On-the-fly Raman microscopy guaranteeing the accuracy of discrimination.

Accelerating the measurement for discrimination of samples, such as classification of cell phenotype, is crucial when faced with significant time and cost constraints. Spontaneous Raman microscopy offers label-free, rich chemical information but suffers from long acquisition time due to extremely small scattering cross-sections. One possible approach to accelerate the measurement is by measuring necessary parts with a suitable number of illumination points. However, how to design these points during measurement remains a challenge. To address this, we developed an imaging technique based on a reinforcement learning in machine learning (ML). This ML approach adaptively feeds back "optimal" illumination pattern during the measurement to detect the existence of specific characteristics of interest, allowing faster measurements while guaranteeing discrimination accuracy. Using a set of Raman images of human follicular thyroid and follicular thyroid carcinoma cells, we showed that our technique requires 3,333 to 31,683 times smaller number of illuminations for discriminating the phenotypes than raster scanning. To quantitatively evaluate the number of illuminations depending on the requisite discrimination accuracy, we prepared a set of polymer bead mixture samples to model anomalous and normal tissues. We then applied a home-built programmable-illumination microscope equipped with our algorithm, and confirmed that the system can discriminate the sample conditions with 104 to 4,350 times smaller number of illuminations compared to standard point illumination Raman microscopy. The proposed algorithm can be applied to other types of microscopy that can control measurement condition on the fly, offering an approach for the acceleration of accurate measurements in various applications including medical diagnosis.

Correction for Extrinsic Background in Raman Hyperspectral Images.

Raman hyperspectral microscopy is a valuable tool in biological and biomedical imaging. Because Raman scattering is often weak in comparison to other phenomena, prevalent spectral fluctuations and contaminations have brought advancements in analytical and chemometric methods for Raman spectra. These chemometric advances have been key contributors to the applicability of Raman imaging to biological systems. As studies increase in scale, spectral contamination from extrinsic background, intensity from sources such as the optical components that are extrinsic to the sample of interest, has become an emerging issue. Although existing baseline correction schemes often reduce intrinsic background such as autofluorescence originating from the sample of interest, extrinsic background is not explicitly considered, and these methods often fail to reduce its effects. Here, we show that extrinsic background can significantly affect a classification model using Raman images, yielding misleadingly high accuracies in the distinction of benign and malignant samples of follicular thyroid cell lines. To mitigate its effects, we develop extrinsic background correction (EBC) and demonstrate its use in combination with existing methods on Raman hyperspectral images. EBC isolates regions containing the smallest amounts of sample materials that retain extrinsic contributions that are specific to the device or environment. We perform classification both with and without the use of EBC, and we find that EBC retains biological characteristics in the spectra while significantly reducing extrinsic background. As the methodology used in EBC is not specific to Raman spectra, correction of extrinsic effects in other types of hyperspectral and grayscale images is also possible.

Differentiability of cell types enhanced by detrending a non-homogeneous pattern in a line-illumination Raman microscope.

A line illumination Raman microscope extracts the underlying spatial and spectral information of a sample, typically a few hundred times faster than raster scanning. This makes it possible to measure a wide range of biological samples such as cells and tissues - that only allow modest intensity illumination to prevent potential damage - within feasible time frame. However, a non-uniform intensity distribution of laser line illumination may induce some artifacts in the data and lower the accuracy of machine learning models trained to predict sample class membership. Here, using cancerous and normal human thyroid follicular epithelial cell lines, FTC-133 and Nthy-ori 3-1 lines, whose Raman spectral difference is not so large, we show that the standard pre-processing of spectral analyses widely used for raster scanning microscopes introduced some artifacts. To address this issue, we proposed a detrending scheme based on random forest regression, a nonparametric model-free machine learning algorithm, combined with a position-dependent wavenumber calibration scheme along the illumination line. It was shown that the detrending scheme minimizes the artifactual biases arising from non-uniform laser sources and significantly enhances the differentiability of the sample states, i.e., cancerous or normal epithelial cells, compared to the standard pre-processing scheme.

Label-Free Evaluation of Maturation and Hepatotoxicity of Human iPSC-Derived Hepatocytes Using Hyperspectral Raman Imaging.

