Oxaliplatin, a platinum-based anticancer drug, is an important agent in the standard chemotherapy for colorectal, gastric, and pancreatic cancers, but it also often induces peripheral neuropathy. Oxaliplatin-induced peripheral neuropathy is classified into acute (cold hypersensitivity) and chronic (numbness, pain, and dysesthesia) neuropathie. The neuropathy affects quality of life and leads to dose reduction or discontinuation of chemotherapy. However, there are no prophylactic or therapeutic agents strongly recommended for the peripheral neuropathy in guidelines from ASCO, ESMO, and JASCC.

In our basic research, mechanisms underlying the acute and chronic neuropathies have been clarified. Furthermore, we have identified several potential prophylactic and therapeutic agents from approved drugs by targeting the mechanisms in basic research, adverse drug reaction database analysis research, and clinical research. In particular, we have recently identified a proton pump inhibitor as a potential prophylactic agent for oxaliplatin-induced neuropathy. In rats, omeprazole, a proton pump inhibitor, prevented neuropathic pain and peripheral nerve degeneration caused by oxaliplatin. In addition, in an analysis of FAERS, a database of adverse drug reactions, the rate of peripheral neuropathy was lower in patient reports of concomitant use of proton pump inhibitors. Furthermore, in a retrospective cohort study of 1,038 patients starting oxaliplatin treatment at Kyushu University Hospital, the concomitant proton pump inhibitor group had significantly lower incidence of peripheral neuropathy (any grade) and treatment discontinuation due to peripheral neuropathy than the non-concomitant group.

This symposium will present the findings of our drug repositioning studies and future perspectives.

According to American Society of Clinical Oncology, no agents are recommended for the prevention of chemotherapy-induced peripheral neuropathy (CIPN), and duloxetine is the only agent that has appropriate evidence to support its use for patients with established CIPN. We have shown that extracellular high mobility group box 1 (HMGB1), a nuclear protein, released from certain cells including macrophages (MMφ) participates in the pathogenesis of CIPN, and that CIPN is prevented by thrombomodulin (TM) alfa (TMα), an anti-DIC agent, capable of causing thrombin-dependent HMGB1 degradation. The anti-CIPN effect of TMα was also demonstrated in humans by a clinical study, and is being intensively investigated in a Japan-U.S. joint clinical trial. In our study, anticoagulants did not only reverse the anti-CIPN effect of TMα, but also aggravated CIPN in mice, possibly through inhibition of thrombin-dependent HMGB1 degradation by endothelial TM, the latter evidence ascertained in humans by our clinical study. Conversely, our clinical study in breast cancer patients treated with paclitaxel (PCT) identified postmenopausal patients as a high risk group for CIPN, and our basic study showed that ovariectomy aggravated CIPN in PCT-treated mice, an effect reversed by estrogen that inhibited PCT-induced HMGB1 release from MMφ. We thus recommend a preventive intervention using TMα for CIPN-prone groups like postmenopausal patients. Collectively, an integrated clinical and basic study is beneficial to investigate the pathogenesis and prevention strategy of CIPN.

As the pharmaceutical industry faces high R&D costs and low productivity, there is a growing demand for the development and validation of physiological human cell models that can be implemented in the early stages of drug discovery phase. Therefore, to effectively and efficiently reproduce the human condition and ensure safety, there is a need for a paradigm shift from existing cell models that do not meet our research questions. There is also a need to properly qualify these Complex in vitro models（CIVMs） to make them suitable for various applications in advanced drug discovery and translational research. In addition, to utilize the output obtained from CIVMs in compound evaluation strategies, it is essential to use not only numerical data but also artificial intelligence and machine learning, and an interdisciplinary perspective is considered necessary.

This symposium is focused on the evaluation for preclinical safety in early phase of drug discovery from the hit/lead discovery stage to lead optimization. And then, I will overview the current status and issues of safety evaluation in early phase of drug discovery.

In recent years, with the establishment of the FDA Modernization Act 2.0, a large number of microphysiological systems (MPS) with more human relevancy in terms of organizational structure and function, such as three-dimensional culture, spheroids, and co-culture models with immune crosstalk has been developed. Furthermore, in pharmaceutical companies, especially in Europe and the United States, its implementation into drug discovery research and practical application are rapidly accelerating.

Although various types of MPS devices and assay systems combining cells have been developed to date, there are not many that can be used universally to meet various needs and issues. On the other hand, when considering the use of MPS in drug discovery research, especially in the field of toxicity, there are many types of research questions such as an extrapolation of side effects observed in animals to humans, identification of mechanisms of side effects, and reverse translational research based on adverse effects in clinical trial. In either situation, research questions arise suddenly, and we need to tackle quickly, especially in early phase in drug discovery.

In my presentation, I will first explain the characteristics of MPS vendor services by dividing them into 1) device supply type, 2) assay supply type that combines cells and devices, 3) contract research type, and 4) joint research and development type. Then, I will introduce the importance of clarifying the needs (context of use) that researchers should first address. Furthermore, I will introduce our challenge to establish MPS assay system quickly that combines automation of culture and other operations, and also introduce case studies using MPS in the nonclinical safety evaluation.

Various in vitro assay systems have been used in the predictive toxicology field and are essential to mitigate safety concerns in drug discovery. (Loiodice et al, 2019). However, conventional many in vitro assay systems are conventionally applied single cells cultured on a flat surface for evaluation, making it difficult to evaluate the complex toxicities contributed by multiple cell types. Recently, microphysiological systems (MPS), which are advanced in vitro models such as Organ-on-a-Chip and Organoids, have been developed, and it was expected that they would make it possible to evaluate complex toxicity, beyond conventional in vitro assay systems. In our company, MPS is increasingly being used in safety evaluation for various areas such as hepatotoxicity and neurotoxicity. In this presentation, we will introduce in-house application examples of MPS such as Organ-on-a-Chip and Organoid in the early safety evaluation, and share the benefit with the example of platelet aggregation assay. MPS was expected to be a game changer in the early safety evaluation within five years as of 2020, whereas the use of MPS is still limited, and it is difficult to say that it has achieved the expected results. In this presentation, we focus on the gap between the ideal and reality required for advanced in vitro models, and discuss the factors limiting the use of MPS in the early safety evaluation.

Gastrointestinal toxicity (GI-tox) is a common clinical adverse event. It is very important to clarify GI-tox concerns in the nonclinical stage because of reduction of patients' QOL.

Monolayer culture of single cell line has been the main method to evaluate GI-tox in vitro. These methods are impossible to detect the effects on stem cells or interactions between cells, and therefore have poor detection sensitivity, reproducibility and predictivity for clinical GI-tox. Intestinal organoids have a three-dimensional structure including undifferentiated cells, thus GI-tox can be evaluated in a closer environment to the body. However, assays using complex in vitro model such as organoids are often difficult to operate. Therefore, it is necessary to develop a new evaluation system for GI-tox with high accuracy and throughput.

We found that GI-tox induced compounds reduced organoids size. Therefore, we considered evaluating GI-tox by organoids size quantification. Then, we developed an application that can recognize organoids from bright-field images and extract feature values. The size of intestinal organoids was to be a sensitive marker to detect GI-tox. Furthermore, this application is superior in convenience because analysis can be performed only using bright-field images, leading to a reduction in the burden on experimenters.

In this presentation, we will introduce the developed organoid image analysis application and the GI-tox evaluation system, and discuss the future development in nonclinical safety evaluation.

