This PDF file contains the front matter associated with SPIE Proceedings Volume 11648, including the Title Page, Copyright information, and Table of Contents.

Introduction to SPIE Photonics West BiOS conference 11648: Multiphoton Microscopy in the Biomedical Sciences XXI

Photodamage in nonlinear microscopy of transparent tissues starts at irradiances 1.5 times above the autofluorescence imaging level. Although the free-electron density is low, their energy suffices to break bonds in water, DNA and the backbone and side chains of proteins. We explored photodamage kinetics using physical indicators (hyperfluorescence, plasma luminescence, bubble formation). By plotting threshold values in (irradiance/radiant exposure) space, we identified a “safe” region for microscopy. Thermomechanical effects become relevant in melanin-containing tissue. Two-photon excitation of retinal fluorophores allows monitoring metabolic transformations. We analyze the thermomechanical damage pathways in retinal imaging, and discuss strategies for mitigating such damage.

Current methods to assess immune cell function use labels that are limiting for time-course studies of immune cell behavior in tumors. Here, we use multiphoton imaging of NAD(P)H and FAD, co-enzymes of metabolism, in T cells and macrophages within the tumor microenvironment. T cells alter their metabolism in response to tumor-like pH, glucose, and lactic acid levels, while macrophages alter their metabolism during tumor-stimulated migration. These results indicate that multiphoton autofluorescence imaging is a powerful label-free method to monitor immune cell metabolism within single cells in the tumor microenvironment.

TCSPC FLIM is the most direct way to obtain molecular information from live cells and tissues. Data analysis in molecular FLIM applications is usually complicated by the need of multi-exponential decay analysis in presence of low photon budget. Other issues are uncertainty in the instrument response function (IRF) and extremely long data processing times. bh's new generation SPCImage NG software addresses these problems by a combination of time-domain and frequency-domain analysis, image segmentation, MLE fitting, GPU processing, and IRF modelling. We demonstrate the performance of SPCImage NG for a number of advanced FLIM applications.

A clinical multicenter study has been performed on patients with suspicious, potentially malignant, pigmented skin lesions to evaluate the femtosecond fiber laser based compact multiphoton tomograph MPTcompact for high-resolution clinical imaging. Label-free high-resolution optical skin biopsies have been obtained based on two-photon autofluorescence excited at 780 nm, second harmonic generation, dermoscopy with an additional white light imager, and confocal reflectance with the NIR ultrashort laser pulses at 80 MHz. Furthermore, fluorescence lifetime imaging (FLIM) based on time-correlated single photon counting has been performed. Here we report on first clinical results. Multimodal compact multiphoton 3D/4D tomographs may become important label-free clinical highresolution imaging tools to replace in part conventional skin histology based on physically taken tissue samples, fixation, slicing, staining, and 2D light microscopy.

Förster resonance energy transfer (FRET) microscopy is a technique that provides the spatial and temporal resolution necessary to investigate protein-protein interactions in live specimens. One of the most robust ways of quantifying FRET is to measure the fluorescence lifetime of the donor fluorophore. Fluorescence lifetime imaging microscopy (FLIM) is routinely used in the biological sciences to monitor dynamic signaling events inside living cells. However, the demands of studying complex signaling networks in living cells using FRET-based biosensor probes requires the ability to track more than one donor fluorophore at the same time. Here, we demonstrate a novel (Pulsed Interleave Excitation) PIE-FastFLIM technique that can be used to simultaneously measure two FRET-based biosensor probes at the same pixel. The PIE-FastFLIM configuration uses tunable wavelengths and synchronized gating detection, which can be tailored and optimized for each biosensor probe. When combined with the model-free phasor plots analysis, the PIEFastFLIM approach enables quantitative multiplexed FRET measurements for tracking multiple signaling events simultaneously in live cells.

