Fluorescence measurements of environmental relaxation at the lipid-water interface region of bilayer membranes

Nanosecond time-resolved emission spectroscopy is used to characterize the complex fluorescence behavior of the probe 2-p-toluidinonaphthalene 6-sulfonate (2,6 p-TNS) when adsorbed to several bilayer membrane system. These include egg phosphatidylcholine vesicles with and without added cholesterol as well as erythrocyte ghost membranes. In each case a nanosecond time-dependent shift of the fluorescence emission to lower energy follows pulsed photoexcitation. The properties of the time-resolved surfaces obtained are consistent with a non-exponential decay law which describes a continuous interaction process of 2,6 p-TNS with its local environment in the membrane. This environment consists in part of polar residues (water plus polar head region) undergoing nanosecond motions. The pure phosphatidylcholine bilayer system was studied at four temperatures and electronic and spectral relaxation contributions to the total fluorescence decay were separated. Temperature coefficients for empirical rate parameters derived for the separated processes were obtained. It appears that a treatment of the fluorescence behavior of amphiphilic probes such as 2,6 p-TNS adsorbed to bilayer membranes at temperatures near ambient in which a single lifetime and radiative decay channel have been assumed is inappropriate.     

Proof of nanosecond timescale relaxation in apomyoglobin by phase fluorometry

Apomyoglobin was labeled with the fluorescent probe 2-p-toluidinylnaphthalene-6-sulfonic acid (TNS). Apparent phase shift and demodulation lifetimes of bound TNS were measured at various emission wavelengths. The lifetimes increased with increasing wavelength. Similar results were obtained for TNS in the viscous solvent glycerol at 10°C but not for TNS in vitrified or fluid solvent. The wavelength-dependent lifetimes suggest apomyoglobin is relaxing around the TNS molecule during its fluorescent lifetime. Importantly, the apparent phase lifetimes exceeded the apparent modulation lifetimes on the long wavelength side of the emission for TNS in apomyoglobin at 3°C and for TNS in glycerol at 10°C. This result proves the increasing lifetimes are a result of an excited state reaction during the lifetime of the excited state and are not a result of heterogeneity in the fluorescence emission. From the lifetimes on the short wavelength side of the emission the relaxation time of apomyoglobin was estimated to be 18 nsec.     

Nanosecond time-resolved emission spectroscopy of a fluorescence probe adsorbed to L-alpha-egg lecithin vesicles.

Nanosecond time-resolved emission spectra (TRES) are fluorescence emission spectra obtained at discrete times during the fluorescence decay. The complete data-set obtainable is a surface representing the intensity at all wavelengths and times during the emission decay time. When 2-p-toluidinonaphthalene-6-sulfonate (2,6 p-TNS) is adsorbed to egg lecithin vesicles, an excited-state reaction associated with energetic changes of the emitting species occurs on the nanosecond time scale. Convolution of the fluorescence decay with the excitation response introduces an artifact in the time-dependent spectra. A precedure is described by which this artifact can be eliminated. The data for the generation of time-resolved emission spectra are obtained with a computer-interfaced instrument based on the single-photon counting method.     

Nanosecond time-resolved emission spectroscopy of a fluorescence probe bound to L- -egg lecithin vesicles

The monophoton counting technique was used to measure nanosecond time-resolved fluorescence emission spectra of 2-p-toluidino-naphthalene-6-sulfonate (2,6 p-TNS) adsorbed to lipid bilayer vesicles. A time-dependent red shift in the emission maximum was observed and the rate of this red shift was shown to be temperature dependent. Analysis of fluorescence decay curves obtained at different emission wavelengths indicates that the time-dependent spectral shifts are due to an excited-state reaction such as solvent relaxation or an excited-state interaction between the chromophore and a polar residue of the phospholipid.     

