A Fully Automated Artificial Intelligence System to Assist Pathologists' Diagnosis to Predict Histologically High-grade Urothelial Carcinoma from Digitized Urine Cytology Slides Using Deep Learning

BACKGROUND

Urine cytology, although a useful screening method for urothelial carcinoma, lacks sensitivity. As an emerging technology, artificial intelligence (AI) improved image analysis accuracy significantly.

OBJECTIVE

To develop a fully automated AI system to assist pathologists in the histological prediction of high-grade urothelial carcinoma (HGUC) from digitized urine cytology slides.

DESIGN, SETTING, AND PARTICIPANTS

We digitized 535 consecutive urine cytology slides for AI use. Among these slides, 181 were used for AI development, 39 were used as AI test data to identify HGUC by cell-level classification, and 315 were used as AI test data for slide-level classification.

OUTCOME MEASUREMENTS AND STATISTICAL ANALYSIS

Out of the 315 slides, 171 were collected immediately prior to bladder biopsy or transurethral resection of bladder tumor, and then outcomes were compared with the histological presence of HGUC in the surgical specimen. The primary aim was to compare AI prediction of the histological presence of HGUC with the pathologist's histological diagnosis of HGUC. Secondary aims were to compare the time required for AI evaluation and concordance between the AI's classification and pathologist's cytology diagnosis.

RESULTS AND LIMITATIONS

The AI capability for predicting the histological presence of HGUC was 0.78 for the area under the curve. Comparing the AI predictive performance with pathologists' diagnosis, the AI sensitivity of 63% for histological HGUC prediction was superior to a pathologists' cytology sensitivity of 46% (p = 0.0037). On the contrary, there was no significant difference between the AI specificity of 83% and pathologists' specificity of 89% (p = 0.13), and AI accuracy of 74% and pathologists' accuracy of 68% (p = 0.08). The time required for AI evaluation was 139 s. With respect to the concordance between the AI prediction and pathologist's cytology diagnosis, the accuracy was 86%. Agreements with positive and negative findings were 92% and 84%, respectively.

CONCLUSIONS

We developed a fully automated AI system to assist pathologists' histological diagnosis of HGUC using digitized slides. This AI system showed significantly higher sensitivity than a board-certified cytopathologist and may assist pathologists in making urine cytology diagnoses, reducing their workload.

PATIENT SUMMARY

In this study, we present a deep learning-based artificial intelligence (AI) system that classifies urine cytology slides according to the Paris system. An automated AI system was developed and validated with 535 consecutive urine cytology slides. The AI predicted histological high-grade urothelial carcinoma from digitized urine cytology slides with superior sensitivity than pathologists, while maintaining comparable specificity and accuracy.

Myofibroblasts impair myocardial impulse propagation by heterocellular connexin43 gap-junctional coupling through micropores

Aim: Composite population of myofibroblasts (MFs) within myocardial tissue is known to alter impulse propagation, leading to arrhythmias. However, it remains unclear whether and how MFs alter their propagation patterns when contacting cardiomyocytes (CMs) without complex structural insertions in the myocardium. We attempted to unveil the effects of the one-sided, heterocellular CM-MF connection on the impulse propagation of CM monolayers without the spatial insertion of MFs as an electrical or mechanical obstacle. Methods and results: We evaluated fluo8-based spatiotemporal patterns in impulse propagation of neonatal rat CM monolayers cultured on the microporous membrane having 8-μm diameter pores with co-culture of MFs or CMs on the reverse membrane side (CM-MF model or CM-CM model, respectively). During consecutive pacing at 1 or 2 Hz, the CM monolayers exhibited forward impulse propagation from the pacing site with a slower conduction velocity (θ) and a larger coefficient of directional θ variation in the CM-MF model than that in the CM-CM model in a frequency-dependent manner (2 Hz >1 Hz). The localized placement of an MF cluster on the reverse side resulted in an abrupt segmental depression of the impulse propagation of the upper CM layer, causing a spatiotemporally non-uniform pattern. Dye transfer of the calcein loaded in the upper CM layer to the lower MF layer was attenuated by the gap-junction inhibitor heptanol. Immunocytochemistry identified definitive connexin 43 (Cx43) between the CMs and MFs in the membrane pores. MF-selective Cx43 knockdown in the MF layer improved both the velocity and uniformity of propagation in the CM monolayer. Conclusion: Heterocellular Cx43 gap junction coupling of CMs with MFs alters the spatiotemporal patterns of myocardial impulse propagation, even in the absence of spatially interjacent and mechanosensitive modulations by MFs. Moreover, MFs can promote pro-arrhythmogenic impulse propagation when in face-to-face contact with the myocardium that arises in the healing infarct border zone.

