Integrated Systems Toxicology Division, National Health and Environmental Effects Research Laboratory, US EPA, MD105-05, Element Genomics, Inc, Durham, NC, Research Triangle Park, North Carolina 27711.
Toxicol Sci. 2017 Nov 1;160(1):121-135. doi: 10.1093/toxsci/kfx169.
Less than 1% of environmental chemicals have been evaluated for developmental neurotoxicity (DNT). Current guideline DNT studies are resource intensive and not amenable to screening large numbers of compounds for hazard. As part of evaluating a battery of more rapid and scalable in vitro assays for DNT hazard, 86 compounds were screened for their ability to alter function during cortical network development. Developing rat cortical networks were treated with a concentration series (usually 0.03-30 µM) of 86 compounds, 60 of which have known in vivo DNT effects ("DNT Reference Set"). Spontaneous network activity was monitored by microelectrode array recordings over 12 days in vitro, and 17 measures of network activity and synchrony were quantified. Following recordings on days in vitro 12, in-well cell assessment of metabolic activity (Alamar blue) and total cellular content (lactase dehydrogenase) were conducted. Of the 86 compounds tested, 64 perturbed cortical network function in a concentration-dependent manner; 49 of the 60 DNT Reference Set compounds (81.7%) altered network formation. Compounds were ranked by potency (network effect EC50) and selectivity (separation of network and cell viability EC50) for hazard prioritization. Machine learning indicates a combination of an overall network activity metric with a measure of network coordination is key in distinguishing network-disruptive from benign treatments. These data demonstrate that this microelectrode array-based assay for developing cortical network function is amenable to medium-throughput evaluation of environmental substances for DNT hazard and further prioritization. For comprehensive identification of compounds of concern, this assay will be a useful component of a battery of assays targeting independent neurodevelopmental processes.
环境化学物质中仅有不到 1% 经过了发育神经毒性 (DNT) 的评估。目前的指南性 DNT 研究需要耗费大量资源,且不适合用于筛查大量化合物的危害。作为评估一系列用于 DNT 危害的快速和可扩展的体外检测方法的一部分,对 86 种化合物进行了筛选,以评估它们在皮质网络发育过程中改变功能的能力。用浓度系列(通常为 0.03-30µM)处理发育中的大鼠皮质网络,其中 86 种化合物具有已知的体内 DNT 效应(“DNT 参考集”)。通过微电极阵列记录在体外 12 天内监测自发网络活动,并对 17 种网络活动和同步性进行了量化。在体外第 12 天记录后,在每个孔中进行代谢活性(Alamar blue)和总细胞含量(乳糖酶脱氢酶)的细胞内评估。在测试的 86 种化合物中,有 64 种以浓度依赖性方式干扰皮质网络功能;60 种 DNT 参考集化合物中的 49 种(81.7%)改变了网络形成。根据效力(网络效应 EC50)和选择性(网络和细胞活力 EC50 的分离)对化合物进行危害优先级排序。机器学习表明,整体网络活动指标与网络协调指标的结合是区分网络破坏与良性处理的关键。这些数据表明,基于微电极阵列的皮质网络发育功能检测方法适用于环境物质的高通量 DNT 危害评估和进一步的优先级排序。为了全面识别关注的化合物,该检测方法将是针对独立神经发育过程的一系列检测方法的有用组成部分。