Fluorescence correlation spectroscopy (FCS) is a sensitive research tool for studying molecular dynamics at the single molecule level. Photophysical dynamics often dramatically influence FCS measurements, as we show here in characterizing the role of excitation saturation in two-photon fluorescence correlation measurements. We introduce a physical model that characterizes the influence of excitation saturation on the two-photon fluorescence observation volume, and derive an analytical expression for the correlation function that includes the influence of saturation. With this model, we can accurately describe both the temporal decay and the amplitude of measured fluorescence correlation functions over a wide range of illumination powers.

Using fluorescence correlation spectroscopy we measured the apparent mobility of a nuclear transport cargo (a streptavidin labeled with a nuclear localization signal) both in the cytoplasm and the nucleus of living cells, and we compared it to the mobility of a streptavidin labeled with mutations of the nuclear localization signal known not to support nuclear import, and with the mobility of a set of inert molecules (dextrans) of different sizes. In the cytoplasm, the mobility of the transport cargo is found to be significantly reduced compared to its mobility in the nucleus, or to the mobility of the streptavidins labeled with a mutant nuclear localization signal. This can be partly explained by the fact that the transport cargo forms a complex with two nuclear import mediator proteins (importin α and importin β) in the cytoplasm, but could also be partly due to specific interactions of this cargo with the cell cytoskeleton.

The combination of fluorescence correlation spectroscopy and two-photon excitation provides us with a powerful spectroscopic technique. Its submicron resolution and single molecule sensitivity make it an attractive technique for in vivo applications. Experiments have demonstrated that quantitative in vivo fluorescence fluctuation measurements are feasible, despite the presence of autofluorescence and the heterogeneity of the cellular environment. I will demonstrate that molecular brightness of proteins tagged with green fluorescent protein (GFP) is a useful and robust parameter for in vivo studies. Knowledge of photon statistics is crucial for the interpretation of fluorescence fluctuation experiments. I will describe photon counting histogram (PCH) analysis, which determines the molecular brightness and complements autocorrelation analysis. Non-ideal detector effects and their influence on the photon statistics will be discussed. The goal of in vivo fluorescence fluctuation experiments is to address functional properties of biomolecules. We will focus on retinoid X receptor (RXR), a nuclear receptor, which is crucial for the regulation of gene expression. The fluorescence brightness of RXR tagged with EGFP will be used to probe the oligomerization state of RXR.

Collagen is known to be a very effective generator of the second harmonic of incident light from 700 to 1100nm, and second harmonic generation (SHG) microscopy is coming into use as a tool for studying the distribution of collagen in tissue. It also shows promise as a technique for characterizing collagen - both in distinguishing different collagen types and their packing and in identifying degradation of collagen in pathologic conditions. However many aspects of image formation in SHG microscopy of collagen remain imperfectly understood, and we have commenced a rigorous study of these factors. The present paper presents the first results from this program.

The effects of structural perturbation on second harmonic generation in collagen were investigated. Type I collagen fascicles obtained from rat tails were structurally modified by increasing nonenzymatic cross-linking, by thermal denaturation, by collagenase digestion, or by dehydration. Changes in polarization dependence were observed in the dehydrated samples. Surprisingly, no changes in polarization dependence were observed in highly crosslinked samples, despite significant alterations in packing structure. Complete thermal denaturation and collagenase digestion produced samples with no detectable second harmonic signal. Prior to loss of signal, no change in polarization dependence was observed in partially heated or digested collagen.