Deep UV-excited fluorescence microscopy installed with CycleGAN-assisted image translation enhances precise detection of lymph node metastasis towards rapid intraoperative diagnosis

Rapid and precise intraoperative diagnosing systems are required for improving surgical outcomes and patient prognosis. Because of the poor quality and time-intensive process of the prevalent frozen section procedure, various intraoperative diagnostic imaging systems have been explored. Microscopy with ultraviolet surface excitation (MUSE) is an inexpensive, maintenance-free, and rapid imaging technique that yields images like thin-sectioned samples without sectioning. However, pathologists find it nearly impossible to assign diagnostic labels to MUSE images of unfixed specimens; thus, AI for intraoperative diagnosis cannot be trained in a supervised learning manner. In this study, we propose a deep-learning pipeline model for lymph node metastasis detection, in which CycleGAN translate MUSE images of unfixed lymph nodes to formalin-fixed paraffin-embedded (FFPE) sample, and diagnostic prediction is performed using deep convolutional neural network trained on FFPE sample images. Our pipeline yielded an average accuracy of 84.6% when using each of the three deep convolutional neural networks, which is a 18.3% increase over the classification-only model without CycleGAN. The modality translation to FFPE sample images using CycleGAN can be applied to various intraoperative diagnostic imaging systems and eliminate the difficulty for pathologists in labeling new modality images in clinical sites. We anticipate our pipeline to be a starting point for accurate rapid intraoperative diagnostic systems for new imaging modalities, leading to healthcare quality improvement.

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.

Generation of myocyte agonal Ca2+ waves and contraction bands in perfused rat hearts following irreversible membrane permeabilisation

Although irreversible cardiomyocyte injury provokes intracellular Ca²⁺ ([Ca²⁺]_i) overload, the underlying dynamics of this response and its effects on cellular morphology remain unknown. We therefore visualised rapid-scanning confocal fluo4-[Ca²⁺]_i dynamics and morphology of cardiomyocytes in Langendorff-perfused rat hearts following saponin-membrane permeabilisation. Our data demonstrate that 0.4% saponin-treated myocytes immediately exhibited high-frequency Ca²⁺ waves (131.3 waves/min/cell) with asynchronous, oscillatory contractions having a mean propagation velocity of 117.8 μm/s. These waves slowly decreased in frequency, developed a prolonged decay phase, and disappeared in 10 min resulting in high-static, fluo4-fluorescence intensity. The myocytes showing these waves displayed contraction bands, i.e., band-like actin-fibre aggregates with disruption of sarcomeric α-actinin. The contraction bands were not attenuated by the abolition of Ca²⁺ waves under pretreatment with ryanodine plus thapsigargin, but were partially attenuated by the calpain inhibitor MDL28170, while mechanical arrest of the myocytes by 2,3-butanedione monoxime completely attenuated contraction-band formation. The depletion of adenosine 5'-triphosphate by the mitochondrial electron uncoupler carbonyl cyanide 4-trifluoromethoxy phenylhydrazone also attenuated Ca²⁺ waves and contraction bands. Overall, saponin-induced myocyte [Ca²⁺]_i overload provokes agonal Ca²⁺ waves and contraction bands. Contraction bands are not the direct consequence of the waves but are caused by cross-bridge interactions of the myocytes under calpain-mediated proteolysis.

5-Aminolevulinic Acid-Induced Protoporphyrin IX Fluorescence Imaging for Tumor Detection: Recent Advances and Challenges

5-Aminolevulinic acid (5-ALA) is a natural amino acid and a precursor of heme and chlorophyll. Exogenously administered 5-ALA is metabolized into protoporphyrin IX (PpIX). PpIX accumulates in cancer cells because of the low activity of ferrochelatase, an enzyme that metabolizes PpIX to heme. High expression of 5-ALA influx transporters, such as peptide transporters 1/2, in cancer cells also enhances PpIX production. Because PpIX radiates red fluorescence when excited with blue/violet light, 5-ALA has been used for the visualization of various tumors. 5-ALA photodynamic diagnosis (PDD) has been shown to improve the tumor removal rate in high-grade gliomas and non-muscular invasive bladder cancers. However, 5-ALA PDD remains a challenge as a diagnostic method because tissue autofluorescence interferes with PpIX signals in cases where tumors emit only weak signals, and non-tumorous lesions, such as inflammatory sites, tend to emit PpIX fluorescence. Here, we review the current outline of 5-ALA PDD and strategies for improving its diagnostic applicability for tumor detection, focusing on optical techniques and 5-ALA metabolic pathways in both viable and necrotic tumor tissues.

Accumulation of Uroporphyrin I in Necrotic Tissues of Squamous Cell Carcinoma after Administration of 5-Aminolevulinic Acid

5-aminolevulinic acid (5-ALA)-induced protoporphyrin IX (PpIX) fluorescence is widely used for the intraoperative detection of malignant tumors. However, the fluorescence emission profiles of the accompanying necrotic regions of these tumors have yet to be determined. To address this, we performed fluorescence and high-performance liquid chromatography (HPLC) analyses of necrotic tissues of squamous cancer after 5-ALA administration. In resected human lymph nodes of metastatic squamous cell carcinoma, we found a fluorescence peak at approximately 620 nm in necrotic lesions, which was distinct from the PpIX fluorescence peak at 635 nm for viable cancer lesions. Necrotic lesions obtained from a subcutaneous xenograft model of human B88 oral squamous cancer also emitted the characteristic fluorescence peak at 620 nm after light irradiation: the fluorescence intensity ratio (620 nm/635 nm) increased with the energy of the irradiation light. HPLC analysis revealed a high content ratio of uroporphyrin I (UPI)/total porphyrins in the necrotic cores of murine tumors, indicating that UPI is responsible for the 620 nm peak. UPI accumulation in necrotic tissues after 5-ALA administration was possibly due to the failure of the heme biosynthetic pathway. Taken together, fluorescence imaging of UPI after 5-ALA administration may be applicable for the evaluation of tumor necrosis.

