The light microscope is an icon of the sciences, but its sophistication and usefulness have had a far from linear evolution. During the last decade great progress in fluorescent probe design, light source and detector technology, and computers have contributed to a revolutionary increase in the capabilities of light microscopy, turning it from a static, 2D tool into a dynamic, 3D means of studying live biological specimens. We believe that the next wave in light microscopy is based on automation and improved spatio-temporal resolution. Our work on multimode robotic microscopy and standing wave-excited fluorescence microscopy is presented here. The former allows simultaneous imaging with several modes of light microscopy and good specimen control (temperature, displacement, chemistry), while the latter affords an axial resolution significantly improving on confocal microscopy and, for the first time, exceeding transverse resolution. Examples of biological information obtainable solely by these advances are also given.

A spectrograph based imaging system attached to a microscope is described for use in monitoring the fluorescence of multiple probes loaded into single cells. Movements of ions, important in regulating cell function, are interrelated under many conditions. Since we are interested in the mechanisms underlying coupling between ion movements and metabolism, we required a microscope based system to analyze these parameters at the level of a single cell. The spectral imaging system was implemented for two purposes. (1) To resolve heterogeneities which exist in the response of individual cells in a population, and within any given cell due to loading of dye into more than one subcellular compartment, and (2) to study the temporal relations between several physiological parameters from a single cell. Since spectral imaging resolves signal from multiple probes simultaneously, the concentrations of several ions or metabolic factors can be followed in a single cell. Moreover, spatial information is acquired along one axis, providing information from probes located in individual compartments. The spectral imaging microscope and some specific applications are described.

We have used digital imaging microscopy to evaluate possible guidance mechanisms of mouse neuroblastoma cells cultured on microfabricated glass substrates. Substrates were bound with either small charged amine or uncharged alkane molecules using silane-coupling chemistry or covered with proteins (laminin, collagen, fibronectin, albumin) by adsorption. Photolithographic procedures were used to pattern substrates with amine - alkane, amine - protein and alkane - protein combinations. Interference reflection microscopy (IRM) was used to visualize sites of cellular attachment on substrates (focal and close contacts), and image processing techniques were used to quantify the results in three different ways: total cellular area in contact with a substrate, area corresponding to focal and close contacts, and the percentage of focal and close contacts on each substrate. Cells attached to and extended neurites on each of the substrates were tested. IRM images of growth cones displayed similar gray levels on amine, alkane, fibronectin, and albumin substrates, whereas images on laminin and collagen were brighter. Brightness on laminin substrates was correlated with less area of focal contact and greater area with no contact. When cells were provided with choices on patterned substrates, they displayed the following preference of attachment: laminin, fibronectin, collagen>amine>alkane>albumin. This hierarchy reflected greater total surface area on preferred substrates. There was, however, no correlation between the hierarchy and the area corresponding to focal and close contacts or the percentage of contacts on different substrates. Our results support several recent observations which demonstrated that guidance was not correlated with differential adhesivity. These results are more in line with the theory that guidance is controlled by a signal transduction mechanism that couples locomotion with activation of membrane receptors.

In order to identify neuronal networks, it is generally required to fix tissue followed by some specific staining procedure. A new procedure is described in this manuscript that labels brain slices that are routinely used for electrophysiological analyses. Fluorescently-labeled lipids can be incorporated into brain slices via passive exchange from exogenously applied vesicles. The labeled lipid is distributed throughout distinct cellular structures of the hippocampus and cerebellum. High resolution images of cells can be obtained and as the labeling process does not affect the electrical properties of the labeled cells, further electrophysiological analyses can be made of identifiable cells. The distribution of the lipid depends on the labeled phospholipid species. One of the lipids analyzed has been previously used for in vitro phospholipase analyses. Addition of phospholipase activating agents resulted in identification with high spatial and temporal resolution of activation of this enzyme in specific cell types. The cells affected correlated with previously identified regions of relevant pharmacological activity. This procedure shows considerable promise for monitoring biochemical changes due to physiological, toxicological, or pathological changes in intact neuronal networks.

Numerical techniques are described which allow quantitative interpretation of 2D and 3D histological imagery obtained from tissue samples which are stained with the neuronal tracer Fast Blue. The methods illustrated in this investigation are readily extensible to many different types of micro neuro-anatomy, and are particularly well suited for investigating pharmacological or environmental perturbations on neuron development. These techniques were used to quantitatively describe neurological development of the optic nerve in the pigmented rat. This analysis was used to show that most of the significant refinement in the retinotectal map occurs during the second to fourth week of life and with coincident with eye opening.

We show that the confocal fluorescence lifetime imaging method is a powerful tool for the quantitative determination of pH on a microscopic scale. This method is easily implemented using a conventional confocal microscope and moreover, utilizes currently available fluorescent probes. It is shown that both the intensity probe DM-NERF and the ratio probe BCECF are suitable for this purpose, albeit with different useful pH ranges. In addition it is shown that the fluorescence decay of both probes are independent of the probe-concentration. Furthermore, the fluorescence lifetime behavior of BCECF is found to be insensitive to changes in the hydrophobicity and protein content of the buffer solution. The intracellular pH was imaged using BCECF since this probe is sensitive in the physiological pH range. A realistic pH value of about 7.3 was found throughout CHO cells.

Recent technological advances have provided an opportunity for the development of `fluorescence lifetime imaging microscopy' (FLIM). FLIM is an extremely important advance, as it allows for the first time, the sensitivity of the fluorescence lifetime to environmental parameters to be monitored in a spatial manner in single living cells. FLIM can be developed both on conventional and confocal fluorescence microscopes. Fluorescence lifetime detection can be performed using either time- or frequency-domain methods. In this paper, we report on the development of conventional and confocal FLIM systems currently underway in our laboratory. We also present three examples of current biological research projects in which we employ FLIM.

The ability to quantitate physiological parameters in single living cells using fluorescence spectroscopic imaging has expanded our understanding of many cell regulatory processes. Previous studies have focussed on the measurement of single parameters, such as the concentration of calcium, and more recently two parameters, such as calcium and pH using fluorescence ratio imaging. The complexity of the interrelationships among cell biochemical reactions suggests a need to extend the measurement scheme to several parameters. Expansion of the number of parameters involves several complexities associated with fluorescent probe selection and instrumentation design as well as the processing and management of the data. A system has been assembled which provides maximum flexibility in multiparameter fluorescence imaging measurements. The system provides multiple combinations of excitation, dichroic mirror, and emission wavelengths. It has automatic acquisition of any number of parameters. The number of parameters is primarily limited by the selection of fluorescent probes with nonoverlapping spectra. We demonstrate the utility of the system by the coordinated monitoring of stimulated changes in the concentrations of calcium, magnesium, and pH using fluorescence ratio imaging coupled with a conventional transmitted light image of single smooth muscle cells. The results demonstrate coordinated changes in some instances but uncoordinated changes in others.

Time-resolved fluorescence imaging can enhance the contrast of microscope images and it can also provide important information about the micro-environment in cellular systems. We have developed a fluorescence microscope which can measure fluorescence lifetimes over an entire image. Fluorescence lifetimes are measured by using heterodyne frequency domain techniques. Heterodyning is accomplished by using an intensity modulated laser light source and a fast scan CCD camera coupled with a gain modulated microchannel plate as the detector. The high duty cycle of this method allows us to generate a phase resolved image with about five seconds integration time. Operating in the fast scan mode, the systematic uncertainties in lifetime determination caused by photobleaching are less severe than those of slow-scan cameras. The microchannel plate can be modulated at frequencies up to 300 MHz, which allows us to measure lifetimes as short as 500 ps with resolution of 50 ps. The modulation of the microchannel plate only slightly degrades the spatial resolution of the image from the diffraction limit; 0.8 micron resolution is maintained with 500 nm laser excitation.

We have designed an epifluorescence video microscope for simultaneous dual excitation of indo 1 (for [Ca²⁺]_i) at 350 nm and SNARF 1 (for pH_i) at 540 nm. The microscope will simultaneously capture all four emission images at 405, 475, 575, and 640 nm from the two ratio dyes at video frame or field rates. Popular dyes for measuring [Ca²⁺]_i, such as indo 1 and fura 2, have pH- dependent Kd's; thus changes in pH_i can be misinterpreted as changes in [Ca²⁺]_i. For any pixel (or region of interest), we use the pH value to generate the appropriate Kd. The corrected Kd is then used to calculate a corrected calcium value. Using the imaging system, we show that, for the peptide secreting melanotropes of the pituitary intermediate lobe, changes in intracellular calcium ([Ca²PLU)]_i) produce changes in intracellular pH (pH_i). Melanotropes grown in primary explant culture and double-loaded with indo-1 acetoxy methyl ester (AM) and SNARF-1 AM, were examined for Ca²⁺/pH interactions. Following experimentation, cells were positively identified by (beta) -endorphin fluorescence immunohistochemistry. K-induced depolarization of melanotropes produced increases in [Ca²PLU)]_i due to activation of L-type Ca-channels. Ca²⁺ entry was closely coupled to reductions in pH_i. Effects were dependent upon entry of extracellular Ca²⁺ rather than release from intracellular stores. The close association between increases in intracellular Ca²⁺ and H⁺ suggest that the pH_i changes are due to release of H⁺ upon binding of Ca²⁺ to intracellular buffers. Although the pH drop is `passive,' it will be sensed by all cytoplasmic components. Thus it represents a second messenger pathway, akin to the generation of cAMP or inositol trisphosphate, which cannot fail to influence numerous pH-dependent cell activities.

Fluorescence lifetime imaging is a relatively new technique for acquiring directly the nanosecond temporal characteristics of the fluorescence emission of a spatially extended object, and for capturing the dynamic features at every pixel of an image simultaneously. In general, the applications of fluorescence lifetime imaging have been mainly in the microscope, but other diverse imaging situations can benefit from the technology. Our instrument employs periodically modulated excitation light, synchronous modulation of the amplification stages of a microchannel plate intensifier, and subsequent digital recording of the image with a charge-coupled device camera. The digitized images can be subsequently analyzed with a variety of different ways. A short description of the lifetime resolved fluorescence imaging instrumentation is given together with typical applications depicting lifetime spatial distributions, multiple lifetime analysis, statistical analysis of the image data, and suppression or enhancement of particular fluorescent species.