We find that several key endogenous structural proteins including collagen, acto-myosin, and tubulin give rise to intense second harmonic generation (SHG) and that these structures can be imaged in intact tissues on a laser-scanning microscope. Because SHG is a non-resonant process, this modality suffers little inherent photobleaching or toxicity. In this study we demonstrate the clarity of SHG optical sectioning within unfixed, unstained thick specimens, including fish scales, C. elegans, and mouse muscle, where penetration into tissue upwards of 600 microns was achieved. The simultaneous use of SHG and two-photon excited GFP fluorescence allows for the inference of the molecular isoform that gives rise to SHG from the myofilament lattice in C. elegans. The physical origin of SHG within these tissues is addressed and is attributed to the laser interaction with dipolar protein structures. SHG polarization anisotropy is also used to determine the extent of dipolar order and radial symmetry in the helical structures. Comparisons are drawn between SHG and other forms of microscopy including polarization and fluorescence microscopy, highlighting the advantages and disadvantages.

We demonstrate localized photo-induced flip-flop of stilbazolium markers in model lipid bilayer membranes. The flip-flop mechanism and dynamics are determined by combined two-photon excited fluorescence and second harmonic generation microscopy. Upon illumination of labeled membranes with a femtosecond laser beam, two-photon absorption induced photo-isomerization provokes a significant increase in the cis-marker population whose flip-flop rate was determined to be at least a thousand times faster than that for trans-markers.

We present a fast scanning transmission-mode confocal scanning laser microscope system based on the use of a second harmonic generation (SHG) crystal for signal detection. The quadratic intensity dependence of SHG is exploited to preferentially reveal unscattered signal light and reject out-of-focus scattered background. The SHG crystal plays the role of a virtual pinhole that remains self-aligned without a need for de-scanning. We demonstrate that this new microscope method produces images with higher contrast and less speckle than transmission scanning microscopy with linear detection.

The problem of weak signal intensity due to the low incident average intensity limited by photodamage probability in common nonlinear light microscopy and spectroscopy can be fundamentally solved by increasing the repetition rate of the excitation light source. Since the possibility of nonlinear photodamage is determined by the incident peak intensity (or pulse energy), increasing the repetition rate of the excitation light source while keeping its peak intensity (or pulse energy) well below than damage threshold will not provoke any optical damage but will augment the average nonlinear signals. We used a femtosecond Ti:sapphire laser with a 2-GHz repetition rate as the light source of a second-harmonic-generation (SHG) microscope and strongly enhanced SHG signal was observed while no photodamage could be identified. Compared with the common 80-MHz Ti:sapphire laser, the microscopic images taken with the 2-GHz laser require shorter acquisition time and exhibit higher contrast, resulting in real-time SHG imaging capability.

Targeted transfection of cells is an important technique for gene therapy and related biomedical applications. We delineate how high-intensity (10¹² W/cm²) near-infrared (NIR) 80 MHz nanojoule femtosecond laser pulses can create highly localised membrane perforations within a minute focal volume, enabling non-invasive direct transfection of mammalian cells with DNA. We suspended Chinese hamster ovarian (CHO), rat kangaroo kidney epithelial (PtK2) and rat fibroblast cells in 0.5 ml culture medium in a sterile miniaturized cell chamber (JenLab GmbH, Jena, Germany) containing 0.2 μg plasmid DNA vector pEGFP-N1 (4.7 kb), which codes for green fluorescent protein (GFP). The NIR laser beam was introduced into a femtosecond laser scanning microscope (JenLab GmbH, Jena, Germany) and focussed on the edge of the cell membrane of a target cell for 16 ms. The integration and expression efficiency of EGFP were assessed in situ by two-photon fluorescence-lifetime imaging using time-correlated single photon counting. The unique capability to transfer foreign DNA safely and efficiently into specific cell types (including stem cells), circumventing mechanical, electrical or chemical means, will have many applications, such as targeted gene therapy and DNA vaccination.

We report on a novel laser source, emitting high energy (20 nanoJoule) femtosecond pulses, in a broad spectrum (250 nanometers). This source is easily tuned from 950 to 1200 nanometers, without any laser adjustment, and delivers sub-300 femtosecond pulses with a 10 nanometers spectral width.