Urine cell image recognition using a deep-learning model for an automated slide evaluation system

OBJECTIVES

To develop a classification system for urine cytology with artificial intelligence (AI) using a convolutional neural network algorithm that classifies urine cell images as negative (benign) or positive (atypical or malignant).

PATIENTS AND METHODS

We collected 195 urine cytology slides from consecutive patients with a histologically confirmed diagnosis of urothelial cancer (between January 2016 and December 2017). Two certified cytotechnologists independently evaluated and labelled each slide; 4637 cell images with concordant diagnoses were selected, including 3128 benign cells (negative), 398 atypical cells, and 1111 cells that were malignant or suspicious for malignancy (positive). This pathologically confirmed labelled dataset was used to represent the ground truth for AI training/validation/testing. Customized CutMix (CircleCut) and Refined Data Augmentation were used for image processing. The model architecture included EfficientNet B6 and Arcface. We used 80% of the data for training and validation (4:1 ratio) and 20% for testing. Model performance was evaluated with fivefold cross-validation. A receiver-operating characteristic (ROC) analysis was used to evaluate the binary classification model. Bayesian posterior probabilities for the AI performance measure (Y) and cytotechnologist performance measure (X) were compared.

RESULTS

The area under the ROC curve was 0.99 (95% confidence interval [CI] 0.98-0.99), the highest accuracy was 95% (95% CI 94-97), sensitivity was 97% (95% CI 95-99), and specificity was 95% (95% CI 93-97). The accuracy of AI surpassed the highest level of cytotechnologists for the binary classification [Pr(Y > X) = 0.95]. AI achieved >90% accuracy for all cell subtypes. In the subgroup analysis based on the clinicopathological characteristics of patients who provided the test cells, the accuracy of AI ranged between 89% and 97%.

CONCLUSION

Our novel AI classification system for urine cytology successfully classified all cell subtypes with an accuracy of higher than 90%, and achieved diagnostic accuracy of malignancy superior to the highest level achieved by cytotechnologists.

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.

5-ALA-assistant automated detection of lymph node metastasis in gastric cancer patients

BACKGROUND

5-aminolevulinic acid (5-ALA) has been utilized for cancer diagnosis as a fluorescence probe. We have reported the feasibility of 5-ALA-induced protoporphyrin IX (PpIX) fluorescence for detecting lymph node (LN) metastasis in gastrointestinal malignancies. However, a major barrier to the fluorescence diagnosis has been that the evaluation has been highly dependent on the observers. In this study, we examined the validity of a developed device for automated detection without subjectivity.

METHODS

Gastric cancer patients who received oral administration of 5-ALA (20 mg/kg) prior to surgery were enrolled. For a total of 323 LNs obtained from 64 patients, the diagnostic results of the device were compared to those of conventional histopathological examination based on hematoxylin-and-eosin-stained slides. The accuracy with the device was compared to that of stereoscopic detection with conventional fluorescence microscopy for 211 LNs from 42 patients. We used two types of image processing that we previously developed to eliminate autofluorescence of background tissues: differential and ratio methods.

RESULTS

For detection of metastasis in 323 LNs, the areas under the receiver operating characteristic curves with the differential method and ratio method were 0.921 and 0.909, respectively. The sensitivity, specificity, and accuracy with the differential method were 78.0%, 96.8%, and 94.4%; while those with the ratio method were 78.0%, 96.1%, and 93.8%, respectively. In 211 LN analysis, the diagnostic accuracy with the device was comparable to that of stereoscopic examination.

CONCLUSION

Our device for automated detection of LN metastasis using 5-ALA can be a useful tool for intraoperative diagnosis.

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.

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.

Spatiotemporally Non-Uniform Ca2+ Dynamics of Cardiac Purkinje Fibers in Mouse Myocardial Infarct

Surviving Purkinje fibers in myocardial infarct are regarded as an important substrate in arrhythmogenesis. However, poorly understood are functional properties of Purkinje fibers in the infarcted heart. We sought to visualize intracellular Ca²⁺ ([Ca²⁺]_i) dynamics of Purkinje fiber networks in the mouse myocardial infarct. Using 3- to 4-day-old or 7- to 9-day-old infarcted hearts after the left coronary-artery ligation corresponding, respectively, to acute or healing phase, we conducted rapid fluo4-fluorescence imaging on the endocardial surface of the left ventricular septum by macro-zoom fluorescence microscopy and rapid-scanning confocal microscopy. In contrast with the intact heart, where uniform Ca²⁺ transients propagated rapidly, the infarcted heart exhibited slow, non-uniform impulse propagations. On confocal microscopy, Purkinje fibers in the peri-infarct zone exhibited non-uniform [Ca²⁺]_i dynamics: beat-to-beat alternans of the Ca²⁺ transient amplitude in and among the individual fibers, whereas the intact fibers exhibited uniform Ca²⁺ transients. Such non-uniform [Ca²⁺]_i dynamics were more conspicuous in the acute infarcted hearts than in the healing ones. In accordance with [Ca²⁺]_i dynamics, fixed fluo4-loaded heart preparations exhibited definitive connexin-40 plaques in the peri-infarct Purkinje fibers, whereas the subjacent myocardium presented coagulative necrosis and granulation tissues, respectively. The surviving Purkinje fibers in the peri-infarct zone exhibited non-uniform [Ca²⁺]_i dynamics, which may lead to arrhythmogenesis.