Fluorescence lifetime imaging microscopy (FLIM) and phosphorescence lifetime imaging microscopy (PLIM) often require complex and computationally intensive processes for analysis. For time-domain based measurements, computation of fluorescence and phosphorescence lifetimes conventionally involves nonlinear curve fitting techniques to model the time-resolved profiles as mono- or multi-exponential decays. The phasor, or “polar plot”, analysis method has recently gained attention as a simple method to characterize variations in fluorescence lifetime. The technique involves calculations of the intensity-normalized Fourier transform of the fluorescence profiles. The phasor can be visualized by plotting the real and imaginary components on a 2-dimensional plot. We have adapted the phasor analysis method for absolute quantitation of phosphorescence lifetimes of oxygen-sensitive phosphors. We utilize the phasor-derived lifetime values to quantify oxygen partial pressure in cortical micro vessels of awake mice. Here, we describe the modifications to adapt the technique for longer-duration phosphorescence decays. Our results demonstrate that oxygen measurements obtained from phasor analysis are in strong agreement with traditional curve fitting calculations. Using simulated phosphorescence decays, we also compare the effectiveness of the phasor method to nonlinear curve fitting techniques. To our knowledge, these findings constitute the first application of the phasor analysis method for characterizing phosphorescence measurements on the microsecond time scale. The method shows promise for monitoring cerebral metabolism and pathological changes in preclinical rodent models.

Fluorescence Lifetime Imaging (FLIM) is an essential tool in Life Sciences, but up to now users had to chose between high timing precision or fast data acquisition when using Time-Correlated Single Photon Counting (TCSPC) electronics. Our approach, named rapidFLIMHiRes, allows recording several FLIM images per second with a temporal resolution of 10 ps. The method combines advances in fast scanning, hybrid photomultiplier detectors, TCSPC modules, and correction algorithms to reduce decay curve distortions. Thus fast processes can be observed with the high optical and temporal resolution achievable in confocal microscopy at a rate of several frames per second.

Time-resolved analysis of periodically excited luminescence decays by the phasor method in the presence of time-gating or binning is revisited. Analytical expressions for discrete configurations of square gates are derived and the locus of the phasors of such modified periodic single-exponential decays is compared to the canonical universal semicircle. The effects of IRF offset, decay truncation and gate shape are also discussed. Finally, modified expressions for the phase and modulus lifetimes are provided for some simple cases. A discussion of a modified phasor calibration approach is presented.

Fluorescence lifetime imaging (FLIM) has proved to be ideal for non-invasive and very sensitive investigations in clinical applications. In contrast to pure intensity-based imaging, the usage of the timing information lifts FLIM on another level enabling functional imaging. The most salient example is metabolic imaging e.g. to investigate melanoma lesion. Here we show the capability of metabolic FLIM of monitoring melanoma growth in a very early stage. We also present the results of a multiphoton FLIM study on tracing nanoparticles after topical application on human skin in vivo for safety aspects of cosmetics.

Quantifying intestinal microvascular oxygenation may shed light on origin and progression of intestinal diseases as well as the role of the gut microbiota. In this work, we developed a method for depth-resolved microscopic measurements of partial pressure of oxygen (pO2) and Red Blood Cell (RBC) flux in the capillaries of intestinal wall in mouse cecum. Capillary pO2 and RBC flux were measured in healthy mice as a function of depth during normoxia and hyperoxia. Our results show low mean capillary RBC flux and relatively uniform capillary pO2 across the cecum wall depth.

A common property during tumor development and other diseases is altered energy metabolism, which could lead to a switch between oxidative phosphorylation (OXPHOS) and glycolysis. FLIM of metabolic coenzymes, as NAD(P)H and FAD, is now widely accepted to be the most reliable method to determine cell metabolism and different algorithms are actually proved to get reproducible results. However, it is questionable if redox state and cell metabolism correlate. For this new algorithms has to be developed to improve the diagnostic procedure. We will discuss the different approaches which were published so far and present new results in metabolic imaging taking into account the FLIM based NADH metabolic index as well as FLIRR based calculations including NADH, FAD and FMN. In addition, the phosphorescence lifetime of newly developed drugs will be used to demonstrate local oxygen changes. Simultaneous imaging of phosphorescence and fluorescence lifetime parameters enables complex analysis of bioenergetic alterations.