Horse liver alcohol dehydrogenase is a homodimeric protein; each subunit has two tryptophan residues that are in distinctly different microenvironments. Trp-15 is located on the surface and Trp-314 is buried at the intersubunit interface. Steady-state and time-resolved fluorescence and phosphorescence studies have enabled the assignment of parameters, e.g., quantum yield, emission maximum, decay times, to the individual tryptophan residues of the protein. We have prepared, by site-directed mutagenesis, the mutated W15F protein and have characterized its fluorescence properties. We show that the Trp-314 of the mutant experiences an apolar microenvironment, but that the fluorescence decay and exposure to solute quenchers of the mutant are somewhat different than was expected from the assignments for the wild type.

Minimum perturbation mapping reveals that five side chain conformations of tryptophan-59 can potentially pack in the hydrophobic core of ribonuclease T1. In the crystallographic structure tryptophan-59 is in the trans perpendicular conformation. The other four wells in the tryptophan-59 (chi) ¹ x (chi) ² torsion space minimum perturbation mapping are trans antiperpendicular, gauche⁺ antiperpendicular, gauche⁺ perpendicular, and guache^- perpendicular. The point mutations V33I and V33L are predicted to stabilize the trans antiperpendicular over the perpendicular conformation by 1.3 and 0.6 kcal/mol, respectively. The three gauche conformations require the creation of a new pocket within the protein core. The Mutation F80G creates the pocket for the gauche⁺ antiperpendicular tryptophan-59 conformation. The additional mutation V78A makes a more generous pocket for this conformation. The three mutation A19G, V78G and F80L creates the pocket for the gauche⁺ perpendicular conformation. A slightly more generous mutation at residue 80 may be required to fully stabilize this conformation. It does not appear possible to create a pocket that fits the gauche^- perpendicular conformation better than the other two gauche conformations. The expression of ribonuclease-T1 mutants with tryptophan-59 in a gauche conformation probably requires filling in the crystallographic tryptophan-59 pocket. The mutations A22F, V33F, or V67F are all promising possibilities for this purpose.

Our recent linear dichroism study of transition moment directions for protoporphyrin derivatives [1,2] demonstrate that heme cannot be considered a planar oscillator when it acts as an acceptor of radiationless excitation energy transfer from tryptophan. The linear nature of the heme absorption transition moment implies a strong dependence of the transfer rate factors on the relative angular position of the heme and tryptophan, i.e. on the k² orientation parameter of the Forster equation. Using the atomic coordinates of human hemoglobin and taking into account the direction of the transition moment of the near UV (300-380 nm) heme absorption band we have estimated the rate of energy transfer from tryptophan to heme in the isolated a chains, which are a single tryptophan protein. It appears that the rate of energy transfer is very sensitive to the orientation of the transition moment of the heme and similarly to myoglobin [3] natural heme disorder significantly reduces the transfer efficiency in isolated a subunits. On this basis we were able to predict very accurately the two lifetimes detectable in the systems, of 32 and 1050 ps respectively, where the amplitude of the longer lifetime is very consistent with the amount of disordered hemes found by La Mar [4,5] for the a subunits of hemoglobin.

Dynamic fluorescence measurements of radiationless energy transfer between tryptophan donor groups located within the regulatory peptides Phk13 and Phk5 of glycogen phosphorylase kinase and nitrotyrosine acceptor groups within calmodulin or Phk13 have provided a set of separations and separation distributions which place definite restrictions upon the geometry of the peptide-calmodulin complexes. Phk13 appears to be bent into a hairpin-shaped (beta) -structure. The distributions of separations are suggestive of substantial internal mobility of the complex species.

The phosphorescence decays of horse liver alcohol dehydrogenase (LADH) Trp314, E. coli alkaline phosphatase (AP) Trp109, and B. stearothermophilus phosphofructokinase (PFK) Trp179 are decidedly nonexponential at room temperature. When the data is analyzed using the maximum entropy method (MEM) the AP phosphorescence decay is dominated by a single gaussian distribution while for LADH and PFK the data reveals at least two amplitude packets. The MEM lifetime-normalized widths for these proteins are significantly larger than obtained for the model monoexponential chromophore terbium suggesting that the complex kinetics is intrinsic to the protein. Since the phosphorescence lifetime of a tryptophan residue is related to its microviscosity, the nonexponential decay behavior may imply that the phosphorescing tryptophan residue in each of these samples is best described as existing in at least two states of different local rigidity which interconvert more slowly than the time scale of the phosphorescence decay (0.1 to 1.0 sec). The existence of multiple, long-lived, conformers is further supported by the observation that the phosphorescence lifetime in the LADH sample is excitation wavelength dependent.

The fluorescence intensity decays of the two tryptophans in rabbit skeletal myosin rod were fit to double exponential functions at six temperatures ranging from 5 to 62 degree(s)C. The radiative decay rate (kf) was estimated from the ratio of the average lifetime and quantum yield (0.11) at 23 degree(s)C. This value (0.024 ns^-1 was used to calculate the non-radiative decay rates (equals knr) for each lifetime species at each temperature. Plots of log (knr) versus 1/T for each species showed distinctive breaks at 45 degree(s)C. The Arrhenius activation energies (Ea) for non-radiative quenching of the two lifetime species were significantly different below 45 degree(s)C, with values of 85 J/mole for the long and 31 J/mole for the short lifetime species, while the values were similar, 210 J/mole for the long and 230 J/mole for the short species, above 45 degree(s)C. The thermal unfolding of rod is known to be complex with transitions at 43 degree(s), 47 degree(s), and 53 degree(s)C; only the 43 degree(s) and 53 degree(s)C transitions involve the tryptophan-containing light meromyosin region (LMM) (King & Lehrer (1989) Biochemistry 28, 3498). Our results indicate that the two lifetime components are quenched by different mechanisms below, and similar mechanisms above, 45 degree(s)C. Although the environment of the two pairs of tryptophans differ in the folded protein, they have similar environments in the unfolded protein.

Ga contains two tryptophans. Tryptophan fluorescence decay of Ga at 25°C in 10 mM Tris-HC1 buffer pH=7.3 was triexponential with lifetimes of 3.62, 1.06 and 0.l7ns. Contrary to expectation, the shorter lived (lower quantum yield) components were blue-shifted relative to the longer Denaturation with 8M urea diminished the 1.O6ns component and increased net intensity which became dominated by a lifetime of 5. iOns. Emission of a tryptophan species was identified at 450nm with lifetime of 20-4Ons, similar to that reported earlier by Vanderkooi J. et al' . This long-wavelength emission was also eliminated by 8M urea denaturation, making it seem possible that the low intensity of the 1 .O6ns blue shifted component may result from competing excited state processes that give rise to the 450nm emission. Transient absorbance spectra of Ga show no evidence for triplet states in denatured protein while the native protein showed phosphorescence at 442nm of lifetime between 0. 1 and 1 .Oms.

The time-resolved fluorescence decay of single tryptophan (Trp) proteins is typically described using either a distribution of lifetimes or a sum of two or more exponential terms. A possible interpretation for this fluorescence decay heterogeneity is the existence of different isomeric conformations of Trp about its (chi) +1) and (chi) +2) dihedral angles. Are multiple Trp conformations compatible with the remainder of the protein in its crystallographic configuration or do they require repacking of neighbor side chains? It is conceivable that isomers of the neighbor side chains interconvert slowly on the fluorescence timescale and contribute additional lifetime components to the fluorescence intensity. We have explored this possibility by performing minimum perturbation mapping simulations of Trp 28 and Trp 31 in thioredoxin (TRX) using CHARMm 22. Mappings of Trp 29 and Trp 31 give the TRX Trp residue energy landscape as a function of (chi) +1) and (chi) +2) dihedral angles. Time-resolved fluorescence intensity and anisotropy decay of mutant TRX (W28F and W31F) are measured and interpreted in light of the above simulations. Relevant observables, like order parameters and isomerization rates, can be derived from the minimum perturbation maps and compared with experiment.

The fluorescence intensity decay of ribonuclease T1 is biexponential at neutral pH. The lifetimes in nsec and preexponential factors of the exponential components are 3.9 (81%) and 1.7 (19%). The mutations A22L, G23A, L26A, V67G, and V67D, which all neighbor tryptophan-59, have a fairly small effect on this biexponential decay. The lifetime of the long lived component varies from 3.7 to 4.2 nsec and the preexponential varies from 75% to 92%. The emission maximum varies from 319 to 328 nm and the acrylamide quenching rate constant varies from 2.0 to 4.0 x 10⁸ M₁s₁ for these mutations. Minimum perturbation mapping simulations of the tryptophan-59 side chain in wild type ribonuclease T1 show that the (chi) ² side chain dihedral angle may adopt either a perpendicular or an antiperpendicular conformation. These computational and spectroscopic results lead us to propose a two- way packing model for tryptophan-59. This model predicts the relative free energies of the perpendicular and antiperpendicular conformations and acrylamide interaction site to indole ring distances for the wild type and mutant ribonucleases.

The Center for Fluorescence Spectroscopy (CFS) is a multi-user facility providing state of the art time-resolved fluorescence instrumentation and software for scientists, whose research can be enhanced by such experimental data. The CFS is a national center, supported by the National Center for Research Resources Division of the National Institutes of Health, and in part by the National Science Foundation. Both time-domain (TD) and frequency- domain (FD) measurements (10 MHz to 10 Ghz) are available, with a wide range of excitation and emission wavelengths (UV to NIR). The data can be used to recover distances and site-to-site diffusion in protein, interactions between macromolecules, accessibility of fluorophores to quenchers, and the dynamic properties of proteins, membranes and nucleic acids. Current software provides for analysis of multi-exponential intensity and anisotropy decays, lifetime distribution, distance distributions for independent observation of fluorescence donors and acceptors, transient effects in collisional quenching, phase-modulation spectra and time-resolved emission spectra. Most programs provide for global analysis of multiple data sets obtained under similar experimental conditions. Data can be analyzed on-site by connection with the CFS computers through the internet. During six years of operation we have established scientific collaborations with over 30 academic and industrial groups in the United States. These collaborations have resulted in 63 scientific papers.