Efficient fluorescence detection of protoporphyrin IX in metastatic lymph nodes of murine colorectal cancer stained with indigo carmine

Protoporphyrin IX (PpIX), a biochemical converted from 5-aminolevulinc acid (5-ALA) in living cells, is useful for intraoperative fluorescent detection of cancer metastasis in lymph nodes (LNs). However, unknown is whether the fluorescence of PpIX can be detected in the LNs when they coexist with indigo carmine, a blue dye commonly used for identification of sentinel LNs during surgery. To address this issue, we sought to evaluate the diagnostic usefulness of PpIX fluorescence in the presence of indigo carmine in a mouse LN metastasis model of rectal cancer after administration of 5-ALA. Spectral analysis of pure chemicals revealed that the absorption spectrum of indigo carmine widely overlapped with the fluorescence spectrum of PpIX specifically at the peak of 632nm, a common emission wavelength for detecting PpIX, but not at the other peak of 700nm. Due to such spectral overlap, the PpIX fluorescence intensity was significantly attenuated by mixture with indigo carmine at 632nm, but not at 700nm. Accordingly, fluorescent measurements of the mouse metastatic LN revealed more intense presentation of PpIX at 700nm than at 632nm, indicating that the diagnostic usefulness is greater at 700nm than at 632nm for the indigo carmine-dyed LNs after administration of 5-ALA. From these observations, we propose that the fluorescence measurement is more efficient at 700nm than at 632nm for detection of PpIX in metastatic LNs stained with indigo carmine.

Simplified and optimized multispectral imaging for 5-ALA-based fluorescence diagnosis of malignant lesions

5-aminolevulinic acid (5-ALA)-based fluorescence diagnosis is now clinically applied for accurate and ultrarapid diagnosis of malignant lesions such as lymph node metastasis during surgery. 5-ALA-based diagnosis evaluates fluorescence intensity of a fluorescent metabolite of 5-ALA, protoporphyrin IX (PPIX); however, the fluorescence of PPIX is often affected by autofluorescence of tissue chromophores, such as collagen and flavins. In this study, we demonstrated PPIX fluorescence estimation with autofluorescence elimination for 5-ALA-based fluorescence diagnosis of malignant lesions by simplified and optimized multispectral imaging. We computationally optimized observation wavelength regions for the estimation of PPIX fluorescence in terms of minimizing prediction error of PPIX fluorescence intensity in the presence of typical chromophores, collagen and flavins. By using the fluorescence intensities of the optimized wavelength regions, we verified quantitative detection of PPIX fluorescence by using chemical mixtures of PPIX, flavins, and collagen. Furthermore, we demonstrated detection capability by using metastatic and non-metastatic lymph nodes of colorectal cancer patients. These results suggest the potential and usefulness of the background-free estimation method of PPIX fluorescence for 5-ALA-based fluorescence diagnosis of malignant lesions, and we expect this method to be beneficial for intraoperative and rapid cancer diagnosis.

Highly sensitive fluorescence detection of metastatic lymph nodes of gastric cancer with photo-oxidation of protoporphyrin IX

BACKGROUND

The establishment of a precise and rapid method to detect metastatic lymph nodes (LNs) is essential to perform less invasive surgery with reduced gastrectomy along with reduced lymph node dissection. We herein describe a novel imaging strategy to detect 5-aminolevulinic acid (5-ALA)-induced protoporphyrin IX (PpIX) fluorescence in excised LNs specifically with reduced effects of tissue autofluorescence based on photo-oxidation of PpIX. We applied the method in a clinical setting, and evaluated its feasibility.

METHODS

To reduce the unfavorable effect of autofluorescence, we focused on photo-oxidation of PpIX: Following light irradiation, PpIX changes into another substance, photo-protoporphyrin, via an oxidative process, which has a different spectral peak, at 675 nm, whereas PpIX has its spectral peak at 635 nm. Based on the unique spectral alteration, fluorescence spectral imaging before and after light irradiation and subsequent originally-developed image processing was performed. Following in vitro study, we applied this method to a total of 662 excised LNs obtained from 30 gastric cancer patients administered 5-ALA preoperatively.

RESULTS

Specific visualization of PpIX was achieved in in vitro study. The method allowed highly sensitive detection of metastatic LNs, with sensitivity of 91.9% and specificity of 90.8% in the in vivo clinical trial. Receiver operating characteristic analysis indicated high diagnostic accuracy, with the area under the curve of 0.926.

CONCLUSIONS

We established a highly sensitive and specific 5-ALA-induced fluorescence imaging method applicable in clinical settings. The novel method has a potential to become a useful tool for intraoperative rapid diagnosis of LN metastasis.