The fluorescence lifetime of NADH in live cells is an indicator of the metabolic state and, consequently, a discrimination parameter between normal and tumor cells. However, the fluorescence lifetime of NADH also depends on the cell type and on the mitochondrial pH. The task is therefore to find a FLIM signature that describes the metabolic state without being influenced by other parameters. We show that the ratio of the amplitudes of the decay components, a1/a2, of NADH is an absolute parameter for discriminating normal cells from tumor cells, and outline the technical requirements for metabolic FLIM for clinical use.

Autofluorescence lifetime imaging is a useful tool to quantify features of cellular metabolism. Here, we use multiphoton fluorescence lifetime imaging to measure NADH and FAD fluorescence lifetimes. We compared fluorescence intensity and lifetime features of T cells treated with a panel of metabolic inhibitors to correlate imaging features with metabolic pathways. Differences between autofluorescence features of T cells and cancer cells allow robust classification of cell type within simulations of complex tumor tissues. Autofluorescence lifetime imaging combined with automated image segmentation, analysis, and classification enables robust and label-free determination of cell type and function.

Urinary tract infections (UTI) are one of the frequently encountered infections in clinical practice. As there are different strains of bacteria responsible for UTI, the identification of types of bacterial is necessary to administer a proper antibiotic. Conventional staining and biochemical methods for the identification of bacteria are time-consuming and it usually leads to administer patients with broad-spectrum antibiotics which are less effective and expensive. In this regard, Multiphoton fluorescence imaging based on the distribution of NADH and FAD in several bacterial species isolated from UTI is carried out. Metabolic imaging based on fluorescence enables to analyze both biochemical distribution and their conformation. Spectral deconvolution method is used to isolate fluorescence emission from the coenzymes NADH and FAD to generate redox imaging. Further, redox imaging of bacteria was analyzed using different machine learning algorithms to improve the accuracy of classification. The results of this study revealed that the proposed technique of redox imaging was found to discriminate bacterial species. As the proposed method is both effective and less time consuming, the proposed method may be considered for real-time classification of bacterial species in the clinical setup.

Autofluorescence-based optical imaging methods have advanced cellular metabolism research as a technique to non-invasively measure metabolic shifts temporally and spatially. In this work, we present metabolic investigations using fluorescence anisotropy in two well-defined plant and animal models: 1) A set of flavonoid-deficient Arabidopsis thaliana mutants which shows unique fluorescence polarization properties with varying degrees of flavonoid accumulation. 2) The cancer metabolic deformities manifested in animal cells using real-time fluorescence anisotropy based metabolic readout (4fps). These studies demonstrated the potential of using fluorescence anisotropy in a systematic way to study metabolism and cellular heterogeneity.

The aim of our work was to investigate energy metabolism of patient’s tumor ex vivo using fluorescence lifetime imaging (FLIM). Tumor metabolism was examined by detecting the fluorescence lifetime of metabolic cofactor NAD(P)H. In glioma tumor samples was shown a different energy metabolism with white matter. Colorectal tumors displayed marked intertumor heterogeneity and shift towards oxidative energy metabolism. Thereby, the possibility of assessing the energy metabolism in postoperative tumor samples of patients was shown.

Fluorescence labelling is one of the significant properties for optical bioimaging and biosensing applications. The development in the field of nanotechnology has created enormous conveniences in biomedical research. The fluorescence enhancement of mammalian cell in presence of zinc nitride colloidal nanoparticle was imaged using multiphoton confocal microscope. The statistical significance in the presence and absence of nanoparticle evidenced the fluorescence intensity enhancement in the cells. In order to suppress the background signal and increase the penetration depth, the multiphoton technique is utilized by 740 nm excitation to examine the enhancement potential of Zn₃N₂ colloidal nanoparticles towards cell structures. Hence, we suggest that, the unique feature of inorganic nanomaterial is a promising material for probing cells for future diagnostic applications.

By the interaction with femtosecond-laser pulses, a cell membrane can be made transiently permeable for macro molecules which enables cell transfection and cell reprogramming. While adherent cells can be effectively targeted subsequently with high precision, we have previously shown that the realization of a continuous flow of cells through a laser focus region increases the number of addressable cells tremendously, although, at the cost of precise individual cell targeting. We describe, here, our improved new setup, which employs microfluidic-flow control and online monitoring of the flow using a high-speed camera, and steps towards optimizing the flow and laser parameters. The capabilities of the setup, which combines femtosecond-laser excitation, control of microfluidic cell flow, and high-speed imaging, are illustrated.