The technical requirements for a small, rugged, and moderately- priced device for measuring fluorescence lifetimes have been investigated. The suitability and performance of various lifetime measuring schemes were compared. Based on these investigations a compact time-domain instrument was developed allowing measurement of fluorescence decays with a time resolution well below 1 ns. A semiconductor laser (frequency-doubled, if necessary) is used as a light source. Detection is done with a miniaturized photomultiplier. In favorable cases measurement of a fluorescent decay curve is accomplished within less than one minute.

We have constructed a fluorescence decay instrument with computer-automated optical components and data acquisition. Stepping motors control polarizer orientations, filter holders, retarder for incident intensity control, and four position sample changer. The changer uses a mechanical indexer for rapid, precise positioning, and has thermoelectric temperature control. Nitrogen flush and magnetic stirring are provided for all four cuvette positions. An automatic shutter protects the photomultiplier tube, closing automatically when the instrument is opened or the measured photon flux exceeds a predetermined limit. Optical sensors relay position information to the computer for all moving components. The instrument is controlled by a Windows-based program designed to accommodate users of widely varying ability. An inexperienced student can automatically run a complex anisotropy decay experiment with careful sensitivity corrections. Using simple editing functions, a more experienced user, on the other hand, can vary an experiment in minute detail. Automatic algorithms are used to home the instrument at the beginning of an experiment, to increase incident laser intensity until a specified count rate is achieved, and to maintain the count rate during a measurement. We also summarize here some instrumental artifacts common to time-resolved fluorescence data as well as approaches we have used to minimize their effects on recovered decay parameters.

A new versatile system for the measurement of time-resolved fluorescence emission spectra of biomolecules is presented. Frequency doubling and tripling of a Ti:Sapphire laser allows excitation over a wide wavelength range. The influence of increasing the spectral resolution on the time resolution has been investigated. System performance can be optimized for best resolution in the spectral or time domain, respectively. System performance can be optimized for best resolution in the spectral or time domain, respectively. The currently achieved temporal resolution is 6 psec, and the best spectral resolution is 3 nm. Long fluorescence decays can be resolved with optimal time resolution by way of taking into account the flyback of the streak camera. With the system described, the core complex ((alpha) (beta) )₃^APCL_C^8.9 of the phycobilisome from the photosynthetic cyanobacteria Mastigocladus laminosus has been analyzed. Lifetime analysis clearly demonstrated the influence of the linker polypeptide on the phycobiliprotein complex and the identity of native and reconstituted complex.

The goal of `smart photoacoustics' is to rapidly analyze the rate and amplitude of photoinduced volume changes (ns to microsecond(s) time window), as well as to diagnose the source of the measured volume change: heat release, molecular conformational change and/or electrostriction (solvation or desolvation of charged groups). In contrast, the older technique of `photoacoustic calorimetry' tacitly assumes that only heat release from photoexcited molecules contributed to signals. Smart photoacoustics uses hardware, software, and chemical protocols to distinguish the different contributors to the observed signals. The method is demonstrated using two systems: the proton reactions of photoexcited pyranine (8-hydroxpyrene-1,3,6-trisulfonic acid) and the conformational changes in sperm whale carboxymyoglobin upon cleavage of the Fe-CO bond. Smart photoacoustics is the basis for the pulsed-laser photoacoustic instrument we are developing, which includes a sample chamber having thermoelectric control of temperature, magnetic stirring, control of transducer mounting tension, nitrogen flush of cuvette surfaces to eliminate condensation, argon degassing of samples, and two ports for monitoring optical properties of the sample. Central to the instrument is a Windows-based instrument control program which uses a script language to completely automate the operation of the instrument, the collection of data, and the analysis of the photoacoustic waveforms.

Blood clots are illuminated with a focused HeNe laser beam at different power levels. The time-varying reflected speckle pattern (which is recorded with a video camera and filed in a computer memory) rapidly changes in the early heating stage and then remains progressively at rest. The difference between pairs of time-consecutive images is measured by adding the squared differences of intensities at corresponding pixels. This allows one to define a speed of variation D(t) of the reflected image. It is shown that the time-law D(t) is similar to the function describing the derivative of sample temperature with respect to heating time (t). This supports the hypothesis that the observed effect is due to thermal dilatation of biological material forming the clot.

The Confocal Synchrotron Radiation Microscope at Daresbury was used in a study of the transport and distribution of the steroid Coumestrol in single Leydig cells. The broad spectrum of synchrotron radiation in combination with UV compatible microscope optics affords the extension of confocal microscopy from the visible to the UV region down to about 200 nm. Consequently fluorescent molecules with absorption bands in the UV can be imaged. In addition the pulsed nature of the light source allows us to perform time-resolved fluorescence spectroscopy experiments on microscopic volumes. Coumestrol is a naturally fluorescing plant steroid exhibiting estrogenic activity. In physiological environments it has an absorption peak in the UV at 340 nm and it emits around 440 nm. First results indicate that the Coumestrol transport through the cell membrane is diffusion limited. The weak fluorescence observed in the nuclei of the Leydig cells may be due to fluorescence quenching arising from the interaction of the Coumesterol with nuclear components. However, micro-volume time-resolved fluorescence spectroscopy experiments on cell nuclei have revealed the same decay behavior for Coumesterol in both the cytoplasm and nucleus of the cells.

Low density lipoprotein (LDL) and lipoprotein (a) [Lp(a)] were covalently labeled with the fluorescent dyes BODIPY succinimidyl ester (green) or Rhodamine iodoacetamide (red). The interaction of the fluorescent lipoproteins with HepG2 cells was visualized by means of a confocal laser scanning fluorescence microscope operating in the dual wavelength mode. If LDL or Lp(a) were incubated with the cells both lipoproteins bound to the cell surface at 4 degree(s)C or were internalized by the cells at 37 degree(s)C. In all cases larger amounts of LDL interacted with the cells compared with Lp(a). When mixtures of LDL and Lp(a), each labeled with a different dye, were incubated with cells again both lipoproteins bound to the cell surface (4 degree(s)C) or were internalized by the cells (37 degree(s)C). In addition, the major part of the lipoproteins colocalized either on the cell surface or inside the cells. thus, we conclude that interactions of Lp(a) with cells is mediated by LDL, probably via the LDL receptor, to a large extent.

Fluorescence polarization anisotropy can be used to determine the rotational mobility of a fluorescent analog, detect anisotropic orientation distributions, or measure the fluorescence lifetime of a fluorophore. Steady state fluorescence anisotropy can be simply measured in a standard fluorescence microscope equipped with excitation and emission polarizers and therefore, two dimensional maps of fluorescence anisotropy can be easily acquired. We are using steady state fluorescence anisotropy imaging microscopy to study the biochemistry of cell motility. The optimum fluorophore for fluorescence polarization measurements has a fluorescence lifetime that is comparable to the rotational correlation time of the molecule of interest. In order to make imaging measurements with high sensitivity and reasonable time resolution, however, this general rule has to be adjusted, and we have found that FITC-calmodulin has a useful combination of the above features. Calmodulin is a key regulatory protein, that is proposed to be involved in the regulation of the actin-myosin II based force generation in non-muscle cells. The rotational mobility of macromolecules is very sensitive to molecular interactions, yet is relatively insensitive to any surrounding gel matrix. We have taken advantage of this feature to map FITC-calmodulin interactions in the complex cytomatrix in living cells by steady state Fluorescence Anisotropy Imaging Microscopy (FAIM). In addition, we are investigating the use of FAIM for mapping variations in molecular orientation distributions, and fluorescence lifetime distributions.

We have observed the polarization resolved two-photon fluorescence excitation spectra of jet cooled (2 to 8 K) indole. ¹L_a lines are distinguished from ¹L_b lines by their reduced intensity under excitation with circularly polarized light and by their sharp Q branch using linear polarization. Extensive one-photon excitation spectra of complexes of indole with methanol, H₂O, and D₂O have been obtained. The pair of '¹L_a' lines at 455 and 480 cm^-1 shift only a few wavenumbers more than the ¹L_b lines in the type I complexes (where indole is thought to be a H-bond donor) but undergo large intensity redistribution, casting doubt on their assignment as the ¹L_a 'origin'. In the methanol complex I, the 480 cm^-1 line does retain its ¹L_a character (as the origin at -160 cm^-1 from the bare origin also retains all of its ¹L_b character). Previous work on the D₂O complex revealed that the Franck-Condon active intermolecular vibrations involve translation of the water center-of-mass by 0.4 angstroms in the case of the type II ('(pi) ') complex at -450 cm^-1. Except for a few transitions in the complex indole + methanol, type II peaks are not seen at higher energies in the spectra of the complexes due to their low intensity.

We have recently demonstrated that fluorescence emission can be quenched by laser light pulses from modern high-repetition rate lasers, a phenomenon we call 'light quenching'. We now describe some of the possible effects of light quenching on the time- resolved intensity and anisotropy decays of fluorophores. We show that light quenching can decrease or increase the time-zero anisotropy, and that the initial anisotropies can increase to unity under selected conditions. Quenching with time-delayed light pulses is predicted to result in oscillations in the frequency-domain intensity and anisotropy decays. The increasing available and use of pulsed laser sources of light offers the opportunity for a new class of one- two- or multiple-pulse time- resolved experiments.

We studied the steady state and time-resolved fluorescence spectral properties of the DNA stain Hoechst 33342 for one-photon (OPE) and two-photon (TPE) excitation. Hoechst 33342 was found to display a large cross-section for two-photon excitation within the fundamental wavelength range of pyridine 2 and rhodamine 6G dye lasers, 690 to 770 nm and 560 to 630 nm, respectively. The time-resolved measurements show that intensity decays are similar for one- and two-photon excitation. The anisotropy decay measurements of bis-benzimide, 2,5'-bi-1H-benzimidazole, 2'-(4- ethoxphenyl)-5-(4-methyl-1-piperazinyl) (HOECHST 33342) in ethanol revealed the same correlation times for two-photon excitation as observed for one-photon excitation. However, the zero-time anisotropies recovered from anisotropy decay measurements are 1.4-fold higher for two-photon excitation than for one-photon excitation. The anisotropy spectra of Hoechst 33342 was examined in glycerol at -20 degree(s)C, revealing limiting values close to the theoretical limits for one-photon (0.4) and two-photon (0.57) excitation. The steady-state anisotropy for one-photon excitation decreases in the shorter wavelength region (R6G dye laser, 280 to 315 nm), but the two-photon anisotropy for 560 to 630 nm excitation remains as high as in the long- wavelength region (690 to 770 nm). This result suggests that one- photon absorption is due to two electronic transitions, but only one transition contributes to the two-photon absorption over the wavelength range from 580 to 770 nm.