Dynamic fluorescence of extrinsic fluorophores as a tool for studying protein conformational substates

Summary The fluorescence lifetime distribution of 2-p-toluidinyl-6-naphthalene sulfonic acids (TNS) bound to the heme site of apomyoglobin has been examined. The results were compared to those observed for the free fluorophore in isotropic nonviscous solvent. Two different excitation wavelengths were used, i.e. 290 and 350 nm. The results showed that the distribution of TNS bound to apomyoglobin is wider than that of the free fluorophore, thus indicating the existence of a large number of conformational substates originating from the interaction between TNS and the protein matrix. The comparison of the distribution obtained at two different excitation wavelengths allowed the emission arising from conformational substates, in which the excited state of fluorophore moiety has a higher probability to be populated by Forster energy transfer mechanism, to be distinguished.     

Probing pH and pressure effects on the apomyoglobin heme pocket with the 2''-(N,N-dimethylamino)-6-naphthoyl-4-trans-cyclohexanoic acid fluorophore.

The environmentally sensitive fluorophore 2'-(N,N-dimethylamino)-6-naphthoyl-4-trans-cyclohexanoic acid (DANCA) has been used to probe the apomyoglobin heme pocket. The unexpected polarity of this domain is generally interpreted as arising from dynamic dipolar relaxation of the peptide dipoles surrounding the heme pocket. In the present work we reexamine the photophysical properties of DANCA in a variety of solvents and complexed with apomyoglobin (apoMb) to further probe the heme pocket environment as a function of external solvent conditions. Absorption and excitation spectra in a number of solvents are consistent with the well-known pi*<--pi (LE) and pi*<--n (CT) electronic absorption transitions observed for naphthylamine derivatives. Dual emission is also a well-documented property of such derivatives. Based on the time scale of the heterogeneity in the decay of the DANCA fluorophore observed in a series of solvents, we propose that the emission properties of DANCA in apoMb are not uniquely attributable to dynamic relaxation events, but also reflect dual emission from both a long-lived, red CT state and the shorter-lived, blue LE state. The pH studies in the range of pH 5-9 of the emission properties of DANCA in apoMb support this hypothesis. They also suggest a specific interaction of DANCA with one or both of the pocket histidyl residues, which leads to a drastic static quenching and red shift of the bound DANCA fluorescence upon protonation. Similar effects are observed with increasing pressure, indicating that these two perturbations alter the DANCA-apoMb complex in a similar fashion. The pressure-induced form of the protein is distinct both energetically and structurally from the previously characterized acid intermediate, in that it is populated above pH 5 and retains a significant degree of integrity of the heme pocket.     

Ligand-apomyoglobin interactions. Configurational adaptability of the haem-binding site.

1. The interaction of the haem-binding region of apomyoglobin with different ligands was examined by ultrafiltration, equilibrium dialysis and spectrophotometry, to study unspecific features of protein-ligand interactions such as they occur in, for example, serum albumin binding. 2. Apomyoglobin, in contrast with metmyoglobin, binds at pH 7, with a high affinity, one molecule of Bromophenol Blue, bilirubin and protoporphyrin IX, two molecules of n-dodecanoate and n-decyl sulphate and four molecules of n-dodecyl sulphate and n-tetradecyl sulphate. 3. The number of high-affinity sites and/or association constants for the alkyl sulphates are enhanced by an increase of hydrocarbon length, indicating hydrophobic interactions with the protein. 4. Measurements of the temperature-dependence of the association constants of the high-affinity sites imply that the binding processes are largely entropy-driven. 5. Binding studies in the presence of two ligands show that bilirubin plus Bromophenol Blue and dodecanoate plus Bromophenol Blue can be simultaneously bound by apomyoglobin, but with decreased affinities. By contrast, the apomyoglobin-protoporphyrin IX complex does not react with Bromophenol Blue. 6. Optical-rotatory-dispersion measurements show that the laevorotation of apomyoglobin is increased towards that of metmyglobin in the presence of haemin and protoporphyrin IX. Small changes in the optical-rotatory-dispersion spectrum of apomyoglobin are observed in the presence of the other ligands. 7. It is concluded that the binding sites on apomyoglobin probably do not pre-exist but appear to be moulded from predominantly non-polar amino acid residues by reaction with hydrophobic ligands. 8. Comparison with data in the literature indicates that apomyoglobin on a weight basis has a larger hydrophobic area avaialble for binding of ligands than has human serum albumin. On the other hand, the association constants of serum for the ligands used in this study are generally somewhat larger than those of apomyoglobin.     

