Developmental Effects of the ToxCast™ Phase I and Phase II Chemicals in Caenorhabditis elegans and Corresponding Responses in Zebrafish, Rats, and Rabbits

Abstract

Background: Modern toxicology is shifting from an observational to a mechanistic science. As part of this shift, high-throughput toxicity assays are being developed using alternative, nonmammalian species to prioritize chemicals and develop prediction models of human toxicity.

Methods: The nematode Caenorhabditis elegans (C. elegans) was used to screen the U.S. Environmental Protection Agency's (EPA's) ToxCast™ Phase I and Phase II libraries, which contain 292 and 676 chemicals, respectively, for chemicals leading to decreased larval development and growth. Chemical toxicity was evaluated using three parameters: a biologically defined effect size threshold, half-maximal activity concentration (AC50), and lowest effective concentration (LEC).

Results: Across both the Phase I and Phase II libraries, 62% of the chemicals were classified as active ≤ 200 μM in the C. elegans assay. Chemical activities and potencies in C. elegans were compared with those from two zebrafish embryonic development toxicity studies and developmental toxicity data for rats and rabbits. Concordance of chemical activity was higher between C. elegans and one zebrafish assay across Phase I chemicals (79%) than with a second zebrafish assay (59%). Using C. elegans or zebrafish to predict rat or rabbit developmental toxicity resulted in balanced accuracies (the average value of the sensitivity and specificity for an assay) ranging from 45% to 53%, slightly lower than the concordance between rat and rabbit (58%).

Conclusions: Here, we present an assay that quantitatively and reliably describes the effects of chemical toxicants on C. elegans growth and development. We found significant overlap in the activity of chemicals in the ToxCast™ libraries between C. elegans and zebrafish developmental screens. Incorporating C. elegans toxicological assays as part of a battery of in vitro and in vivo assays provides additional information for the development of models to predict a chemical's potential toxicity to humans.

Citation: Boyd WA, Smith MV, Co CA, Pirone JR, Rice JR, Shockley KR, Freedman JH. 2016. Developmental effects of the ToxCast™ Phase I and II chemicals in Caenorhabditis elegans and corresponding responses in zebrafish, rats, and rabbits. Environ Health Perspect 124:586-593; http://dx.doi.org/10.1289/ehp.1409645.

Figure 1

Association between C. elegans size and developmental stage. Nematode developmental stages (L1 larva–adult) were determined after direct observation by microscope, and then size characteristics (EXT) were measured using COPAS Biosort flow cytometry. The mean log(EXT) of the nematodes in each well for a treatment group, which contained nematodes at a single larval stage, are presented. The log(EXT) of L4s and young adults were all greater than 5.665 (dotted line); nematodes that had not developed to the L3 stage were all less than 5.138 (dashed line). Each point represents the mean size [log(EXT)] of the nematodes in an individual well.

Figure 2

Comparison between t-test and effect size threshold. The histogram presents the number of chemicals in each size class [mean log(EXT)]. The dark gray bars indicate the number of inactive compounds in each size class according to the weighted t-test, and the light gray bars indicate the number of compounds determined to be active in each size class. The two vertical lines indicate the maximum log(EXT) for nematodes ≤ L3 (5.138) and the minimum log(EXT) (5.665) for L4 and young adult nematodes (see Figure 1). Chemicals between the vertical lines had weighted mean Log(EXT) values consistent with a mixture of L3s and L4s.

Figure 3

Concordance between C. elegans larval development and zebrafish embryonic development assays for ToxCast™ Phase I chemical activity. Venn diagram illustrating the concordance between the effects of chemicals on C. elegans development and two zebrafish development assays, Zebrafish^P (Padilla et al. 2012) and Zebrafish^T (Truong et al. 2014).

Figure 4

Hierarchical clustering of chemical activity on C. elegans development. Blue corresponds to inactive chemicals with responses similar to controls, and yellow to red indicates decreasing nematode size with increasing toxicity. The histogram illustrates the size distribution of matched negative controls. Upper panel: activity of 959 unique chemicals from ToxCast™ Phase I and Phase II libraries clustered according to mean log(EXT). Lowerpanel: activity and chemical names of the 50 chemicals with the greatest effects on C. elegans growth. Lists and descriptions of chemicals in the lower panel are presented in Table S4.