To promote the clinical application of human induced pluripotent stem cell (hiPSC)-derived hepatocytes, a method capable of monitoring regenerative processes and assessing differentiation efficiency without harming or modifying these cells is important. Raman microscopy provides a powerful tool for this as it enables label-free identification of intracellular biomolecules in live samples. Here, we used label-free Raman microscopy to assess hiPSC differentiation into hepatocyte lineage based on the intracellular chemical content. We contrasted these data with similar phenotypes from the HepaRG and from commercially available hiPSC-derived hepatocytes (iCell hepatocytes). We detected hepatic cytochromes, lipids, and glycogen in hiPSC-derived hepatocyte-like cells (HLCs) but not biliary-like cells (BLCs), indicating intrinsic differences in biomolecular content between these phenotypes. The data show significant glycogen and lipid accumulation as early as the definitive endoderm transition. Additionally, we explored the use of Raman imaging as a hepatotoxicity assay for the HepaRG and iCell hepatocytes, with data displaying a dose-dependent reduction of glycogen accumulation in response to acetaminophen. These findings show that the nondestructive and high-content nature of Raman imaging provides a promising tool for both quality control of hiPSC-derived hepatocytes and hepatotoxicity screening.

High-throughput line-illumination Raman microscopy with multislit detection.

Raman microscopy is an emerging tool for molecular imaging and analysis of living samples. Use of Raman microscopy in life sciences is, however, still limited because of its slow measurement speed for spectral imaging and analysis. We developed a multiline-illumination Raman microscope to achieve ultrafast Raman spectral imaging. A spectrophotometer equipped with a periodic array of confocal slits detects Raman spectra from a sample irradiated by multiple line illuminations. A comb-like Raman hyperspectral image is formed on a two-dimensional detector in the spectrophotometer, and a hyperspectral Raman image is acquired by scanning the sample with multiline illumination array. By irradiating a sample with 21 simultaneous illumination lines, we achieved high-throughput Raman hyperspectral imaging of mouse brain tissue, acquiring 1108800 spectra in 11.4 min. We also measured mouse kidney and liver tissue as well as conducted label-free live-cell molecular imaging. The ultrafast Raman hyperspectral imaging enabled by the presented technique will expand the possible applications of Raman microscopy in biological and medical fields.

Raman imaging of rat nonalcoholic fatty liver tissues reveals distinct biomolecular states.

An essential challenge in diagnosing states of nonalcoholic fatty liver disease (NAFLD) is the early prediction of progression from nonalcoholic fatty liver (NAFL) to nonalcoholic steatohepatitis (NASH) before the disease progresses. Histological diagnoses of NAFLD rely on the appearance of anomalous tissue morphologies, and it is difficult to segment the biomolecular environment of the tissue through a conventional histopathological approach. Here, we show that hyperspectral Raman imaging provides diagnostic information on NAFLD in rats, as spectral changes among disease states can be detected before histological characteristics emerge. Our results demonstrate that Raman imaging of NAFLD can be a useful tool for histopathologists, offering biomolecular distinctions among tissue states that cannot be observed through standard histopathological means.

Lipid droplet accumulation and adipophilin expression in follicular thyroid carcinoma.

Follicular neoplasms of the thyroid include follicular thyroid carcinoma (FTC) and follicular thyroid adenoma (FTA). However, the differences in cytological findings between FTC and FTA remain undetermined. Here, we aimed to evaluate the accumulation of lipid droplets (LDs) and the expression of adipophilin (perilipin 2/ADRP/ADFP), a known LD marker, in cultured FTC cells. We also immunohistochemically compared adipophilin expression in the FTC and FTA of resected human thyroid tissues. Cultured FTC (FTC-133 and RO82W-1) possessed increased populations of LDs compared to thyroid follicular epithelial (Nthy-ori 3-1) cells. In vitro treatment with phosphatidylinositol-3-kinase (PI3K)/Akt/mammalian target of rapamycin (mTOR) signaling inhibitors (LY294002, MK2206, and rapamycin) in FTC-133 cells downregulated the PI3K/Akt/mTOR/sterol regulatory element-binding protein 1 (SREBP1) signaling pathway, resulting in a significant reduction in LD accumulation. SREBP1 is a master transcription factor that controls lipid metabolism. Fluorescence immunocytochemistry revealed adipophilin expression in the LDs of FTC-133 cells. Immunohistochemical analysis of surgically resected human thyroid tissues revealed significantly increased expression of adipophilin in FTC compared with FTA and adjacent non-tumorous thyroid epithelia. Taken together, LDs and adipophilin were abundant in cultured FTC; the evaluation of adipophilin expression can help distinguish FTC from FTA in surgical specimens.

