- Full (HTML) - Full (PDF) - Supplemental Material - Related EHP Articles - Purchase This Issue

Hao Zhu,¹ Lin Ye,¹ Ann Richard,² Alexander Golbraikh,¹ Fred A. Wright,³ Ivan Rusyn,^4,* and Alexander Tropsha^1,*

¹Laboratory for Molecular Modeling, Division of Medicinal Chemistry and Natural Products, School of Pharmacy, University of North Carolina at Chapel Hill, Chapel Hill, North Carolina, USA; ²National Center for Computational Toxicology, Office of Research and Development, U.S. Environmental Protection Agency, Research Triangle Park, North Carolina, USA; ³Department of Biostatistics, and ⁴Department of Environmental Sciences and Engineering, School of Public Health, University of North Carolina at Chapel Hill, Chapel Hill, North Carolina, USA

Abstract
Background: Accurate prediction of in vivo toxicity from in vitro testing is a challenging problem. Large public–private consortia have been formed with the goal of improving chemical safety assessment by the means of high-throughput screening.

Objective: A wealth of available biological data requires new computational approaches to link chemical structure, in vitro data, and potential adverse health effects.

Methods and results: A database containing experimental cytotoxicity values for in vitro half-maximal inhibitory concentration (IC₅₀) and in vivo rodent median lethal dose (LD₅₀) for more than 300 chemicals was compiled by Zentralstelle zur Erfassung und Bewertung von Ersatz- und Ergaenzungsmethoden zum Tierversuch (ZEBET ; National Center for Documentation and Evaluation of Alternative Methods to Animal Experiments) . The application of conventional quantitative structure–activity relationship (QSAR) modeling approaches to predict mouse or rat acute LD₅₀ values from chemical descriptors of ZEBET compounds yielded no statistically significant models. The analysis of these data showed no significant correlation between IC₅₀ and LD₅₀. However, a linear IC₅₀ versus LD₅₀ correlation could be established for a fraction of compounds. To capitalize on this observation, we developed a novel two-step modeling approach as follows. First, all chemicals are partitioned into two groups based on the relationship between IC₅₀ and LD₅₀ values: One group comprises compounds with linear IC₅₀ versus LD₅₀ relationships, and another group comprises the remaining compounds. Second, we built conventional binary classification QSAR models to predict the group affiliation based on chemical descriptors only. Third, we developed k-nearest neighbor continuous QSAR models for each subclass to predict LD₅₀ values from chemical descriptors. All models were extensively validated using special protocols.

Conclusions: The novelty of this modeling approach is that it uses the relationships between in vivo and in vitro data only to inform the initial construction of the hierarchical two-step QSAR models. Models resulting from this approach employ chemical descriptors only for external prediction of acute rodent toxicity.

Key words: acute toxicity, computational toxicology, IC₅₀, LD₅₀, LOAEL, NOAEL, QSAR. Environ Health Perspect 117:1257–1264 (2009) . doi:10.1289/ehp.0800471 available via http://dx.doi.org/ [Online 3 April 2009]

Address correspondence to A. Tropsha, 327 Beard Hall, University of North Carolina, Chapel Hill, NC 27599-7360 USA. Telephone: (919) 966-2955. Fax: (919) 966-0204. E-mail: alex_tropsha@unc.edu

*These authors contributed equally to this work.

Supplemental Material is available online (doi:10.1289/ehp.0800471.S1 via http://dx.doi.org/)

We thank T. Martin [U.S. Environmental Protection Agency (EPA) ], and J. Strickland and M. Jackson (ILS, Inc., Durham, NC) for providing some of the data used in this study. We also thank W. Setzer (U.S. EPA) for his interest in this study and valuable comments on the moving M-regression method.

This work was supported, in part, by grants from the National Institutes of Health (GM076059 and ES005948) and the U.S. EPA (RD83272001 and RD83382501) .

The research described in this article has not been subjected to each funding agency’s peer review and policy review and therefore does not necessarily reflect their views, and no official endorsement should be inferred. The manuscript has been reviewed by the U.S. EPA National Center for Computational Toxicology 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.

The authors declare they have no competing financial interests.

Received 26 November 2008 ; accepted 3 April 2009.