Establishment and Assessment of a New Human Embryonic Stem Cell‐Based Biomarker Assay for Developmental Toxicity Screening

Abstract

A metabolic biomarker-based in vitro assay utilizing human embryonic stem (hES) cells was developed to identify the concentration of test compounds that perturbs cellular metabolism in a manner indicative of teratogenicity. This assay is designed to aid the early discovery-phase detection of potential human developmental toxicants. In this study, metabolomic data from hES cell culture media were used to assess potential biomarkers for development of a rapid in vitro teratogenicity assay. hES cells were treated with pharmaceuticals of known human teratogenicity at a concentration equivalent to their published human peak therapeutic plasma concentration. Two metabolite biomarkers (ornithine and cystine) were identified as indicators of developmental toxicity. A targeted exposure-based biomarker assay using these metabolites, along with a cytotoxicity endpoint, was then developed using a 9-point dose-response curve. The predictivity of the new assay was evaluated using a separate set of test compounds. To illustrate how the assay could be applied to compounds of unknown potential for developmental toxicity, an additional 10 compounds were evaluated that do not have data on human exposure during pregnancy, but have shown positive results in animal developmental toxicity studies. The new assay identified the potential developmental toxicants in the test set with 77% accuracy (57% sensitivity, 100% specificity). The assay had a high concordance (≥75%) with existing in vivo models, demonstrating that the new assay can predict the developmental toxicity potential of new compounds as part of discovery phase testing and provide a signal as to the likely outcome of required in vivo tests.

The ChemScreen project to design a pragmatic alternative approach to predict reproductive toxicity of chemicals

“…Significant advances have been made particularly in the further simplification of the test battery, while still being able to identify reproductive toxic chemicals efficiently, either in isolation [25], or in a grouping context [41,43]. We found that even with a relatively small number of tests, either apical tests combined with mechanistic tests or an relatively small number of mechanistic tests predictivities ranging from 74 to 94% can be reached [20,25], which is comparable to that obtained with much larger ToxCast screening panels [21], the ReProTect battery [7], or the zebrafish ELS tests [49], a human embryonic stem cell test [50]. This is remarkable, since the concordance between different test species, like rabbit and rat has been estimated to be not more than 60% [6].…”

“…On the other hand, it seems also likely that chemicals often address multiple pathways of toxicity, thereby reducing the need to have all possible mechanism of action included in a final battery. This is supported by the very good performance of our battery, even when compared with much larger batteries [21] or more integrative lower throughput batteries and tests [7,49,50]. At present the current combination of apical and HTS tests seems a pragmatic and effective choice that can be further improved when more mechanisms of reproductive toxicity become known.…”

Structural and functional screening in human induced-pluripotent stem cell-derived cardiomyocytes accurately identifies cardiotoxicity of multiple drug types

“…While our in vitro panel was not 100% accurate at detecting cardiotoxicity, the lack of false positives and the high sensitivity generated from our xCELLigence testing (88.9%) suggest that these tests provides a robust method for early evaluation of cardiac liability. To date, other in vitro preclinical testing methods also show b100% statistical correlation for drug-induced toxicity (Pointon et al, 2013;Guo et al, 2013;Sirenko et al, 2013;Palmer et al, 2013;Morton et al, 2014;van de Water et al, 2011), and it is unlikely that any preclinical test would have 100% accuracy. A variety of differences (i.e., metabolites generated or patient-specific genetic predispositions) in exposed patient populations that cannot be recapitulated fully in vitro make exact prediction difficult.…”

A Developmental and Reproductive Toxicology Program for Chemical Registration

“…Other screening tools that can be used include in vitro estrogen receptor/androgen receptor transactivation assays and the in vitro aromatase assay, which can help identify potential endocrine disrupting compounds. For developmental toxicity prediction, a number of predictive assays have been promulgated over the years including transcriptome profiling of in vitro systems, mammalian whole embryo culture, zebrafish embryo exposures, and embryonic stem cell metabolomics [9,10]. These predictive tools can be used in a screening program to identify compounds of potential concern, not only just for pesticides but for chemicals in general.…”