Key players in cholesterol regulation are the members of a family of transcription factors known as the Sterol Regulatory Binding Proteins or SREBPs. The cellular redundancy of these proteins is under investigation, and our findings suggest that where these proteins reside may provide evidence for differences in the molecular dynamics of their transcriptional activity. Specifically, we have found that GFP-tagged SREBP-2 in contrast to SREBP-1 resides in discrete nuclear foci. To further explore functional differences between SREBP-1 and SREBP-2 we have developed an approach to monitor hetero- and homodimer formation by two-photon imaging and spectroscopy of fluorescence resonance energy transfer (TPIS-FRET). TPIS-FRET results will be presented. Collectively, these findings support the possibility that differences in function between SREBP family members may be governed by their localization within the cell.

We report the development, calibration and biomedical application of a multiphoton fluorescence lifetime imaging system (FLIM) using a streak camera. The present system is versatile with high spatial (approximately 0.2 μm) and temporal (approximately 50 psec) resolution and allows rapid data acquisition and reliable and reproducible lifetime determinations. The system was calibrated with standard fluorescent dyes and the lifetime values obtained were in very good agreement with values reported in literature for these dyes. We also demonstrate the applicability of the system to FLIM studies in cellular specimens in the context of quantitative measurement of fluorescence resonance energy transfer (FRET).

Time-correlated single photon counting (TCSPC) fluorescence lifetime imaging in laser scanning microscopes can be combined with a multi-detector technique that allows to record time-resolved images in several wavelength channels simultaneously. The technique is based on a multi-dimensional histogramming process that records the photon density versus the time within the fluorescence decay function, the x-y coordinates of the scanning area and the detector channel number. It avoids any time gating or wavelength switching and therefore yields a near-ideal counting efficiency. We show an instrument that records dual wavelength lifetime images with up to 512 x 512 pixels, and single wavelength lifetime images with up to 1024 x 1024 pixels. It resolves the components of double-exponential decay functions down to 30 ps, and works at the full scanning speed of a two-photon laser scanning microscope. The performance of the instrument is demonstrated for simultaneous lifetime imaging of the donor and acceptor fluorescence in CFP/YFP FRET systems and for tissue samples stained with several fluorophores.

Here, we have investigated the molecular mechanisms underlying the dynamics of protein distribution within membranes using Fluorescence Resonance Energy Transfer Microscopy (FRET). We have developed a one-photon (1-P) and two-photon (2-P) FRET assay to differentiate between the clustered and random distribution of membrane-bound fluorophore-labeled receptor-ligand complexes. Our results demonstrate that polymeric IgA-receptor-ligand complexes are organized in clusters within apical endocytic membranes of polarized MDCK cells, since energy transfer efficiency (E%) levels are independent from acceptor fluorescence, a standard parameter to confirm clustered distribution. We also describe a second parameter: E% decreases with increasing unquenched donor fluorescence and unquenched donor:acceptor ratios, a phenomenon which we ascribe to some donors preventing others from interacting with an acceptor. We call this effect 'donor geometric exclusion.' Going beyond the determination of clustered vs. random distribution of protein complexes, mathematical models have been developed, tailored to large, tightly packed molecular clusters, estimating their local densities with an adjustable parameter 's.'

Tissues of many marine invertebrates of class Anthozoa contain intensely fluorescent or brightly colored pigments. These pigments belong to a family of photoactive proteins closely related to Green Fluorescent Protein (GFP), and their emissions range from blue to red wavelengths. The great diversity of these pigments has only recently been realized. To investigate the role of these proteins in corals, we have performed an in vivo fluorescent pigment (FP) spectral and cellular distribution analyses in live coral cells using single and multi-photon laser scanning imaging and microspectroscopy. These analyses revealed that even single color corals contain spectroscopically heterogeneous pigment mixtures, with 2-5 major color types in the same area of tissue. They were typically arranged in step-wise light emission energy gradients (e.g. blue, green, yellow, red). The successive overlapping emission-excitation spectral profiles of differently colored FPs suggested that they were suited for sequential energy coupling. Traces of red FPs (emission = 570-660 nm) were present, even in non-red corals. We confirmed that radiative energy transfer could occur between separate granules of blue and green FPs and that energy transfer was inversely proportional to the square of the distance between them. Multi-photon micro-spectrofluorometric analysis gave significantly improved spectral resolution by restricting FP excitation to a single point in the focal plane of the sample. Pigment heterogeneity at small scales within granules suggested that fluorescence resonance energy transfer (FRET) might be occurring, and we confirmed that this was the case. Thus, energy transfer can take place both radiatively and by FRET, probably functioning in photoprotection by dissipation of excessive solar radiation.

