, R.; Chebolu, R.; Bullen, A.; Hoffacker, K. D.; Kerimo, J.; Cohen, L. B.; Zochowski, M. R.; Loew, L. M. Synthesis, Spectra, Delivery and Potentiometric Responses of New Styryl Dyes with Extended Spectral Ranges. J. Neurosci. Meth. 2006, 151, 200−215. (62) Clarke, R. J.; Kane, D. J. Optical Detection of Membrane Dipole Potential: Avoidance of Fluidity and Dye-induced Effects. Biochim. Biophys. Acta 1997, 1323, 223−239. (63) Haidekker, M. A.; Theodorakis, E. A., Environment-Sensitive Behavior of Fluorescent Molecular Rotors. J. Biol. Eng. 2010, 4. (64) Zouni, A.; Clarke, R. J.; Holzwarth, J. F. Kinetics of the Solubilization of Styryl Dye Aggregates by Lipid Vesicles. J. Phys. Chem. 1994, 98, 1732−1738.

10817


Spectral Properties and Orientation of Voltage-Sensitive Dyes in Lipid

Recommend Documents