The developing brain is extremely sensitive to many chemicals. Exposure to neurotoxicants during development has been implicated in various neuropsychiatric and neurological disorders, including autism spectrum disorder, attention deficit hyperactive disorder, schizophrenia, Parkinson's disease, and Alzheimer's disease. Although rodents have been widely used for developmental neurotoxicity testing, experiments using large numbers of rodents are timeconsuming, expensive, and raise ethical concerns. Using alternative non-mammalian animal models may relieve some of these pressures by allowing testing of large numbers of subjects while reducing expenses and minimizing the use of mammalian subjects. In this review, we discuss some of the advantages of using zebrafish in developmental neurotoxicity testing, focusing on central nervous system development, neurobehavior, toxicokinetics, and toxicodynamics in this species. We also describe some important examples of developmental neurotoxicity testing using zebrafish combined with gene expression profiling, neuroimaging, or neurobehavioral assessment. Zebrafish may be a systems toxicology model that has the potential to reveal the pathways of developmental neurotoxicity and to provide a sound basis for human risk assessments.

Hypertrophic cardiomyopathy (HCM) is characterized by left ventricular hypertrophy and is associated with a number of potential outcomes, including impaired diastolic function, heart failure, and sudden cardiac death. Various etiologies have been described for HCM, including pressure overload and mutations in sarcomeric and non-sarcomeric genes. However, the molecular pathogenesis of HCM remains incompletely understood. In this study, we performed comparative transcriptome analysis to identify dysregulated genes common to five mouse HCM models of differing etiology: (i) mutation of myosin heavy chain 6, (ii) mutation of tropomyosin 1, (iii) expressing human phospholamban on a null background, (iv) knockout of frataxin, and (v) transverse aortic constriction. Gene-by-gene comparison identified five genes dysregulated in all five HCM models. Glutathione S-transferase kappa 1 (Gstk1) was significantly downregulated in the five models, whereas myosin heavy chain 7 (Myh7), connective tissue growth factor (Ctgf), periostin (Postn), and reticulon 4 (Rtn4) were significantly upregulated. Gene ontology comparison revealed that 51 cellular processes were significantly enriched in genes dysregulated in each transcriptome dataset. Among them, six processes (oxidative stress, aging, contraction, developmental process, cell differentiation, and cell proliferation) were related to four of the five genes dysregulated in all HCM models. GSTK1 was related to oxidative stress only, whereas the other four genes were related to all six cell processes except MYH7 for oxidative stress. Gene–gene functional interaction network analysis suggested correlative expression of GSTK1, MYH7, and actin alpha 2 (ACTA2). To investigate the implications of Gstk1 downregulation for cardiac function, we knocked out gstk1 in zebrafish using the clustered regularly interspaced short palindromic repeats/Cas9 system. We found that expression of the zebrafish homologs of MYH7, ACTA2, and actin alpha 1 were increased in the gstk1-knockout zebrafish. In vivo imaging of zebrafish expressing a fluorescent protein in cardiomyocytes showed that gstk1 deletion significantly decreased the end diastolic volume and, to a lesser extent, end systolic volume. These results suggest that downregulation of GSTK1 may be a common mechanism underlying HCM of various etiologies, possibly through increasing oxidative stress and the expression of sarcomere genes.

Sleep-wake states are impaired in various neurological disorders. Impairment of sleep-wake states can be an early condition that exacerbates these disorders. Therefore, treating sleep-wake dysfunction may prevent or slow the development of these diseases. Although many gene products are likely to be involved in the sleep-wake disturbance, hypnotics and psychostimulants clinically used are limited in terms of their mode of action and are not without side effects. Therefore, there is a growing demand for developing new hypnotics and psychostimulants with high efficacy and few side effects. Toward this end, animal models are indispensable for use in genetic and chemical screens to identify sleep-wake modifiers. As a proof-of-concept study, we performed behavioral profiling of zebrafish treated with chemical and genetic sleep-wake modifiers. We were able to demonstrate that behavioral profiling of zebrafish treated with hypnotics or psychostimulants from 9 to 10 days post-fertilization was sufficient to identify drugs with specific modes of action. We were also able to identify behavioral endpoints distinguishing GABA-A modulators and hypocretin (hcrt) receptor antagonists and between sympathomimetic and non-sympathomimetic psychostimulants. This behavioral profiling can serve to identify genes related to sleep-wake disturbance associated with various neuropsychiatric diseases and novel therapeutic compounds for insomnia and excessive daytime sleep with fewer adverse side effects.