A novel excitation and detection scheme has been proposed to measure time-resolved fluorescence. A train of continuous excitation pulses modulated by a pseudo-random bit sequence, instead of periodical short pulses of a very low duty factor, is used as the light source to illuminate a fluorescent sample. The cross-correlation between the emission from the sample and the modulation sequence yields the temporal profile of fluorescence decay. We have developed a primitive PC-based time-resolved fluorometer, in which an ultraviolet LED is modulated by a 4095-bit long maximal length pseudo-random bit sequence. The temporal resolution of the fluorometer is up to 1 microsecond. A long lifetime (more than 300 microseconds) europium chelate was used to test our instrument. Its fluorescence decay profile was successfully retrieved, and the measured lifetime is compatible with the specification.

Index mismatch induced spherical aberration is studied by comparing imaging ability of different immersion objectives in both uniformly fluorescent solutions and fluorescently labeled skin samples using scanning two-photon fluorescence microscopy. We investigated the performances of the objectives (air, water, glycerin, and oil immersion) by measuring the fluorescence profiles at different depths. In homogeneous fluorescent samples, we found that immersion medium with compatible refractive index as the samples yields better results, and small differences in refractive indices did not cause noticeable effects. Similar results were found in skin samples. Except for the air objective, we found that the choice of immersion medium did not have significant effects for in-depth imaging in the fluorescent solutions or skin samples.

xThe genus Saprolegnia in the phylum Oomycetes contains a number of parasitic species that can cause a range of important animal diseases. The aim of this study was to measure the calcium gradient, one of the growth regulating mechanisms, in Saprolegnia ferax. The two-photon laser scanning microscope allowed for detailed physiological measurements of calcium levels along the fungus-like hyphae of S.ferax. Calcium concentration was determined by making ratiometric calculation of emission levels of the calcium-sensitive fluorochrome Indo-1 at 485nm to 405nm. The calculated values were compared to the intracellular calibration values. The advantage of the two-photon laser scanning microscope is that it allows minor changes in concentration to be detected in highly localized regions of the hyphae. The technique used in this study minimized background and autofluorescence and therefore allowed for more accurate changes in intracellular Ca₂₊ concentration to be detected. The calcium concentration at the hyphal tip and 5, 10 and 40μm distal to the tip were calculated to be 65, 17, 38 and 20nM respectively, confirming other studies that suggest a tip-high calcium gradient.

It is well known that topical exposure to sulfur mustard (SM) produces persistent, incapacitating blisters of the skin. However, the primary lesions effecting epidermal-dermal separation and disabling of mechanisms for cutaneous repair remain uncertain. Immunofluorescent staining plus multiphoton imaging of human epidermal tissues and keratinocytes exposed to SM (400 μM x 5 min)have revealed that SM disrupts adhesion-complex molecules which are also disrupted by epidermolysis bullosa-type blistering diseases of the skin. Images of keratin-14 showed early, progressive, postexposure collapse of the K5/K14 cytoskeleton that resulted in ventral displacement of the nuclei beneath its collapsing filaments. This effectively corrupted the dynamic filament assemblies that link basal-cell nuclei to the extracellular matrix via α₆β₄-integrin and laminin-5. At 1 h postexposure, there was disruption in the surface organization of α₆β₄ integrins, associated displacement of laminin-5 anchoring sites and a concomitant loss of functional asymmetry. Accordingly, our multiphoton images are providing compelling evidence that SM induces prevesicating lesions that disrupt the receptor-ligand organization and cytoskeletal systems required for maintaining dermal-epidermal attachment, signal transduction, and polarized mobility.