Histopathological Characteristics of Post-inflamed Coronary Arteries in Kawasaki Disease-like Vasculitis of Rabbits

Kawasaki disease (KD) is a systemic vasculitis in infants that develops predominantly in the coronary arteries. Despite the clinically transient nature of active inflammation in childhood albeit rare complications (e.g., coronary artery aneurysm), KD has recently been suggested to increase the incidence of ischemic heart diseases in young adulthood. However, little is known about the histopathology of the coronary artery long after development of the acute KD vasculitis. To address this, we conducted histological studies of rabbit coronary arteries in adolescent phase after induction of the KD-like vasculitis induced by horse serum administration. After a transmural infiltration of inflammatory cells in acute phase at day 7, the artery exhibited a gradual decrease in the number of inflammatory cells and thickening of the intima during the chronic phase up to day 90, where proteoglycans were distinctly accumulated in the intima with abundant involvement of α-smooth muscle actin (α-SMA)-positive cells, most of which accompanied expression of VCAM-1 and NF-κB. Distinct from classical atherosclerosis, inflammatory cells, e.g., macrophages, were barely detected during the chronic phase. These observations indicate that the KD-like coronary arteritis is followed by intimal thickening via accumulation of proteoglycans and proliferation of α-SMA-positive cells, reflecting aberrant coronary artery remodeling.

Recent advances in photodynamic diagnosis of gastric cancer using 5-aminolevulinic acid

Photodynamic diagnosis based on 5-aminolevulinic acid-induced protoporphyrin IX has been clinically applied in many fields based upon its evidenced efficacy and adequate safety. In order to establish a personalized medicine approach for treating gastric cancer patients, rapid intraoperative detection of malignant lesions has become important. Feasibility of photodynamic diagnosis using 5-aminolevulinic acid for gastric cancer patients has been investigated, especially for the detection of peritoneal dissemination and lymph node metastasis. This method enables intraoperative real-time fluorescence detection of peritoneal dissemination, exhibiting higher sensitivity than white light observation without histopathological examination. The method also enables detection of metastatic foci within excised lymph nodes, exhibiting a diagnostic accuracy comparable to that of a current molecular diagnostics technique. Although several complicating issues still need to be resolved, such as the effect of tissue autofluorescence and the insufficient depth penetration of excitation light, this simple and rapid method has the potential to become a useful diagnostic tool for gastric cancer, as well as urinary bladder cancer and glioma.

Ex vivo peripheral nerve detection of rats by spontaneous Raman spectroscopy

Nerve-sparing surgery is increasingly being applied to avoid functional deficits of the limbs and organs following surgery. Peripheral nerves that should be preserved are, however, sometimes misidentified due to similarity of shape and color to non-nerve tissues. To avoid misidentification of peripheral nerves, development of an in situ nerve detection method is desired. In this study, we report the label-free detection of ex vivo peripheral nerves of Wistar rats by using Raman spectroscopy. We obtained Raman spectra of peripheral nerves (myelinated and unmyelinated nerves) and their adjacent tissues of Wistar rats without any treatment such as fixation and/or staining. For the identification of tissue species and further analysis of spectral features, we proposed a principal component regression-based discriminant analysis with representative Raman spectra of peripheral nerves and their adjacent tissues. Our prediction model selectively detected myelinated nerves and unmyelinated nerves of Wistar rats with respective sensitivities of 95.5% and 88.3% and specificities of 99.4% and 93.5%. Furthermore, important spectral features for the identification of tissue species were revealed by detailed analysis of principal components of representative Raman spectra of tissues. Our proposed approach may provide a unique and powerful tool for peripheral nerve detection for nerve-sparing surgery in the future.

Spectral fingerprinting of individual cells visualized by cavity-reflection-enhanced light-absorption microscopy

The absorption spectrum of light is known to be a "molecular fingerprint" that enables analysis of the molecular type and its amount. It would be useful to measure the absorption spectrum in single cell in order to investigate the cellular status. However, cells are too thin for their absorption spectrum to be measured. In this study, we developed an optical-cavity-enhanced absorption spectroscopic microscopy method for two-dimensional absorption imaging. The light absorption is enhanced by an optical cavity system, which allows the detection of the absorption spectrum with samples having an optical path length as small as 10 μm, at a subcellular spatial resolution. Principal component analysis of various types of cultured mammalian cells indicates absorption-based cellular diversity. Interestingly, this diversity is observed among not only different species but also identical cell types. Furthermore, this microscopy technique allows us to observe frozen sections of tissue samples without any staining and is capable of label-free biopsy. Thus, our microscopy method opens the door for imaging the absorption spectra of biological samples and thereby detecting the individuality of cells.

Highly efficient direct conversion of human fibroblasts to neuronal cells by chemical compounds

Direct conversion of mammalian fibroblasts into induced neuronal (iN) cells has been attained by forced expression of pro-neural transcriptional factors, or by combining defined factors with either microRNAs or small molecules. Here, we show that neuronal cells can be converted from postnatal human fibroblasts into cell populations with neuronal purities of up to >80% using a combination of six chemical compounds. The chemical compound-induced neuronal cells (CiNCs) express neuron-specific proteins and functional neuron markers. The efficiency of CiNCs is unaffected by either the donor’s age or cellular senescence (passage number). We propose this chemical direct converting strategy as a potential approach for highly efficient generation of neuronal cells from human fibroblasts for such uses as in neural disease modeling and regenerative medicine.