Using saturated absorption for superresolution laser scanning transmission microscopy.

We improved the three-dimensional spatial resolution of laser scanning transmission microscopy by exploiting the saturated absorption of dye molecules. The saturated absorption is induced by the high-intensity light irradiation and localises the signal within the centre of the focal spot. Our numerical calculation indicates that the spatial resolution in transmission imaging is significantly improved for both lateral and axial directions using nonlinear transmitted signals induced by saturated absorption. We experimentally demonstrated the improvement of the three-dimensional resolution by observing fine structures of stained rat kidney tissues, which were not able to be visualised by conventional laser scanning transmission microscopy.

Confocal laser scanning microscopy is a powerful technique for three-dimensional imaging to study structures in a specimen. The use of confocal pinhole provides three-dimensional spatial resolution in various types of optical microscopes, such as fluorescence, reflection and scattering. However, in transmission microscopy, the confocal pinhole cannot provide the same effect because the spatial information on the optical axial is not transferred in the imaging system. Therefore, the three-dimensional distribution of light absorbers cannot be observed by laser scanning transmission microscopy. In this paper, we propose the use of saturated absorption to image the three-dimensional distribution of light absorbers in a sample by laser scanning transmission microscopy. The saturated absorption is induced by the high-intensity light irradiation and occurs prominently at the centre of a focal spot. The information of the saturated absorption signal within the focal spot is transferred to the transmitted light, providing the capability of optical sectioning in transmission imaging. In our research, we theoretically and experimentally confirmed that light absorption by dye molecules is saturable at the high illumination intensity, and the saturated absorption signal can be extracted by harmonic demodulation. We obtained the images of a stained rat kidney tissue by selectively detecting the nonlinear transmission signals induced by saturable absorption of the dye molecules. We confirmed the high depth discrimination capability of our technique clearly visualised the fine structures in the specimen that cannot be observed by a conventional laser scanning absorption microscope.

Raman Spectroscopic Assessment of Myocardial Viability in Langendorff-Perfused Ischemic Rat Hearts.

Spontaneous Raman spectroscopy, which senses changes in cellular contents of reduced cytochrome c, could be a powerful tool for label-free evaluation of ischemic hearts. However, undetermined is whether it is applicable to evaluation of myocardial viability in ischemic hearts. To address this issue, we investigated sequential changes in Raman spectra of the subepicardial myocardium in the Langendorff-perfused rat heart before and during ligation of the left coronary artery and its subsequent release and re-ligation. Under 532-nm wavelength excitation, the Raman peak intensity of reduced cytochrome c at 747 cm-1 increased quickly after the coronary ligation, and reached a quasi-steady state within 30 min. Subsequent reperfusion of the heart after a short-term (30-min) ligation that simulates reversible conditions resulted in quick recovery of the peak intensity to the baseline. Further re-ligation resulted in resurgence of the peak intensity to nearly the identical value to the first ischemia value. In contrast, reperfusion after prolonged (120-min) ligation that assumes irreversible states resulted in incomplete recovery of the peak intensity, and re-ligation resulted in inadequate resurgence. Electron microscopic observations confirmed the spectral findings. Together, the Raman spectroscopic measurement for cytochrome c could be applicable to evaluation of viability of the ischemic myocardium without labeling.

Detecting nitrile-containing small molecules by infrared photothermal microscopy.

The use of infrared (IR) photothermal microscopy (IR-PTM) is emerging for imaging chemical substances in various samples. In this research, we demonstrated the use of a nitrile group as a vibrational tag to image target molecules in the low water-background region. We performed IR photothermal imaging of trifluoromethoxy carbonyl cyanide phenylhydrazone (FCCP) in cells and confirmed the high spatial resolution by photothermal detection using visible light as a probe beam. We imaged FCCP-treated HeLa cells and confirmed that the photothermal signal was indeed produced from the vibrational tag in lipid droplets. We also compared the results with nitrile imaging by stimulated Raman scattering (SRS) microscopy. From both the calculated and experimental results, IR-PTM demonstrated a signal-to-noise ratio (SNR) several tens of times better than that of SRS microscopy on the basis of the same power input.