Complexation of the extracellular domain of tissue factor to the serine protease factor VIIa is a critical step in the process of blood coagulation following tissue damage. To study the structure and function of the extracellular domain of tissue factor (soluble tissue factor, sTF), we have used site-directed mutagenesis to replace each of sTF's four tryptophans (Trp) with phenylalanine (Phe) or tyrosine (Tyr). Replacement of any one of the four Trps reduced the protein stability against denaturation by guanidinium chloride in a similar manner, indicating that each residue has important structural interactions within the protein. Replacement of Trps 25, 45, and 158 resulted in reduced cofactor activities, indicating that these residues are located in regions important for biological activity. The activities of mutants with Trp 14 or both Trps 14 and 158 replaced were comparable to sTF. From the combination of absorbance and fluorescence spectra of the individual Trps, information is obtained showing that all the Trps are buried in the protein matrix, and Trps 14 and 25 are in highly constrained environments compared to Trps 45 and 158. To directly monitor interactions of sTF with factor VIIa and its substrate factor X, we have undertaken a program to generate spectrally enhanced protein (SEP) analogs of sTF and the sTF Trp mutants by in vivo incorporation of Trp analogs with absorbance and fluorescence distinct from Trp. Attempts to incorporate the Trp analogs 5-hydroxytryptophan (5-OHTrp) and 7-azatryptophan (7- ATrp) into sTF have provided further information on the structural significance of the Trp residues in sTF.

In vivo proteins fold to form one active structure in minutes or seconds, ruling out the possibility that a polypeptide samples all possible conformational space during folding. We have used site directed mutagenesis to produce 15 single tryptophan containing mutants of Bacillus stearothermophilus lactate dehydrogenase (BS LDH) thus enabling the equilibria of unfolding to be seen from 15 defined positions. These mutant versions of BS LDH have the same X-ray structure as the wild type protein⁸. Previously Smith et al.¹¹ had detected and assigned structures to 4 folding states. The first intermediate, a monomer with secondary and super secondary structure largely intact, is formed after the dimer dissociates at 0.55 M guanidinium hydrochloride (GuHCl). The second intermediate on the unfolding pathway is stable at 2.2 M GuHCl. It had been assumed previously that the transition from this molten-globule structure to the fully denatured form occurred as a single process. We have now identified a core folding motif. In this, helix (alpha) -1F forms a helix-sheet interaction with (beta) -K and (beta) -K has interactions with both (alpha) -2G and (alpha) -3G. This super secondary interaction forms the most stable folding motif in BS LDH and is lost at 2.8 M GuHCl leaving helix (alpha) -1F, (alpha) -2G, and (alpha) -3G which are stable until 3 M GuHCl.

The fluorescence properties of tryptophan derivatives in dioctyl sodium sulfosuccinate (AOT)/iso-octane/buffer reverse micelles were studied using the intrinsic fluorescence of the indole group. The fluorescence decay is more complex for both N-acetyl- 1-tryptophan-amide (NATA) and Human Serum Albumin (HSA) in AOT reverse micelles than in aqueous solution. Time-resolved anisotropy and fluorescence quenching studies using carbon tetrachloride suggest one species of NATA lies on the internal micellar interface and another lies embedded in the interfacial region. The fluorescence decay of HSA in AOT reverse micelles has three unchanging exponential components over the (omega) ₀ (the ratio of the concentration of water to AOT) range 9 to 51, indicating the environment of the tryptophan residue does not change in this range of waterpool size. Fluorescence quenching experiments of HSA in reverse micelles using acrylamide and carbon tetrachloride show that, like NATA, HSA also lies in the interfacial region. There is a minimum in the static component of quenching by CCL₄ of HSA in reverse micelles at (omega) - 0)$AP21. This may be due to conformational stability around this waterpool size, and providing this is not an HSA-specific effect, may correlate with the enhancement of enzyme activity often observed in reverse micelles at a particular waterpool size.

The spectral overlap of tryptophan containing proteins and DNA has limited the use of luminescence methods to investigate protein-nucleic acid interactions. However, the steady-state and time-dependent phosphorescence of wild-type and 5-hydroxy-L-tryptophan (5-O1-ITrp) X ci repressor are spectroscopically distinct such that we can selectively excite the 5-OHTrp-?. ci in the presence of DNA, or even in the presence of a 15-fold molar excess of N-acetyl-tryptophanamide (NATrpA). The phosphorescence of wild-type X ci is red-shifted by 3 nm relative to NATrpA, characteristic of buried tryptophan, and the phosphorescence of the spectrally enhanced protein (SEP), 5-OHTrp-X ci repressor is also red shifted relative to the model, 5-OHTrp, providing spectroscopic evidence that the modified repressor is structurally equivalent to the native repressor. Although the phosphorescence decays of NATrpA and 5-OHTrp are simple exponentials, the decay of either wild-type or 5-OHTrp-X ci repressors requires three exponentials whose fractional contributions to the phosphorescence are similar. Since the 5-OFITrp phosphorescence can be excited without interference from tryptophan or DNA, we measured the phosphorescence of the SEP/DNA complex. Th emission characteristics of SEP alone and the SEP/DNA complex are indistinguishable, showing that during binding, the C-terminal domain of the protein, believed to be involved in protein dimer stabilization, is structurally conserved in the vicinity of the three modified trytptophans.

Mesoporphyrin IX was used as a fluorescent analogue of heme in cytochrome c peroxidase (CcP). Details of the fluorescence spectra of CcP obtained under conditions of energy selection revealed interactions of the porphyrin with the heme pocket. It was shown that the energy of a 0,0 transition shifted with pH in parallel with changes in the electric field of the protein.

The effect of solvent in the first two (pi) ^*$IMP(pi) excited states of indole and 3-methylindole, (¹L_a and ¹L_b), has been studied using a hybrid theoretical method that couples molecular dynamics and a semiempirical molecular orbital procedure. This method yields information about the mechanism and the time scales involved in bulk solvent reorganization after excitation of the indole chromophore to the ¹L_a, (or ¹L_b), state, as well as the solvent induced inhomogeneous broadening. The fluorescence red shifts predicted in several solvents, (water, methanol, butanol, and dimethyl ether), agree reasonably well with experimental values. These time resolved calculations also predict inversion of the excited states with respect to vacuum and indicate that the solvent relaxation has two components: the first one is inertial in character, with a Gaussian shape and a half-width at half-maximum of approximately 15 fs for water, and of 100 to 300 fs for the other solvents. The second component shows an exponential decay behavior and seems related to the longitudinal relaxation time of the solvent. The correlation times for this component are approximately 170 fs for water and a few picoseconds for the other solvents. Calculations performed in butanol at 0 K indicate that we can expect a fast, (1 to 5 ps), red shift in the fluorescence, as large as 1000 to 2000 cm^-1, even when the solvent is rigid, because the inertial response is still possible in this environment. Absorption shifts are not calculated well using this method because it does not take into account the electronic polarizability of the solvent, which would be responsible for half of the absorption shift.

Fluorescence lifetime measurements can be made on the decay of individual 'frozen' conformers of tryptophan analogs in supersonic gas expansions. These measurements have shown differences in single exponential lifetimes among conformers of a given analog, results which are strongly consistent with the presence of conformation-dependent charge transfer quenching. For jet cooled tryptophan, however, one of the conformers emits from ¹L_a, thereby complicating any interpretation of the results. To remove this problem we considered tryptophan analogs with C-5 substituents, in which all the jet cooled conformers emit from ¹L_b. We observed differences in conformer lifetimes similar to cases considered earlier. In the course of this work we discovered that, in contrast to tryptophan, 5- methoxytryptophan shows single exponential decay in solution. Based on additional data for jet cooled samples, we propose the most likely explanation to be substituent induced changes in conformer geometries and populations.

There are two dominant intramolecular quenching reactions for the deactivation of the excited singlet state of tryptophan and its related chromophores. These mechanisms are charge transfer to an electrophilic side chain group and proton transfer from a side chain to the indole ring. When occurring in D₂O, the photoinduced proton transfer reaction results in an exchange of the hydrogen isotope at position 4 of the benzenoid ring of indole, which can be quantitated by ¹H-NMR studies. We have determined the rate constant for this photoinduced H-D exchange for tryptophan and some of its analogs as a function of temperature. We show that the proton transfer quenching reaction makes a large contribution to the quenching of tryptophan, whereas for other chromophores, charge transfer quenching must dominate.

In bulk water at neutral pH, a bimodal excited state lifetime distribution characterizes the fluorescence decay of tryptophan (Trp) in the blue edge of the fluorescence emission spectrum (centers 'ri =0.58 ns, t2 = 3.15 ns, proportions: cq= 0.33, a = 0.67 at 320 nm). A monomodal distribution is found in the red edge (center: t= 3.18 ns at 390 nm) as described in the litterature. The decay of N-acetyl-tryptophanamide (NATA) is also biexponential at short emission wavelength with similar decay values as Trp but with different proportions (centers: ti =0.61±0.13 ns, t2 = 3.01±0.05 ns, proportions: cq= 0.05, a = 0.95 at 320 nm) and monomodal at long emission wavelength (center: t = 3.00±0.05 ns at 390 nm). No evidences for a fast build-up at long emission wavelength can be observed. The fluorescence decays of indole, N-methyl indole and NATA in ethanol are described in the blue edge of the emission spectrum by the existence of a bimothl distribution of time constants (centers: t—2O30 ps and t2 3.7-4.3 ns) and in the red edge of the fluorescence emission spectrum, by a bimodal distribution with the same centers as in the blue edge and identical preexponential values but opposite in sign. In isobutanol, the rising time value increased to —90 ps. The Trp fluorescence decay in ethanol shows three separated time constant populations in the blue edge of the emission spectrum (centers ti =20 p5; 'c = 0.43 ns and t3 = 2.47 ns). In the red edge, a fast rising of the fluorescence (center: —70 ps) is observed, together with a long decaying component (2.67 ns) and a shorter one (0.46 ns). For NATA in reverse micelles, similar observations are obtained in the red edge with a rising time of —30 ps, a value independent of the degree of immobilization of the water molecules. These observations support the existence of excited state reactions involving interactions with the solvant molecules (exciplex formation or solvent-cage relaxation ). The example of annexin V is proposed to illustrate the case of a protein: the data support the existence of a fast relaxation process (—30 ps) observed at low temperature for the lone buried tryptophan residue. According to the tridimensionnal structure of the protein, this can be interpreted as resulting from the existence of a H-bonding involving the E-OH group of a threonine residue (Thr224) and the indole nitrogen.

