Coherent anti-Stokes Raman scattering microscopy.

This thesis describes a new technique that improves specificity, selectivity and sensitivity in coherent anti-Stokes Raman scattering (CARS) microscopy. CARS microscopy is a nonlinear optical technique that utilizes specific bonds of molecules, sometimes referred to as the `fingerprint' of a molecule, to obtain chemically specific imaging of a sample. A single `unique' vibrational resonance within a molecule can be addressed and the appearance of this resonance within a sample can be imaged. This nonlinear vibrational microscopy technique provides a noninvasive method to identify structures of a specific chemical composition. CARS features high collection efficiency and absence of interference from one-photon autofluorescence. However, CARS suffers from a frequency-independent background signal, the `non-resonant background'. In mixtures, the resonant CARS signal of less abundant constituents may be overwhelmed by the non-resonant background, preventing detection of the resonant molecules. A cascaded phase preserving chain is used for heterodyne detection and yields amplitude and phase. The vibrational phase of the oscillators in the focal volume can be obtained by the subtraction of the local excitation phase from the heterodyne phase. We call this technique ``Vibrational Phase Contrast CARS'' and can be regarded as a vibrational extension of the linear (refractive index) phase contrast microscopy introduced by Zernike in 1933. The phase detection allows for rejection of the non-resonant background signal, without reduction of the resonant signal. CARS microscopy is applied to biology, pharmacology and food, where chemical selectivity and label-free detection are essential. Other nonlinear signals, generated simultaneously, can be used to obtain more information from the sample; CARS, second harmonic generation microscopy and spontaneous Raman spectroscopy are used side-by-side. Two in-depth studies are presented on the dissolution behavior of a drug and on early bone growth in cells.

Hyperspectral coherent anti-Stokes Raman scattering (CARS) microscopy has provided an imaging tool for extraction of 3-dimensional volumetric information, as well as chemically-sensitive spectral information. These techniques have been used in a variety of different domains including biophysics, geology, and material science. The measured CARS spectrum results from interference between the Raman response of the sample and a non-resonant background. We have collected four dimensional data sets (three spatial dimensions, plus spectra) and extracted Raman response from the CARS spectrum using a Kramers-Kronig transformation. However, the three dimensional images formed by a CARS microscope are distorted by interference, some of which arises because of the Gouy phase shift. This type of interference comes from the axial position of the Raman resonant object in the laser focus. We studied how the Gouy phase manifests itself in the spectral domain by investigating microscopic diamonds and nitrobenzene droplets in a CARS microscope. Through experimental results and numerical calculation using finite-diference time-domain (FDTD) methods, we were able to demonstrate the relationship between the spatial configuration of the sample and the CARS spectral response in three dimensional space.

In a coherent anti-Stokes Raman scattering (CARS) microscope, when samples with different shapes and dimensions are excitated by collinearly introduced and tightly focused Gaussian beams, the microscopic structure will be determined by the spatial distributions of generated CARS signals. Therefore, we build a theoretical model for CARS signals from spherical sample under the tightly focused condition. The intensity and phase distributions of tightly focused linear polarization Gaussian beams are analyzed with vector wave equations. The vector wave equation of CARS signals is derived from Green's function. The far-field CARS radiation patterns of spherical scatters with different diameters are simulatively calculated. Theoretical analysis and simulative calculation results show that the intensities of forward and backward CARS signals from the small spherical sampler are similar. The images with high contrast can be obtained by backward detection method from an objective with a high numerical aperture. For big spherical samplers, intensities of CARS signals are greatly increased. The emission direction is mainly concentrated in a spatial angle. The forward CARS signals can be effectively collected by an objective with low numerical aperture.