Deep-UV excitation fluorescence microscopy for detection of lymph node metastasis using deep neural network.

Deep-UV (DUV) excitation fluorescence microscopy has potential to provide rapid diagnosis with simple technique comparing to conventional histopathology based on hematoxylin and eosin (H&E) staining. We established a fluorescent staining protocol for DUV excitation fluorescence imaging that has enabled clear discrimination of nucleoplasm, nucleolus, and cytoplasm. Fluorescence images of metastasis-positive/-negative lymph nodes of gastric cancer patients were used for patch-based training with a deep neural network (DNN) based on Inception-v3 architecture. The performance on small patches of the fluorescence images was comparable with that of H&E images. Gradient-weighted class activation mapping analysis revealed the areas where the trained model identified metastatic lesions in the images containing cancer cells. We extended the method to large-size image analysis enabling accurate detection of metastatic lesions. We discuss usefulness of DUV excitation fluorescence imaging with the aid of DNN analysis, which is promising for assisting pathologists in assessment of lymph node metastasis.

Fluorescence-based discrimination of breast cancer cells by direct exposure to 5-aminolevulinic acid.

Protoporphyrin IX-fluorescence measurement is a powerful in situ approach for cancer detection after oral/topical administration of 5-aminolevulinic acid. However, this approach has not been clinically established for breast cancer, probably due to insufficient delivery of 5-aminolevulinic acid to the mammary glands. In the present study, we directly exposed breast cancer cells to 5-aminolevulinic acid to assess their discrimination via protoporphyrin IX-fluorescence. Fluorescence intensity (FI) was measured in the human breast cancer cell lines MCF7 and MDA-MB-231 and breast epithelial cell line MCF10A by confocal microscopy and flow cytometry. After 5-aminolevulinic acid exposure for 2 hours, protoporphyrin IX-FI in MCF7 and MDA-MB-231 cells significantly increased with marked cell-to-cell variability, whereas that in MCF10A cells increased moderately. Combined exposure of the cancer cells to 5-aminolevulinic acid and Ko143, a specific inhibitor of ATP-binding cassette transporter G2, further increased protoporphyrin IX-FI and alleviated the cell-to-cell variability in MCF7 and MDA-MB-231 cells, indicating improvement in the reproducibility and accuracy for fluorescence-based cancer detection. The increased FI by combined administration of these two drugs was also demonstrated in cells obtained via fine needle aspiration from mouse xenograft models inoculated with MDA-MB-231 cells. Furthermore, a cutoff value for increased protoporphyrin IX-FI ratio, before and after exposure to these drugs, clearly discriminated between cancer and noncancer cells. Taken together, direct exposure to 5-aminolevulinic acid and Ko143 may be a promising strategy for efficient fluorescence-based detection of breast cancer cells ex vivo using fine needle aspiration.

Terbium ion as RNA tag for slide-free pathology with deep-ultraviolet excitation fluorescence.

Deep-ultraviolet excitation fluorescence microscopy has enabled molecular imaging having an optical sectioning capability with a wide-field configuration and its usefulness for slide-free pathology has been shown in recent years. Here, we report usefulness of terbium ions as RNA-specific labeling probes for slide-free pathology with deep-ultraviolet excitation fluorescence. On excitation in the wavelength range of 250-300 nm, terbium ions emitted fluorescence after entering cells. Bright fluorescence was observed at nucleoli and cytoplasm while fluorescence became weak after RNA decomposition by ribonuclease prior to staining. It was also found that the fluorescence intensity at nucleoplasm increased with temperature during staining and that this temperature-dependent behavior resembled temperature-dependent hypochromicity of DNA due to melting. These findings indicated that terbium ions stained single-stranded nucleic acid more efficiently than double-stranded nucleic acid. We further combined terbium ions and DNA-specific dyes for dual-color imaging. In the obtained image, nucleolus, nucleoplasm, and cytoplasm were distinguished. We demonstrated the usefulness of dual-color imaging for rapid diagnosis of surgical specimen by showing optical sectioning of unsliced tissues. The present findings can enhance deep-ultraviolet excitation fluorescence microscopy and consequently expand the potential of fluorescence microscopy in life sciences.

