Time Resolved Fluorescence in Doped Aerogels and Organosilicate Glasses

Influence of lipid composition and membrane curvature on fluorescence and solvent relaxation kinetics in unilamellar vesicles

Time-resolved fluorescence on unilamellar vesicles shows that increasing amounts of anionic, natural lipid lead to a larger increase in polarity close to the headgroups than in the hydrophobic core of the bilayer. The region close to the headgroups is less polar in vesicles containing phosphatic acid rather than phosphatidylserine. A greater membrane curvature increases the mobility of the hydrated headgroups.

Solvent relaxation in phospholipid bilayers: principles and recent applications

Although there exist a number of methods, such as NMR, X-ray, e.g., which explore the hydration of phospholipid bilayers, the solvent relaxation (SR) method has the advantage of simple instrumentation, easy data treatment and possibility of measuring fully hydrated samples. The main information gained from SR by the analysis of recorded "time-resolved emission spectra" (TRES) is micro-viscosity and micro-polarity of the dye microenvironment. Based on these parameters, one can draw conclusions about water structure in the bilayer. In this review, we focus on physical background of this method, on all the procedures that are needed in order to obtain relevant parameters, and on the requirements on the fluorescence dyes. Furthermore, a few recent applications (the effect of curvature, binding of antibacterial peptides and phase transition) illustrating the versatility of this method are mentioned. Moreover, limitations and potential problems are discussed.

Probing Ethanol-Induced Phospholipid Phase Transitions by the Polarity Sensitive Fluorescence Probes Prodan and Patman

The emission behaviour of the two polarity sensitive probes Prodan and Patman in phospholipid vesicles was studied as a function of the concentration of ethanol. Comparing the spectral shifts in both the symmetric lipid 1,2-dipalmitoyl-sn-glycero-3-phosphatidylcholine (DPPC) showing a phase transition from a normal to a fully interdigitated gel phase and the strongly asymmetric lipid 1-stearoyl-2-lauroyl-sn-glycero-3-phosphatidylcholine (C(18):C(12)-PC) favouring a mixed interdigitated gel phase we show that the huge red shifts of Prodan in presence of higher ethanol concentrations cannot be easily attributed to a specific lipid phase transition. Rather, probe relocation and a pronounced increase in solvent relaxation (SR) as monitored by time-resolved emission spectra (TRES) in presence of ethanol contribute to the large shifts observable in both lipid systems in case of Prodan. While Patman exhibits a red shift caused by increased SR due to the ethanol induced formation of a fully interdigitated phase in DPPC, hardly any shift occurs in C(18):C(12)-PC, which is supposed not to undergo an ethanol-induced phase transition.

Calcium-induced conformational change in fragment 1-86 of factor X

Time-resolved fluorescence of the single tryptophan residue Trp41 in fragment 1-86 of factor X (FX F1-86) is studied using a time-correlated single photon counting technique with synchrotron radiation as the excitation source. Calcium ions are believed to induce a conformational change in the N-termini of the activated factor X and other vitamin K dependent proteins, which is accompanied by a decrease in fluorescence intensity. The titration with calcium yields a sigmoidal fluorescence titration curve with a transition midpoint concentration of 0.44 mM. The wavelength-dependent tryptophan fluorescence decays of the apo-FX F1-86 (in the absence of calcium) and Ca-FX F1-86 are characterized by conventional multiexponential analysis and fluorescence lifetime distribution analysis. In the absence of calcium there are three significant classes of fluorescence lifetimes (ns) that are nearly wavelength independent: 0.55 +/- 0.08 (component A), 2.6 +/- 0.1 (component B), and 5.3 +/- 0.3 (component C). However, their preexponential amplitudes vary with wavelength. The decay associated emission spectra of the individual components show that components B and C contribute over 85% to the total fluorescence for all examined wavelengths. However, in the presence of calcium, the analysis of the time-resolved fluorescence data of Ca-FX F1-86 yields four wavelength-independent lifetimes (ns) of 0.30 +/- 0.09 (component D), 0.65 +/- 0.10 (component A), 2.7 +/- 0.2 (component B), and 5.4 +/- 0.3 (component C). Calcium addition to the apo-FX F1-86 leads to a decrease in the fluorescence intensities of components B and C while their decay times remain unaffected. In Ca-FX F1-86 an additional component D arises that has a decay time of 0.30 ns and that contributes up to 35% to the total fluorescence intensity. A comparison with a previous investigation of prothrombin fragment 1 demonstrates the extensive structural and functional homology between the N termini of prothrombin and factor X(a).