This report gives evidence that the biexponential fluorescence decay of tryptophan zwitterion in H₂O solution is due to the occurrence of a reversible two-state excited-state process whereby the corresponding ground-state species are excited. The rate constants are within the intervals: 0<k₀₁<0.57(ns)^-1, 0.76(ns)^-1<k₂₁<1.33(ns)^-1, 0<k₀₂<0.58(ns)^-1, 0.77(ns)^-1<k₁₂<1.35(ns)^-1. These limits were calculated using the values for S₁ equals k₀₁ + k₂₁ (1.33+/- 0.01(ns)^-1), S₂ equals k₀₂ + k₁₂(1.35+/- 0.01(ns)^-1), and P equals k₂₁k₁₂(1.03+/- 0.01(ns)^-2). The emission spectra of the two excited-state species can be uniquely determined and are different from those associated with the decay times. These results were obtained by repetitive global compartmental analyses of the fluorescence decay surface of tryptophan zwitterion measured over the entire emission spectrum as a function of quencher concentration. This new and powerful analysis method is applicable to all biexponential protein fluorescence decays.

Frequency-domain fluorometry was used to investigate the time- dependent intensity decays of N-acetyl-L-tryptophanamide (NATA) when collisionally quenched by acrylamide and iodide in propylene glycol at 20 degree(s)C. The intensity decays of NATA became increasingly heterogeneous in the presence of quenching, and the steady-state Stern-Volmer plots show significant upward curvature. These frequency-domain and steady-state data clearly indicate that the rate constants for quenching of NATA by acrylamide and iodide depend exponentially on the fluorophore- quencher separation distance. The NATA intensity decays were not consistent with the Smoluchowski-Collins-Kimball radiation boundary condition for quenching. The upward curvature of the Stern-Volmer plots provide a sensitive method to determine the characteristic distances for fluorophore-quencher interaction. The rate constant for quenching are discussed based on the mechanism of interaction between tryptophanyl fluorophore and quencher molecules which appear to involve electron transfer for acrylamide and heavy atom effects for iodide. These results suggest that the distance-dependent interaction and the effects of solvent dynamics need to be considered in the interpretations of data for quenching of proteins.

At present there is a great deal of interest in the study of the transference of energy in biological systems. For example, electron transfer is of major importance in many synthetic and biological processes and in nature is mediated by proteins. Information regarding this process is therefore useful in leading to a greater understanding of phenomena such as photosynthesis and respiration. Previous work on protein systems has shown the electron transfer process to be complex to analyze because of the presence of competing pathways. This has led to the use of model systems to simplify the kinetics. We have synthesized novel model systems using peptides containing both a fluorescent methoxy- naphthalene donor and a dicyanoethylene group as a potential electron acceptor and observed fluorescence quenching for both dipeptide and oligopeptide systems. Biexponential fluorescence decay behavior was observed for all donor acceptor systems, with an increase in the amount of the shorter fluorescence decay component on increasing temperature.

In order to study experimentally the ultrafast (<1 picosecond to several hundreds of picoseconds) molecular dynamics of protein-protein interactions, an optical probe is required. Tryptophan has been the most widely used intrinsic optical probe of protein structure and dynamics. There are, however, two major problems attendant to the use of tryptophan, especially in fluorescence measurements. First, since tryptophan is a naturally-occurring amino acid there are often several tryptophans in a protein molecule whose emission must be distinguished. Second, the fluorescence decay of tryptophan itself in aqueous solution is nonexponential. We have consequently investigated alternatives to tryptophan. Our work has led us to the amino acid analog, 7-azatryptophan, and its chromophoric moiety, 7-azaindole.

We have used steady-state measurements of polarized emission to determine the orientation of a xanthene probe covalently attached to cys374 in F-actin oriented by flow. The probe orientation was described in terms of the angular disorder of the filament axis about the laboratory vertical ((Theta) ) and the angle between the emission dipole and the long filament axis ((phi) ). Measurements of the prompt fluorescence from erythrosin indicated that (Theta) equals 16+/- 1 degree(s) and (phi) equals 42.1+/- 0.2 degree(s) at 20 degree(s)C while similar analyses of the delayed fluorescence gave (Theta) equals 6.4+/- 0.1 degree(s) and (Phi) equals 48.9+/- 0.1 degree(s). The probe dipole orientation ((Phi) ) was not affected by the binding of the fungal toxin phalloidin, the thin filament regulatory protein tropomyosin, or isolated myosin heads (S1) with or without nucleotide, indicating that changes in the filament anisotropy due to interaction with these ligands reflect changes in filament rotational motion. Although the angular disorder of the filaments ((Theta) ) was not influenced by ligand binding on the prompt fluorescence time scale ((tau) f is approximately 0.2 ns), it was modulated on the delayed fluorescence time scale ((tau) p is approximately 275 microsecond(s) ); the latter changes appeared to correlate with isotropic measurements of filament dynamics that indicated that phalloidin, tropomyosin, or S1 binding increased the rigidity of F-actin.

Four-way branched DNA structures (Holliday junctions) are the key intermediates formed during genetic recombination between two duplex DNA molecules. The outcome of recombination at specific DNA sequences may depend on the 3D structure at the junction site. We have used time-resolved fluorescence resonance energy transfer methods to measure interarm distance distributions in a series of related four-way DNA junctions. The aim of these studies is to establish how the global structure and flexibility of the junction are influenced by the base sequence flanking the branch point. Energy transfer donor and acceptor dyes were conjugated to the 5' termini of the DNA strands via a C₆ linker. Six pairwise combinations of labeled arms have been examined for each junction, yielding six interarm distance distributions. All the junctions we examined formed `X' shape structures by pairwise stacking of duplex arms, although there were significant differences in the range of interarm distances, depending on the junction sequence. Certain sequences appear to produce a mixture of the two possible stacking arrangements. These studies help to define the molecular interactions that dictate the crossover bias at a specific junction sequence.

7-Aminoactinomycin D is a potent inhibitor of RNA polymerase and it has been demonstrated to bind to a single stranded DNA. Wadkins and Jovin showed that there is a 20 to 40 fold enhancement in the fluorescence of 7-aminoactinomycin D upon binding to certain single stranded DNA sequences. We report studies of the thermodynamics of the binding of the parent drug, actinomycin D, to single and double stranded DNA. A five-fold range of association constants is found for the different binding sequences that we have investigated. We will correlate these thermodynamic studies with measurements of the fluorescence enhancement, lifetime, and rotational correlation time of the drug bound to the various single and double stranded oligonucleotides.

Time-resolved fluorescence anisotropy decay measurements are useful for characterizing differences in the local structural environment of probes attached to DNA molecules. We synthesized oligonucleotides with dansyl probes covalently attached to uridine near the center of a 17 base oligonucleotide. The fluorescence properties of these DNAs hybridized to complementary DNA were examined using time-correlated single photon counting. We present preliminary data indicating changes in the probe anisotropy in the presence of trifluoroethanol, divalent zinc cations, and under conditions promoting the formation of a hairpin junction at the site of probe incorporation. The results of these perturbations indicate that this approach will be useful for investigating site-specific and global structural phenomena in DNA.

A 16-mer deoxy oligonucleotide was labeled at the 5' end with x- rhodamine and at the 3' end with fluorescein. The fluorescence lifetime of the donor, fluorescein, under conditions for resonance energy transfer, was studied using the SLM 4850 multiharmonic frequency phase fluorometer in order to obtain information on the end-to-end distance distribution P(R) in the oligomer. When this doubly labeled oligonucleotide was hybridized to its 16-mer complement, the fluorescein fluorescence decay could be very well described by a P(R) that was a symmetric shifted Gaussian with center at 68.4 angstrom and (sigma) equals6.4 angstrom. Simulations suggested that part of the width might be attributable to a distribution in (kappa) ². In the single- stranded labeled oligomer, there was enhanced energy transfer from the fluorescein to the rhodamine and the best fitting symmetrical shifted Gaussian representation of P(R) was centered at 53.8 angstrom with (kappa) equals6.9 angstrom. There was significant lack of fit with this model, however. A model independent procedure was developed for extracting P(R) as a sum of weighted Hermite polynomials. This procedure gave a P(R) with a large negative region at R<20 angstrom, suggesting that rotational averaging for (kappa) ² was not quite complete prior to significant decay of the donor excited state.

Highly hydrophobic peptides in small unilamellar vesicles can be used to model membrane-embedded proteins such as the dopamine D₂ receptor. The transmembrane domains of the dopamine D₂ receptor are known to contain residues corresponding to the binding sites for natural receptor ligands. We have developed a model system consisting of a peptide whose sequence was taken from the transmembrane region of the dopamine D₂ receptor and incorporated it into phospholipid bilayers. This polypeptide sequence, NH₂-D-V-L-Y-S-A-F-T-W-L-G-Y-V-N-S-A-V-N-P-I-I-Y-T- T-F-N-V-CO₂H, contains a single tryptophan residue, whose fluorescence properties provides an intrinsic probe of the microenvironment of the peptide within the bilayer. Purification of this highly hydrophobic peptide required the development of a novel alcohol-based reversed-phase HPLC solvent system. The vesicles were produces by cosonication of the peptide with dimyristoylphosphatidylcholine lipid and were characterized by electron microscopy and fluorescence spectroscopy. Time- correlated single photon counting was sued to measure the fluorescence anisotropy of the system as a function of temperature across the lipid phase transition range and as a function of the peptide/lipid ratio.