In drug development, it is important to mitigate toxicity risks detected in nonclinical studies or estimate human risk. In vitro evaluation is widely used for toxicity risk assessment because it has some advantages including required less amount of drug substance, high throughput, and the ability to use human cell/tissue. However, they also have several challenges such as reproducing the toxic phenotypes observed in vivo studies. In recent years, various complex in vitro models (CIVM), including organ-on-chip and organoid systems have been developed. These models are utilized for evaluating toxic responses which used to be difficult to assess using simple in vitro systems alone. In addition, need for animal alternative methods is expected under the FDA Modernization Act 2.0. Whereas, we have observed several issues in their application during the early stage of drug discovery. Additionally, since CIVM is based on a "phenotype" approach like in vivo toxicity studies, it is difficult to elucidate the mechanisms underlying toxic responses in some cases. To solve this problem, one possible approach would be to understand molecular initiating events (MIE), which can be applied to relatively simple in vitro evaluation systems. In this presentation, we would like to discuss the strategic use and future prospects of complex/simple in vitro models to efficiently understand and mitigate toxicity risks.

Recently drug modalities have been diversified, and new modalities including oligonucleotide drugs, gene therapy, and protein degraders have emerged in addition to small molecule drugs and antibody drugs. In general, risk management for small molecule drugs involves treatment cessation, monitoring by biomarkers and imaging, ensuring a safety margin, and dose reduction. However, some new drug modalities such as gene therapy that assumes only one treatment or oligonucleotide drugs with a long half-life after tissue distribution are considered to be therapeutic modalities where it is difficult to manage risks by "treatment cessation". As an alternative, use of an antidote to control the side effects (including exaggerated pharmacology) may be an option.

In this study, we conducted a survey of public information, and a questionnaire survey of JPMA member organizations with the aim of understanding the current status the development of antidotes in new modalities such as gene therapy and oligonucleotide drugs. In this presentation, we will introduce the results and discuss them based on the results of the survey.

Harnessing the ubiquitin proteasome system to degrade proteins of interest offers potential to modulate previously undruggable targets. The field of protein degraders has been generating much interest in recent years with many molecules entering the clinic, both molecular glues and heterobifunctional degraders that contain an E3 ligase binding domain and a target binding domain connected by a linker. Heterobifunctional degraders have some unique properties that will be discussed, including such as the catalytic nature, physical properties that can impact drug discovery testing, and the potential for decreased degradation at high drug concentrations. In addition, while the nonclinical toxicology assessment for protein degraders (either molecular glues or heterobifunctional targeted protein degraders) shares the same principles as for classical small molecules, there are some additional considerations due to the modality. This presentation will discuss the challenges and opportunities for these exciting molecules and current practices across the industry for in vitro and in vivo nonclinical safety assessments for heterobifunctional protein degraders. Approaches for selecting a toxicologically relevant species for in vivo studies will also be discussed.

Antibody therapeutics are now well-established drugs for the treatment of various diseases. Factors that give rise to safety concerns arise through interactions between the body and the target protein, its biological　properties, etc., which in turn depends on a variety of factors. One of these factors, the evolution of antibody engineering, tends to increase the efficacy and specificity of drug treatments, while increasing the complexity and difficulty of nonclinical safety studies. Therefore, pharmaceutical companies are trying to identify potential toxicities to patients by structuring nonclinical safety studies with creative ideas for each drug molecule. One of the gaps that tend to occur between reactions observed in patients and those detectable in nonclinical evaluations are immunological response differences. Because antibody drugs frequently involve reactions of the immune system, nonclinical safety studies must take such immunological differences into account to characterize toxicity. However, the estimation of the immunological responses and gaps often requires unique evaluation to each drug molecule because there are few standardized methods. In this presentation, I would like to focus on in vitro and in vivo immune-related assessments among the components of nonclinical safety studies. I would also like to discuss how we should make use of the information obtained in nonclinical immune-related assessments, despite the limitations of such assessments, and what challenges we should take up in the future.

The PMDA evaluates new drugs from various angles, including quality, non-clinical, and clinical, in accordance with the "Act on Quality, Efficacy and Safety Assurance of Drugs and Medical Devices" (Pharmaceutical Affairs Law). Our purpose is to ensure the prompt delivery of more effective and safer products to patients and medical facilities, while taking into account the risk-benefit ratio. In recent years, PMDA has focused on evaluating new modalities such as antisense nucleic acids, siRNA, and CAR-T cells, which are expected to receive marketing approval as soon as possible due to their unique mechanisms of action. However, since these products have limited development experience, their safety in humans must be carefully evaluated.

Based on PMDA's experience in regulatory strategy consultation, clinical trial consultation, and regulatory submission, this presentation will discuss the concept of non-clinical safety studies for various new modalities and aim to deepen mutual understanding on non-clinical safety evaluation of these modalities.

Primary human kidney cells or kidney-derived cell lines do not maintain the gene expression of drug transporters. Conventional kidney cells have not yet been used in drug discovery. In animal studies, low predictivity to the clinical trial, species difference and animal welfare are always issues. Therefore, it is desired to commercialize the human kidney cells that can be used for drug discovery screening and analysis of the mechanism. Recently, highly functional kidney cells have been investigated, including the iPS cell-derived kidney organoids and the application to the Micro Physiological System. In this study, we investigated the usefulness of evaluating drug-induced kidney injury (DIKI) using 3D-cultured renal proximal tubule epithelial cells (3D-RPTEC, Nikkiso), whose expression of drug transporters and endocytosis receptors are comparable to those of human kidney cortex.

The intracellular ATP content in 3D-RPTEC was reduced by the exposure of tenofovir that is a substrate of OAT. On the other hand, ATP reduction by tenofovir was not observed in the 2D-cultured RPTEC. These results suggested that the expression of drug transporters was required for the evaluation of DIKI. In addition, over 30 drugs were exposed to 3D-RPTEC for 7 days, and the deviation between the concentration causing 20% reduction in intracellular ATP level and the effective plasma concentration was calculated as the safety margin of DIKI. The prediction of DIKI was judged if a safety margin is less than 100 times. As a result, the sensitivity was about 80% and the specificity was more than 90%. Furthermore, we investigated the usefulness of High Content Analysis using a confocal image cytometer that is used to analyze toxicity mechanisms.

In this study, we would like to introduce the application of 3D-RPTEC for drug discovery and the evaluation of oligonucleotide therapeutics.

In small molecule drug discovery, evaluating both on-target and off-target toxicities at early stages is critical to preventing dropout in later phases. When dealing with neurotoxicity, determining the safety margin and risk-to-benefit ratio is pivotal for go/no-go decisions in drug projects. When a safety margin is inadequate, the mechanism of neurotoxicity should be elucidated to potentially avoid it. Here, we present an evaluation method using human iPSC-derived neurons and describe its application in a project at Chugai Pharmaceutical. In that project, we observed neurotoxicity in a preliminary canine toxicity study at the expected human effective exposure level. Given the low likelihood of on-target toxicity based on the target safety assessment, we focused on off-target possibilities. We began by using an in vitro functional assay to identify a suspicious target: monoamine transporter (MAT) inhibition. Subsequent in vitro experiments used a microelectrode array (MEA) with iPSC-derived neurons to investigate whether MAT inhibition mediates neurotoxicity. The principal component analysis of MEA indicated that the neurotoxic compound had a profile similar to known MAT inhibitors, suggesting that MAT inhibition may have caused the neurotoxicity. Conversely, in our in vivo approach examining the effects of metabolites, we did not find a correlation with neurotoxicity. In this presentation, I will share our integrated approach to avoiding compound-mediated neurotoxicity and to elucidating neurotoxic mechanisms using human iPSC-derived neurons. The limitations of in vitro nonclinical safety studies will also be discussed.