The field of multiphoton microscopy has undergone significant advances since its beginning just over a decade ago. One of these was the development of coherent multiphoton techniques, which typically image intrinsic properties of the sample, without resorting to staining or labeling. Here we describe some recent technological developments in the two leading coherent multiphoton techniques: third-harmonic generation and coherent anti-Stokes Raman spectroscopy. These techniques are applied for visualization of a variety of biological and other samples.

Singular value decomposition was applied to the set of the non-resonant Raman spectra, recorded during Raman imaging of the single apoptotic cell. The basis vectors and the corresponding singular values were assessed in terms of their statistical significance. The noise-containing basis vectors were rejected while keeping the meaningful ones. In this way the Raman images of the apoptotic cell were successfully reconstructed from the spectrally-filtered data matrix. The results are demonstrated on the spatial distribution of the DNA, protein and phospholipids, present in the fragments of the apoptotic cell.

We have observed fluorescence from Coumarin 334 in solution using light of twice an actinic wavelength for excitation. The effect is demonstrated to be biphotonic, but is observed at optical powers below those required for resonant biphotonic excitation, and with an incoherent source. We infer operation of HRS. By extrapolation we propose that this effect should contribute to fluorescence under usual conditions for multiphoton fluorescence microscopy, and that it might be exploited in fluorescence microscopy under conditions where usual mechanisms are inapplicable.

Multiphoton microscopy is becoming an increasingly popular modality of laser scanning microscopy for imaging living specimens. In order to improve the signal to noise ratio under the challenging imaging conditions typical of biomedical research, researchers may be forced to resort to measures which substantially increase the amount of energy to which specimens are exposed. Several mechanisms of damage to living cells compete to limit the power window for minimally invasive imaging in multiphoton microscopy; these mechanisms include heating due to linear absorption, phototoxicity related to multiphoton absorption, and optical breakdown due to multiphoton ionization. The relative contribution of each of these factors may change significantly depending on the specimen and imaging parameters. The present study investigates the dominant factors in limiting cell viability at moderate to high-energy fluence levels that may be necessary under non-ideal imaging conditions. The results of this study suggest that heating associated with optical breakdown is an important factor in limiting cell viability under difficult imaging conditions, and that this heating scales in a manner proportional to the energy fluence associated with a given set of scanning parameters. Also significant is the finding that cell viability does not appear to scale proportionally to pulse width in a manner consistent with 2-photon absorption at ultrashort pulse widths and moderate energy fluence levels. These results suggest that the use of a pre-compensation scheme to offset the positive group velocity dispersion due to laser scanning microscope optics is a practical means to increase the signal to noise ratio while minimally impacting cell damage thresholds at ultrafast (sub 200 femtosecond) pulse widths.

A theoretical model of a two-photon fluorescence microscopy imaging in a scattering medium has been developed. Unlike the existing approaches, this model, based on the small-angle approximation of the radiative transfer equation, takes into account the effects of multiple small-angle scattering of probing and fluorescent light in the medium. The expression for a detected signal from an object observed by its fluorescence at the two-photon absorption of pulsed IR focused light beam, and also at the one-photon absorption, has been obtained. The model has been applied for constructing the images of a) a continuous uniformly fluorescent turbid medium and b) a thin fluorescent layer buried under a thick layer of non-fluorescent scattering medium. On the basis of the proposed model, general properties of the calculated images, such as the registered fluorescence signal dependence on the focusing depth and the influence of the scattering in the medium, have been studied.