We have used phosphorescence anisotropy from erythrosin-5- iodoacetamide covalently attached to rabbit skeletal muscle tropomyosin (Tm) to characterize the molecular dynamics of Tm on the surface of F-actin. Although the probe was loosely bound to Tm, it was able to monitor the molecular dynamics of the F- actin/Tm complex on the microsecond time scale. The steady-state phosphorescence anisotropy of Tm in the complex was 0.025+/- 0.005 at 20 degree(s)C. Since the anisotropy of actin in a similar complex was 0.11+/- 0.01, Tm may be moving independently on the filament surface. Studies of the effect of temperature and solution viscosity provided additional evidence for independent tropomyosin motions. A modified Perrin plot of 1/r versus (tau) T/(eta) (where r is the steady-state phosphorescence anisotropy, (tau) is the phosphorescence lifetime, T is the absolute temperature, and (eta) is solution viscosity) provided evidence for two distinct modes of motion for Tm on F-actin; these modes corresponded to the rotation of segments with significantly different volumes. While the volume of the fast component was comparable to that of a single tropomyosin molecule, providing evidence that Tm moves on the surface of F- actin, the volume of the slow component corresponded to only a small portion of the entire complex.

Theoretical descriptions of excitation energy transfer between chromophores in allophycocyanin are presented, including bilateral energy transfer paths between chromophores, and are expressed, based on Foster interaction mechanism, as Pauli master equations. Group analysis in C₃ symmetry is performed to carry out analytic expressions for fluorescence decays which is generally of triexponential with effects of chromophore coupling and exciton splitting taken account. It is pointed out that the time constant of each decay component contains mixed information of different energy transfer paths, and therefore show its dependence on subtle configuration of chromophores, probably related to site heterogeneity and thus to inhomogeneous broadening previously observed.

We describe a method to improve the resolution of donor-to- acceptor distance distributions in molecules which are flexing on the timescale of the fluorescence lifetime. We measured the time- dependent donor decays of two donor (D)-acceptor (A) pairs, where the donor lifetimes were substantially different. The donors were an indole residue (5.7 ns) and a naphthalene residue (24.4 ns). The same dansyl acceptor was used for both D-A pairs. The donor decays are complex due to both a distribution of D-A distances and D-A diffusion. Using the donor decay data for each D-A pair alone, it is difficult to resolve both the distance distribution and the D-to-A diffusion coefficient. However, these values are unambiguously recovered from global analysis of the data from both D-A pairs. Simulations were also used to demonstrate the increased reduction of global analysis with different lifetime donors to obtain distance distribution parameters in the presence of D-A diffusion.

Single ps pulse spectroscopic techniques have been employed to investigate excited-state dynamics of R-phycoerythrin (R-PE) isolated from two red algae Polysiphonia urceolata and Porphyra yezoensis, respectively, at various excitation photon densities in a range of 8 X 10¹⁴ $AP 1 X 10¹⁷ photons/cm². The fluorescence quantum yields of the R-PE from the two algae are found to decrease in logarithm with increasing of excitation intensity due to S₁-S₁ exciton collision. A fast component in addition to a slow one recorded at low excitation density and a large initial anisotropy are observed at higher excitation density. Different dynamical characteristics of exciton collision between the cases of Polysiphonia urceolata and Porphyra yezoensis are shown and proposed to be owing to different numbers of excited pigments available and chromophore architectrue in R-PE from different biological sources.

Jet-cooled spectroscopy has been carried out on non-sulfonate containing derivatives of the fluorescence probe molecules 1,8 ANS and 2,6 TNS. The absence of the sulfonate groups makes for straightforward sample volatilization without decomposition, yet fundamental features in the excitation spectra should be largely unchanged in ANS/TNS. The analogs studied can be viewed as naphthalenes with significant perturbations due to group substituents, the most important element of which is the amino nitrogen. The ultimate interest of this work will be ANS/TNS solvent complexation experiments, where photophysical changes may be induced by specific solvent-solute interactions. An intermediate point in the project will be analogous complexation experiments with anilinonaphthalenes, where there are indications that the results might mirror the ones seen with ANS/TNS.

Both micellar and lipid membrane systems have been used as models to provide further information regarding peptide properties in biological systems. Peptides are basic architectural units in nature and the study of their properties in membranes and other non-homogeneous media is of fundamental importance. We present the results for a time-resolved and steady state fluorescence study of two 4-methoxy-naphthalene labelled twelve amino acid residue model peptides that we have synthesised. One peptide, with a N-tert-Butoxycarbonyl (BOC) modified N-terminal, was incorporated into small unilamellar vesicles of L-(alpha) dipalmitoylphosphatidycholine (DPPC) and the other, with a free amino group, into inverse micelles of sodium bis (2-ethylhexyl) sulfosuccinate (AOT) in 2,2,4-trimethylpentane. Steady state fluorescence and time-resolved fluorescence lifetime and anisotropy measurements show the first peptide to be situated in the lipid bilayer. In the inverse micelle system there is evidence for the peptide being situated at the surfactant waterpool interface.

The room temperature phosphorescence decay of Horse Liver Alcohol Dehydrogenase (LADH) was analyzed with continuous lifetime distribution models such as the Exponential Series and Maximum Entropy Methods, revealing the existence of a broad distribution of phosphorescence lifetimes. Possibly reflecting the existence of two or more conformational species that do not rapidly interconvert on a time scale shorter than seconds. In order to gain insight into the underlying reason for the lifetime distribution, we performed a series of quenching experiments on LADH phosphorescence. When quenching data is presented in terms of a distribution of decay rate constants (rather than lifetimes) it is easy to show that quenching of the phosphorescence by mechanisms that do not distinguish between protein species will result in a uniform increase in the decay rate constant without affecting the width of the distribution. An example would be a Forster quenching mechanism if the components within the distribution have identical overlap integrals with the energy transfer partner. Conversely, if the species responsible for the distribution have a differential susceptibility to the quencher, and increase in the mean rate constant and a change in the distribution width will occur. Thus, a quencher that diffuses differentially into various protein conformers is expected to cause a change in the width of the phosphorescence distribution. This change in width provides information on the relative efficiency of quenching of conformers. Using a number of quenchers, one may resolve components within the distribution of conformational states by analyzing the dependence of the width of the phosphorescence lifetime distribution on quencher concentration.

Both theoretical and experimental studies are presented on chromophore excited-state coupling in linker-free allophycocyanin (APC), one of the antenna phycobiliproteins in algal photosynthesis. A three-site-coupling model has been introduced to describe the exciton interaction mechanism amoung the excited (beta) chromophore in APC, and the exciton energy splitting is estimated. Picosecond polarized fluorescence experiments both on monomeric and trimeric APC isolated from alga Spirulina platensis have been performed. The experimental results show that APC monomer and trimer exhibit remarkedly different spectropic characteristics, and satisfy the suggestion of strong excited- state coupling among chromophores in APC.

With the advent and development of time-resolved spectroscopic technique and substantial progress in understanding of photophysical and photochemical events in condensed phases, a new goal may be achieved: modeling of biochemical reaction or its elementary step by a photochemical event occurring within the probe, bound to protein molecule. The probe may be located in a well-determined site of the protein matrix and report on the modulation of the reaction rate by the matrix and by the surrounding solvent, or by interactions in multiprotein complexes and in biomembranes. The advantage of this approach is obvious: in contrast to ordinary biochemical reaction, the excited-state reaction may be started by a short light pulse, and its development may be observed directly with high resolution in time. If the reaction rate is influenced by the dynamics of the protein matrix, these dynamics may be studied simultaneously with the reaction, by the same or similar probe, and within the same time range. In this report we discuss shortly the prospects for application of probes exhibiting electron transfer, twisted intramolecular charge transfer, and isomerizations. In the studies of electron-transfer dynamics of bianthryl in the complex with albumin and anti-bianthryl antibodies we show that the reaction is controlled by the dynamics of protein matrix which is found to be slower than nanosecond. The general problem of photochemical modeling of biochemical reactions is discussed.

Nonsymmetric cationic polymethine dyes (tricarbocyanines) demonstrate dual emission which depends upon the dynamical properties of the solvent. In this study the symmetric and nonsymmetric polymethine dyes were investigated comparatively by light-absorption and fluorescence spectroscopy. In liquid solvents the narrow emission band at 758 to 762 nm is observed only, while in solid media the major emission component is at 720 to 740 nm, which depends strongly upon the solvent and the excitation wavelength. Both components are observed in the intermediate case of highly viscous media. These features become more significant with the increase of electronic asymmetry of the dye. The mechanism involving cis-trans isomerization in the exited dye molecule and reorganization of its solvate shell is proposed for the explanation of this phenomenon. The viscosity- dependent dual emission and convenient spectral range, which does not interfere with the spectra of cellular pigments, makes these dyes very promising in the studies of microscopic viscosity of biomembranes and the whole cells.

We synthesized two conjugatable long-wavelength fluorescence probes. They consist of a squaraine moiety, which is a cyanine- type chromophore with a central squarate bridge and a reactive N- hydroxysuccinimide group for coupling with amino functions. One form is water soluble due to the presence of a sulfobutyl group, the other is water insoluble. The water insoluble form was reacted with taurine to achieve water solubility and this squaraine-taurine conjugate displayed a very high binding affinity to BSA. The squaraines exhibit desirable properties of short lifetimes and low quantum yields in water, with a significant increase of lifetime and quantum yields when bound to proteins. Their absorption maxima around 635 nm in water and 640 nm when bound to proteins allow excitation with the newly commercially available diode lasers sources at 635, 645, and 650 nm. The spectral properties and photostabilities of the water soluble squaraine probes are compared with those of the commercially available CY5-NHS-ester.

The accurate determination of binding constants for protein:DNA complexes is highly desirable. In the biological sciences such binding constants are often determined by gel methods that may introduce severe interactions and therefore might strongly influence the value of the physical constants. Optical methods that are also very often used are limited in protein:DNA complexes because of the overlap of the DNA and the protein absorption. To be able to use absorption or fluorescence spectroscopy it is therefore necessary to introduce labels either on the protein, the DNA, or on both. In this study an approach based on time-resolved fluorescence spectroscopy is used to determine the binding of the antennapedia homeodomain from drosophila to its operator sequence. The increase of the anisotropy decay time of fluorescently labeled DNA upon protein binding can be used to monitor binding. Oligonucleotides which include a fluorescent base analog or which are fluorescently labeled at the 5' end can be used for this purpose.