In recent years, the modalities in drug discovery are being diversified. The lipid nanoparticle (LNP) have attracted attention as the drug delivery vehicles for oligonucleotide therapeutics, which are not stable in the blood, to target organs. However, several LNPs are known to induce the complement activation-related pseudoallergy (CARPA). In order to reduce the risk of CARPA induction by LNP, animal models are used in non-clinical safety studies, but extrapolation to human safety evaluation is low due to species differences. On the other hand, it was assumed that the degree of complement activation by LNP using human samples could be greatly affected by the difference between donors. To address this issue, we attempted to establish in vitro complement activation assessment using multiple human plasmas in order to discover safer LNPs.

First, we investigated the LNP-induced complement activation by measuring the increase of two complement fragments, C3a and C5b-9, using human plasma from approximately 100 donors. As a result, it was observed a difference in the degree of increase in C3a and C5b-9 by LNP between donors. Therefore, we are investigating several factors that led to the donor difference, for example, involvement of anti-polyethylene glycol (PEG) antibody. PEG is one of the components of LNPs, and previous reports have shown that higher concentrations of anti-PEG IgM in the CARPA-symptomatic subjects group compared to the asymptomatic subjects after LNP administration, suggesting that the formation of LNP and anti-PEG IgM complexes is one of the factors related to the induction of CARPA (Kozma et al., Vaccines, 2023).

In this workshop, we will introduce the current status of LNP-mediated CARPA assessments using human plasma and will discuss the expected prospects and challenges for the future.

Cytokine release syndrome (CRS), which is caused by excessive cytokine release, is known as a side effect of antibody drugs. Symptoms of CRS include fever and headache, but in severe cases, tachycardia, hypotension, and dyspnea may appear. A well-known example is an anti-CD28 superagonist antibody TGN1412, which induced a multiple organ failure in all subjects of the clinical trial in 2006. The side effects of TGN1412 were not observed in toxicity studies in monkeys, and later it became clear that species differences were involved. After this tragedy, various in vitro CRAs using human blood cells have been investigated to detect cytokine release by antibody drugs. However, an in vitro CRA method with high clinical predictability has not been established even 18 years after the TGN1412 incident. This is due to a different characteristics of each method, multiple mechanisms of CRS, and diversity of the formats and mode of action (MoA) of antibody drugs.

Currently, considering the characteristics of each method, we are conducting our research with the following two objectives:

1) To establish an assay method that dose not overlook the cytokine release potential of antibody drugs.

2) To make a strategy for selecting an appropriate assay method considering the MoA of antibody drugs.

In this presentation, we will introduce the results of several existing in vitro CRAs and our newly developed in vitro CRAs using some antibody drugs, including TGN1412. We hope that our presentation will provide a good opportunity to think about the strategy for selecting assay methods.

Many pharmaceutical companies take the strategy to screen out compounds that form highly reactive metabolites to avoid drug-induced liver injury (DILI). Reactive metabolites could bind to carrier proteins and show immunogenicity as haptens. Idiosyncratic DILI is not detectable in conventional animal studies of drug development and does not occur in most patients indicating the species- and interindividual-differences in DILI onset. Retrospective Genome Wide Association Study (GWAS) have identified human leukocyte antigens (HLAs) as a human specific genetic risk marker. HLA is necessary for antigen presentation system so the clinical associations provide persuasive evidence to hypothesize that the reactions involve the drug-specific activation of the adaptive immune system. Mechanistic studies using T-cells from skin-hypersensitivity patients who express risk HLA provide direct evidence to support this adaptive immune system and the hapten-mediated reaction. With regard to DILI, I will show mechanistic study results using isoniazid specific T-cell from patients with anti-tuberculosis drug-related liver injury or skin reaction and the challenges for DILI mechanism/prediction will be discussed.

The history of non-clinical safety evaluation for pharmaceuticals dates back to the early 20th century when animal experiments were first conducted to provide information on the toxicity of chemical substances and drugs.

In the 1950s, regulatory authorities and international organizations engaged in discussions regarding the essential toxicity evaluation parameters necessary for drug development and ensuring safety for humans. These discussions led to the creation of various toxicity testing guidelines.

While standardized toxicity evaluation methods have been established, the need for continuous improvement in toxicity detection, including refinement of evaluation criteria, has arisen due to adverse events observed in humans. Additionally, the diversification of drug modalities has highlighted the limitations of existing evaluation approaches, prompting the development of novel technologies from a more scientific perspective.

Furthermore, efforts have been made to develop evaluation methods and technological innovations aimed at reducing the number of animals from the perspective of animal welfare. In this introduction, we aim to provide an overview of the objectives of this workshop and explore the historical evolution of toxicity assessment methods and technologies.

CNS toxicity is one of the primary reasons for discontinuing the development of new pharmaceuticals. Particularly, risks such as seizure might lead to development termination or dose restrictions during clinical trials, acting as significant hurdles. Neurotoxicity assessment has traditionally relied on methods like modified Irwin’s method and FOB, along with in vivo evaluation systems specific to individual neurological symptoms. However, practical in vitro evaluation options have been lacking. Nowadays, human iPSC-derived neural cells have emerged as in vitro experimental materials alongside rat neuronal primary cultures, breaking down species barriers and offering new possibilities for drug evaluation. Additionally, advancements in measurement devices, such as MEA and imaging tools, have demonstrated the usefulness of novel evaluation platforms worldwide. Furthermore, there are ongoing efforts to construct neural functional evaluation systems that combine simplicity and throughput using techniques like Ca-transient assays with fluorescent sensors. I will introduce experiments conducted by the CSAHi Ca-transient team. Beyond the conventional in vivo neurotoxicity assessments, I will comprehensively discuss other potential experimental systems, including Secondary Pharmacology and the utilization of EEG, considering central nervous system toxicity. Additionally, I will provide information on in vitro evaluation systems for peripheral neurotoxicity. I aim to explore existing possibilities and discuss how to effectively utilize them in the drug development.

Error-corrected next generation sequencing (ecNGS) is attracting attention as a new technique for genotoxicity testing. It is a method that can detect low-frequency mutations through direct sequencing without relying on phenotypes, by reducing the error of the NGS to the ultimate level. Traditionally, in vivo mutation tests have been performed using transgenic rodents (TGR) in which target genes have been introduced. ecNGS is applicable to all species, including humans, and is expected to replace these tests in the near future. In fact, the OECD TGR test guideline (TG488) is currently undergoing accelerated movement toward revision.

In Europe and the U.S., the Duplex Sequencing method has already become popular as ecNGS, and the Nano-Seq and HiFi-Seq methods have also been developed. In Japan, our original PECC-Seq, and the HAWK-seq are also developed. The characteristics and current status of these ecNGS methods will be introduced.

The ecNGS method can be applied to all in vitro assays, including bacteria, and allows comparison of in vitro and in vivo test results using a common indicator. Conventional in vitro genotoxicity tests have high sensitivity but difficulty in quantitative evaluation. With ecNGS, direct comparison with in vivo test results can lead to the development of more quantitative in vitro tests. The development of test systems using novel human cells such as organoids and MPS combined with the ecNGS method is already underway, and it is hoped that this will lead to the development of test systems that enable the evaluation of real effects in humans. The ecNGS method can be directly incorporated into general toxicity studies such as 28-day repeated-dose tests in rodents, and is expected to be widely used as a multi-endpoint toxicity test that incorporates genotoxicity testing into repeated-dose studies in the near future.

Currently, particularly in the field of drug discovery, the development of microphysiological systems (MPS) is actively being investigated with the aim of improving the extrapolation of preclinical studies to humans [1-3]. Among various organs, the structure and function of the placenta vary greatly depending on the animal species. Therefore, the development of MPS with human placental function has high biomedical value, and can be expected to have a significant ripple effect in the evaluation system not only for pharmaceuticals but also for supplements and cosmetics. We recently discovered culture conditions for organoids with the villous structure of the placenta, using recently established human placenta-derived trophoblast stem cells (TS cells) [4,5]. Furthermore, we have also succeeded in creating more versatile planar organoids for quantitatively evaluating placental permeability. In this lecture, the human placental MPS that we are developing will be presented, and the possibility of its application to reproductive and developmental toxicity studies will be discussed.