Coherent anti-Stokes Raman scattering (CARS) spectroscopy and microscopy have many potential applications in biology and medicine. Among many variants of the technique itself, the method of single-pulse CARS spectroscopy and microscopy is attractive for its simplicity and quick implementation. Single-pulse CARS microscopy can be performed by shaping the excitation spectrum using a notch filter, yet the resonant signal rides on a large background caused by a non-resonant signal, a background which is usually removed by lock-in detection. Here, we show that the background can be reduced significantly by adding a small chirp to the pulse and can even be made smaller than the resonant signal. In order to enhance the CARS signal and thus the contrast further, double-notch shaping is introduced. The double-notch induces two sets of CARS features shifted by the frequency difference between the two notches, thereby coherently enhancing a particular CARS feature. The experimental results agree well with theoretical simulations. We applied this scheme to perform lock-in free CARS microscopy of bone tissue with enhanced contrast.

A picosecond time-resolved multiplex coherent anti-Stokes Raman scattering (CARS) spectrometer based on a new principle has been constructed using a femtosecond laser, nanosecond lasers, and a streak camera. The second harmonic of the output from the femtosecond Ti:sapphire oscillator/regenerative-amplifier system is used for the photoexcitation. The second harmonic of a nanosecond Nd:YLF laser is used as CARS probing ω1 radiation, and the output of a nanosecond broadband dye laser is used as ω2 radiation. CARS signals, which are generated from the sample during the nanosecond probing pulses, are analyzed in the frequency domain by a spectrograph and are resolved in the time domain by a streak camera. This enables us multiplex CARS measurements not only in the frequency domain but also in the time domain (termed two-dimensional multiplex CARS). Consequently, CARS spectra at various delay times are obtainable with a single measurement. Owing to the short duration and high stability of the femtosecond pumping pulses, time resolution as high as 15 ps has been achieved. Picosecond time-resolved CARS spectra of all-trans retinal and β-carotene are presented to demonstrate the high capability of the constructed system. The factors which determine the time resolution of this CARS spectrometer are discussed.

BackgroundCoherent anti-Stokes Raman scattering (CARS) microscopy provides fine resolution imaging and displays morphochemical properties of unstained tissue. Here, we evaluated this technique to delineate and identify brain tumors.MethodsDifferent human tumors (glioblastoma, brain metastases of melanoma and breast cancer) were induced in an orthotopic mouse model. Cryosections were investigated by CARS imaging tuned to probe C-H molecular vibrations, thereby addressing the lipid content of the sample. Raman microspectroscopy was used as reference. Histopathology provided information about the tumor's localization, cell proliferation and vascularization.ResultsThe morphochemical contrast of CARS images enabled identifying brain tumors irrespective of the tumor type and properties: All tumors were characterized by a lower CARS signal intensity than the normal parenchyma. On this basis, tumor borders and infiltrations could be identified with cellular resolution. Quantitative analysis revealed that the tumor-related reduction of CARS signal intensity was more pronounced in glioblastoma than in metastases. Raman spectroscopy enabled relating the CARS intensity variation to the decline of total lipid content in the tumors. The analysis of the immunohistochemical stainings revealed no correlation between tumor-induced cytological changes and the extent of CARS signal intensity reductions. The results were confirmed on samples of human glioblastoma.ConclusionsCARS imaging enables label-free, rapid and objective identification of primary and secondary brain tumors. Therefore, it is a potential tool for diagnostic neuropathology as well as for intraoperative tumor delineation.