Absorption and fluorescence spectra of the DNA duplex decamer d[CTGA(2AP)TTCAG]₂ have been measured as a function of temperature. Emission is dominated by the 2-aminopurine (2AP) base. A fluorescence excitation band near 260 to 270 nm disappears near the duplex melting temperature but reappears at higher temperatures, indicating singlet-singlet energy transfer occurs between normal DNA bases and the 2AP base in the B-helical conformation and to a lesser extent in the structurally-mobile melted conformation. The efficiency of transfer is 4 to 5% at 4 degree(s)C under the assumption of a single acceptor; preferential transfer to a low-fluorescence conformation of 2AP would imply no more than a two-times greater transfer efficiency. Emission spectra excited at 260 nm (energy transfer band), 330 nm (basepaired conformation of 2AP) and 305 nm (nonpaired 2AP conformation) appear to indicate transfer occurs from normal DNA bases to both 2AP conformers.

Arachidonic acid (AA) samples and mixtures of arachidonic acid - cholesterol (AA-C) were manufactured and subjected to external electric cw or pulsatory optical laser fields. The polarizing microscope showed a pseudohomeotropic smectic C texture for AA at 10 degree(s)C and smectic C textures with fibers for mixtures AA-C at room temperature. The dependences I equals I(U) at t degree(s)C equals const. and I equals I(t) at U equals const., t being time, showed that samples are dielectrics with a weak conduction (I approximately equals 10^-9 A, U equals 0 - 40V). Pure acid samples present an electric hysteresis which disappears when cholesterol is added. The current occurred when the electric field is applied, decreases in time, and tends to a constant value. When the temperature was 19 degree(s)C, a cw focussed laser radiation from a He-Ne laser ((lambda) equals 632.8 nm, P equals 20 mW) determined a lenselike effect in AA, but no effect in mixtures. At the same temperature, the samples were crossed by some green pulses ((lambda) equals 532 nm, (tau) equals 200 microsecond(s) ) furnished by an experimental setup containing a Nd³⁺ glass laser and a KDP crystal for the second harmonic generation. The shape and extension of the pulse in the time domain, displayed on a sampling oscilloscope, are some indications of the cholesterol percentage.

Chlorophyll a (Chl) in most model systems (monolayers, fluid solvents, adsorbed layers, and polymer films) occurs in three forms: `dry monomers' (isolated from interaction with water), hydrated dimers and oligomers built from such dimers; but in nematic liquid crystal (LC) cell `dry monomers' are predominant. It is a competition between the various paths of deactivation of excited Chl. Excitation energy can be emitted as fluorescence, or delayed luminescence, transferred to other forms of Chl (ET), thermally deactivated or used for photochemical reactions. In order to compare the efficiency of these various paths the following measurements were done and analyzed: absorption, fluorescence, fluorescence excitation, photoacoustic spectra, and lifetime of fluorescence in ps range. There are some important differences between Chl in LC cell and polymer films: Chl in LC cell has a much lower concentration of aggregated forms and the pigment molecules are more uniformly distributed as compared to the Chl in polymer samples. To explain ET in polymer films the fractals model has to be used, whereas mean distances between solvated Chl molecules in LC can be obtained from dye concentration. In order to establish the interaction between Chl a and (beta) -carotene the LC cell with both pigments were also investigated.

Incorporation of 2-aminopurine (2AP) for adenine in DNA gives an optical probe of local and global DNA conformation, since 2AP does not seriously disturb the B-DNA structure. The temperature dependence of the absorption of the double-stranded d[CTGA[2AP]TTCAG]₂ DNA decamer shows that the helix melts in an all-or-none fashion. Absorbance at wavelengths of 260 nm and 330 nm monitor, respectively, the `average' normal base conformation and the 2AP base local conformation. The 2AP base locally melts at a temperature 0.5 degree(s)C or less below the other bases. Temperature dependence of the longest fluorescence decay time of the decamer also shows the melting transition, with the transition curve shifted a few degrees higher. Temperature dependent fluorescence spectral shifts, on the other hand, suggest the environment of the 2AP base changes about 10 degree(s)C below the primary melting temperature. The data suggest a premelting transition purely dynamic in nature--transient 2AP exposure to water increases but the average conformation remains B helical. A summary of conformational information extractable from 2AP fluorescence measurements is presented.

Time-resolved fluorescence measurements of optical probes incorporated at specific sites in DNA provides a new approach to studies of DNA structure and DNA:protein interactions. This approach can be used to study complex multi-state behavior, such as the folding of DNA into alternative higher order structures or the transfer of DNA between multiple binding sites on a protein. In this study, fluorescence anisotropy decay of an internal dansyl probe attached to 17/27-mer oligonucleotides was used to monitor the distribution of DNA 3' termini bound at either the polymerase of 3' to 5' exonuclease sites of the Klenow fragment of DNA polymerase I. Partitioning of the primer terminus between the two active sites of the enzyme resulted in a heterogeneous probe environment, reflected in the associative behavior of the fluorescence anisotropy decay. Analysis of the anisotropy decay with a two state model of solvent-exposed and protein-associated dansyl probes was used to determine the fraction of DNA bound at each site. We examined complexes of Klenow fragment with DNAs containing various base mismatches. Single mismatches at the primer terminus caused a 3-fold increase in the equilibrium partitioning of DNA into the exonuclease site, while two or more consecutive G:G mismatches caused the DNA to bind exclusively at the exonuclease site, with a partitioning constant at least 250- fold greater than that of the corresponding matched DNA sequence. Internal single mismatches located up to four bases from the primer terminus produced larger effects than the same mismatch at the primer terminus. These results provide insight into the recognition mechanisms that enable DNA polymerases to proofread misincorporated bases during DNA replication.

We have studied the steady-state fluorescence properties of the polynucleotides poly(dA)poly(dT) and poly(dT) as well as of the 20-mers (dA)₂₀(dT)₂₀ and (dT)₂₀ at room temperature for excitation at 293 nm where T is selectively excited. The fluorescence spectrum of the double-stranded 20-mer is shifted to longer wavelengths relative to that of poly(dA)poly(dT), which may be the result of reduced stacking interactions. The fluorescence spectrum of the single-stranded 20-mer is much broader than that of poly(dT); this casts doubt on the validity of the assumption made in the literature that the crystallographic results from a dimer can be used to develop a model for poly(dT). The fluorescence anisotropies of the double- stranded systems are unexpectedly lower than those of the single- stranded ones, which suggests that cooperative interactions operate in the dynamics of the former. The analysis of the results of these systems was considerably simplified because of the fact that T can be selectively excited at the long-wavelength edge of the absorption spectrum. For the more general case for which such a spectral window does not exist, we have developed a methodology that employs a global nonlinear least-squares analysis of the dependence of the fluorescence intensity on the polarization angle and have estimated, for a simulated mixture of the C and G bases, their individual fluorescence spectra in the presence of energy transfer from C to G as well as from G to C.

Complexes of methylene blue with DNA are characterized by time- resolved fluorescence spectroscopy and transient photobleaching methods. At least four, and probably five, spectroscopically distinct binding sites have been identified. Three of these (components 1, 2, and 3B) dominate the fluorescence decay at low ionic strength and have fluorescence lifetimes significantly different from that of the free dye. With increasing ionic strength a fourth component (3A) appears at the expense of components 1 and 3B. Component 3A exhibits two subcomponents with different degrees of shielding from O₂ quenching of its triplet state. The relative amplitudes of the components at low ionic strength are strongly dependent on the composition of the DNA, and independent of superhelix density. Hence, it is inferred that components 1, 2, and 3B represent binding to different base pair steps, and that all of these components represent intercalation sites that unwind the DNA to the same degree. Component 3A is apparently not intercalated. From plots of the relative photobleach amplitudes versus the relative fluorescence intensities, we infer that the triplet yield and photobleach amplitude are dominated by components 3A and/or 3B under nearly all conditions. Our results are used to discuss the suitability of methylene blue as the extrinsic probe in transient photodichroism experiments.

Regulation of transcription involves a complex interplay between protein-ligand, protein-DNA, and protein-protein interactions. Fluorescence probes seem to be very well suited to study such complex systems since the selectivity and sensitivity of fluorescence makes possible to select only a part of the system for observation leaving the rest of it transparent to the technique. We have used fluorescence spectroscopy to study the activation of E.coli RNA polymerase by cAMP receptor protein (CRP). The cAMP interactions with CRP, domain flexibility in CRP molecule, the structure of CRP-DNA complex, and interaction of CRP with RNA-polymerase have been studied. Here we report the preparation and properties of 5-OH-Trp derivative of the sigma subunit of E.coli RNA polymerase. This subunit is responsible for specific promoter recognition. The obtained results show that the biological activities of the derivative are identical as observed for the native protein. Comparison of fluorescence properties of the 5-OH-Trp sigma derivative free and bound to the core RNA polymerase suggests a conformational change in the sigma protein induced by this interaction. These data show that replacement of Trp residues with 5-OH-Trp can be a very useful approach to prepare specific fluorescence derivatives of multimeric proteins.

The intercellular lipid lamellae of mammalian stratum corneum (SC) constitute the major barrier to percutaneous penetration of drugs and other solute molecules. In order to understand the barrier property of skin on a molecular level, we have initiated fluorescence spectroscopic investigation of the membranous structures of the SC and related model systems using the lipophilic probe 1,6-diphenyl-1,3,5-hexatriene (DPH). Incorporated into distearoylphosphatidylcholine and stratum corneum bilayers, DPH fluorescence reflected the change in lipid structure under thermal and chemical perturbations. Using a multiharmonic frequency approach, we measured the fluorescence lifetime and rotational correlation times for DPH in these systems. Our data indicated that a biexponential decay ((tau) ₁ approximately equals 9 ns, (tau) ₂ approximately equals 1.5 ns) described the intensity decay, while a hindered rotor model ((phi) approximately equals 5 ns, r_(infinity ) approximately equals 0.3) described the anisotropy decay. These parameters reported the known thermotropic phase transition in porcine stratum corneum, and the influence of the penetration enhancer oleic acid in human epidermis. Thus, we have shown frequency- domain fluorescence spectroscopy to be a facile and powerful tool for monitoring the permeability of a solid tissue such as the SC.

The motional properties of DPH labelled diacylglycerol (DG) in vesicles have been investigated in the absence and presence of its biological target: protein kinase C (PKC). In the absence of PKC the extent of ordering and rotational dynamics of DPH-DG turned out to be considerably different from those of DPH labelled phosphatidylcholine (DPH-PC). When DPH-DG was dispersed in membranes containing 10 mole % of phosphatidylserine (PS), addition of PKC led to an immobilization as judged from a slower fluorescence anisotropy decay. This effect was not seen when PS was replaced by PC or in the absence of calcium indicating that negatively charged lipids and calcium are required for interaction between PKC and DPH-DG. Furthermore, the specificity of the interaction of PKC with DPH-DG was compared with that of the control choline lipid DPH-PC.