[1] I. M. Goncalves et al., Cancers 14, 935 (2022).

[2] V. Carvalho et al., Sensors 21, 3304 (2021).

[3] L.-J. Chen, H. Kaji, Lab Chip 17, 4186-4219 (2017).

[4] H. Okae et al., Cell Stem Cell 22, 50-63 (2018).

[5] T. Hori et al., Nat. Commun. 15, 962 (2024).

The structure of hERG, an ion channel to which drugs should not bind, has been analysed by Roderick MacKInnon using cryo-electron microscopy. The hERG example suggests that structural information can contribute to toxicity avoidance. However, the resolution of all these hERG structures published to date is 3.8 Å, which is insufficient for structure-based calculations aimed at avoiding toxicity. The instrumental resolution of electron microscopes is as high as better than 1 Å, and the resolution for observing biological samples is limited by radiation damage but not by instrumental resolution. Since radiation damage to biological samples is significantly reduced by lowering the specimen temperature, cryo-electron microscopy is essential for SGDD: Structure-Guided Drug Development. Cryo-electron microscopy also allows high-resolution images to be taken of target molecules by embedding them in a thin layer of ice. For high-resolution data collections, 8th generation cryo-electron microscope was developed.

We are analysing drug target molecules including membrane proteins by single particle analysis utilising cryo-electron microscopy, and also challenging drug development based on SGDD. I believe SGDD may effectively help to avoid toxicity. Although there are still few examples of a drug rescuing strategy, where compounds that could not be developed into drugs due to toxicity or other reasons are rescued from the shelves of pharmaceutical companies, I would like to present an example of the rapid development of a potent infectious disease drug candidate based on SGDD.

To obtain accurate information from the living body, the validity of the analytical method and the validity of the biological evaluation of the biological components to be measured are a prerequisite. Laboratory testing in the non-clinical field involve the analysis of animal samples, most of which use clinical laboratory devices or in vitro diagnostics. Regarding the optimal analytical method selected by each laboratory facility, the validity of the analytical method will be confirmed with reference to guidance such as "Consideration for the validation of clinical laboratory methods in nonclinical field". However, the difference in the reactivity between reference materials and animal samples and the understanding of the metrological traceability are issues for clinical pathology expert.

On the other hand, in non-clinical experiments, relative analytical values with precision may be used if comparison with the control group or initial values is possible, but these may cause intra- or inter-facility differences. In addition, factors that affect laboratory test values include physiological variation of animals and sample collection, and standardization of the pre-analytical process within a laboratory facility is an issue.

In this topic, I hope to share the background and issues of laboratory tests in the non-clinical field from the viewpoint of clinical pathology expert and discuss with other presenters and participants in this workshop leading to future activities.

Pharmaceutical companies outsource some or all of the safety studies, including GLP, to a non-clinical CRO (Contract Research Organization) to evaluate new drug candidates. The number of outsourced studies is increasing year by year, making non-clinical CROs indispensable partners for pharmaceutical companies.

In the case of outsourced studies, the final interpretation regarding the evaluation of the obtained study results is made by the study director of the contractor (non-clinical CRO), but in the process, the study director has scientific discussions with the clients (e.g., pharmaceutical companies). Although regulations regarding compliance monitoring such as GLP-related Japanese regulations and OECD Series No. 21 exist in discussions between the contractor and the clients, the details are not described, so that the way to proceed differs depending on each facility.

Furthermore, an increasing number of non-clinical CRO companies are able to measure new toxicity biomarkers that have been used in recent years, in addition to the traditional clinical pathology parameters normally measured in toxicity studies. In that case, the non-clinical CRO may also provide information necessary for toxicity evaluation regarding biomarkers (such as background values of biomarkers and study results using model animals).

In light of the situation, this presentation will introduce the difference in viewpoints between the contractor and the client in scientific discussions regarding clinical pathology results, based on the speaker's experience.

I hope the whole audience will participate in the comprehensive discussion during the final session of the symposium.

Non-clinical studies are conducted to establish appropriate dosing regimens for candidate compounds to ensure their safety in clinical trials and to evaluate them for toxicological changes.

In non-clinical studies, clinical pathology is performed on laboratory animals using human in vitro diagnostics and analysis equipment. The clinical pathology results are used in conjunction with the results of anatomic pathology and other examinations to evaluate the toxicity of the candidate compounds and make a comprehensive judgment on their clinical significance while taking into account the effects of species differences and animal-specific artifacts.

Since effects specific to animal experiments are seen in clinical pathology in non-clinical studies, knowledge on human clinical pathology cannot be directly applied. Nonetheless, there are few specialized texts on clinical pathology in laboratory animals, and because there is no systematic organization of the latest findings, the knowledge, techniques, and experience accumulated in each laboratory are passed on to new entrants to the field by word of mouth.

In this presentation, I will discuss the initiatives we have taken to pass on knowledge and the outlook for the future, while introducing actual examples of training we use in the clinical pathology laboratories at Shin Nippon Biomedical Laboratories.

The automated histopathology using whole slide images (WSIs) and deep learning-based algorithms (AI pathology) should be useful for efficient toxicity detection of candidate compounds. However, there are few reports on the prediction accuracy when dealing with a large number of findings or variations in hematoxylin and eosin staining of specimens. Additionally, while AI pathology generally makes predictions on small-sized images obtained by cropping WSIs, toxicological pathologists make comprehensive diagnoses at the whole slide level. Here, investigation was conducted to determine the extent to which the automated histopathology using AI pathology is possible. First, an AI pathology models using EfficientnetV2 were constructed for 18 findings using images obtained from WSIs, which were stored in Open TG-GATEs and were scanned from rat liver specimens created in multiple facilities with staining variations. The models achieved an average F1 score exceeding 0.946. Furthermore, when this model was used to automatically diagnose 11,299 rat liver WSIs stored in Open TG-GATES, it showed a high level of consistency with the diagnoses of pathologists (sensitivity, 0.913; recall, 0.870). Based on these results that our AI pathology models could predict generally consistent diagnoses with pathologists in numerous WSIs showing diverse staining, it was suggested that an AI pathology model capable of automating histopathological examination of rat livers can be constructed.

The Life Intelligence Consortium (LINC) is a non-profit organization aiming to promote the development of the life science field through AI, big data, and IoT applications. Researchers from life science companies, IT companies, and academia who agree with LINC's policies have gathered and taken on the challenge of solving various social issues by utilizing these technologies.

Histopathology of non-clinical research fields in the development of pharmaceuticals and agrochemicals is crucial for evaluating effects on organs⁄tissues; however, these often require significant effort, which becomes a bottleneck in the development process. In addition, since histopathology is a qualitative evaluation, different results may be obtained by different evaluators. Therefore, we had formed a working group to improve the efficiency and accuracy of pathological evaluation and have dedicated to developing a high-performance AI-driven image analysis⁄pathological diagnosis tool.

Our group have focused on the kidney as the main target organ. Pathologists have provided pathological images annotated with region and lesions and AI engineers have tried to develop AI models for region classification and anomaly detection. To account for variations in staining techniques and imaging devices, preprocessing methods have been explored to minimize the impact of color differences across different institutions. Our presentation will share the progress and discuss the benefits and challenges of developing pathology AI across multiple institutions and explore the potential for AI in non-clinical research fields.

In recent years, there has been growing expectation for the practical implementation of AI in the field of toxicologic pathology to alleviate the burden on pathologists reading a lot of tissue slides.