Ultrafast lasers have enabled a number of advanced imaging modalities, including two-photon fluorescence excitation (TPFE), Coherent Anti-Stokes Raman Scattering (CARS), Stimulated Raman Scattering (SRS), Second Harmonic Generation (SHG) imaging, and Fluorescence Lifetime Imaging Microscopy (FLIM). The core principle that governs this important body of research is multiphoton absorption of a femtosecond laser pulse that allows for 2 or more photons being simultaneously absorbed. Images are produced from the resulting fluorescence. Over the years, multiphoton light sources have become more powerful, more tunable, and more accessible to more users. We will present the latest light source technologies with special focus on the experiments and images enabled by these technologies including neuroscience, cell biology, and optogenetics.

We discuss the key trends in two and three photon microscopy driving new laser technologies. When distilled, these trends are moving down three paths, for mainstream, tunable 2 photon excitation sources, advanced high energy sources for especially for brain imaging and compact, fixed wavelength OEM friendly lasers.

Label-free phenotypic classification at a single-cell level is a challenging yet important task in cell biology. Stimulated Raman scattering (SRS) microscopy provides high chemical selectivity and sensitivity for label-free imaging of biological samples. With the capability to record hyperspectral SRS images with high-speed, mapping of biomolecules inside living cells enables label-free phenotyping. However, like all high-dimensional data, it remains challenging to fully exploit the excessive amount of information contained in hyperspectral data for single-cell analysis. Here, we developed and compared two machine-learning-based methods - the convolutional neural network and support vector machine - to automatically extract important features from high dimensional data and achieve a high-accuracy label-free single-cell classification. These methods serve as a robust approach to classify cells based on their molecular features, allowing unbiased, high-throughput data analysis.

The routine pathology workflow relies on cutting tissue into single-cell layer thick slices using paraffin or frozen sectioning. We propose a fast method to obtain sections of equivalent quality optically using the strong sectioning capabilities of two-photon microscopy (TPM). Hematoxylin and eosin (HE) equivalent staining of the tissue is achieved using acridine orange and sulforhodamine 101. We improved our previously presented pulsed fiber laser to deliver adjustable pulse durations of ~30ps at repetition rates of up to 16MHz and kW peak power. We can now image up to one square centimeter of tissue with sub-micrometer resolution within 15 minutes.

Metastases are the leading cause of breast cancer-related deaths. Tumor metastasis is affected by the microenvironment, including the extracellular matrix (ECM). Fibrillar collagen in the ECM produces an intrinsic optical signal called secondharmonic generation (SHG). The ratio of forward- to backward-scattered SHG photons (F/B) is sensitive to the collagen fiber internal structure and has been shown to be an independent prognostic indicator of metastasis-free survival time. We evaluated the effects of heterogeneity in the tumor ECM on F/B’s prognostic ability. Using SHG imaging we identified two distinct regions within 95 untreated primary tumor excisions: the tumor bulk and surrounding collagenous tumorstroma interface. We found that F/B measured in the tumor-stroma interface, but not tumor bulk, is prognostic of metastasis-free survival time (MFS) using both an intensity threshold selected by a blinded observer and adaptive thresholding. We calculated a 21-gene recurrence score (surrogate OncotypeDX®, or S-ODX) for each patient and using a Random Survival Forest method we found that F/B from the tumor-stroma interface, but not tumor bulk, and S-ODX contribute to predicting MFS in this patient cohort. These results suggest that F/B from the tumor-stroma interface of primary tumor excisions may provide information independent of genomic methods to stratify patients by metastatic risk and identify need for post-operative treatment.

Third and second harmonic generation (THG/SHG) and two-photon autofluorescence (2PEF) microscopy is a novel imaging technique that has been suggested as a promising clinical tool, mainly for cancer diagnosis. Here, we test a compact, portable ex-vivo THG/SHG/2PEF microscope (Femto Diagnostics B.V.) in the hospital. We will present the first results of an ongoing study in which we image lung tumor tissue immediately after excision, compare the images to the golden standard histopathology, assess the quality of diagnosis by independent pathologists, and examine the clinical impact of the THG/SHG/2PEF microscope. We also focus on identification and distinction of different immune cells.