Evidence for heterogeneity in DNA-associated solvent mobility from acridine phosphorescence spectra

The mobility of solvent associated with native DNA in comparison with that of the bulk solvent is monitored from the temperature-dependent red shift in the phosphorescence spectra of acridines bound to DNA and free in glycol–buffer mixtures. Over the temperature range for which the red shift occurs the phosphorescence decay changes with emission wavelength, indicating the time-dependent nature of the process. Moreover, at these temperatures, emission anisotropy measurements establish that motions of the dye itself are not involved. Correspondence between perturbations to the solvent that influence the temperature at which the red shift occurs for free acridine with those for the DNA-bound dye confirm that “bound solvent” is responsible for the spectral changes. For the DNA-bound acridines the extent of the red shift is smaller and the midpoint T_1/2 of the transition is warmer. The reduction in the red shift reveals that the bound dye is less exposed to solvent and varies as 9-aminoacridine < acridine orange ～ proflavin, i.e., 9-amino-acridine is less exposed to solvent. On the other hand, the warmer T_1/2 indicates that DNA-associated solvent is considerably less mobile than bulk solvent. T_1/2 varies for proflavin bound to DNA, poly[d(AT)], poly[d(GC)], and poly(dG): poly(dC), and for proflavin, acridine orange, and 9-aminoacrine bound to DNA. These observations suggest that there is a heterogeneity in the mobility of DNA-associated solvent.     

Determination of time-resolved fluorescence emission spectra and anisotropies of a fluorophore-protein complex using frequency-domain phase-modulation fluorometry

We report the first time-resolved fluorescence emission spectra and time-resolved fluorescence anisotropies obtained using frequency-domain fluorescence spectroscopy. We examined the fluorophore p-2-toluidinyl-6-naphthalenesulfonic acid (TNS) in viscous solvents and bound to the heme site of apomyoglobin using multifrequency phase fluorometers. Fluorescence phase shift and modulation data were obtained at modulation frequencies ranging from 1 to 200 MHz. For time-resolved emission spectra, the impulse response for the decay of intensity at each emission wavelength was obtained from the frequency response of the sample at the same emission wavelength. The decays have negative pre-exponential factors, consistent with a time-dependent spectral shift to longer wavelengths. These multiexponential decays were used to construct the time-resolved emission spectra, which were found to be in good agreement with earlier spectra obtained from time-domain measurements. Additionally, time-resolved anisotropies were obtained from the frequency-dependent phase angle differences between the parallel and perpendicularly polarized components of the emission. The rotational correlation times of TNS bound to apomyoglobin are consistent with those expected for this probe rigidly bound to the protein. TNS in propylene glycol also displayed a single exponential decay of anisotropy. These results, in conjunction with the previous successful resolution of multiexponential decays of fluorescence intensity (Lakowicz, J. R., Gratton, E., Laczko, G., Cherek, H., and Limkeman, M. (1984) Biophys. J., in press; Gratton, E., Lakowicz, J. R., Maliwal, B. P., Cherek, H., Laczko, G., and Limkeman, M. (1984) Biophys. J., in press) demonstrate that frequency-domain measurements provide information which is, at a minimum, equivalent to that obtainable from time-domain measurements.     

The effect of the medium on the functional and structural properties of serum albumins. III. Relation between the N-F1-, F1-F2- and F2-E transitions of human serum albumin and temperature and ionic strength

