Previously a standard toxicological test termed as DarT (Danio rerio Teratogenic assay) using wild type zebrafish embryos has been established and it is widely applied in toxicological and chemical screenings. As an increasing number of fluorescent transgenic zebrafish lines with specific fluorescent protein expression specifically expressed in different organs and tissues, we envision that the fluorescent markers may provide more sensitive endpoints for monitoring chemical induced phenotypical changes. Here we employed Tg(nkx2.2a:mEGFP) transgenic zebrafish which have GFP expression in the central nervous system to investigate its potential for screening neurotoxic chemicals. Five potential neurotoxins (acetaminophen, atenolol, atrazine, ethanol and lindane) and one neuroprotectant (mefenamic acid) were tested. We found that the GFP-labeled ventral axons from trunk motoneurons, which were easily observed in live fry and measured for quantification, were a highly sensitive to all of the five neurotoxins and the length of axons was significantly reduced in fry which looked normal based on DarT endpoints at low concentrations of neurotoxins. Compared to the most sensitive endpoints of DarT, ventral axon marker could improve the detection limit of these neurotoxins by about 10 fold. In contrast, there was no improvement for detection of the mefenamic acid compared to all DarT endpoints. Thus, ventral axon lengths provide a convenient and measureable marker specifically for neurotoxins. Our study may open a new avenue to use other fluorescent transgenic zebrafish embryos/fry to develop sensitive and specific toxicological tests for different categories of chemicals.

Previously we have developed a transgenic zebrafish line (LiPan) with liver-specific red fluorescent protein (DsRed) expression under the fabp10a promoter. Since red fluorescence in the liver greatly facilitates the observation of liver in live LiPan fry, we envision that the LiPan zebrafish may provide a useful tool in analyses of hepatotoxicity based on changes of liver red fluorescence intensity and size. In this study, we first tested four well-established hepatotoxins (acetaminophen, aspirin, isoniazid and phenylbutazone) in LiPan fry and demonstrated that these hepatotoxins could significantly reduce both liver red fluorescence and liver size in a dosage-dependent manner, thus the two measurable parameters could be used as indicators of hepatotoxicity. We then tested the LiPan fry with nine other chemicals including environmental toxicants and human drugs. Three (mefenamic acid, lindane, and arsenate) behave like hepatotoxins in reduction of liver red fluorescence, while three others (17β-estradiol, TCDD [2,3,7,8-tetrachlorodibenzo-p-dioxin] and NDMA [N-nitrosodimethylamine]) caused increase of liver red fluorescence and the liver size. Ethanol and two other chemicals, amoxicillin (antibiotics) and chlorphenamine (pain killer) did not resulted in significant changes of liver red fluorescence and liver size. By quantitative RT-PCR analysis, we found that the changes of red fluorescence intensity caused by different chemicals correlated to the changes of endogenous fabp10a RNA expression, indicating that the measured hepatotoxicity was related to fatty acid transportation and metabolism. Finally we tested a mixture of four hepatotoxins and observed a significant reduction of red fluorescence in the liver at concentrations below the lowest effective concentrations of individual hepatotoxins, suggesting that the transgenic zebrafish assay is capable of reporting compound hepatotoxicity effect from chemical mixtures. Thus, the LiPan transgenic fry provide a rapid and convenient in vivo hepatotoxicity assay that should be applicable to high-throughput hepatotoxicity test in drug screening as well as in biomonitoring environmental toxicants.