Two-photon microscopy enables high-resolution biomedical imaging as well as photoluminescence lifetime investigations in photovoltaics. In order to provide a compact, highly flexible and cost-effective system, we present a compact two-photon platform box with a kagome hollow core photonics crystal fiber for laser pulse delivery. This fiber is suitable for the transmission of ultrashort IR laser pulses within the two optical ranges (i) 700 nm to 1000 nm and (ii) 1400 nm to 1700 nm. Furthermore, two-beam arrangements with parallel operation at two different wavelengths are possible. Our previous measurements of the transmission indicated values above 70% without distortion of the spectral properties or pulse durations using (i) a tunable femtosecond titanium:sapphire laser of 100 fs pulse width as well as a (ii) femtosecond fiber laser at 780 nm (80 fs) and at 1560 nm as well as a (iii) 10 fs titanium:sapphire laser. Subsurface recombination mechanisms in solar cells can be studied by multiphoton excitation and time-correlated single photon counting (TCSPC). Furthermore, deep tissue imaging can be performed. However, bending of the fiber may induce transmission losses. Investigations on the bending properties are presented. Time-resolved measurements on standard samples and plant tissue (Convallaria) is provided by using fluorescence lifetime analysis. Bulk solar cell impurity investigations are possible by analyzing photoluminescence signals.

Spectral phasors are currently based on fluorescence spectra in the wavelength domain. However, this is a problem when using spectral phasors to quantify emitters since wavelength represented spectra are not proportional to emitter population. Here we show the effect of unmixing using spectra in their transition dipole moment representation for a theoretical two component system and on a real seven component system.

Laser scanning microscopy is widely used to measure blood hemodynamics with line-scans. Analysis of red blood cell (RBC) velocity involves two types of algorithms. The algorithms of the first kind are the image-processing algorithms such as the shear [1], the LSPIV [2], the radon-transform [3] and the Fourier transform [4, 5] algorithms. Each of them provides the apparent velocity (V_RBCapp) of RBCs in the image, which does not consider the scanning conditions. The algorithms of the second kind give the real RBC velocity from V_{RBC app} and the scanning conditions [6]. Here we developed a predictive model that provide the accuracy of one image-processing algorithms. We find that the accuracy of the image-processing algorithm depends on several parameters such as the design of the scanned line, the scanning velocity and the velocity of RBCs. Last we validate our model by testing various types of artificial linescan images. The same strategy will be followed for various image-processing algorithms. Overall, our analysis will allow accurate comparisons of blood velocity from all vessel types in control and pathological animal models.

Exposure of tissues to sugar lead to the formation of advanced glycation end products (AGEs), contributing to diabetic complications. In human physiology, the vasculature is in direct contact with blood, thus the effect of diabetes is expected to be most severe with the vasculature. In this study, we incubated excised porcine aorta in D-glucose, Dgalactose, and D-fructose solutions. Multiphoton microscopy shows that for Days 4 to 48 incubation, autofluorescence is constant along the aorta sections, suggesting that monosaccharide diffusion is rapid when compared to the rate of fluorescent AGE formation.

Advanced glycation end products (AGEs) contribute to diabetic complications. Proteins undergo a series of complex and chain of biochemical actions in producing AGEs, some of which are fluorescent. In our study, we investigated the effects of glucose, fructose, and galactose in forming AGEs. Porcine tissues such as corneas, tendons, skin, aortas, and kidney were treated in the D-glucose, D-galactose, and D-fructose solutions, and multiphoton microscopy was performed on the glycated tissue specimens. We found that the rate of fAGEs formation from fructose was the highest.