How Does the Ca(2+)-paradox Injury Induce Contracture in the Heart?-A Combined Study of the Intracellular Ca(2+) Dynamics and Cell Structures in Perfused Rat Hearts

The calcium (Ca²⁺)-paradox injury of the heart, induced by restoration of extracellular Ca²⁺ after its short-term depletion, is known to provoke cardiomyocyte contracture. However, undetermined is how the Ca²⁺-paradox provokes such a distinctive presentation of myocytes in the heart. To address this, we imaged sequential intracellular Ca²⁺ dynamics and concomitant structures of the subepicardial ventricular myocytes in fluo3-loaded, Langendorff-perfused rat hearts produced by the Ca²⁺ paradox. Under rapid-scanning confocal microscopy, repletion of Ca²⁺ following its depletion produced high-frequency Ca²⁺ waves in individual myocytes with asynchronous localized contractions, resulting in contracture within 10 min. Such alterations of myocytes were attenuated by 5-mM NiCl₂, but not by verapamil, SEA0400, or combination of ryanodine and thapsigargin, indicating a contribution of non-specific transmembrane Ca²⁺ influx in the injury. However, saponin-induced membrane permeabilization of Ca²⁺ showed no apparent contracture despite the emergence of high-frequency Ca²⁺ waves, indicating an essential role of myocyte-myocyte and myocyte-extracellular matrix (ECM) mechanical connections in the Ca²⁺ paradox. In immunohistochemistry Ca²⁺ depletion produced separation of the intercalated disc that expresses cadherin and dissipation of β-dystroglycan located along the sarcolemma. Taken together, along with the trans-sarcolemmal Ca²⁺ influx, disruption of cell-cell and cell-ECM connections is essential for contracture in the Ca²⁺-paradox injury.

Photoacoustic microscopy using ultrashort pulses with two different pulse durations

We propose photoacoustic microscopy using ultrashort pulses with two different pulse durations in the range from femtoseconds to picoseconds. The subtraction of images for longer-pulse excitation from those for shorter-pulse excitation extracts two-photon photoacoustic images effectively, based on observation that the intensity ratio of two-photon to one-photon absorption-induced photoacoustic signals depends on the pulse duration in the same manner as the intensity ratio of two-photon and one-photon fluorescence signals. Two-photon photoacoustic microscopy using this subtraction method enables precise observation of the cross-sections of silicone hollows filled with the mixture of one-photon and two-photon absorption solutions.

Label-free evaluation of myocardial infarction and its repair by spontaneous Raman spectroscopy

Raman spectroscopy, which provides information about molecular species and structures of biomolecules via intrinsic molecular vibrations, can analyze physiological and pathological states of tissues on the basis of molecular constituents without staining. In this study, we analyzed Raman spectra of myocardial infarction and its repair in rats using the hypothesis that the myocardium in the course of myocardial infarction and its repair could be recognized by spontaneous Raman spectroscopy on the basis of chemical changes in myocardial tissues. Raman spectra were acquired from unfixed frozen cross sections of normal and infarcted heart tissues upon excitation at 532 nm. Raman spectra of the infarcted tissues were successfully obtained at characteristic time points: days 2, 5, and 21 after coronary ligation, at which the main components of the infarcted region were coagulation necrosis, granulation tissue, and fibrotic tissue, respectively. The latent variable weights calculated by a multivariate classification method, partial least-squares-discriminant analysis (PLS-DA), revealed fundamental information about the spectral differences among the types of tissues on the basis of molecular constituents. A prediction model for the evaluation of these tissue types was established via PLS-DA. Cross-validated sensitivities of 99.3, 95.3, 96.4, and 91.3% and specificities of 99.4, 99.5, 96.5, and 98.3% were attained for the discrimination of normal, necrotic, granulation, and fibrotic tissue, respectively. A two-dimensional image of a marginal area of infarction was successfully visualized via PLS-DA. Our results demonstrated that spontaneous Raman spectroscopy combined with PLS-DA is a novel label-free method of evaluating myocardial infarction and its repair.

Neutrophil Phagocytosis of Platelets in the Early Phase of 2,4,6-trinitro-1-chlorobenzene (TNCB)-induced Dermatitis in Mice

Activated platelets form platelet–leukocyte aggregates in the circulation in inflammatory diseases. We investigated whether activated platelets in inflamed skin tissues are phagocytized and removed by neutrophils. To investigate the kinetics of platelets and neutrophils, we immunohistochemically examined the spatiotemporal distribution of them in a murine model of 2,4,6-trinitro-1-chlorobenzene (TNCB)-induced dermatitis by using confocal and structured illumination microscopy. Four hours after elicitation, aggregates of CD41-positive platelets were adhered to CD31-positive endothelial cells within the vessels, and CD62P and PF4, markers of activated platelets, were expressed on platelet aggregates. At 8 hour post-elicitation, fragmented CD41-positive platelets were located both inside and outside vessels. Twenty-four hours after elicitation, the number of Ly-6G-positive neutrophils ingesting fragmented CD41-positive platelets outside vessels was increased, and CD62P and PF4 expression on the phagocytosed platelets was no longer observed. Disc-shaped CD41-positive platelets were not found outside vessels at any time during the experiment. Our data revealed that aggregates of activated platelets inside vessels were ingested and removed by neutrophils in the early stage of TNCB-induced dermatitis, suggesting that the process of removal of activated platelets by neutrophils may play an important role not only in the early phase of skin inflammation but also in other types of acute inflammation.