Fujifilm is conducting research on AI for screening of a lot of tissue slides. We have confirmed that a model created without training on real lesion images can detect 12 types of rat-liver lesions based on deviations from normal morphological characteristics.

Building a dataset with a sufficient scale and quality is crucial for the successful implementation of AI. Particularly, in toxicologic pathology, there exist challenges in both (1) collecting tissue images and (2) annotating them. (1) It is difficult to obtain lesion images, especially for rare lesions. Additionally, there are morphological variations in normal tissues due to factors such as sex and aging. Furthermore, technical variations in slide preparation are reflected in images, so there are various attributes to consider in collecting tissue images. (2) Lesions that are spatially and morphologically ambiguous introduce subjectivity in the annotation process. Ideally, the involvement of multiple pathologists would enhance the reliability of annotations, but this is often impractical due to their demanding schedules. Consequently, the current implementation is functionally restricted to in-house slides and specific lesions.

To implement a versatile AI, it is important to ensure diversity in images and reduce the burden of annotation for each institution. Cross-institution programs will accelerate these efforts.

In recent years, AI technology innovation has been advancing at a rapid pace, beginning to permeate various fields. Efforts to adapt AI technology to pathological histological evaluations in non-clinical research fields are likely being carried out by pharmaceutical companies and research institutions. Chugai Pharmaceutical is no exception, and has been implementing AI technology for non-clinical pathological histological evaluations since around 2018. They have established a database of pathological images from toxicity and efficacy tests, created their own evaluation AI models for several organs and disease animals, collaborated with external AI companies, used commercial software, and established hybrids of these, which are being effectively utilized. However, challenges have also emerged during the creation process. In particular, toxicological pathological evaluations often observe unknown findings, unlike efficacy test pathological evaluations. This makes it difficult, or even impossible, to define the findings that the AI model should detect. In addition, because there are many organs to evaluate, it is difficult for one model to adapt to all organs, and a number of models corresponding to the number of organs are required. Furthermore, countless models are needed for different ages, strains, and animal species. From another perspective, as software that excels in extracting specific observation areas is gradually entering the market, when creating models in-house or in collaboration with external AI companies, it is necessary to consider whether the actual utilization image and the corresponding cost are appropriate. In this presentation, I would like to discuss the challenges of implementing AI technology in non-clinical pathological histological analysis, as well as practical methods of utilizing digital pathology and AI pathology, along with case studies.

The advancement of digital technology has enabled high-resolution scanning of pathological tissue specimens, producing digital whole slide images (WSI) which can be evaluated in lieu of histological glass slides. The combination of artificial intelligence (AI) technology has accelerated the development of `AI-assisted digital pathology´, which is expected to have a wide range of applications from drug discovery to clinical diagnosis. However, there are several challenges such as regulatory acceptance for development and implementation under GLP. In our presentation, we will introduce the European BIGPICTURE project to create a large-scale database of WSI and our in-house efforts to implement digital pathology in GLP compliant toxicity studies.

[The BIGPICTURE Project] The BIGPICTURE project was established in 2021 to foster rapid development of AI in pathology. This project is a European public-private partnership co-funded by the Innovative Medicines Initiative (IMI) consortium, a joint undertaking between the European Union and the European Federation of Pharmaceutical Industries and Associations (EFPIA). Its goal is to build a large-scale platform for sharing clinical and non-clinical WSI and AI algorithms. Other important objectives of the BIGPICTURE project are to develop regulatory, legal, and ethical frameworks for the optimal use of WSI and AI algorithms for non-clinical safety studies and clinical use. This project is the first European initiative to connect veterinary (toxicologic) and clinical pathologists, researchers, AI developers, patients, industry, and regulatory authorities in an ethical, quality-controlled platform, fully compliant with the EU General Data Protection Regulation (GDPR). It is being developed in a sustainable and comprehensive manner, and its achievements and challenges will be presented.

[GLP Digital Pathology] Boehringer Ingelheim is a German pharmaceutical company dedicated to the global development of innovative medicines and treatments. Through the digitalization of toxicological pathology, timely scientific discussions between pathologists across countries and regions have become possible. Therefore, we are working towards the implementation of digital pathology in GLP-compliant toxicity tests to accelerate the research and development of new drugs.

Spatial transcriptome is an innovative technology that comprehensively measures gene expression levels while preserving the spatial information of tissues. Commercially available spatial transcriptome methods, such as Visium (10x Genomics), allow for overlaying measurement data with H&E-stained images, thereby elucidating the relationship between cellular morphological features and gene expression levels. However, challenges such as a shortage of transcriptomics researchers capable of extracting information from H&E-stained images, high experimental costs, partial missing regions measuring expression levels, and insufficient measurement resolution have impeded the full utilization of spatial transcriptome data. To address these issues, we developed a deep learning method, DeepSpaCE, which uses H&E-stained images as input data and gene expression levels (or cluster information) as output data (Monjo, Koido, et al. Sci Rep. 2022). DeepSpaCE offers solutions to the challenges above and further enhances the value of spatial transcriptomics. For instance, it enables the prediction of gene expression levels within tens of micrometers on H&E-stained images of the same tissue with just a minimum of one experiment of spatial transcriptome. DeepSpaCE is available on GitHub, facilitating users to easily learn and predict with their spatial transcriptomics data (https://github.com/tmonjo/DeepSpaCE). In this presentation, I would like to discuss the benefits and limitations of DeepSpaCE's predictive capabilities and its potential applications in toxicity research.

Artificial intelligence (AI) contributes to various fields of science and　technology, such as improving the efficiency of research and development and creating new methods. In the field of toxicology, AI such as machine learning are also expected to find toxicity-related information from accumulated large-scale experimental data, which will be useful in evaluating the safety of chemical substances. Machine learning is a technology that allows computers to discover patterns hidden in data and make predictions about new data. Technological progress in machine learning, including deep learning, has been remarkable, and it is now possible to handle unstructured data such as images and text. In addition, generative AI and explainable AI are also attracting attention. In the field of toxicology, there are situations in which it is difficult to utilize AI. Still, the foundations for accelerating the use of AI are being laid, such as the publication of high-throughput screening data and approach to few data. However, when utilizing AI technology, it is important to consider data quality, model reproducibility, and interpretability. Our laboratory focuses on toxicity-related databases and machine learning from the perspective of regulatory science. In this workshop, I will introduce skin sensitization prediction models and side effect prediction models in humans.

In vitro evaluation of pharmacokinetics and drug safety is an important factor in drug discovery and development. However, conventional methods including primary culture cells have low cellular responsiveness in terms of drug metabolism and toxicity. To overcome the concern, microphysiological systems such as organoids are attracting attention. The authors investigate construction of novel safety evaluation systems using microphysiological systems in liver and kidney focusing on the regulatory factors for pharmacokinetics. Hepatic drug disposition consists of uptake, metabolism, and biliary excretion; however in primary cultured hepatocytes, the bile canaliculi is closely formed between the cells, leading to difficulty to collect biliary excreted compounds. The authors found that open hemi-luminal bile canaliculi were formed on the cell culture insert coated with the adhesion protein claudins. This hepatocyte culture system icHep (induced open-form canalicular hepatocyte) can be applied to a permeability assays for quantitative evaluation of drug bile excretion¹⁾. The authors also investigate the development of an evaluation system for drug-induced kidney injury (DIKI). In three-dimensional cultured human proximal tubular epithelial cells (3D-RPTEC^®), the expression levels of major drug transporters such as the organic anion transporter OAT1 are comparable to those of human kidney cortex²⁾. The combination of ATP content measurement and high content analysis showed high sensitivity of detecting DIKI compared with conventional 2D-cultured RPTECs. In conclusion, the microphysiological systems considering pharmacokinetic properties improve prediction ability for drug biliary disposition and DIKI.