The nematode Caenorhabditis elegans has been employed as a model organism to study human obesity due to the conservation of the pathways that regulate energy metabolism. To assay for fat storage in C. elegans, a number of fat-soluble dyes have been employed including BODIPY, Nile Red, Oil Red O, and Sudan Black. However, dye-labeled assays produce results that often do not correlate with fat stores in C. elegans. An alternative label-free approach to analyze fat storage in C. elegans has recently been described with coherent anti-Stokes Raman scattering (CARS) microscopy. Here, we compare the performance of CARS microscopy with standard dye-labeled techniques and biochemical quantification to analyze fat storage in wild type C. elegans and with genetic mutations in the insulin/IGF-1 signaling pathway including the genes daf-2 (insulin/IGF-1 receptor), rict-1 (rictor) and sgk-1 (serum glucocorticoid kinase). CARS imaging provides a direct measure of fat storage with unprecedented details including total fat stores as well as the size, number, and lipid-chain unsaturation of individual lipid droplets. In addition, CARS/TPEF imaging reveals a neutral lipid species that resides in both the hypodermis and the intestinal cells and an autofluorescent organelle that resides exclusively in the intestinal cells. Importantly, coherent addition of the CARS fields from the C-H abundant neutral lipid permits selective CARS imaging of the fat store, and further coupling of spontaneous Raman analysis provides unprecedented details including lipid-chain unsaturation of individual lipid droplets. We observe that although daf-2, rict-1, and sgk-1 mutants affect insulin/IGF-1 signaling, they exhibit vastly different phenotypes in terms of neutral lipid and autofluorescent species. We find that CARS imaging gives quantification similar to standard biochemical triglyceride quantification. Further, we independently confirm that feeding worms with vital dyes does not lead to the staining of fat stores, but rather the sequestration of dyes in lysosome-related organelles. In contrast, fixative staining methods provide reproducible data but are prone to errors due to the interference of autofluorescent species and the non-specific staining of cellular structures other than fat stores. Importantly, both growth conditions and developmental stage should be considered when comparing methods of C. elegans lipid storage. Taken together, we confirm that CARS microscopy provides a direct, non-invasive, and label-free means to quantitatively analyze fat storage in living C. elegans.

We demonstrate theoretically and experimentally a novel epidetection scheme for coherent anti-Stokes Raman scattering (CARS) microscopy that significantly improves the detection sensitivity. Calculations show that epidetected CARS (E-CARS) signals are present for scatterers smaller than the wavelength of light, whereas the large background signals from the surrounding bulk solvent are suppressed by destructive interference. E-CARS microscopy is capable of revealing small intracellular features that are otherwise buried by the strong water CARS signal.

We demonstrate coherent anti‐Stokes Raman scattering (CARS) microscopy in a wide‐field geometry and discuss the reasons why this approach is capable of becoming a complementary method to confocal scanning CARS microscopy. By means of nanosecond laser pulses and a non‐confocal excitation geometry, the CARS signal is generated over a sample area of approximately 50 µm in diameter. As an example, we recorded images of oil drops within finely sliced sunflower seeds by tuning the laser frequency to the strong Raman transition of linoleic acid around 2870 cm−1. Meaningful images with spectral selectivity could already be recorded using only a few nanosecond pulses. Copyright © 2006 John Wiley & Sons, Ltd.

By integrating sum-frequency generation (SFG), and two-photon excitation fluorescence (TPEF) on a coherent anti-Stokes Raman scattering (CARS) microscope platform, multimodal nonlinear optical (NLO) imaging of arteries and atherosclerotic lesions was demonstrated. CARS signals arising from CH2-rich membranes allowed visualization of endothelial cells and smooth muscle cells in a carotid artery. Additionally, CARS microscopy allowed vibrational imaging of elastin and collagen fibrils which are rich in CH2 bonds in their cross-linking residues. The extracellular matrix organization was further confirmed by TPEF signals arising from elastin's autofluorescence and SFG signals arising from collagen fibrils' non-centrosymmetric structure. The system is capable of identifying different atherosclerotic lesion stages with sub-cellular resolution. The stages of atherosclerosis, such as macrophage infiltration, lipid-laden foam cell accumulation, extracellular lipid distribution, fibrous tissue deposition, plaque establishment, and formation of other complicated lesions could be viewed by our multimodal CARS microscope. Collagen percentages in the region adjacent to coronary artery stents were resolved. High correlation between NLO and histology imaging evidenced the validity of the NLO imaging. The capability of imaging significant components of an arterial wall and distinctive stages of atherosclerosis in a label-free manner suggests the potential application of multimodal nonlinear optical microscopy to monitor the onset and progression of arterial diseases.© (2008) COPYRIGHT SPIE--The International Society for Optical Engineering. Downloading of the abstract is permitted for personal use only.