Automated methods are described for in vivo quantitation of changes in tumor vasculature. The tumor subsurface is imaged non-invasively over time with two-photon confocal microscopy aided by a variety of chronic animal window preparations. This results in time series of three-dimensional (3-D) image stacks for each specimen at high resolution (768x512x32 voxels, 8 bits/voxel, every 24 hours for 7 days), imaging depth and signal-to-background ratio. Next, automated image analysis allows detection and quantitation of vascular changes in a rapid and objective manner without manual tedium. We describe a fast new algorithm for fully automated 3-D tracing (50 seconds to trace a 10 MB stack on a Dell 1 GHz Pentium III personal computer). A variety of measurements including tortuosity, length, thickness, and branching order are generated and analyzed. Quantitative validation of the performance of the tracing algorithm against manual tracing resulted in 81% concordance. This enables a broader set of change analysis studies including testing the efficacy of anti-angiogenic therapies and deriving vessel growth parameters that may be correlated with physiological and gene expression profiles in tumor.

Microspectroscopic measurements in plant cells are complicated by the presence of dense cellular structures such as the cell wall that causes severe light scattering. In addition, the low penetration depth of the excitation light limits the fluorescence signal originating from deeper cell layers in thick multi-cellular plant preparations when single-photon excitation (SPE) is applied. However, two-photon excitation (TPE) can overcome these problems. We report on two-photon microscopy studies of Histone 2B-YFP, a nuclear-expressed protein involved in chromatin packaging. In contrast to SPE, TPE allows imaging throughout the whole root. Therefore by using TPE it was also possible to visualize the root quiescent centers using SCARECROW-EGFP localized in the middle of the root. The interactions between various members of the Arabidopsis thaliana embryogenesis receptor kinase family (AtSERK) have been studied by monitoring Forster resonance energy transfer (FRET) between AtSERK-ECFP and -EYFP fusion proteins using fluorescence lifetime imaging microscopy (FLIM) of the two-photon excited ECFP component.

The extremely small (femtoliter) excitation volume of multiphoton (MP) microscopy renders all emitted photons useful in detecting fluorescence signals. Hence, multiphoton laser scanning microscopy (MPLSM) systems can collect fluorescence through the objective (epi-fluorescence), as well as the condenser (trans-fluorescence). For maximal collection efficiency, both optical paths can be used concurrently (4π detection). Most MPLSM systems incorporate photodetectors directly in or adjacent to the epi- and trans-fluorescence optical paths of the microscope, generally photomultiplier tubes with associated optics. These arrangements are optically straightforward, but are often bulky and difficult to reconfigure. Here, we demonstrate that all fluorescence from the specimen can be efficiently coupled into two multimode optical fibers -- one each for the epi- and trans-fluorescence pathways. Fiber-coupled detection enables a modular detection paradigm where light can be routed to easily reconfigurable and interchangeable detection module(s). A novel MPLSM system was constructed, which is readily switched between the original de-scanned detection path for confocal microscopy, and the newly added pathways supporting fiber-coupled non-descanned 4π detection for MP microscopy. Sample MP images of fluorescent beads and fluorescent-labeled hippocampal neurons are presented, demonstrating the viability of fiber-coupled detection.

Single molecule spectroscopy often requires the immobilization of the molecules onto solid or quasi-solid substrates and the use of relatively high excitation intensity. We have studied the fluorescence emission of four common dyes used for bio-imaging studies, rhodamine 6G, fluorescein, pyrene and indo-1 at the single molecule level under two-photon excitation regime. We focus on two-photon excitation thermal effects on the stability of the single molecules, influencing the internal photo-dynamics and the total duration of the fluorescent emission. Single dye molecules, spread on a glass substrate by spin coating, show a constant fluorescence output till a sudden transition to a dark state. The bleaching time varies in the series pyrene, indo-1, fluorescein and rhodamine 6g from the fastest to the slowest one respectively, has a gaussian distribution suggesting that bleaching is not due to photo-bleaching. These observations are interpreted as thermal bleaching where the temperature increase is induced by the two-photon excitation process and the thermal bleaching is correlated to the amount of absorbed power that is not re-irradiated as fluorescence.