1) Arakawa et al., Commun Biol, 6:866 (2023).

2) Ishiguto et al., Drug Metab Dispos, 51:1177-87 (2023).

We are surrounded by a variety of fragrances such as perfumes and fabric softeners. While these fragrances have been reported to have relaxing effects, in recent years, an increasing number of people suffer from multiple chemical sensitivity (MCS), in which headaches and discomfort are caused by these fragrance substances. However, the mechanism of their action on the brain remain unknown. We hypothesized that not only the olfactory information of fragrance substances, but also the transfer of fragrance substances to the brain via the nasal route, may show direct toxic effects on the brain. Since genetic polymorphisms of drug-metabolizing enzymes are involved in MCS, we also focused on drug-metabolizing enzymes in the brain and attempted to elucidate the toxicity of fragrant substances on the brain.

Mice were exposed intranasally to several fragrance substances commonly contained in perfumes and fabric softeners, and unpleasant behavior was confirmed by behavioral tests and heart rate measurements. After exposure of mice to these fragrance substances, we analyzed the fragrance substances in the brain using GC-MS, and found that the fragrance substances accumulated in various parts of the brain, and that the accumulation of these substances in the brain was greater than in other organs such as the liver. In addition, metabolites of these fragrance substances were detected in the brain, suggesting that the fragrance substances are metabolized by drug-metabolizing enzymes after entering the brain. These fragrance substances or their metabolites also affect the function of neurons and astrocytes. The present study found that fragrance substances may have direct toxic effects on the brain by migrating to the brain and undergoing metabolism.

Recent studies have demonstrated a close relationship between the gut microbiota in the intestinal tract and the host's health status. Gut bacteria are not just "symbionts" but also play an important role in regulating various physiological functions of the host, including metabolism, immunity, and neurology. These diverse physiological regulatory functions are attributed to the intestinal bacteria’s ability to produce various metabolites, including short-chain fatty acids, branched-chain amino acids, aromatic amino acids, trimethylamines, secondary bile acids, and sulfur metabolites. Gut bacteria metabolize carbohydrates, proteins, fats, and many other compounds not digested and absorbed by the host, transferring them into the bloodstream as a variety of metabolites, which act as bioactive substances for the host. Recent innovations in metabolomics and metagenomics have led to the discovery of many metabolites derived from intestinal bacteria and genes related to their production, and the intestinal microbiota is considered a "metabolic organ" with a metabolic capacity equivalent to that of the liver. On the other hand, because they reside on the luminal side of the intestinal tract (outside the body), various environmental factors can easily affect gut bacteria. Furthermore, feces ultimately expel intestinal bacteria from the body, making it simple and noninvasive to collect samples of intestinal bacteria that reflect the body's environment. Therefore, the gut microbiota is a new target area for environmental chemical exposure effects. In this presentation, the speaker will discuss the findings on the toxicological effects of exposure to environmental chemicals on the intestinal microbiota and the possibility of a new indicator of effects, i.e., qualitative changes in the intestinal microbiota, based on adduct science, in addition to the conventional compositional (quantitative) changes in the intestinal microbiota.

As can be imagined from the example of dressing, we often see liquids like water and oil separated in our daily lives. In the mid-to-late 2010s, membrane-less compartmentalization within cells was discovered as "intracellular phase separation" or “liquid-liquid phase separation (LLPS)” and it has been attracting attention as a new framework for understanding biological phenomena. Such intracellular phase separation phenomena occur within cells, in which the environment mainly uses water as a medium. The phase separation of the aqueous medium has been known since the 1960s for water-soluble polymer solutions. In this presentation, I would like to outline in vitro experiments of LLPS using solutions of water-soluble polymers and discuss potential connections with toxicology.

Although sleep is a ubiquitous behavior in animal species with a nervous system, many aspects in the neurobiology of sleep remain mysterious. Our discovery of orexin, a hypothalamic neuropeptide involved in the maintenance of wakefulness, has triggered intensive research examining the exact role of the orexinergic and other neuronal pathways in the regulation of sleep/wakefulness. Orexin receptor antagonists, which specifically block the endogenous waking system, have been approved as a new drug to treat insomnia. Also, since the sleep disorder narcolepsy-cataplexy is caused by orexin deficiency, orexin receptor agonists are expected to provide mechanistic therapy for the disease; they will likely be also useful for treating excessive sleepiness due to other etiologies.

Even though the executive neurocircuitry and neurochemistry for sleep/wake switching, including the orexinergic system, has been increasingly revealed in recent years, the mechanism for homeostatic regulation of sleep, as well as the neural substrate for "sleepiness" (sleep pressure), remains unknown. To crack open this black box, we have initiated a large-scale forward genetic screen of sleep/wake phenotype in mice based on true somnographic (EEG/EMG) measurements. We have so far screened >10,000 heterozygous ENU-mutagenized founders and established several pedigrees exhibiting heritable and specific sleep/wake abnormalities. By combining linkage analysis and the next-generation whole exome sequencing, we have molecularly identified and verified the causal mutation in several of these pedigrees. Since these dominant mutations cause strong phenotypic traits, we expect that the mutated genes will provide new insights into the elusive pathway regulating sleep/wakefulness. Indeed, through a systematic cross-comparison of the SIK3 Sleepy mutants and sleep-deprived mice, we have found that the cumulative phosphorylation state of a specific set of mostly synaptic proteins may represent the molecular substrate of sleep pressure. We have also found that the neuronal molecular pathway LKB1-SIK3-HDAC4/5 may represent the level of sleep pressure, regulating the amount, depth, and timing of sleep by acting in different brain regions, respectively (Kim et al. Nature 612: 512-518, 2022; Zhou et al. Nature 612: 519-527, 2022).

The liver is a primary target of toxicity of chemical substances such as pharmaceuticals and pesticides, and species differences are often observed in the hepatotoxicity of chemicals. Therefore, understanding the mechanisms of hepatotoxicity is essential for developing mechanism-based evaluation systems for risk assessment of chemicals, considering species differences.

The nuclear receptor CAR has been identified as essential for phenobarbital-dependent enzyme induction and hepatocarcinogenesis, but the carcinogenic mechanism remained unclear for a long time. In addition, the molecular mechanism for the species differences in this CAR-mediated hepatocarcinogenesis also remained unknown. We have demonstrated that CAR induces hepatocyte proliferation in mice by activating the transcriptional co-activator YAP, an effector molecule of the Hippo pathway involved in cell and organ size control. Furthermore, identifying the YAP-binding motif of CAR has revealed a one amino acid difference between the motifs of human and rodent CAR, which determines the rodent specificity in the CAR-mediated hepatocyte proliferation.

We have also demonstrated that although the nuclear receptor PXR does not show liver cancer-promoting activity, it enhances hepatocyte proliferation induced by CAR, PPARα, or hepatocellular injury. In contrast, PXR markedly suppressed the progression of CAR-dependent liver cancer in a two-stage liver carcinogenesis model. These results indicate that CAR and PXR have similar functions in inducing drug-metabolizing enzymes but have different roles in hepatocyte proliferation and hepatocarcinogenesis.

Our findings provide valuable information not only from the basic toxicology of carcinogenesis but also from a regulatory science perspective.

In the award lecture, I will outline the history of my international contributions to JSOT.

Reactive oxygen species (ROS) are produced in various pathways and induce oxidative stress in the body. On the other hand, there are also protective and eliminating mechanisms against such oxidative stress, and the imbalance between them leads to many diseases. In this presentation, I will share two studies that I have been involved in related to oxidative stress.