High-Resolution Raman Microscopic Detection of Follicular Thyroid Cancer Cells with Unsupervised Machine Learning.

We use Raman microscopic images with high spatial and spectral resolution to investigate differences between human follicular thyroid (Nthy-ori 3-1) and follicular thyroid carcinoma (FTC-133) cells, a well-differentiated thyroid cancer. Through comparison to classification of single-cell Raman spectra, the importance of subcellular information in the Raman images is emphasized. Subcellular information is extracted through a coarse-graining of the spectra at high spatial resolution (∼1.7 µm2), producing a set of characteristic spectral groups representing locations having similar biochemical compositions. We develop a cell classifier based on the frequencies at which the characteristic spectra appear within each of the single cells. Using this classifier, we obtain a more accurate (89.8%) distinction of FTC-133 and Nthy-ori 3-1, in comparison to single-cell spectra (77.6%). We also infer which subcellular components are important to cellular distinction; we find that cancerous FTC-133 cells contain increased populations of lipid-containing components and decreased populations of cytochrome-containing components relative to Nthy-ori 3-1, and that the regions containing these contributions are largely outside the cell nuclei. In addition to increased classification accuracy, this approach provides rich subcellular information about biochemical differences and cellular locations associated with the distinction of the normal and cancerous follicular thyroid cells.

High-Throughput Cell Imaging and Classification by Narrowband and Low-Spectral-Resolution Raman Microscopy.

We investigated the use of narrowband Raman spectra for rapid label-free molecular imaging aimed at cell classification using principal component regression and linear discriminant analysis. In the classification of breast nontumorigenic epithelial and cancer cell lines, the classification accuracies using a spectral range of 100 cm-1 were equivalent to or better than that with using the fingerprint and high-wavenumber regions. Narrowing the Raman spectral range for analysis allows reduction of the charge-coupled device (CCD) pixels required for spectrum detection, resulting in the improvement of image acquisition speed with adequate classification accuracy. Our measurements revealed that the wavenumber region at 1397-1501 cm-1 can provide molecular information sufficient for cell classification without causing notable errors in the baseline-correction. A spectral resolution of ∼9 cm-1 was found to be sufficient to provide high accuracy in cell classification, which allowed us to apply pixel binning at the CCD readout for further acceleration of the imaging speed. As a result, the acquisition time for a 1200 × 1500 pixels Raman hyperspectral image at 1397-1501 cm-1 was reduced to 21 min. Under this condition, different cell lines were classified at accuracies higher than 90%. The presented approach will improve throughput of cell and tissue analysis and classification using Raman spectroscopy and extend practical uses of Raman imaging in biology and medicine.

Bioconjugation strategy for cell surface labelling with gold nanostructures designed for highly localized pH measurement.

Regulation of intracellular pH is critically important for many cellular functions. The quantification of proton extrusion in different types of cells and physiological conditions is pivotal to fully elucidate the mechanisms of pH homeostasis. Here we show the use of gold nanoparticles (AuNP) to create a high spatial resolution sensor for measuring extracellular pH in proximity of the cell membrane. We test the sensor on HepG2 liver cancer cells and MKN28 gastric cancer cells before and after inhibition of Na+/H+ exchanger. The gold surface conjugation strategy is conceived with a twofold purpose: i) to anchor the AuNP to the membrane proteins and ii) to quantify the local pH from AuNP using surface enhanced Raman spectroscopy (SERS). The nanometer size of the cell membrane anchored sensor and the use of SERS enable us to visualize highly localized variation of pH induced by H+ extrusion, which is particularly upregulated in cancer cells.

Deep-ultraviolet Raman scattering spectroscopy of monolayer WS2.

