Toward toxicity testing of nanomaterials in the 21st century: a paradigm for moving forward


  • David Y. Lai

    Corresponding author
    1. U.S. Environmental Protection Agency, Office of Pollution Prevention and Toxics, Risk Assessment Division, Washington, DC, USA
    • U.S. Environmental Protection Agency, Office of Pollution Prevention and Toxics, Risk Assessment Division, Washington, DC, USA
    Search for more papers by this author


A challenge facing hazard identification and safety evaluation of engineered nanomaterials being introduced to market is the diversity and complexity of the types of materials with varying physicochemical properties, many of which can affect their toxicity by different mechanisms. In general, in vitro test systems have limited usefulness for hazard identification of nanoparticles due to various issues. Meanwhile, conducting chronic toxicity/carcinogenicity studies in rodents for every new nanomaterial introduced into the commerce is impractical if not impossible. New toxicity testing systems which rely on predictive, high-throughput technologies may be the ultimate goal of evaluating the potential hazard of nanomaterials. However, at present, this approach alone is unlikely to succeed in evaluating the toxicity of the wide array of nanomaterials and requires validation from in vivo studies. This article proposes a paradigm for toxicity testing and elucidation of the molecular mechanisms of reference materials for specific nanomaterial classes/subclasses using short-term in vivo animal studies in conjunction with high-throughput screenings and mechanism-based short-term in vitro assays. The hazard potential of a particular nanomaterial can be evaluated by conducting only in vitro high-throughput assays and mechanistic studies and comparing the data with those of the reference materials in the specific class/subclass—an approach in line with the vision for ‘Toxicity Testing in the 21st Century’ of chemicals. With well-designed experiments, testing nanomaterials of varying/selected physicochemical parameters may be able to identify the physicochemical parameters contributing to toxicity. The data so derived could be used for the development of computer model systems to predict the hazard potential of specific nanoparticles based on property–activity relationships. WIREs Nanomed Nanobiotechnol 2012, 4:1–15. doi: 10.1002/wnan.162

For further resources related to this article, please visit the WIREs website.

This manuscript has been reviewed by the U.S. EPA and approved for publication. Approval does not signify that the contents necessarily reflect the views and policies of the agency, nor does mention of trade names or commercial products constitute endorsement or recommendation for use.