Pacing-induced non-uniform ca(2+) dynamics in rat atria revealed by rapid-scanning confocal microscopy

Intracellular Ca(2+) ([Ca(2+)]i) dynamics in isolated myocytes differ between the atria and ventricles due to the distinct t-tubular distributions. Although cellular aspects of ventricular [Ca(2+)]i dynamics in the heart have been extensively studied, little is known about those of atrial myocytes in situ. Here we visualized precise [Ca(2+)]i dynamics of atrial myocytes in Langendorff-perfused rat hearts by rapid-scanning confocal microscopy. Of 16 fluo-4-loaded hearts imaged during pacing up to 4-Hz, five hearts showed spatially uniform Ca(2+) transients on systole among individual cells, whereas no discernible [Ca(2+)]i elevation developed during diastole. In contrast, the remaining hearts showed non-uniform [Ca(2+)]i dynamics within and among the cells especially under high-frequency (4 Hz) excitation, where subcellular cluster-like [Ca(2+)]i rises or wave-like [Ca(2+)]i propagation occurred on excitation. Such [Ca(2+)]i inhomogeneity was more pronounced at high-frequency pacing, showing beat-to-beat Ca(2+) transient alternans. Despite such non-uniform dynamics, cessation of burst pacing of the atria was not followed by emergence of spontaneous Ca(2+) waves, indicating minor Ca(2+)-releasing potentials of the sarcoplasmic reticulum (SR). In summary, rat atria display a propensity to show non-uniform [Ca(2+)]i dynamics on systole due to impaired Ca(2+)-release from the SR and paucity of t-tubules. Our results provide an important basis for understanding atrial pathophysiology.

The involvement of the vasa vasorum in the development of vasculitis in animal model of Kawasaki disease

BACKGROUND

Kawasaki Disease (KD) involves a diffuse and systemic vasculitis of unknown etiology that mainly affects infants and children. Although a considerable number of analyses of the clinical, histopathological and molecular biological details underlying the mechanism responsible for the development of coronary arterial lesions, it is still poorly understood.The purpose of this study was to analyze the state of angiogenesis, vasculogenesis and the distribution of blood vessels using an animal model of KD like vasculitis. We investigated the involvement of the vasa vasorum from the adventitia in the vascular involvement and the development of the disease state by performing sequential histopathology, scanning electron microscopy (SEM) and micro computed tomography (CT) studies using a murine model of vasculitis induced by the Candida albicans water-soluble fraction (CAWS).

METHODS

To prepare the animal model of KD like vasculitis, CAWS was intraperitoneally injected into C57BL/6N mice for five consecutive days as reported by Ohno et al. We observed the changes of the vasa vasorum at the aorta and the orifices of the coronary arteries by SEM and micro CT, and also compared the neovascularization at the media and adventitia of the aorta by an immunohistochemical analysis.

RESULTS

As previously reported, obvious inflammation was detected two weeks after the injection of CAWS, and also intimal thickening was observed three weeks after the injection. We found that the vasa vasorum in the adventitia of the aorta was increased in the model mice. The vasa vasorum started increasing one week after the injection of CAWS, before any obvious vasculitis was microscopically detected.

CONCLUSION

The present results indicate that the vasculitis in Kawasaki disease starts as a disorder of the vasa vasorum.

In Vivo Detection of Rat Colorectal Cancers by using a Dual-Wavelength Excitation Method

Hypoxia is a characteristic feature of solid neoplasms, and insufficient oxygen supply increases cellular nicotinamide adenine dinucleotide (NADH) fluorescence, which is a main component of autofluorescence of the colorectal mucosa. We investigated whether a dual-wavelength excitation method which is optimized for sensing mucosal NADH fluorescence could be applicable to the detection of rat colorectal cancers in vivo. Rat colorectal adenocarcinomas were studied by using fluorescence stereomicroscopy. After autofluorescence images at 470 nm irradiated with dual-wavelength excitation at 365 nm (F365 ex) and 405 nm (F405 ex) were acquired, ratio images were produced by dividing F365 ex by F405 ex: The excitation-emission wavelength pairs in F365 ex and F405 ex were adjusted for acquisition of NADH fluorescence and reference fluorescence. Based on observations from the luminal surface in vivo, F365 ex/F405 ex ratio images indicated a 1.57-fold higher signal value in the cancers than in the surrounding normal mucosa. The signal values in F365 ex/F405 ex ratio images were less mutually related with the hemoglobin concentration index. Small adenocarcinomas (less than 4 mm) could be detected on F365 ex/F405 ex ratio images. The results showed that NADH fluorescence measurement with little interference from tissue hemoglobin is efficient for visualizing rat colorectal cancers in vivo, suggesting that the dual-wavelength excitation method has potential for label-free endoscopic detection of diminutive colorectal neoplasms.