Binding of prothrombin and its fragment 1 to phospholipid membranes studied by the solvent relaxation technique

The phospholipid headgroup mobility of small unilamellar vesicles composed of different mixtures of phosphatidyl-L-serine (PS) and phosphatidylcholine is characterized by the solvent relaxation behavior of the polarity sensitive dyes 6-propionyl-2-(dimethylamino)naphthalene (Prodan) and 6-palmitoyl-2-[trimethylammoniumethyl]-methylamino]naphthalene chloride (Patman). If the PS content exceeds 10%, the addition of calcium leads to a substantial deceleration of the solvent relaxation of both dyes, indicating the formation of Ca(PS)2 complexes. Addition of prothrombin and its fragment 1 leads to a further decrease of the headgroup mobility, as explained by the binding of more than two PS-molecules by a single protein molecule. Prodan monitors the outermost region of the bilayer and it clearly distinguishes between the binding of prothrombin and its fragment 1. The deeper incalated Patman does not distinguish between both proteins. The validity of the solvent relaxation technique for the investigation of the membrane binding of peripheral proteins is demonstrated by the studies of prothrombin induced changes in the steady-state fluorescence anisotropies of 1,6-diphenyl-1,3, 5-hexatriene.

Picosecond tryptophan fluorescence of membrane-bound prothrombin fragment 1

The wavelength-dependent tryptophan (Trp) fluorescence decays of Ca-prothrombin fragment 1 (Ca-BF1), which contains three tryptophan residues, in the presence of pure phosphatidylcholine (PC) small unilamellar vesicles (SUV) and PC-SUV containing either 25% phosphatidyl-l-serine (PS), and 25% or 40% phosphatidylglycerol (PG) are characterized, using fluorescence lifetime distribution, conventional multiexponential, and global analysis. In analogy to previous investigations on apo- and Ca-BF1 (M. Hof, G.R. Fleming, V. Fidler, Proteins Struct. Func. Genet. 24 (1996) 485-494), the analysis resulted in a five exponential decay model in all investigated systems, where the five fluorescence lifetimes (e.g. 0. 04+/-0.02 ns (component A), 0.24+/-0.02 ns (B), 0.66+/-0.03 ns (C), 2.4+/-0.3 ns (D), and 5.4+/-0.4 ns (E) for Ca-BF1 in the presence of PC-SUV) are wavelength-independent. The fluorescence lifetimes and the corresponding amplitudes of the 'Gla-Trp' (components D and E) and of the two 'kringle-Trp' (components B, C, and D) remain unchanged when bound to the PS-containing vesicles. Saturation binding to PG-containing membranes leads to a prolongation of the Gla component E from 5.3 in solution to 7.5 ns, indicating a change in the Gla-domain conformation. The results represent the first experimental evidence of a lipid-specific conformational change in the N-terminal 'Gla domain' of a vitamin K-dependent protein.

Photochemical and photobiological studies with acridine and phenanthridine hydroperoxides in cell-free DNA

The acridine and phenanthridine hydroperoxides 3 and 7 were synthesized as photochemical hydroxyl radical sources for oxidative DNA damage studies. The generation of hydroxyl radicals upon UVA irradiation (lambda = 350 nm) was verified by trapping experiments with 5,5-dimethyl-1-pyrroline N-oxide and benzene. The enzymatic assays of the damage in cell-free DNA from bacteriophage PM2 caused by the acridine and phenanthridine hydroperoxides 3 and 7 under near-UVA irradiation revealed a wide range of DNA modifications. Particularly, extensive single-strand break formation and DNA base modifications sensitive to formamidopyrimidine DNA glycosylase (Fpg protein) were observed. In the photooxidation of calf thymus DNA, up to 0.69 +/- 0.03% 8-oxo-7,8-dihydroguanine was formed by the hydroperoxides 3 and 7 on irradiation, whose yield was reduced up to 40% in the presence of the hydroxyl radical scavengers mannitol and tert-butanol. The acridine and phenanthridine hydroperoxides 3 and 7 also induce DNA damage through the type I photooxidation process, for which photoinduced electron transfer from 2'-deoxyguanosine to the singlet states of 3 and 7 was estimated by the Rehm-Weller equation to possess a negative Gibb's free energy of ca -5 kcal/ mol. Control experiments with the sensitizers acridine 1 and the acridine alcohol 4 in calf thymus and PM2 DNA confirmed the photosensitizing propensity of the UVA-absorbing chromophores. The present study emphasizes that for the development of selective and efficient photochemical hydroxyl radical sources, chromophores with low photosensitizing ability must be chosen to avoid type I and type II photooxidation processes.

Photochemical and photobiological studies of a furonaphthopyranone as a benzo-spaced psoralen analog in cell-free and cellular DNA

Photobiological activities of the benzo-spaced psoralen analog furonaphthopyranone 3 have been investigated in cell-free and cellular DNA. The molecular geometry parameters of 3 suggest that it should not form interstrand crosslinks with DNA. With cell-free DNA no evidence for crosslinking but also not for monoadduct formation was obtained; rather, the unnatural furocoumarin 3 induces oxidative DNA modifications under near-UVA irradiation. The enzymatic assay of the photosensitized damage in cell-free PM2 DNA revealed the significant formation of lesions sensitive to formamidopyrimidine DNA glycosylase (Fpg protein). In the photooxidation of calf thymus DNA by the furonaphthopyranone 3, 0.29 +/- 0.02% 8-oxo-7,8-dihydroguanine (8-oxoGua) was observed. With 2'-deoxyguanosine (dGuo), the guanidine-releasing photooxidation products oxazolone and oxoimidazolidine were formed predominately, while 8-oxodGuo and 4-HO-8-oxodGuo were obtained in minor amounts. The lack of a significant D2O effect in the photooxidation of DNA and dGuo reveals that singlet oxygen (type II process) plays a minor role; control experiments with tert-butanol and mannitol confirm the absence of hydroxyl radicals as oxidizing species. The furonaphthopyranone 3 (Ered = -1.93 +/- 0.03V) should act in its singlet-excited state as electron acceptor for the photooxidation of dGuo (delta GET ca -6 kcal/mol), which corroborates photoinduced electron transfer (type I) as a major DNA-oxidizing mechanism. A comet assay in Chinese hamster ovary (CHO) AS52 cells demonstrated that the psoralen analog 3 damages cellular DNA upon near-UVA irradiation; however, no photosensitized mutagenicity was observed in CHO AS52 cell cultures.