Multimodal imaging combining various imaging approaches can provide complementary information about tissue in a single imaging session. Here, we introduce a multimodal approach combining three-photon microscopy (3PM) and spectral-domain optical coherence microscopy (SD-OCM) for the first time. We demonstrate the use of an optical parametric chirped-pulse amplification (OPCPA) laser source, commonly used for three-photon fluorescence excitation and thirdharmonic generation (THG), for simultaneous OCM and three-photon (3P) imaging. We validated the system in deep mouse brain in vivo with neuronal imaging. We visualized small structures such as myelinated axons, neurons, and large fiber tracts in white matter with high spatial resolution in a fast and non-invasive manner using linear and nonlinear contrast in the deep mouse brain (>1 mm) with an OPCPA source operating at 1620 nm central wavelength. Our method demonstrates the potential of the system for simultaneous OCM and 3PM with the same laser source.

Light-sheet fluorescence microscopy is a method of choice for multiscale live imaging. Indeed, its orthogonal geometry results in high acquisition speed, large field-of-view and low photodamage. Its combination with multiphoton excited fluorescence improves its imaging depth in biological tissues. However, it appears femtosecond laser sources commonly used in multiphoton microscopy at an 80 MHz repetition rate may not be optimized to take full advantage of light-sheet illumination during live imaging. Hence, we investigated the nature of induced photodamage in multiphoton light-sheet microscopy and the influence of laser parameters on the signal-to-photodamage ratio. To this end, we used zebrafish embryonic heart beat rate and fluorophore photobleaching as sensitive reporters of photoperturbations. We characterized linear and nonlinear disruptions depending on laser parameters such as laser mean power, pulse frequency or wavelength, and determine their order and relative impact. We found an optimal pulse frequency of ~10 MHz for imaging mCherry labeled beating hearts at 1030 nm excitation wavelength. Thus, we achieved high-speed imaging without inducing additional linear heating or reaching nonlinear photodamage compared to previous implementation. We reach an order-of-magnitude enhancement in two-photon excited fluorescence signal by optimizing the laser pulse frequency while maintaining low both the laser average power and its peak irradiance. It is possible to reach even larger enhancement of 3- photon excited fluorescence using such laser parameters. More generally, using low laser pulse frequency in multiphoton light-sheet microscopy results in a drastic improvement in signal level without compromising live sample, which opens new opportunities for fast in vivo imaging.

An optical parametric chirped-pulse amplifier (OPCPA) system is presented, capable of providing ultrashort laser pulses at all relevant central wavelengths, from 920 to 1700 nm, for 2- and 3-photon imaging applications with repetition rates up to 10 MHz for ultrafast image acquisition.

Multiphoton microscopy (MPM) techniques provide unprecedented deep access into live scattering tissue. Three-photon microscopy (3PM) has the advantage of the highly confined excitation volume, and its excitation wavelength falls in the optimal wavelength window for deep imaging of common fluorophores in scattering media. These features enable deep high-resolution imaging in complex biological tissues. Here, we provide a theoretical prediction for the depth limit in 3PM and experimentally verify our results. We show both theoretically and experimentally that the depth limit of 3PM depends on the labeling density of the sample, and 3PM can penetrate beyond 10 effective attenuation lengths (EAL) with close to diffraction-limited spatial resolution in a tissue phantom.

Fluorescence lifetime imaging microscopy (FLIM) systems are limited by their slow processing speed, low signal- to-noise ratio (SNR), and expensive and challenging hardware setups. In this work, we demonstrate applying a denoising convolutional network to improve FLIM SNR. The network will integrated with an instant FLIM system with fast data acquisition based on analog signal processing, high SNR using high-efficiency pulse-modulation, and cost-effective implementation utilizing off-the-shelf radio-frequency components. Our instant FLIM system simultaneously provides the intensity, lifetime, and phasor plots in vivo and ex vivo. By integrating image de- noising using the trained deep learning model on the FLIM data, provide accurate FLIM phasor measurements are obtained. The enhanced phasor is then passed through the K-means clustering segmentation method, an unbiased and unsupervised machine learning technique to separate different fluorophores accurately. Our experimental in vivo mouse kidney results indicate that introducing the deep learning image denoising model before the segmentation effectively removes the noise in the phasor compared to existing methods and provides clearer segments. Hence, the proposed deep learning-based workflow provides fast and accurate automatic segmentation of fluorescence images using instant FLIM. The denoising operation is effective for the segmentation if the FLIM measurements are noisy. The clustering can effectively enhance the detection of biological structures of interest in biomedical imaging applications.

