Shuya Satoh

Label-free visualization of acetaminophen-induced liver injury by high-speed stimulated Raman scattering spectral microscopy and multivariate image analysis

We recently established a high‐speed, label‐free, spectral imaging method based on stimulated Raman scattering (SRS). This method enables examination of cellular features within relatively short periods, thus enabling new imaging applications in pathology. Previously, we reported on label‐free visualization of mouse tissue using SRS spectral microscopy combined with multivariate image analysis, but the feasibility of applying this approach to diseased tissues with diverse morphology and irregular chemical species has not been examined. We, therefore, assessed acetaminophen‐induced liver injury to evaluate the potential use of Raman spectral microscopy for visualizing histopathologic specimens. Acetaminophen‐overdosed mouse liver was prepared and the pathologic changes including centrilobular necrosis were confirmed. Multi‐colored images were reconstructed through principal component analysis (PCA) of a multi‐band SRS dataset, which provided rich information compared with a monochrome single‐band SRS dataset. A wide view of the multi‐colored principal component (PC) images showed the distribution of cellular constituents, which was similar to that observed by fat staining. In addition, different types of cells in liver parenchyma were also demonstrated. In conclusion, the combination of SRS spectral microscopy and PCA has the potential to reveal both the morphological and chemical features of specimens and therefore has potential utility in diagnostic pathology.

Visualization of cancer-related chemical components in mouse pancreas tissue by tapping-mode scanning probe electrospray ionization mass spectrometry

Mass spectrometry imaging is an informative approach for the comprehensive analysis of multiple components inside biological specimens. We used novel tapping-mode scanning probe electrospray ionization mass spectrometry method to visualize cancer-related chemical components in the mouse pancreas tissue section at a sampling pitch of 100 µm. Positive ion mode measurements from m/z 100 to 1500 resulted in the visualization of multiple components that are tentatively assigned as polyamines, lipids and proteins. Their signal intensities inside the cancerous and the non-cancerous regions were found to be significantly different by the two-sample t-test.

High-image quality, high-resolution camera with high sensitivity up to 1,100 nm

We have developed three types of prototype cameras based on multi-purpose and DSLR models equipped with full-frame (35 mm) CMOS sensors, i.e. modified versions of the Canon ME20F-SH, EOS-1D X, and EOS 5DS. We have also made a prototype interchangeable lens that is highly transmissive in the visible and near infrared (NIR) region, based on a Canon EF 100-400 mm lens. High signal-to-noise (S/N) ratio spectral images up to 1,100 nm were observed using the following tools: the modified ME20F-SH, which features a monochromatic sensor with a 19 μm pixel pitch; the modified EF 100-400 mm lens; NIR optical bandpass filters ranging from 1,000 to 1,200 nm with 25 nm bandwidth; and a short wavelength cut-off filter that blocks light below 840 nm. This camera and lens, with a liquid crystal tunable filter (LCTF) attached, was able to clearly observe absorption bands attributed to hydroxyl groups near 960 nm in aqueous sucrose solutions. This experimental set was also applied to discrimination between dairy cream and whipped topping in combination with multivariate analysis; multi spectral data set obtained by this experimental set was processed by principal component analysis followed by a modified algorithm of independent component analysis. The Canon EOS-1D X’s 18.1 million pixel sensor was modified by introducing NIR pixels in addition to its RGB pixels. The modified EOS-1D X and the lens were able to acquire both high-image-quality photography and high-resolution NDVI (Normalized Difference Vegetation Index) in the observation of a lawn with a single press of the shutter button. This camera and lens set was also applied to observe hair in dried seaweed, producing a high-resolution NIR image in which each hair was visualized. Furthermore, a sharper image of the hair was obtained using the modified EOS 5DS and the LCTF.

Visualization of acetaminophen-induced liver injury by time-of-flight secondary ion mass spectrometry

Time-of-flight secondary ion mass spectrometry (MS) provides secondary ion images that reflect distributions of substances with sub-micrometer spatial resolution. To evaluate the use of time-of-flight secondary ion MS to capture subcellular chemical changes in a tissue specimen, we visualized cellular damage showing a three-zone distribution in mouse liver tissue injured by acetaminophen overdose. First, we selected two types of ion peaks related to the hepatocyte nucleus and cytoplasm using control mouse liver. Acetaminophen-overdosed mouse liver was then classified into three areas using the time-of-flight secondary ion MS image of the two types of peaks, which roughly corresponded to established histopathological features. The ion peaks related to the cytoplasm decreased as the injury became more severe, and their origin was assumed to be mostly glycogen based on comparison with periodic acid-Schiff staining images and reference compound spectra. This indicated that the time-of-flight secondary ion MS image of the acetaminophen-overdosed mouse liver represented the chemical changes mainly corresponding to glycogen depletion on a subcellular scale. In addition, this technique also provided information on lipid species related to the injury. These results suggest that time-of-flight secondary ion MS has potential utility in histopathological applications.

High-speed stimulated Raman spectral imaging for digital staining of mouse cancer tissues

We previously reported the combination of the high-speed stimulated Raman scattering (SRS) microscope and the multivariate analysis (principal component analysis and independent component analysis) for the tissue imaging. The results indicated the visualization of tissue components without chemical staining. Here, we report the multi-area observation of the tumor-grafted mouse tissue based on the same approach. The tumor-grafted mouse (balb/cAcJ nu/nu) was prepared by injection of human pancreatic carcinoma cell line (SUIT-2) into the tail of pancreas. Both of the pancreas cancer and the liver metastasis were harvested and fixed in formalin. Tissues were cryo-sectioned with a thickness of 100 μm and observed. The multispectral images (130 μm square, 500 x 500 pixels) of C-H vibration range from 2800 to 3100cm-1 (91 different Raman shift images) were obtained at a frame rate of 30 frames/sec. The data acquisition both of pancreas and liver were continued for the 48 adjacent areas for the observation both of cancerous and non-cancerous region, respectively. All the datasets were combined to analyze for multivariate analysis. We propose a protocol for drastic data reduction, which we found to give reproducible results and allows fast calculation of ICA. The independent component images indicated the different shapes and compositions between the cancerous region and the non-cancerous region.

Toward digital staining using stimulated Raman scattering and statistical machine learning

Stimulated Raman scattering (SRS) spectral microscopy is a promising imaging method, based on vibrational spectroscopy, which can visualize biological tissues with chemical specificity. SRS spectral microscopy has been used to obtain two-dimensional spectral images of rat liver tissue, three-dimensional images of a vessel in rat liver, and in vivo spectral images of mouse ear skin. Various multivariate analysis techniques, such as principal component analysis and independent component analysis, have been used to obtain spectral images. In this study, we propose a digital staining method. This method uses SRS spectra and statistical machine learning that makes use of prior knowledge of spectral peaks and their two-dimensional distributional patterns corresponding to the composition of tissue samples. The method selects spectral peaks on the basis of Mahalanobis distance, which is defined as the ratio of inter-group variation to intragroup variation. We also make use of higher-order local autocorrelations as feature values for two-dimensional distributional patterns. This combination of techniques allows groups corresponding to different intracellular structures to be clearly discriminated in the multidimensional feature space. We investigate the performance of our method on mouse liver tissue samples and show that the proposed method can digitally stain each intracellular structure such as cell nuclei, cytoplasm, and erythrocytes separately and clearly without time-consuming chemical staining processes. We anticipate that our method could be applied to computer-aided pathological diagnosis.