1. Suppression of cytochrome P450 (P450, CYP) via protein-protein interaction

P450 is one of the major drug-metabolizing enzymes that receives electrons from NADPH-P450 reductase and activates molecular oxygen to catalyze the oxidation of various hydrophobic chemicals. In addition to such involvement in detoxication, P450 is a source of ROS in the endoplasmic reticulum (ER) since electrons not used for substrate oxidation are released as superoxide and hydrogen peroxide. Our team found that another drug-metabolizing enzyme, UDP-glucuronosyltransferase (UGT), forms a complex with CYP3A4 and suppresses CYP3A4-mediated oxidation for the first time. It was suggested that UGT suppresses the whole catalytic cycle of CYP3A4, including not only substrate oxidation but also the release of hydrogen peroxide, by dragging the P450 into the interior of the ER membrane and inhibiting the substrate-binding to it. Such suppression of CYP3A by UGT was also supported by in vivo studies using rat liver. The results are expected to contribute to the elucidation of inter- and intra-individual differences in drug-metabolizing enzyme activities and the prevention of oxidative stress caused by P450.

2. Elucidation of the mechanism underlying cytotoxicity caused by Maillard reaction products

Maillard reaction products are produced by the non-enzymatic reaction of reducing sugars, such as glucose, with amino acids, and accumulate in the body with age, leading to diseases such as diabetes. Although a group of compounds called "advanced glycation end products (AGEs)" has been well studied, the effects of other Maillard reaction products have not been elucidated. We have studied 3-hydro-2,2,5,6-tetramethylpyrazine (DHP-3), a highly cytotoxic dihydropyrazine formed by the dimerization of glucosamine or 5-aminolevulinic acid, comparing it with other Maillard reaction products to investigate the molecular mechanism of its cytotoxicity. First, we generated HeLa cells transiently expressing the plasma membrane receptor for AGEs (RAGE), which recognizes AGEs and transmits inflammatory signals into the cell to investigate the possibility that the receptor is involved in the cytotoxicity of DHP-3. However, overexpression of RAGE did not affect the DHP-3 sensitivity of HeLa cells, suggesting that the cytotoxicity of DHP-3 is independent of RAGE. We also found that DHP-3 induced more oxidative stress than the typical Maillard reaction products, N^ε-(carboxymethyl)-L-lysine and acrylamide, and that this difference led to pronounced cytotoxicity. Maillard reaction products are known to be generated in the cooking and producing processes of food. I would like to continue the toxicological analysis of Maillard reaction products for the safety of food.

Cleft palate (CP) is one of the most common birth defects worldwide (affects approximately 1:500 live births). The etiology of CP is multifactorial and involves environmental and genetic factors; for example, disruption of signaling pathways by gene mutation induce CP. Environmental factors such as medication usage, cigarette smoking, and alcohol consumption are well-known risk factors for CP since they lead to suppression of the expression of specific genes or signaling pathways. Although novel gene mutations associated with CP have been reported, it remains unclear how environmental and genetic factors affect CP.

MicroRNAs (miRNAs) are small endogenous RNA (18–24 nucleotides long) that regulate post-transcriptional gene expression. Recent studies suggested that miRNAs play crucial roles in palate development in human and mice. However, it is still unclear how and which miRNAs play important roles in CP.

To address this black box, I searched the miRNAs associated with mouse and human CP by combination with systematic review, bioinformatics, and cell proliferation assays. Using above combination test, I identified several miRNAs associated with CP both in human and mice. Additionally, I demonstrated that medicine (all-trans retinoic acid, phenobarbital)-induced CP induced specific miRNAs and miRNA inhibitor restored medicine-induced toxicity. The understanding of miRNA dysregulation by medicine will shed light on the link between environmental factor and genetic factor in CP.

Degranulation of mast cells is caused by two distinct pathways: IgE-dependent and IgE-independent pathway. IgE-independent reaction is referred to as "pseudo-allergic" reaction, distinguishing it from IgE-dependent allergic reactions. In recent years, Mas-related G-protein coupled receptor X2 (MRGPRX2) has been identified as a crucial receptor for pseudo-allergic reactions. Dogs, commonly used in nonclinical safety studies, are known to be highly sensitive to pseudo-allergic reactions. However, the functional ortholog of human MRGPRX2 in dogs has not been identified, and the mechanism behind the high sensitivity in dogs remains unknown. Since there are rarely high-sensitive subjects which show severe pseudo-allergic reactions in humans, while research on human individual susceptibility differences is being investigated, little is known about the factors underlying human hypersensitivity. Therefore, elucidating the mechanism of high sensitivity in dogs would contribute to the advancement of research on hypersensitivity mechanisms in humans. Additionally, species differences in sensitivity to pseudo-allergic reactions make it challenging to extrapolate histamine release reactions observed in nonclinical studies to humans, highlighting the need to establish an assay system that can evaluate species differences.

The aim of this study was to identify the mechanism of sensitivity differences in MRGPRX2-related reactions and establish an assay system. First, we attempted to determine the functional ortholog in dogs. Through canine tissue analysis, canine MRGPRX2 was found to be expressed in limited tissues including the skin, and connective tissue-type mast cells residing in the skin expressed canine MRGPRX2. In evaluations using receptor-expressing cells, histamine-releasing agents activated canine MRGPRX2-expressing cells. Based on these results, canine MRGPRX2 was identified as the functional ortholog of human MRGPRX2. Next, we investigated the mechanism of high sensitivity in dogs from the perspective of receptor structure. The important residues involved in receptor binding were estimated by homology modeling and docking simulations, and reactivity of the mutant variants against MRGPRX2 ligands was evaluated with the variant-expressing cells. As a result, we identified residues M109 and F78 in human MRGPRX2 as contributing to high sensitivity to fluoroquinolones. This led to the first identification of gain-of-function variants in the transmembrane region of human MRGPRX2. Finally, we verified the usefulness of an assay system with MRGPRX2-expressing cells for evaluation of species differences using a launched drug, valemetostat, which induced histamine release in dogs during nonclinical safety studies. As a result, the sensitivity of human MRGPRX2-expressing cells to valemetostat was significantly lower compared to canine MRGPRX2-expressing cells, indicating the presence of a species difference in sensitivity to valemetostat-induced histamine release. Indeed, no increase in blood histamine levels was observed in clinical trials with valemetostat. Notably, the EC₅₀ obtained in this study using MRGPRX2-expressing cells showed a correlation with the plasma concentration at which histamine release occurred in dogs, and it was also consistent with the results in clinical trials. This assay system would be useful for mechanistic analysis of histamine release reactions observed in nonclinical studies and extrapolation to humans.

Firefly bioluminescence is the most widely used bioluminescence system in the field of bioluminescence imaging. However, the yellow-green light of firefly bioluminescence is not able to penetrate deeper through living organisms. To overcome the problem, we have developed three longer wavelength luciferin analogues, "AkaLumine", "TokeOni" and "SeMpai" based on the substrate of the firefly bioluminescence system. The luminescence induced by these analogues is extremely transmissive to living organisms. Unexpectedly, even though the luminescent enzyme gene was not introduced, we observed the luminescence from the liver of mice in vivo when AkaLumine and TokeOni were administered. In the process of several analyses to elucidate the mechanism of this unknown luminescence phenomenon, we found that inhibitors for some of cytochrome P450s (CYPs) significantly attenuated the luminescence. To investigate whether this luminescence phenomenon is also observed in other animal species, TokeOni was added to the extracts of blow flies and pill bugs. As a result, luminescence was observed in both species, and it was also attenuated by several CYP inhibitors. These results suggest that TokeOni luminescence may be caused by a mechanism involving CYPs in various animal species. On the other hand, it has been reported that a putative fatty acyl CoA synthetase has luminescent activity against several luciferin analogs in fruit flies. We would like to report the results of our investigation into the possible involvement of fatty acyl CoA synthases in the luminescence phenomena by TokeOni.

