Published papers

Living microalga Cyanidium caldarium was visualized using ultra-multiplex coherent anti-Stokes Raman scattering (CARS) microspectroscopy. The Raman band related to the SO42− ion at 982 cm−1 was detected at the peripheral part of the cells and showed a signal amplitude that was approximately threefold larger than that in the medium, indicating accumulation of SO42− ions. Depending on the incubation conditions, a broad and sharp Raman band at approximately 505 cm−1 was observed at the peripheral part and inside the cells, respectively. Based on comparison of the results with model substances, these broad and sharp bands were assigned to polysulfides and glycogen, respectively.

Spore-forming bacteria accumulate dipicolinic acid (DPA) to form spores to survive in extreme environments. Vibrational spectroscopy is widely used to detect DPA and elucidate the existence of the bacteria, while vegetative cells, another form of spore-forming bacteria, have not been studied extensively. Herein, we applied coherent anti-Stokes Raman scattering (CARS) microscopy to spectroscopically identify both spores and vegetative cells without staining or molecular tagging. The spores were identified by the strong CARS signals due to DPA. Furthermore, we observed bright spots in the vegetative cells in the CARS image at 1735 cm-1. The vegetative cells contained molecular species with C=O bonds because this vibrational mode was associated with the carbonyl group. One of the candidate molecular species is diketopimelic acid (DKP), a DPA precursor. This hypothesis was verified by comparing the spectrum obtained by the vegetative cells with that of the DKP analogue (ketopimelic acid) and with the result obtained by DFT calculation. The results indicate that the observed vegetative cell is in the sporulation process. CARS spectra can be used to monitor the maturation and preformation of spores.

In this study, second harmonic generation (SHG) and third harmonic generation (THG) spectroscopic imaging were performed on biological samples using a femtosecond laser source in the third near-infrared (NIR) optical window (NIR-III). Using a visible-NIR spectrometer, the SHG and THG signals were simultaneously detected and were extracted using spectral analysis. Visualization of biological samples such as cultured cells (HEK293 T), mouse brain slices, and the nematode Caenorhabditis elegans was performed in a label-free manner. In particular, in an SHG image of an entire coronal brain section (8 × 6 mm2), we observed mesh-like and filamentous structures in the arachnoid mater and wall of the cerebral ventricle, probably corresponding to the collagen fibers, cilia, and rootlet. Moreover, the THG images clearly depicted the densely packed axons in the white matter and cell nuclei at the cortex of the mouse brain slice sample and lipid-rich granules such as lipid droplets inside the nematode. The observations and conclusions drawn from this technique confirm that it can be utilized for various biological applications, including in vivo label-free imaging of living animals.

Chemical responsivity in materials is essential to build systems with switchable functionalities. However, polarity-switchable materials are still rare because inducing a symmetry breaking of the crystal structure by adsorbing chemical species is difficult. In this study, we demonstrate that a molecular organic-inorganic hybrid crystal of (NEt4)2[MnN(CN)4] (1) undergoes polarity switching induced by water vapor and transforms into a rare example of proton-conducting second-harmonic-generation-active material. Centrosymmetric 1 transforms into noncentrosymmetric polar 1·3H2O and 1·MeOH by accommodating water and methanol molecules, respectively. However, only water vapor causes a spontaneous single-crystal-to-single-crystal transition. Moreover, 1·3H2O shows proton conduction with 2.3 × 10-6S/cm at 298 K and a relative humidity of 80%.

Ultra-multiplex coherent anti-Stokes Raman scattering (CARS) spectroscopic imaging was used to visualize the distribution of water concentration in human skin ex vivo. The CARS signal of the OH stretching vibrational mode of water was found to coexist with the signal of intercellular lipids such as ceramides, which were visualized by a sharp vibrational band at 2882 cm-1. Depth-resolved CARS spectroscopic imaging of a skin sample revealed that ceramides were localized in the stratum corneum. These findings demonstrate the powerful potential of CARS spectroscopic imaging for probing pathological changes caused by anomalous water concentration distribution in human skin. © 2021 The Japan Society of Applied Physics.

Liquid–liquid phase separation leads to the formation of liquid droplets (LqDs) such as P granules in Caenorhabditis elegans (C. elegans). In this study, we demonstrate the label-free visualization of LqDs using multimodal nonlinear optical imaging both in vitro and in vivo. In vitro measurements with polymerized adenine [poly(A)], we found significantly higher poly(A) concentrations in LqDs compared to surrounding solutions, with the limit of detection (LoD) of 32 mg/mL. In vivo measurements, we performed label-free imaging of C. elegans. Despite efforts to detect P granules within P lineage cells in both wild-type C. elegans and green fluorescent protein (GFP)-tagged strains, no clear RNA-specific signals were observed. This indicates that the RNA concentration in P granules is lower than anticipated and falls below our in vitro LoD. These results underscore the challenges of label-free RNA detection in P granules.