A new vibrational imaging method based on coherent anti-Stokes Raman scattering (CARS) has been used for high-speed, selective imaging of neutral lipid droplets (LDs) in unstained live fibroblast cells. LDs have a high density of C-H bonds and show a high contrast in laser-scanning CARS images taken at 2,845 cm-1, the frequency for aliphatic C-H vibrations. The contrast from LDs was confirmed by comparing CARS and Oil Red O (ORO)-stained fluorescence images. The fluorescent labeling processes were examined with CARS microscopy. It was found that ORO staining of fixed cells caused aggregation of LDs, whereas fixing with formaldehyde or staining with Nile Red did not affect LDs. CARS microscopy was also used to monitor the 3T3-L1 cell differentiation process, revealing that there was an obvious clearance of LDs at the early stage of differentiation. After that, the cells started to differentiate and reaccumulate LDs in the cytoplasm in a largely unsynchronized manner. Differentiated cells formed small colonies surrounded by undifferentiated cells that were devoid of LDs. These observations demonstrate that CARS microscopy can follow dynamic changes in live cells with chemical selectivity and noninvasiveness. CARS microscopy, in tandem with other techniques, provides exciting possibilities for studying LD dynamics under physiological conditions without perturbation of cell functions.

CARS spectra of polystyrene sheet at excitation of fs pulses and chirped pulses

We propose a structured illumination scheme for achieving widefield coherent anti-Stokes Raman scattering (CARS) microscopy with a resolution surpassing the diffraction limit in two dimensions (2D). By acquiring a set of coherent images of a sample with third-order nonlinear susceptibility illuminated by the phase-matched excitation field of square lattice patterns, a 2D super-resolution CARS image can be reconstructed. We derive a theoretical framework to describe the coherent image formation and reconstruction scheme for this structured illumination CARS imaging system and carry out numerical simulations to investigate its imaging performance. The results demonstrate that our method promises a particular benefit on CARS microscopy by adding the super-resolution capability to improve its 2D spatial resolution by a factor of approximately three.

This work describes in detail a wide-field surface-enhanced coherent anti-Stokes Raman scattering (CARS) microscope, which enables enhanced detection of sample structures in close proximity (∼100 nm) of the substrate interface. Unlike conventional CARS microscopy, where the sample is illuminated with freely propagating light, the current implementation uses evanescent fields to drive Raman coherences across the entire object plane. By coupling the pump and Stokes excitation beams to the surface plasmon-polariton mode at the interface of a 30 nm thick gold film, we obtained strong CARS signals from cholesteryl oleate droplets adhered to the surface. The surface-enhanced CARS imaging system visualizes lipid structures with vibrational selectivity using illumination doses per unit area that are more than four orders of magnitude lower than in point-scanning CARS microscopy.

We report the retrieval of the Raman signal from coherent anti-Stokes Raman scattering (CARS) spectra using a convolutional neural network (CNN) model. Three different types of non-resonant backgrounds (NRBs) were explored to simulate the CARS spectra viz (1) product of two sigmoids following the original SpecNet model, (2) Single Sigmoid, and (3) fourth-order polynomial function. Later, 50 000 CARS spectra were separately synthesized using each NRB type to train the CNN model and, after training, we tested its performance on 300 simulated test spectra. The results have shown that imaginary part extraction capability is superior for the model trained with Polynomial NRB, and the extracted line shapes are in good agreement with the ground truth. Moreover, correlation analysis was carried out to compare the retrieved Raman signals to real ones, and a higher correlation coefficient was obtained for the model trained with the Polynomial NRB (on average, ∼0.95 for 300 test spectra), whereas it was ∼0.89 for the other NRBs. Finally, the predictive capability is evaluated on three complex experimental CARS spectra (DMPC, ADP, and yeast), where the Polynomial NRB model performance is found to stand out from the rest. This approach has a strong potential to simplify the analysis of complex CARS spectroscopy and can be helpful in real-time microscopy imaging applications.