Biocidal substances and their environmental relevant metabolites are highly toxic for fish. However, an important scarcity of toxicity data for metabolites is recognised. This article provides new data about the toxicity to fish of these compounds and evaluates the potential use of fish cell lines as screening tools to assess the acute toxicity of these compounds in fish. To this aim, acute toxicity of 7 substances was tested in Oncorhynchus mykiss (OECD TG203) and cytotoxicity of 16 substances was assessed in fish cell lines from two species; Poeciliopsis lucida (PLHC-1) and O. mykiss (RTH-149, RTG-2 and RTgill-W1) performing three cytotoxicity tests: Alamar-Blue, 5-carboxyfluorescein diacetate, acetoxymethyl ester and Neutral Red Uptake. Additionally, in vitro and in vivo data from the LIFE-COMBASE database were included in a dataset finally comprising 33 biocides and 14 metabolites. Hazard data were categorized into 4 toxicity groups, according to the intervals established in Regulation (EC) 1272/2008. Finally, the Spearman correlation test was performed and coincidences between in vitro-in vivo data established. In vitro and in vivo results revealed a high positive correlation, with a complete coincidence for 56.5% of the substances, a 2% of false positives (non-toxic in vivo) and a 13% of false negatives (toxic in vivo) for the 4 toxicity categories. However, when results were grouped in toxic or non-toxic coincidence was obtained for 85% of the substances. In conclusion, although fish denote a greater sensitivity, the use of at least two fish cell lines and three cytotoxicity endpoints appear to be valid approaches for fish acute toxicity screening of biocides and their metabolites.

杀生物物质及其与环境相关的代谢物对鱼类具有剧毒。然而，人们认识到代谢物的毒性数据非常缺乏。本文提供了有关这些化合物对鱼的毒性的新数据，并评估了鱼细胞系作为筛选工具的潜在用途，以评估这些化合物对鱼的急性毒性。为此，在Oncorhynchus mykiss (OECD TG203) 中测试了 7 种物质的急性毒性，并在来自两个物种的鱼细胞系中评估了 16 种物质的细胞毒性；Poeciliopsis lucida (PLHC-1) 和O. mykiss (RTH-149、RTG-2 和 RTgill-W1) 进行三种细胞毒性测试：Alamar-Blue、5-羧基荧光素二乙酸酯、乙酰氧基甲酯和中性红摄取。此外，来自 LIFE-COMBASE 数据库的体外和体内数据包含在最终包含 33 种杀菌剂和 14 种代谢物的数据集中。根据法规 (EC) 1272/2008 中规定的时间间隔，危害数据分为 4 个毒性组。最后，进行了斯皮尔曼相关性检验，并建立了体外-体内数据之间的一致性。体外和体内结果显示高度正相关，56.5% 的物质完全重合，2% 的假阳性（体内无毒）和 13% 的假阴性（体内有毒）) 用于 4 种毒性类别。然而，当结果按有毒或无毒分组时，85% 的物质得到了巧合。总之，尽管鱼表现出更高的敏感性，但使用至少两种鱼细胞系和三个细胞毒性终点似乎是对鱼类杀生物剂及其代谢物进行急性毒性筛选的有效方法。