Solvent relaxation behaviour of n-anthroyloxy fatty acids in PC-vesicles and paraffin oil: a time-resolved emission spectra study

Time-resolved fluorescence measurements were performed for a set of n-anthroyloxy fatty acids (n-AS; n = 2, 3, 6, 9, 12, 16) in both solvent and vesicle systems. The Stokes' shifts and the mean relaxation times calculated from the time-resolved emission spectra (TRES) are shown to be strongly dependent on the position of the fluorophore in small unilamellar vesicles (SUV) composed of phosphatidylcholine (PC), while they are essentially independent of the fluorophore position in isotropic paraffin oil. The concept of an intramolecular relaxation process which had been suggested to explain the wavelength dependence of the emission behaviour of the n-AS dyes in viscous solvents is supported by semiempirical calculations showing that a more planar conformation is favoured in the excited compared to the ground state. However, in order to explain the results in vesicle systems, the concept of intramolecular relaxation is not sufficient. Rather, we show that intermolecular solvent relaxation processes play the dominant role for the wavelength dependent emission behaviour in polar, viscous environments.

Binding and relaxation behaviour of prodan and patman in phospholipid vesicles: a fluorescence and 1H NMR study

The relative location, binding behaviour and the solvent relaxation behaviour of the polarity sensitive membrane probes 6-propionyl-2-(dimethylamino)naphthalene and 6-palmitoyl-2-[[trimethylammoniumethyl]methylamino]naphthalene chloride in vesicles composed of 1,2-dimyristoyl-sn-glycero-3-phosphatidylcholine or egg yolk lecithin have been compared using steady-state and time-resolved fluorescence as well as high resolution NMR measurements. The reconstructed time-resolved emission spectra show unambiguously that the observed spectral shifts in vesicle systems have to be assigned to time-dependent solvent relaxation processes rather than to a probe relocation mechanism. All fluorescence as well as the NMR relaxation data suggest a deeper localization of Patman in the membrane, sensing a less polar and/or more restricted probe environment.

Time-resolved fluorescence study of a calcium-induced conformational change in prothrombin fragment 1

The wavelength dependent fluorescence decay properties of bovine prothrombin fragment 1 have been investigated employing a picosecond time-correlated single photon counting technique. All observations are discussed with using the crystal structure (Soriano-Garcia et al., Biochemistry 31:2554-2566, 1992). Fluorescence lifetimes distribution and conventional multiexponential analysis, as well as acrylamide quenching studies lead to the identification of six distinguishable tryptophan excited-states. Accessibility to the quencher and the known structure are used to associate a fluorescence decay of the tryptophan present in the Gla domain (Trp42) with two red shifted components (2.3 and 4.9 ns). The two kringle domain tryptophans (Trp90 and Trp126) exhibit four decay times (0.06, 0.24, 0.68, and 2.3 ns), which are blue shifted. The calcium-induced fluorescence quenching is a result of static quenching: the five decay times remain unchanged, whereas the fluorescence intensity of Trp42 is decreased. The static quenching process is a consequence of a ground state interaction between the Cys18-Cys23 disulfide bridge and Trp42. The monomolecular equilibrium constant for this disulfide-pi-electron interaction is found as 4.8.

Influence of vesicle curvature on fluorescence relaxation kinetics of fluorophores

The effect of membrane curvature on the fluorescence decay of 2-p-toluidinyl-naphthalene-6-sulfonic acid (TNS), 2-(9-anthroyloxy) stearic acid (2-AS) and 12-(9-anthroyloxy)-stearic acid (12-AS) was investigated for egg lecithin vesicles of average diameter dm = 22 nm and 250 nm. The biexponential fluorescence decay of TNS at the red edge of the emission spectrum was analysed according to the model of Gonzalo and Montoro [1]. Over the entire temperature range (1-40 degrees C) the small TNS labelled vesicles showed significantly shorter solvent relaxation times tau(r) than their larger counterparts (e.g. 1.3 ns compared with 2.1 ns at 5 degrees C), indicating a higher mobility of the hydrated headgroups in the highly curved, small vesicles. The fluorescence decay of both AS derivatives is also biexponential. While the shorter decay times (1-3 ns) are practically identical for small and large vesicles, the longer decay times (5-14 ns) are identical only for 12-AS but not for 2-AS. This indicates that the microenvironment is similar in small and large vesicles deep in the membrane in spite of the differences in curvature.