The Raman fingerprint spectral region provides abundant structural information on molecules. However, analyzing vibrational images within this region using coherent Raman imaging remains challenging due to the small Raman cross section and congested spectral features. In this study, we combined ultrabroadband coherent anti-Stokes Raman scattering (CARS) microspectroscopy across the spectral range of 500-4000 cm-1 with multivariate curve resolution-alternating least-squares (MCR-ALS) to reveal hidden Raman bands in the fingerprint region. Applying this method to mouse brain tissue, we extracted information on cholesterol and collagen, leveraging their distinctive molecular signatures, as well as on key molecules such as lipids, proteins, water, and nucleic acids. Moreover, the simultaneous detection of second harmonic generation facilitated label-free visualization of organelles, including arachnoid membrane and Rootletin filaments.

Second harmonic generation imaging is a powerful tool for visualizing molecular structures in living organisms without the need for exogenous dyes. However, SHG signal lacks molecular specificity in identifying the source. This study aimed at molecular identification of SHG sources in the mouse brain using a multimodal imaging technique combining SHG and multiplex coherent anti Stokes Raman scattering (CARS) spectroscopy. We performed multimodal imaging in two different regions, the surface and dentate gyrus of the brain tissue. For the brain surface, the SHG signal was recognized through CARS spectrum analysis, indicating its origin in collagen. In the dentate gyrus, CARS images did not unveil corresponding molecular origins; however, morphologically, the SHG signal likely originated from Rootletin within neurons. Overall, Multimodal imaging approach to molecular identification of SHG has the potential to contribute to a comprehensive understanding of the molecular and structural features of the mouse brain. These findings advance label-free imaging techniques and have implications for brain tissue analysis and functional mapping research.

We propose a method to measure enzymatic reactions using the C-D stretching vibrational spectra of CD3 groups that exist next to the site where a chemical bond changes. In the proposed method, temporal changes in the concentrations of the substrate and product are measured from changes in the C-D stretching spectra of the CD3 groups. Vibrational spectra are obtained using multiplex coherent anti-Stokes Raman scattering (CARS) spectroscopy, which can obtain vibrational spectra much faster than spontaneous Raman spectroscopy. We tested the proposed method by examining the dehydrogenation reaction of isopropyl alcohol-1,1,1,3,3,3-D6 (IPA-D6) catalyzed by alcohol dehydrogenase and found that the change in concentrations of the substrate and product, IPA-D6 and acetone-D6, respectively, was successfully measured from the C-D stretching spectra.

Raman spectroscopy is a powerful tool that has found broad adoption in characterizing biological and material systems due to the molecular information embedded in Raman spectra. Some ways this spectroscopic method has been applied include monitoring chemical reactions, identifying molecular structures, and measuring the distribution of chemical species throughout a sample. Traditionally, CCD cameras have found popular use as a detector in Raman spectrometers because spectra can be detected in a single acquisition by dispersing the spectrum across the CCD’s pixel array. While CCDs have aided in the development of Raman spectroscopy as a sensitive detector, their readout mechanism can also introduce experimental limitations through slow frame rates, signal smearing, and blooming of strong signals. We discuss the integration of a scientific complementary metal oxide semiconductor (sCMOS) camera into a Raman spectrometer and conduct a side-by-side comparison with CCD and EMCCD cameras under a range of experimental conditions. The sCMOS readout structure provides performance advantages in bright conditions and enables the rapid acquisition of Raman spectra with millisecond exposure times. Additional, recent, experiments are reviewed that highlight the capability of sCMOS cameras to accurately capture spatially resolved Raman spectra under conditions where other detection technologies might struggle.