Recent experimental results from this laboratory have shown that fluorescence emission can be quenched by the light pulses from cavity-dumped dye lasers, a phenomenon we call `light quenching.' In this overview article we describe some of the possible effects of light quenching on the steady state and time-resolved spectral properties of fluorophores. The extent of light quenching was found to depend on the amplitude of the emission spectrum at the quenching wavelength. Different effects are expected and were observed for light quenching by a single laser beam (within a single laser pulse) or for a time-delayed quenching pulse. Light quenching can decrease or increase the time-zero anisotropy. Our calculations indicate that the anisotropies can increase to unity under selected conditions. Remarkably, the light quenching can break the usual z-axis symmetry of the excited state population, and the measured anisotropy (or polarization) depends upon whether the observation axis is parallel or perpendicular to the propagation direction of the light quenching beam. Quenching with time-delayed light pulses is predicted to result in oscillations in the frequency-domain intensity and anisotropy decays. A more complete description of `One and Two-Pulse Theory of Light Quenching,' is presented by Kusba et al. in this same volume. The predicted oscillations have been observed in frequency-domain intensity decay data. Overall, the results suggest a new class of two-pulse or multiple-pulse time-resolved experiments where the sample is prepared by the excitation pulse and subsequent modification of the excited state population by the quenching pulse(s), followed by time- or frequency-domain measurements of the resulting emission.

New fluorescent dyes with characteristic fluorescence lifetimes have been developed for bioanalytical applications. Based on the concept of 'multiplex dyes', we have designed several rhodamine dyes with nearly identical absorption and emission spectral characteristics but different fluorescence lifetimes. In order to influence the excited state lifetime without changing the spectral characteristics we modified rhodamines with non- conjugated substituents that promote non-radiative transitions. First investigations with covalently coupled biomolecules show the potential of multiplex dyes in DNA sequencing and antigen detection. The biomolecules are identified through the intrinsic fluorescence lifetime of the dye. This principle offers the possibility to make use of different fluorescence lifetimes in each wavelength region. Therefore the number of discernable tags is greatly enhanced. Extending this principle to applications with laser diodes, new rhodamines with functional groups for covalent coupling of analytes have been developed. The new labels exhibit absorption and emission beyond 600 nm and have a high fluorescence quantum efficiency, even in aqueous buffer systems. Time-resolved and intensity measurements of the dyes covalently coupled to a synthetic oligonucleotide are presented. The results obtained in different capillary systems using laser diodes as excitation sources show the potential of these dyes in the red region of the spectrum.

Distance measurements using fluorescence resonance energy transfer (FRET) have been extensively applied to structural studies of proteins and protein:protein complexes. Only in a few instances FRET has been used for distance determination in oligonucleotides or oligonucleotide:protein complexes. In most of these studies donor acceptor systems with labels attached to the 5' ends of the individual oligonucleotide strands have been used. Since long flexible linkers are used to attach the dyes to the phosphate backbone of the oligonucleotide it is advantageous to analyze the experimental data in terms of a distribution of donor acceptor distances instead of a single distance. However, the length and the high flexibility of the linker arms result in broad distributions unsatisfactory for high precision distance determinations. In this study intrinsic labels like 2-aminopurine or dimethyllumazine are utilized instead. These fluorescent base analogs are incorporated into the DNA oligonucleotides and their viability for FRET measurements is tested. Time-domain measurements on oligonucleotides with one intrinsic label and one extrinsic label show that the recovered distance distributions are narrower. This improvement may even be enhanced by using intrinsic probes both as donor and as acceptor.

The fluorescent indicators Fura-2 and PBFI are widely used for the determination of intracellular concentrations of Ca²⁺ and K⁺, respectively. To investigate the complex forming reaction between Fura-2 and Ca²⁺, and between PBFI and K⁺ in the ground and excited states, steady-state and time-resolved measurements were performed. The fluorescence decay surfaces were analyzed with global compartmental analysis yielding the following values for the rate constants at room temperature in aqueous solution: (1) for Fura-2: k₀₁ equals 1.2 X 10⁹ s^-1, k₂₁ equals 1.0 X 10¹¹ M^-1x^-1, k₀₂ equals 5.5 X 10⁸s^-1, k₁₂ equals 2.2 X 10⁷s^-1 (2) for PBFI: k₀₁ equals 1.1 X 10⁹s^-1, k₂₁ equals 2.7 X 10⁸M^-1s^-1, k₀₂ equals 1.8 X 10⁹s^-1, k₁₂ equals 1.4 X 10⁹s^-1 k₀₁ and k₀₂ denote the deactivation rate constants of the free and bound forms of the indicator, respectively k₂₁ represents the bimolecular rate constant of binding of the cation by the indicator whereas k₁₂ is the rate constant of dissociation of the cation:indicator complex. For both probes the effect of the excited-state reaction can be neglected in the determination of K_d and/or the ion concentration.

Heterodyning of modulated pulses in flow cytometry in the PMT offers three major advantages over external homodyning techniques: first, intermodulation distortion is reduced; second, the ability to choose a lower intermediate frequency for evaluating phase affords the opportunity to use digital signal processing techniques to measure phase; third, the limiters we used previously in homodyning phase measurement circuitry may be removed, thereby reducing phase distortion. We constructed a heterodyning PMT base capable of providing an intermediate frequency of 1 MHz, and installed it on our phase flow cytometer (PFC). The fluorescence lifetime of 2 micrometers Fluoresbrite yellow beads was approximately 3 to 4 ns on the heterodyning system, which compares well to the 3.5 ns measured with the original homodyning PFC electronics. This lifetime parameter, when coupled with recently-developed lifetime-based probes for Ca²⁺ and pH, provides new opportunities in flow cytometry for study of intensity-independent intracellular physiology.

The time-correlated single photon counting technique in combination with ultrashort pulsed laser excitation is capable to determine picosecond fluorescence relaxation times provided that the data are accurately deconvoluted from the finite instrumental response which is typically tens of picoseconds. Knowledge of very rapid relaxation times in proteins is required to obtain insight into the role of structural fluctuations in protein function and to have an experimental tool to compare molecular dynamics simulations of proteins. We have found a reference compound having a short single fluorescence lifetime of 12 ps which is very useful for picosecond-resolved protein fluorescence spectroscopy. A description of some experimental pitfalls is given and the reference convolution method is illustrated with a few examples using N-acetyl-L-tryptophanamide.

Near-field fluorescence microscopy combines nanometer resolution with the ability to conduct optical measurements with high sensitivity. To illustrate the potential of the near-field technique, we summarize recent results from our laboratory which include the fluorescence imaging of single dye molecules and single proteins and picosecond fluorescent lifetime measurements on intact photosynthetic membranes, all done at room temperature with nanometric spatial resolution.

The individual contribution of the 4 tryptophan residues (at positions 40, 54, 88, and 135) of E.coli SSB to its fluorescence emission was elucidated by deconvolution of the spectral envelope of wild type (wt), singly (-W40F, -W54F, -W88F, -W135F), and doubly (-W88F+W135F) point-mutated SSBs. Upon excitation at 297 nm, Trp 40, Trp 54, Trp 88, and Trp 135 peaked at 342, 347, 344, and 352 nm, respectively, and accounted for 22, 43, 24, and 11% of the SSB integrated emission. Extensive Tyr to Trp energy transfer at the singlet level was detected. Upon binding poly(dT), the emission spectra of all SSBs containing Trp 135 was red-shifted by 3 to 5 nm and their linewidth increased by ca. 4%, whereas for SSBs lacking Trp 135 more pronounced broadening without spectral shift was observed. In poly(dT) complexes, the limiting quenching (Q_max) at saturation (relative to the free protein) for the various SSB mutants was 93, 90, 89, 88, 85, and 79% for SSB-W88F+W135F, SSB-W135F, wt SSB, SSB-W88F, SSB-W40F, and SSB-W54F, respectively. From these Q_max values and the contribution of individual Trp residues to the overall emission of free SSB, the calculated individual Trp Q_max were 92, 95, 81, and 59% for residues 40, 54, 88, and 135, respectively. Trp 40 and Trp 54, shown earlier to be the only two Trp residues involved in stacking interactions with nucleic acid bases in SSB/polynucleotide complexes, exhibit almost complete quenching. In wt SSB/poly(dT) complex, Trp 40, 54, 88, and 135 account for 14, 18, 33, and 36% of the total emission. The dominance of the red-emitting Trp 135 in the fluorescence of SSB/poly(dT) complexes accounts for the spectral red-shift in systems containing this residue. Global analysis of fluorescence lifetime experiments in the frequency domain showed that Trp 54 has the longest average lifetime (8.0 ns). The long lifetime of Trp 54 appears to be due to its particular microenvironment, since that average lifetime for all individual Trp were in the 2 to 4 ns range in 8 M urea.

The energy relaxation mechanism of Au(III) protoporphyrin IX dimethyl ester in glacial acetic acid was determined by means of time-resolved picosecond spectroscopy. The rate of the absorbance change induced by a 355 nm ps pulse was measured. The kinetics of the ground state repopulation and the formation and decay of the excited states were measured at 425 nm, 547 nm, and 600 nm. Based on these data the kinetics of the transient intermediate states and the relaxation mechanism of Au(III) Protoporphyrin were determined.

Noninvasive, lifetime-based sensing in tissues depends upon deconvolving the probe lifetime from the photon `time-of-flights' which arise from the strong scattering properties of tissues. In this work, we investigated the utility of employing long-lived probes in order to minimize the contribution of photon migration to the decay kinetics of re-emitted, near IR light. Employing finite element techniques, we computed the spatial and temporal maps of the emission and excitation fluence for the cases of uniform and nonuniform distributions of embedded optical probe. Computations were performed for an incident impulse of light at the surface of tissues containing optical probes with varying lifetimes, (tau) . The results demonstrate that when the probe lifetime is significantly longer than the times associated with photon migration, the origin of the re-emitted light becomes located closer to the tissue surface and the `apparent' image of a fluorescent heterogeneity may be displaced from the true position.