Alternative Methods for Neurotoxicity

This realization has led to the conclusion that a paradigm shift is necessary. This new testing paradigm must move away from traditional animal-based toxicity testing and be based on the use of emerging technologies and in vitro test systems. This type of approach has been proposed by the NRC Committee on Toxicity Testing and Assessment of Environmental Agents in the publication “Toxicity Testing in the Twenty-first Century: A Vision and a Strategy” (NRC, 2007). As our understanding of the molecular and cellular responses to chemical perturbation increases and key toxicity pathways are elucidated, the need for traditional animal testing can be greatly reduced.

In response to a considerable rise in the incidence of neurodevelopmental disease in children and the large resource requirements of traditional methods, the push for new testing paradigms for neurotoxicity has recently focused on developmental neurotoxicity testing (DNT). Workgroups in the United States and Europe have set forth proposals for the systematic evaluation of alternative methods for DNT, including the use of in vitro and non-mammalian test systems (Lein et al., 2007; Coecke et al., 2007). For both adult and developmental neurotoxicity testing, moving away from whole-animal testing hinges upon recognizing scientific advances in neurobiology that demonstrate the molecular and cellular basis of nervous system function. New assays will use endpoints based on early molecular changes in simplified systems like cell cultures rather than on apical tests performed in animals.

In order to move in this direction, basic research in the area of alternative methods for neurotoxicological testing should focus on: 1) gaining a systems-level understanding of the molecular and cellular networks which regulate normal neural functions, 2) identifying the mechanisms of action by which chemicals perturb these molecular and cellular networks to cause neurotoxicity, and 3) developing high(er) throughput assays that assess the effect of chemicals on toxicity pathways which lead to adverse effects in the nervous system.

There are many challenges associated with using in vitro approaches to assess the potential of a chemical to produce toxicity in humans, and these challenges tend to be magnified when applied to the nervous system (Harry & Tiffany-Castiglioni, 2005). Unique aspects of the nervous system, including cellular heterogeneity, regional specificity, metabolic competency, and elaborate connectivity, can all present barriers to the development of widely applicable in vitro systems for neurotoxicity testing. These barriers, however, should not be considered as insurmountable. Rather, they should be used as signposts to direct resources to the most pressing areas of research.

Based on the heterogeneity of cell types in the nervous system, it can be predicted that no single model system will suffice to predict in vivo neurotoxicity: a battery of models will be needed. Thus, there should be a coordinated and systematic comparison of different neuronal systems focused on determining what and how many cell types will be necessary to encompass the mechanisms of action relevant to the nervous system. The relative lack of metabolic capability of nervous system is well recognized, and leads to the conclusion that some a priori prediction of chemical metabolites reaching the nervous system will be required before in vitro toxicity testing. Promising directions for research include the development of in silico metabolic simulators which predict in vivo metabolites and in vitro models incorporating sequential exposure of chemicals to a metabolic activating system followed by neural cell cultures.

The intricate connectivity of the nervous system is ultimately responsible for higher order functions such as cognition, learning and memory. Recent work has shown that neuronal cultures grown on multi-electrode arrays readily form complex neural networks and can be used to study network physiology, plasticity and learning in vitro. Research focused on understanding how chemicals known to alter cognition, learning and memory in vivo affect neural networks in vitro would facilitate the assessment of these techniques as models of higher order nervous system function.

There are several issues that can be considered in the short term to speed progress in the development of alternative methods for neurotoxicity testing. First, assay development should focus on endpoints that are related to specific mechanisms of neurotoxicity or to specific determinants of nervous system function in addition to commonly used but more general assays of cell viability and cytotoxicity. This is related to the scientific need to identify toxicity pathways relevant to the nervous system. For some classes of chemicals the mechanism of action for neurotoxicity at the molecular level (e.g. effects on receptors, ion channels, or cytoskeletal elements) may be well characterized. For other chemicals the mechanism of action is unknown, and assays of nervous system response at the cellular level (e.g. neurotransmission, cellular morphology, plasticity) will be required.

Second, a common “reference set” of chemicals should be identified to be used as both positive and negative controls in the development and validation of new test methods. To do this, efforts should be made to understand the chemical space (which encompasses all possible molecular structures) that is occupied by chemicals known to be neurotoxic. Careful attention will then need to be paid to those areas of chemical space which are not currently occupied by known neurotoxic compounds.

Finally, in order to facilitate the extrapolation of results obtained using in vitro tests to risk assessment in humans, emphasis should be placed on developing in vitro models based on human cell systems. While there are a number of human neuronal cell lines available, human neural stem cells show particular promise. Recently, neural stem cells of human origin have been immortalized to create clonal neural stem cell lines. Several human stem cell cultures that can be differentiated in mature cultures of neurons and glia are commercially available. Assays for neurotoxicity using cells of human origin should be compared to the more widely used neural cell cultures of animal origin to determine whether there is an improvement in predictive ability.

Acknowledgement

This document has been reviewed by the National Health and Environmental Effects Laboratory, U.S. EPA and approved for publication. Approval does not signify that the contents reflect the views of the Agency.
©2007 William Mundy