Using fluorescence parameters of tryptophanyl and bound ANS, the acid-induced structural transitions of defatted monomeric human serum albumin were measured as pH-dependences from 6 to 2.5 in the wide range of temperature (10 to 45 degrees C) and ionic strength (from 0.001 to 0.2 M NaCl or 0.067 M Na2SO4). Temperature rise and decrease in ionic strength value result in the splitting of the N-F-transition onto two stages, N-F1 and F1-F2. The N-F1-transition is accompanied by the blue shift of tryptophanyl and ANS fluorescence spectra and increase in the ANS emission yield. The F1-F2-stage is manifested in an additional blue spectral shift and a sharp drop of the ANS emission yield, which is shown to be due to the lowering of albumin affinity for the dye. In the acidic-extension stage (F2-E), the spectra undergo a red shift which means that the nanosecond dipole relaxation of protein groups and bound water becomes faster. In the F2 from, the albumin affinity for ANS is significantly lowered; the association constant of the primary binding site is lower by an order of quantity and two secondary sites are practically disappeared. The complex effect of temperature, ionic strength and pH changes on the properties of ANS-binding sites is considered as a model of possible control influences of these factors upon the albumin transport of amphiphilic anions in organism.     

Nanosecond fluorescence microscopy. Emission kinetics of fura-2 in single cells.

A microscope based time-correlated single photon counting instrument has been constructed to measure fluorescence intensity and emission anisotropy decays from fluorophores in single cells on a nanosecond time scale. The sample is excited and the emission collected using epi-illumination optics with frequency-doubled pulses from the cavity-dumped output of a synchronously pumped dye laser serving as an excitation source. Collection of decays from a single cell is possible due to the presence of an iris in the emission path that can be reduced to less than the diameter of a single cell. Using the instrument the decay of 60 nM 1,6-diphenyl-1,3,5-hexatriene was measured, demonstrating that adequate data for lifetime analysis can be recorded from fewer 10(3) molecules of the fluorophore in an illuminated volume of 23 fl. In addition, the intensity and anisotropy decays of fura-2 in single adherent cells and in suspensions of fura-2 loaded cells in suspension, although the relative amplitudes and decay constants vary somewhat from cell to cell. The results indicate that a significant but variable fraction of fura-2 is bound to relatively immobile macromolecular components in these cells.     

On the nanosecond mobility in proteins: Edge excitation fluorescence red shift of protein-bound 2-(p-toluidinylnaphthalene)-6-sulfonate

The fluorescence spectra of 2-(p-toluidinylnaphthalene)-6-sulfonate associated with β-lactoglobulin, β-casein. and bovine and human serum albumins are shown to depend on excitation wavelength. A long-wave shift of the spectra is observed at the long-wave edge excitation, reaching 10 nm and above. A similar phenomenon is found in glucose glass and in glycerol at + 1°C, i.e., in systems with delayed dipolar solvent relaxation, but not in liquid solutions. This phenomenon is proposed to be based on relaxation processes in the excited state. There exists a distribution of chromophore microstates with different interactions with surrounding groups which results in heterogeneous broadening of the electronic spectra and allows photoselection of a part of this distribution, being characterized by a low transition energy. The fast structural relaxation results in an altered distribution and, if this is the case, the effect of edge excitation of fluorescence spectra is not observed. If the structural relaxation during the excited state lifetime is absent, this effect is maximal. This interpretation is in agreement with results on the influence of red edge excitation on the low-temperature fluorescence spectra of dyes and with the data on time-resolved nanosecond fluorescence spectroscopy. The results of this work strongly support the significant dye fluorescence spectral shifts on protein binding, being determined not only by polarity changes in their environment, but also by relaxation properties of protein groups in this environment. These results also indicate that on the nanosecond time scale, the structural relaxation around the excited chromophore in proteins may be incomplete.     

Molecular relaxation spectroscopy of flavin adenine dinucleotide in wild type and mutant lipoamide dehydrogenase from Azotobacter vinelandii.