Cisplatin is used as first-line therapy for some types of solid tumors, although it can cause muscle atrophy. Mitochondrial dysfunction is expected as one of the causes that promote muscle atrophy because muscle weakness is induced during muscle atrophy. This study aimed to investigate the effects of cisplatin in ATP levels and gene expressions of mitochondrial OXPHOS and the related factors of mitochondrial biosynthesis. C57BL/6J mice were injected with cisplatin for four consecutive days. Quadriceps muscles were isolated 24 hours after the last injection of cisplatin. C2C12 myotubes were also treated with cisplatin. Systemic cisplatin administration decreased quadriceps muscle mass. Additionally, skeletal muscle strength and endurance were reduced by cisplatin administration. ATP levels and gene expressions of mitochondrial OXPHOS-related factors in quadriceps muscle were significantly lower in the cisplatin group. The mitochondrial DNA (mtDNA)/nuclear DNA (nDNA) ratio was also reduced in the cisplatin group. PGC-1α was significantly downregulated among the related factors of mitochondrial biogenesis. The PGC-1α expression was reduced in the cisplatin-treated group, as well as other PGC-1α isoforms. Cisplatin was observed to have similar effects in C2C12 myotubes as in mouse skeletal muscle. PGC-1α downregulation in the skeletal muscle may be one of the key factors that play an important role in mitochondrial dysfunction and reduced ATP levels and may be a factor in muscle weakness during cisplatin-induced muscle atrophy.

Aripiprazole is an atypical antipsychotic drug whose main pharmacological effect is partial agonism of the dopamine D2 receptor. Its long-acting injection (LAI) formulation is used to treat schizophrenia and bipolar disorders. Medication adherence is often the key to preventing the recurrence of both diseases, and it has been noted that LAI formulations are more effective than oral formulations in preventing recurrence. Because the LAI formulation is initiated after appropriate dosing, and tolerability has been confirmed with the oral formulation of the same ingredient, it is usually assumed that the only major adverse event to observe is injection site reactions if there are no adverse reaction problems with the oral formulation.

Aripiprazole is mainly metabolized by the drug-metabolizing enzymes cytochrome P450 (CYP) 3A4 and CYP2D6.

The metabolic capacity of CYP2D6 varies among individuals and can be classified into three categories: extensive metabolizers (EM), where enzyme activity is normal; intermediate metabolizers (IM), where activity is moderately reduced; and poor metabolizers (PM). The percentage of IM in Japanese patients has been reported to be approximately 40%.

In this study, we report a case of drug-induced liver injury in a manic patient with bipolar disorder who was started on oral aripiprazole and switched to LAI formulation after confirming that the patient tolerated the drug. We report this case because genetic polymorphism of a drug-metabolizing enzyme may have contributed to the development of drug-induced injury in this patient.

There are a wide variety of food ingredients, including molecules that "enhance" and "inhibit the function of certain bioactive substances" - unknown molecules! Since its physiological activity is exerted in a complex system in which - coexists, the component ratio is the key. In order to optimize the expression of bioactivity, it is necessary to change the component ratio freely, however, there is no such thing. Thus we have developed a bidirectional mass transfer device. In this device, when a solution with different component ratios is sent to the upper and lower flow paths in the filter, a low molecular weight substance moves in both directions in the upper and lower flow paths due to the balance between the diffusion rate constant and the flow velocity. By changing the delivery time, the component composition on both sides can be varied step by step. In this article, we will report on the performance of the device.

We have shown that this device has the ability to change the component ratio of the solution flowing to one side reproducibly by dialyzing the circulating water on the other side.

Rice constitutes half of the world’s staple food, including Malaysia. As a result, rice plantation on a large scale has become a national necessity. Pesticides were applied to meet demand and prevent production loss. The pesticide residue, particularly arsenic pesticides, has the potential to leach into the ground, leading to soil contamination and the presence of inorganic arsenic (iAs) in paddy plants. However, reports on iAs contamination in Malaysian rice remain limited. Therefore, this study was performed (1) to develop and validate a reliable analysis for iAs (As (III), As(V)) in rice by using HPLC-ICP-MS and (2) to determine the iAs levels in locally sold rice using the established and validated method. The sample was analysed for total arsenic before iAs extraction. The coefficient of determination (R2) calculation for the analytes was above 0.99. Also, the analytes exhibited a limit of detection at 0.0015 mg kg -1, including the limit of quantification of 0.005mg kg -1. Calibration standards were established in solution at 1.0µg kg -1 and 10.0 µg kg -1. Recovery studies showed 99% to 119% for As(III) and 109% to 131% for As(V). The method was tested on the FAPAS Proficiency Test sample with a z-score of 1.09. Subsequently, 23 rice varieties were tested to measure the iAs content using the convenience sampling method. None of the samples violated the CODEX permissible level (0.2 mg/kg to 0.35 mg/kg) or the maximum permitted limit of European EU regulation (0.1 mg/kg to 0.3 mg/kg). In short, this study summarised that the level of iAs in rice sold in the Malaysian market is generally low.

Epidemiological and experimental studies have shown that exposure to dioxin impairs cognitive and neurobehavioral functions in humans and rodents. The aryl hydrocarbon receptor (AhR), a ligand-activated transcription factor, binds to dioxin and translocates from the cytoplasm into the nucleus, where it induces expression of multiple target genes. Therefore, it is necessary for identification of AhR-expressing neurons in the brain to clarify the molecular mechanism of dioxin neurotoxicity. Immunohistochemistry was performed to detect AhR in the mouse brain and confirm the specificity of the anti-AhR antibody using Ahr-/- mice. The intracellular distribution of AhR and expression level of the target genes (Cyp1a1, Cyp1b1, and Ahrr) were analyzed by microscopic imaging and quantitative PCR, respectively, using adult mice orally exposed to 2,3,7,8-tetrachlorodibenzo-p-dioxin (TCDD). During developmental period in Ahr+/+ mice, AhR expression was found in neurons of the island of Calleja major (ICjM), but not in Ahr-/- mice. In the TCDD-exposed mice, AhR was significantly translocated into the nucleus in ICjM neurons. Additionally, TCDD exposure significantly upregulated expression of the target genes in the brain, suggesting disruption of AhR-regulated genes and signaling pathways in ICjM neurons. These results provide experimental evidence that oral exposure to dioxin activates AhR in brain neurons. Further studies focusing on the ICjM are needed to discover unknown mechanisms of dioxin neurotoxicity.

The prefrontal cortex (PFC) is central to integrating higher brain activities such as working memory, mnemonic memory, and social cognition. Disruption of its intrahemispheric conduction and interhemispheric connectivity significantly affects the pathology of neuropsychiatric disorders. Thus, PFC function should be a critical target for toxicological effects on the brain. The aim of this study was to generate a functional map of the intra- and interhemispheric connections of the PFC. The interhemispheric connection within the PFC through the corpus callosum (CC) plays a central role in pathology and abnormalities. The CC, particularly at the anterior cingulate cortex (ACC), is involved in interhemispheric bilateral propagation of epileptic discharges and is, therefore, a focus for surgical treatment of epilepsy. However, the mechanism by which ipsilateral neural activity propagates to the CC and spreads to the contralateral hemisphere, particularly the frontal lobe, is poorly understood. Here, we aimed to create a functional map of these neural activities in the PFC using large-scale voltage-sensitive dye (VSD) imaging with high speed (1 ms/frame), high resolution (256 × 256 pixels), and a large field of view (10 mm x 10 mm). Cuprizone is a chemical known to induce demyelination and cause multiple sclerosis (MS) symptoms. We have demonstrated that VSD imaging can clearly show the dysfunction of the CC caused by a 6-week cuprizone-containing diet on PFC function. The results highlight the usefulness of large-scale VSD imaging of the brain cortical function and its modulation by toxic substances.