Oligodendrocytes are major myelin-producing cells and play essential roles in the function of a healthy nervous system. However, they are also one of the most vulnerable neural cell types in the central nervous system (CNS), and myelin abnormalities in the CNS are found in a wide variety of neurological disorders, including multiple sclerosis, adrenoleukodystrophy, and schizophrenia. There is an urgent need to identify small molecular weight compounds that can stimulate myelination. In this study, we performed comparative transcriptome analysis to identify pharmacodynamic effects common to miconazole and clobetasol, which have been shown to stimulate myelination by mouse oligodendrocyte progenitor cells (OPCs). Of the genes differentially expressed in both miconazole- and clobetasol-treated mouse OPCs compared with untreated cells, we identified differentially expressed genes (DEGs) common to both drug treatments. Gene ontology analysis revealed that these DEGs are significantly associated with the sterol biosynthetic pathway, and further bioinformatics analysis suggested that sterol regulatory element binding factors (SREBFs) might be key upstream regulators of the DEGs. In silico screening of a public database for chemicals associated with SREBF activation identified fenofibrate, a peroxisome proliferator-activated receptor α (PPARα) agonist, as a drug that increases the expression of known SREBF targets, raising the possibility that fenofibrate may also stimulate myelination. To test this, we performed in vivo imaging of zebrafish expressing a fluorescent reporter protein under the control of the myelin basic protein (mbp) promoter. Treatment of zebrafish with fenofibrate significantly increased expression of the fluorescent reporter compared with untreated zebrafish. This increase was attenuated by co-treatment with fatostatin, a specific inhibitor of SREBFs, confirming that the fenofibrate effect was mediated via SREBFs. Furthermore, incubation of zebrafish with another PPARα agonist, gemfibrozil, also increased expression of the mbp promoter-driven fluorescent reporter in an SREBF-dependent manner. These results suggest that activation of SREBFs by small molecular weight compounds may be a feasible therapeutic approach to stimulate myelination.

Pulmonary arterial hypertension (PAH) is a heterogeneous disorder associated with a progressive increase in pulmonary artery resistance and pressure. Although various therapies have been developed, the 5-year survival rate of PAH patients remains low. There is thus an important need to identify novel genes that are commonly dysregulated in PAH of various etiologies and could be used as biomarkers and/or therapeutic targets. In this study, we performed comparative transcriptome analysis of five mammalian PAH datasets downloaded from a public database. We identified 228 differentially expressed genes (DEGs) from a rat PAH model caused by inhibition of vascular endothelial growth factor receptor under hypoxic conditions, 379 DEGs from a mouse PAH model associated with systemic sclerosis, 850 DEGs from a mouse PAH model associated with schistosomiasis, 1598 DEGs from one cohort of human PAH patients, and 4260 DEGs from a second cohort of human PAH patients. Gene-by-gene comparison identified four genes that were differentially upregulated or downregulated in parallel in all five sets of DEGs. Expression of coiled-coil domain containing 80 (CCDC80) and anterior gradient two genes was significantly increased in the five datasets, whereas expression of SMAD family member six and granzyme A was significantly decreased. Weighted gene co-expression network analysis revealed a connection between CCDC80 and collagen type I alpha 1 (COL1A1) expression. To validate the function of CCDC80 in vivo, we knocked out ccdc80 in zebrafish using the clustered regularly interspaced short palindromic repeats (CRISPR)/Cas9 system. In vivo imaging of zebrafish expressing a fluorescent protein in endothelial cells showed that ccdc80 deletion significantly increased the diameter of the ventral artery, a vessel supplying blood to the gills. We also demonstrated that expression of col1a1 and endothelin-1 mRNA was significantly decreased in the ccdc80-knockout zebrafish. Finally, we examined Ccdc80 immunoreactivity in a rat PAHmodel and found increased expression in the hypertrophied media and adventitia of the pre-acinar pulmonary arteries (PAs) and in the thickened intima, media, and adventitia of the obstructed intra-acinar PAs. These results suggest that increased expression of CCDC80 may be involved in the pathogenesis of PAH, potentially by modulating the expression of endothelin-1 and COL1A1.