Mitochondrial outer membrane permeabilization (MOMP) is a decisive step in the execution of cell death, and MOMP is often altered in cancer cells. MOMP could provide an early measurement of cancer treatment efficacy, and a method to understand relationships between mitochondrial function and cell death. Cationic dyes such as TMRE label mitochondria in a membrane potential dependent manner and provide insight into mitochondrial function in living cells. Currently, there are no easy to use open-source analysis methods to rapidly quantify bulk mitochondrial membrane potentials on a per cell level. Here, we present a custom CellProfiler pipeline to isolate TMRE intensity in both the mitochondria and nuclei of cells for rapid cell level quantification of bulk membrane potential.

In this study, the efficiency of uni- and bi-directional molecular diffusion in porcine liver tissue sections was studied. Tissue sections on the order of a few hundred microns in thickness were used. We found that bi-directional diffusion is more efficient in delivering probe molecules. Therefore, bi-directional labeling is useful in visualizing tissue architecture in three-dimensions.

In this work, we investigated the effect of formaldehyde-induced autofluorescence using multiphoton and confocal microscopy. Porcine aorta, brain, kidney, and ventricle were processed, sliced, and immersed in 10% neutral buffered formalin at ambient temperature for different length of time. Our preliminary results show that the inducedautofluorescence in brain and ventricle tissue increases through time, whereas for aorta and kidney it remains nearly the same.

With different states of two intrinsic fluorophores, nicotinamide adenine dinucleotide (phosphate) (NAD(P)H) and flavin adenine dinucleotide (FAD), we developed a single-layer autoencoder (AE) for feature extraction, which outputs condensed features representing the full metabolic FLIM information with lower dimensionality. We also described distributions of AE features and fluorescence lifetime redox ratio (FLIRR) from single cells by Gaussian mixture models (GMM), and predicted the values of FLIRR based on feature data from each time point for the HeLa cell lines and Caucasian-American (LNCaP) prostate cancer cell lines by the polynomial regression model and the random forest regression model.

Two-photon excitation of fluorescence (2PEF) is a versatile tool for high-resolution functional microscopy, but few existing fluorophores combine high two-photon brightness with ability to report on key in vivo- and in vitro environmental factors such as pH, ionic strength, or solvent polarity. We describe four novel pH-sensitive derivatives of coumarin 151 with different attached phosphazene moieties, offering high lipophilicity both in their neutral- and protonated forms at near physiological conditions (neutral to mildly basic). We measure the two-photon absorption (2PA) and 2PEF emission spectra, as well as the fluorescence quantum yields in neutral acetonitrile solution, as well as with the addition of small amount of trifluoromethanesulfonic (triflic) or hydrochloric acid. We show that the neutral form has high two-photon brightness in the basic environments, whereas protonation quenches the fluorescence by a factor of ~ 20, while blueshifting the absorption- and emission peaks by 60 nm. By evaluating the ratio of the measured 2PA and 1PA spectra corresponding to the transition to the lowest excited singlet state S1 we show, for the first time, that the permanent molecular dipole moment change is reduced upon the protonation. This conclusion is supported by our quantum-chemical calculations which indicate that protonation leads to reduction of extent of conjugation in the system. We also measure the time-dependent fluorescence and show that in neutral environment the fluorescence displays a single-exponential decay, while upon protonation the decay is distinctly non-single-exponential in character.

We analyzed the structural and functional state of liver tissue in normal state and during regeneration using multiphoton microscopy in combination with the time-of-flight secondary ions mass spectrometry (TOF-SIMS). The induction of regeneration process was performed on 30% and 70% hepatectomy models of rats. The FLIM method has shown an increase in the contribution of the bound form of NADH, as well as the contribution of NADPH in hepatocytes, which indicates a growth in the contribution of oxidative phosphorylation and biosynetic processes during liver regeneration. Using the TOF-SIMS method, it has been shown that the signal from fragments of fats, fatty acids, phosphotidylcholine, and sphingomyelin is increased, combined with a decline in the signal from amino acid’s fragments in the regenerating liver. The results will make it possible to identify criteria for assessing the regenerative potential of the liver during resection in order to reduce the risk of developing of the postoperative liver failure.