The temperature dependence of the fluorescence emission spectra of flavin adenine dinucleotide bound to lipoamide dehydrogenase from Azotobacter vinelandii shows that the protein matrix in the vicinity of the prosthetic group is rigid on a nanosecond time scale in a medium of high viscosity (80% glycerol). The active site of a deletion mutant of this enzyme, which lacks 14 C-terminal amino acids, is converted from a solid-state environment (on the nanosecond time scale of fluorescence) into a state where efficient dipolar relaxation takes place at temperatures between 203 and 303 K. In aqueous solution, fast dipolar fluctuations are observed in both proteins. It is shown from fluorescence quenching of the flavin by iodide ions that the prosthetic groups of the mutant protein are partially iodide accessible in contrast to the wild type enzyme. A detailed analysis of the temperature dependence of spectral energies according to continuous relaxation models reveals two distinct relaxation processes in the deletion mutant, which were assigned to solvent and protein dipoles, respectively. From the long-wavelength shifts of the emission spectra upon red-edge excitation, it is demonstrated that the active site of the wild type enzyme has high structural homogeneity in comparison to the deletion mutant. In combination with results obtained by X-ray diffraction studies on crystals of the wild type enzyme, it can be concluded that the C-terminal polypeptide of the A. vinelandii enzyme interacts with the dehydrolipoamide binding site, thereby shielding the flavins from the solvent.     

Conformational substates of myoglobin detected by extrinsic dynamic fluorescence studies

The extent of conformational substates of two apomyoglobins, i.e., sperm whale and tuna apomyoglobin, was investigated by examining the fluorescence decay in the frequency domain of the extrinsic fluorophore TNS [6-(p-toluidino)-2-naphthalenesulfonic acid] bound to the heme binding site. Data analysis was performed in terms of a continuous, unimodal lifetime distribution having a Lorentzian shape. The results were compared with those for the free fluorophore in an isotropic nonviscous solvent. The incorporation of TNS into the protein matrix resulted in a broadening of the lifetime distribution due to the microenvironmental heterogeneity generated by structural fluctuations. The larger width of lifetime distribution observed for TNS bound to tuna apomyoglobin was related to a more extended conformational space accessible to the fluorophore in this protein compared to sperm whale myoglobin. A temperature increase from 15 to 40 degrees C produced a further broadening of the lifetime distributions of TNS bound to both proteins. This result can be explained by assuming the existence of conformational substates at high energy content or separated by high energy barriers, which are not populated at low temperature. The overall picture emerging from the reported data is that the lifetime distributions of TNS bound to apomyoglobins are determined largely by the number of conformational substates accessible to the protein matrix and, to a lesser extent, by the interconversion rates among these states.     

Defining the interacting regions between apomyoglobin and lipid membrane by hydrogen/deuterium exchange coupled to mass spectrometry

Sperm whale myoglobin can be considered as the model protein of the globin family. The pH-dependence of the interactions of apomyoglobin with lipid bilayers shares some similarities with the behavior of pore-forming domains of bacterial toxins belonging also to the globin family. Two different states of apomyoglobin bound to a lipid bilayer have been characterized by using hydrogen/deuterium exchange experiments and mass spectrometry. When bound to the membrane at pH 5.5, apomyoglobin remains mostly native-like and interacts through alpha-helix A. At pH 4, the binding is related to the stabilization of a partially folded state. In that case, alpha-helices A and G are involved in the interaction. At this pH, alpha-helix G, which is the most hydrophobic region of apomyoglobin, is available for interaction with the lipid bilayer because of the loss of the tertiary structure. Our results show the feasibility of such experiments and their potential for the characterization of various membrane-bound states of amphitropic proteins such as pore-forming domains of bacterial toxins. This is not possible with other high-resolution methods, because these proteins are usually in partially folded states when interacting with membranes.     

Protein surface mapping by chemical oxidation: structural analysis by mass spectrometry

The solvent-accessible surface area of proteins is important in biological function for many reasons, including protein-protein interactions, protein folding, and catalytic sites. Here we present a chemical technique to oxidize amino acid side chains in a model protein, apomyoglobin, and subsequent elucidation of the effect of solvent accessibility on the sites of oxidation. Under conditions of low protein oxidation (zero to three oxygen atoms added per apomyoglobin molecule), we have positively identified five oxidation sites by liquid chromatography-tandem mass spectrometry and high-resolution Fourier transform mass spectrometry. Our results indicate that all oxidized amino acids, with the exception of methionine, have highly solvent-accessible side chains, but the rate of oxidation may not be dictated solely by solvent accessibility and amino acid identity.     