Since its first lab demonstration in 2008, multiplex coherent anti-Stokes Raman scattering (MCARS) microspectroscopy in the “long-pulse” regime (50 ps - 1 ns) has become a mature and straightforward technology for label-free bioimaging, offering the high spectral resolution of conventional Raman spectroscopy with reduced acquisition time. In this paper, we review the last developments relative to this technology, in terms of instrumentation (simplified MCARS), data analysis (unsupervised chemical analysis of hyperspectral big data) and biological applications (cell/tissue imaging, time-lapse imaging). It is reminded that running MCARS microspectroscopy in such long-pulse regime allows to get the temporal overlapping of the pump and all the Stokes spectral components without dispersion compensation, enabling simultaneous and effortless hyperspectral operation in both O-H, C-H and fingerprint regions. This simplification of the experimental setup is consolidated by the use of a dual-fiber-output laser source, for which the synchronization between the pump and Stokes pulses can be adjusted by equalizing the fiber lengths of both arms, without the need for a delay line. Chemometric methods as multivariate curve resolution (MCR) are most appropriate for the unsupervised analysis of MCARS hyperspectral data. MCR is an iterative matrix decomposition method, constructing an approximation of data by means of their projection into a subspace guided by different constraints. In this context, we introduce a new approach of cell/tissue imaging, based on a simple workflow and without any phase retrieval computation.

In this study, we report on experiments of spatio-temporal nonlinear frequency conversion in a periodically poled Lithium Niobate (PPLN) crystal designed for second-harmonic generation (SHG). We demonstrated a novel supercontinuum source based on the mixing of second and third-order nonlinearities. We could adjust the (χ(2), χ(3)) nonlinearities by controlling the input laser polarization orientation, the pulse duration and the PPLN crystal temperature. We obtained an ultra-broadband spectrum, ranging from visible to infrared domains, by pumping a 20-mm-long PPLN crystal with a 3 ps pulse at 1030 nm. This broadband pulse was used to achieve direct multiplex Coherent Anti-Stokes Raman Scattering (M-CARS) imaging, without the need for any optical delay line to temporally synchronize the pump and the Stokes waves. Simultaneous vibrational signatures ranging from -3200 cm-1 to -500 cm-1 were obtained. Several filters were placed on the broadband supercontinuum path, to shape the output spectrum between 1030 nm and 1650 nm, before sending it into the microscope. The output spectral analysis allows for the demonstration of multimodal imaging, by using SHG, M-CARS and multiphoton fluorescence processes.

Neuronal primary cilia play important roles in brain development, sensory perception, and neurogenesis. Rootletin, a fibrous protein composed of coiled-coil motifs, is a key structural component of ciliary rootlets and is essential for understanding ciliary functions. However, the precise mechanisms by which Rootletin influences ciliary dynamics and impacts neuronal function remain largely unknown, primarily due to the challenges associated with visualizing these structures. In this study, we present a label-free, rapid, and highly sensitive method for visualizing Rootletin molecules in brain tissue. This platform integrates a second harmonic generation (SHG) microscope with background reduction by a one-step chemical pretreatment. Additionally, we employed coherent anti-Stokes Raman scattering imaging to simultaneously determine the cellular regions and intracellular locations of SHG signals. Applying this multimodal multiphoton imaging approach to the mouse hippocampus revealed that neuronal ciliary rootlets exhibited highly organized and specific intracellular distributions. Moreover, measurements of cultured neurons revealed that ciliary rootlets were detected even in immature neurons. These findings highlight the utility of our label-free imaging platform in developmental and neuroscience research, providing a powerful tool for characterizing ciliary dynamics and neuronal function.

Coumarin-hemicyanine (CHC) hybrid fluorophores bearing an intramolecular nucleophile have been reported as ratiometric fluorescent probes based on the equilibrium between the π-conjugated open and cyclic closed forms. Here, we report red-shifted CHC derivatives that can serve as scaffold dyes of activated fluorescent probes and activated Raman probes. We prepared a series of cyano-substituted CHC derivatives and found that substitution at C-4 of the coumarin moiety is effective in extending the absorption wavelength. The near-infrared (NIR) emission of these 4CN-CHC derivatives exhibits high tissue penetration. Dual-color imaging with 4CN-CHC enabled ratiometric monitoring of the lysosomal pH changes. Furthermore, the coherent anti-Stokes Raman scattering (CARS) signals of 4CN-CHC derivatives could be regulated by modifying the open/closed equilibrium, and the different patterns of CARS spectra enabled us to perform multicomponent imaging. Thus, these 4CN-CHC derivatives expand the repertoire of functional probes for NIR and multiplexed vibrational imaging.