Detection of metastatic lymph nodes using 5-aminolevulinic acid in patients with gastric cancer

BACKGROUND

Precise diagnosis of lymph node metastases is essential to select therapeutic strategy for patients with gastric cancer, and rapid intraoperative diagnosis is useful for performing less invasive surgery. In this study, we focused on a known photosensitizer, 5-aminolevulinic acid (5-ALA), and examined the feasibility of 5-ALA-induced protoporphyrin IX (PpIX) fluorescence to detect metastatic foci in excised lymph nodes of patients with gastric cancer.

METHODS

A total of 144 lymph nodes obtained from 14 gastric cancer patients were examined. The patients were administered 5-ALA orally before surgery. Excised lymph nodes were cut in half and observed by fluorescence microscopy. The diagnostic results were compared to those of the routine histopathological examination.

RESULTS

Observed red fluorescence of PpIX was identical to the metastatic focus, with 84 % accuracy. Twelve non-metastatic lymph nodes showed unexpected PpIX accumulation to lymphoid follicles, but these could be discriminated based on their characteristic fluorescence patterns. With incorporation of this morphological consideration, this method demonstrated good diagnostic power with 92.4 % accuracy. On the quantitative analysis using the signal intensity ratio of red to the sum of red, green, and blue (R/(R + G + B) ratio) as an index corresponding to red fluorescence of PpIX, metastatic lymph nodes showed significantly higher value than non-metastatic lymph nodes (p < 0.0001). The area under the curve was calculated as 0.832 throughout Receiver operating characteristic analysis.

CONCLUSIONS

Our results demonstrated that 5-ALA-induced fluorescence diagnosis is a simple and safe method and is a potential candidate for a novel rapid intraoperative diagnostic method applicable to clinical practice.

TGF-β Signaling Regulates Pancreatic β-Cell Proliferation through Control of Cell Cycle Regulator p27 Expression

Proliferation of pancreatic β-cells is an important mechanism underlying β-cell mass adaptation to metabolic demands. Increasing β-cell mass by regeneration may ameliorate or correct both type 1 and type 2 diabetes, which both result from inadequate production of insulin by β-cells of the pancreatic islet. Transforming growth factor β (TGF-β) signaling is essential for fetal development and growth of pancreatic islets. In this study, we exposed HIT-T15, a clonal pancreatic β-cell line, to TGF-β signaling. We found that inhibition of TGF-β signaling promotes proliferation of the cells significantly, while TGF-β signaling stimulation inhibits proliferation of the cells remarkably. We confirmed that this proliferative regulation by TGF-β signaling is due to the changed expression of the cell cycle regulator p27. Furthermore, we demonstrated that there is no observed effect on transcriptional activity of p27 by TGF-β signaling. Our data show that TGF-β signaling mediates the cell-cycle progression of pancreatic β-cells by regulating the nuclear localization of CDK inhibitor, p27. Inhibition of TGF-β signaling reduces the nuclear accumulation of p27, and as a result this inhibition promotes proliferation of β-cells.

Precise analysis of the autofluorescence characteristics of rat colon under UVA and violet light excitation

PURPOSE

Tissue autofluorescence study is a promising means of endoscopic detection of colonic neoplasia, but the mechanism of autofluorescence eruption has still not been verified. The purpose of this study was to precisely analyze the autofluorescence characteristics of freshly prepared normal rat colon under UVA and violet light excitation.

METHODS

Excised rat colons were studied by using multichannel spectrophotometry, spectroscopic imaging, confocal microscopy, combined two-photon excited fluorescence and second-harmonic generation (SHG) microscopy, and fluorescence lifetime imaging microscopy.

RESULTS

Spectroscopic analysis of freshly prepared colon sections revealed that the mucosa and the submucosa showed strong autofluorescence under UVA and violet light excitation. The combined images of two-photon and SHG microscopy revealed that the mucosal epithelia are the important source of autofluorescence. Nicotinamide adenine dinucleotide seems to be one of the major substances involved in the autofluorescence of the mucosal layer on 365- nm light excitation. The autofluorescence spectra of the luminal surfaces were identical to those of the mucosa on crosssectional examinations with 365-nm excitation. The main origin of autofluorescence of the luminal surface with 365-nm excitation is the epithelial cells in the mucosa without overlay of submucosal fluorescence.

CONCLUSIONS

The mucosal layer is the important source of the autofluorescence observed under excitation with UVA/violet light in multilayered colonic structures. Illumination of 365-nm wavelength light is a suitable means of analyzing the autofluorescence of mucosal epithelia.

Raman molecular imaging of cells and tissues: towards functional diagnostic imaging without labeling

Raman spectroscopy has long been used as a powerful tool for chemical composition analyses. Raman scattering light measurement has the advantage of being able to examine biomolecules in a nondestructive and non-labeling manner. However, molecular imaging of cells and tissues by using Raman microscopy had been hampered due to weak Raman signals and required long acquisition time. Recent advances in imaging devices, such as development of slitscanning Raman microscopy, have enabled us to acquire high-resolution Raman images of biomedical samples. Thus, Raman molecular imaging now has wide application potential for in-situ functional analysis of biomolecules in living bodies, such as studies of intracellular drug pharmacokinetics and oxygen saturation of blood capillaries. There is a possibility that Raman scattering light measurement can be applied for functional diagnostic imaging of human bodies, taking advantage of its noninvasive merits in the future. This review briefly highlights recent topics in spontaneous Raman molecular imaging of cells and tissues.