Raman scattering measurements of monolayer WS2 are reported as a function of the laser excitation energies from the near-infrared (1.58 eV) to the deep-ultraviolet (4.82 eV). In particular, we observed several strong Raman peaks in the range of 700∼850 cm-1 with the deep-ultraviolet laser lights (4.66 eV and 4.82 eV). Using the first-principles calculations, these peaks and other weak peaks were appropriately assigned by the double resonance Raman scattering spectra of phonons around the M and K points in the hexagonal Brillouin zone. The relative intensity of the first-order [Formula: see text] to A1g peak changes dramatically with the 1.58 eV and 2.33 eV laser excitations, while the comparable relative intensity was observed for other laser energies. The disappearance of the [Formula: see text] peak with the 1.58 eV laser light comes from the fact that valley polarization of the laser light surpasses the [Formula: see text] mode since the [Formula: see text] mode is the helicity-exchange Raman mode. On the other hand, the disappearance of the A1g peak with the 2.33 eV laser light might be due to the strain effect on the electron-phonon matrix element.

Label-free Molecular Imaging and Analysis by Raman Spectroscopy.

Raman scattering of a cell conveys the intrinsic information inherent to chemical structures of biomolecules. The spectroscopy of Raman scattering, or Raman spectroscopy, allows label-free and quantitative molecular sensing of a biological sample in situ without disruption. For the last five decades Raman spectroscopy has been widely utilized in biological research fields. However, it is just within the latest decade that molecular imaging and discrimination of living cells and tissues have become practically available. Here we overview recent progress in Raman spectroscopy and its application to life sciences. We discuss imaging of functional molecules in living cells and tissues; e.g., cancer cells and ischemic or infarcted hearts, together with a number of studies in the biomedical fields. We further explore comprehensive understandings of a complex spectrum by multivariate analysis for, e.g., accurate peripheral nerve detection, and characterization of the histological differences in the healing process of myocardial infarct. Although limitations still remain, e.g., weakness of the scattering intensity and practical difficulty in comprehensive molecular analysis, continuous progress in related technologies will allow wider use of Raman spectroscopy for biomedical applications.

Raman micro-spectroscopy as a viable tool to monitor and estimate the ionic transport in epithelial cells.

The typical response to the lowering of plasma Na+ concentration and blood pressure in our body involves the release of aldosterone from the adrenal glands, which triggers the reabsorption of sodium in the kidney. Although the effects of aldosterone on this physiological mechanism were extensively studied in the past decades, there are still some aspects to be fully elucidated. In the present study, we propose for the first time a new approach based on Raman spectroscopy to monitor the ionic activity in aldosterone-treated A6 renal epithelial cells. This spectroscopic technique is capable of probing the cells through their thickness in a non-destructive and nimble way. The spectroscopic variations of the Raman bands associated to the O-H stretching of water were correlated to the variations of ionic concentration in the intracellular and extracellular fluids. The increase of Na+ concentration gradients was clearly visualized in the cytosol of aldosterone-treated cells. The enhancement of the Na+ current density induced by aldosterone was estimated from the variation of the ionic chemical potential across the intracellular space. In addition, the variation of the O-H Raman bands of water was used to quantify the cell thickness, which was not affected by aldosterone.

Rapid and accurate peripheral nerve imaging by multipoint Raman spectroscopy.

Raman spectroscopy allows label-free, minimally invasive, and accurate detection of peripheral nerves. However, the conventional Raman imaging technique is time-consuming when measuring a large area of a sample. Establishing a method for rapidly acquiring spatial distribution of a bundle of peripheral nerve fibers is an essential step for Raman spectroscopy towards application in clinical surgery. Here we present a multipoint Raman spectroscopic technique for rapid peripheral nerve imaging. In only 5 seconds, spectra at 32 points situated on ex vivo rat peripheral nerve bundles and adjoining connective tissues were acquired. Principal component regression and discriminant analysis of spectra revealed that the sensitivity, specificity and accuracy for nerve detection were 85.8%, 96.0%, and 90.8%, respectively. Of 158 peripheral nerves, 152 (96.2%) showed ratio of the number of nerve-positive prediction points to the total measurement points being 0.4 or larger, whereas 119 (99.2%) connective tissues among 120 showed ratio smaller than 0.4. Based on the ratio and a bright-field image of the sample, accurate visualization of peripheral nerves was implemented. The results indicated that the multipoint Raman spectroscopic technique is capable of rapid and accurate peripheral nerve imaging.