An understanding of the interaction between bright correlated twin beams and matter is important for expanding the applications of optical quantum technology. In particular, it is worthwhile to clarify the advantages of twin-beam correlation in spectroscopic applications, since twin-beam sources that use parametric down-conversion have a wide range of wavelength tunability. Here, we theoretically and experimentally demonstrate that the spectral-focusing technique used in coherent anti-Stokes Raman scattering (CARS) can be implemented by twin beams without the use of ultrashort-pulse lasers. In typical spectral focusing using ultrashort pulses, spectral resolution is improved by narrowing the instantaneous frequency width due to frequency chirping. Similarly, we show theoretically that the formation of a chirp structure within the temporal correlation of twin beams produces an effect similar to typical spectral focusing. To experimentally demonstrate twin-beam spectral focusing, CARS spectroscopy of polyethylene is performed when the time delay between the chirped twin beams is set within the correlation width or it is set outside the correlation width. We observe that the spectral resolution is improved when the correlation is effective. By comparing with theoretical simulations, we confirm that the improvement is caused by frequency chirping.

We visualized β-sheet alignment in intact feathers using polarization-resolved second harmonic generation (SHG) and coherent anti-Stokes Raman scattering (CARS) imaging. SHG signal intensity was strongly dependent on the polarization of the incident laser, while the amide I and III bands in the CARS spectra exhibited moderate polarization-dependent behavior. These results demonstrate that SHG and CARS enable complementary, label-free assessment of structural anisotropy in β-sheet–rich protein fibers, with SHG being especially sensitive to the orientation of the peptide backbone.

Cooled CCD cameras are used widely in spectroscopy, mainly due to their sensitivity, low noise under low-light conditions, and relatively high image and spectral readout rates. Despite their many advantages, CCD cameras have limitations. Particularly under bright-light conditions, such as those encountered with coherent Raman spectroscopies, the finite readout time of CCD chips becomes limiting. Furthermore, when weak signals need to be observed close to intense signals, blooming and smearing limit the signal-to-noise ratios achievable. Scientific complementary mixed oxide (sCMOS)–based sensors are relatively new. Although they still show much higher readout noise than cooled CCDs, their application to spectroscopy is of interest given the higher readout rates and dynamic ranges possible. Here, we evaluate sCMOS sensors for multiplex (50 ps) CARS and spontaneous Raman spectroscopy, making a direct comparison with a state-of-the-art EM-CCD detector. The EM-CCD camera outperforms the sCMOS camera in terms of limits of detection, while the sCMOS camera performs better than the CCD in terms of dynamic range and readout rate. Importantly, an sCMOS camera does not suffer from interference due to blooming and smearing seen with CCD cameras, which enables observation of weak bands (e.g., Raman overtones) close to much more intense signals. Here, we show that, at readout rates > 20–50 Hz, the performances of the two detector types are not substantially different. We anticipate that sCMOS-based cameras will find application for bright spectroscopies, such as multiplex CARS, as well as spontaneous Raman spectroscopy and Raman spectral imaging.

Ceramides, which are recognized as pivotal signaling agents in cell differentiation and proliferation, elicit apoptosis in response to anticancer drugs and stress. Although the 1-hydroxy group of ceramide is considered to play an important role in inducing apoptosis, the detailed mechanism underlying ceramide-induced apoptosis remains elusive. We previously synthesized ceramide derivatives by transforming the 1-hydroxy group into amino and carboxy groups, and assessed their ability to form domains in artificial membranes. In this study, we evaluated the apoptotic activities of these derivatives against living cells. Surprisingly, despite the lack of the 1-hydroxy group, most of the derivatives exhibited apoptotic activity, with some being more active than ceramide. Confocal microscopy using fluorescent-ceramide and coherent anti-Stokes Raman scattering microscopy utilizing deuterated ceramide suggested the formation of large domains on living cell membranes. These findings suggest a potential relationship between domain formation and apoptotic activity. Liquid chromatography-mass spectrometry showed that the differences in apoptotic activity among the derivatives were unlikely attributed to variations in cellular uptake. Consequently, we propose that these ceramide derivatives accumulate and form distinct domains within the cellular membranes, inducing apoptosis. Intracellular ceramides synthesized in response to external stimuli may trigger apoptosis through the same mechanism.

Multiplex coherent anti-Stokes Raman scattering (CARS) microscopy enables label-free spectroscopic imaging over an ultra-broadband range using a supercontinuum light source. However, under high numerical aperture conditions, chromatic aberration causes axial splitting of the illumination spot, leading to missing or significantly weakened peaks. Here, we used Bessel beam illumination to compensate for chromatic aberration without modifying objective lenses. Our numerical and experimental results demonstrate that combining the nonlinear nature of the CARS process with the axial elongation of the Bessel beam spot simultaneously achieves broadband spectroscopy and high spatial resolution imaging. This approach is expected to further expand the applicability of vibrational spectroscopic imaging technique.