Rotational dynamics of 4-aminobutyrate aminotransferase

The fluorescence dye 1-anilinonaphthalene-8-sulfonate (ANS) was used as a probe of non-polar binding sites in 4-aminobutyrate aminotransferase. ANS binds to a single binding site of the dimeric protein with a K_d of 6 μM. Nanosecond emission anisotropy measurements were performed on the ANS-enzyme in an effort to detect independent rotation of the subunits in the native enzyme. The observed rotational correlation time (φ = 65 ns) corresponds to the rotation of a rather rigid dimeric structure. The microenvironment surrounding the natural probe pyridoxal-5-P covalently bound to the dimeric structure was explored using ³¹P-NMR at 72.86 MHz. In the native enzyme, the pyridoxal-5-P ³¹P-chemical shift is pH-independent, indicating that the phosphate group is well protected from the solvent. The correlation time determined from the ³¹P-spectrum of the aminotransferase exceeds the value calculated for the hydrated spherical model (φ = 40 ns). It is concluded that the phosphate of the pyridoxal-5-P molecule is rigidly bound to the active site of 4-aminobutyrate aminotransferase.     

Solvent accessibility of the heme pocket in tuna myoglobin

The accessibility of the heme binding site of two apomyoglobins, i.e. tuna and sperm whale apomyoglobin, has been evaluated by quenching the fluorescence of their ANS-conjugates. The quenching pattern obtained by using charged and uncharged quenchers revealed that the heme pocket of tuna apomyoglobin is more accessible than that of sperm whale. Moreover, a larger number of positively charged groups is present in the heme pocket of tuna apomyoglobin as indicated by comparing the extent of quenching produced by iodide and cesium ion. The relaxation time of ANS bound to tuna apomyoglobin is lower than that of the same chromophore bound to sperm whale globin thus indicating that there is some localized flexibility in the tuna globin.     

Modulation of dual fluorescence in a 3-hydroxyquinolone dye by perturbation of its intramolecular proton transfer with solvent polarity and basicity.

A representative of a new class of dyes with dual fluorescence due to an excited state intramolecular proton transfer (ESIPT) reaction, namely 1-methyl-2-(4-methoxy)phenyl-3-hydroxy-4(1H)-quinolone (QMOM), has been studied in a series of solvents covering a large range of polarity and basicity. A linear dependence of the logarithm of its two bands intensity ratio, log(I(N*)/I(T*)), upon the solvent polarity expressed as a function of the dielectric constant, (epsilon- 1)/(2epsilon + 1), is observed for a series of protic solvents. A linear dependence for log(I(N*)/I(T*)) is also found in aprotic solvents after taking into account the solvent basicity. In contrast, the positions of the absorption and the two emission bands of QMOM do not noticeably depend on the solvent polarity and basicity, indicating relatively small changes in the transition moment of QMOM upon excitation and emission. Time-resolved experiments in acetonitrile, ethyl acetate and dimethylformamide suggest an irreversible ESIPT reaction for this dye. According to the time-resolved data, an increase of solvent basicity results in a dramatic decrease of the ESIPT rate constant, probably due to the disruption of the intramolecular H-bond of the dye by the basic solvent. Due to this new sensor property, 3-hydroxyquinolones are promising candidates for the development of a new generation of environment-sensitive fluorescence dyes for probing interactions of biomolecules.     

Accessibility and multivalency of immobilized Cibacron blue F3GA

The effect of immobilized dye concentration on protein complexation was observed using zonal chromatography. A monomeric protein, octopine dehydrogenase, was retained by a single interaction to a Sepharose CL-6B column containing 11.6 mM immobilized Cibacron blue F3GA. By contrast, a tetrameric protein, lactate dehydrogenase, was retained by the same column by multiple interactions. The degree of multiple interactions was found to systematically increase with increasing immobilized dye concentration. The concentration of immobilized dye accessible to protein was found to be inversely related to the concentration of ionic components in the solvent. Zonal chromatographic measurements of free dye and unconjugated matrix suggest that increasing the concentration of ionic components promotes the adsorption of immobilized dye to the adjacent matrix surface. Such adsorption markedly affects both the capacity of an immobilized dye column and the multiplicity of its interaction with oligomeric proteins.