Peter T. Theunissen

An abbreviated protocol for multilineage neural differentiation of murine embryonic stem cells and its perturbation by methyl mercury.

Alternative assays are highly desirable to reduce the extensive experimental animal use in developmental toxicity testing. In the present study, we developed an improved test system for assessing neurodevelopmental toxicity using differentiating embryonic stem cells. We advanced previously established methods by merging, modifying and abbreviating the original 20-day protocol into a more efficient 13-day neural differentiation protocol. Using morphological observation, immunocytochemistry, gene expression and flow cytometry, it was shown predominantly multiple lineages of neuroectodermal cells were formed in our protocol and to a lower extent, endodermal and mesodermal differentiated cell types. This abbreviated protocol should lead to an advanced screening method using morphology in combination with selected differentiation markers aimed at predicting neurodevelopmental toxicity. Finally, the assay was shown to express differential sensitivity to a model developmental neurotoxicant, methyl mercury.

Time-response evaluation by transcriptomics of methylmercury effects on neural differentiation of murine embryonic stem cells

Current globally harmonized Organisation for Economic Co-operation and Development (OECD) animal test guidelines for developmental toxicity require high numbers of experimental animals. To reduce animal use in this field, alternative developmental toxicity assays are highly desirable. We previously developed a dynamic in vitro model for screening effects of possible neurodevelopmental toxicants, using neural cell differentiation of pluripotent murine embryonic stem cells. To further mechanistically characterize the mouse neural embryonic stem cell test (ESTn) and to improve detection of possible neurodevelopmental toxicants, gene expression patterns were studied describing neural cell differentiation over time, as well as the impact on gene expression of exposure to the well-known neurotoxicant methylmercury (MeHg). A transcriptomics study was performed to examine whole-genome expression changes during the first 7 days of the cell differentiation protocol. Specific gene clusters were identified and enrichment analysis of Gene Ontology (GO) terms and gene sets derived from literature was performed using DAVID and T-profiler. Over time, a decrease of blastocyst and trophectoderm GO terms was observed, which included well-characterized pluripotency genes. Furthermore, an increase in the range of neural development-related GO terms, such as neuron differentiation and the wnt pathway, was observed. Analysis of gene expression using principle component analysis showed a time-dependent track in untreated cells, describing the process of neural differentiation. Furthermore, MeHg was shown to induce deviation from the predefined differentiation track. The compound inhibited general development GO terms and induced neural GO terms over time. This system appears promising for studying compound effects on neural differentiation in a mechanistic approach.

Transcriptomic concentration-response evaluation of valproic acid, cyproconazole, and hexaconazole in the neural embryonic stem cell test (ESTn).

Alternative developmental toxicity assays are urgently needed to reduce animal use in regulatory developmental toxicology. We previously designed an in vitro murine neural embryonic stem cell test (ESTn) as a model for neurodevelopmental toxicity testing (Theunissen et al., 2010). Toxicogenomic approaches have been suggested for incorporation into the ESTn to further increase predictivity and to provide mechanistic insights. Therefore, in this study, using a transcriptomic approach, we investigated the concentration-dependent effects of three known (neuro) developmental toxicants, two triazoles, cyproconazole (CYP) and hexaconazole (HEX), and the anticonvulsant valproic acid (VPA). Compound effects on gene expression during neural differentiation and corresponding regulated gene ontology (GO) terms were identified after 24 h of exposure in relation to morphological changes on day 11 of culture. Concentration-dependent responses on individual gene expression and on biological processes were determined for each compound, providing information on mechanism and concentration-response characteristics. All compounds caused enrichment of the embryonic development process. CYP and VPA but not HEX significantly enriched the neuron development process. Furthermore, specific responses for triazole compounds and VPA were observed within the GO-term sterol metabolic process. The incorporation of transcriptomics in the ESTn was shown to enable detection of effects, which precede morphological changes and provide a more sensitive measure of concentration-dependent effects as compared with classical morphological assessments. Furthermore, mechanistic insight can be instrumental in the extrapolation of effects in the ESTn to human hazard assessment.

Comparison of MeHg-induced toxicogenomic responses across in vivo and in vitro models used in developmental toxicology.

Toxicogenomic evaluations may improve toxicity prediction of in vitro-based developmental models, such as whole embryo culture (WEC) and embryonic stem cells (ESC), by providing a robust mechanistic marker which can be linked with responses associated with developmental toxicity in vivo. While promising in theory, toxicogenomic comparisons between in vivo and in vitro models are complex due to inherent differences in model characteristics and experimental design. Determining factors which influence these global comparisons are critical in the identification of reliable mechanistic-based markers of developmental toxicity. In this study, we compared available toxicogenomic data assessing the impact of the known teratogen, methylmercury (MeHg) across a diverse set of in vitro and in vivo models to investigate the impact of experimental variables (i.e. model, dose, time) on our comparative assessments. We evaluated common and unique aspects at both the functional (Gene Ontology) and gene level of MeHg-induced response. At the functional level, we observed stronger similarity in MeHg-response between mouse embryos exposed in utero (2 studies), ESC, and WEC as compared to liver, brain and mouse embryonic fibroblast MeHg studies. These findings were strongly correlated to the presence of a MeHg-induced developmentally related gene signature. In addition, we identified specific MeHg-induced gene expression alterations associated with developmental signaling and heart development across WEC, ESC and in vivo systems. However, the significance of overlap between studies was highly dependent on traditional experimental variables (i.e. dose, time). In summary, we identify promising examples of unique gene expression responses which show in vitro-in vivo similarities supporting the relevance of in vitro developmental models for predicting in vivo developmental toxicity.

Complementary Detection of Embryotoxic Properties of Substances in the Neural and Cardiac Embryonic Stem Cell Tests

In developmental toxicity testing, in vitro screening assays are highly needed to increase efficiency and to reduce animal use. A promising in vitro assay is the cardiac embryonic stem cell test (ESTc), in which the effect of developmental toxicants on cardiomyocyte differentiation is assessed. Recently, we developed a neural differentiation variant of the stem cell test (neural embryonic stem cell test [ESTn]). In both of these models, we have previously performed a series of transcriptomic studies to characterize gene expression changes (1) across time during normal differentiation and (2) in response to a series of developmental toxicants in the ESTn and ESTc. Here, using the cumulative of these studies, we compared gene expression profiles of ESTn and ESTc over time as well as model-specific changes induced by seven compounds, comprising six known in vivo developmental toxicants and one negative control. Time-related gene expression profiles showed similarities between the two EST systems. However, specific genes could be identified changing over time differently in each model related to the two different lineages of differentiation. Interestingly, compound-induced gene expression changes were generally model specific, especially for methylmercury and flusilazole, which were predicted better in ESTn and ESTc, respectively. Valproic acid-induced gene expression changes were most comparable out of the six developmental toxicants between the ESTn and ESTc. Direct transcriptomic comparisons between the ESTn and ESTc indicate that combined transcriptomic analyses support and complement each other. Therefore, a combined approach incorporating ESTc and ESTn may improve developmental toxicant detection over individual assays.

Compound-specific effects of diverse neurodevelopmental toxicants on global gene expression in the neural embryonic stem cell test (ESTn)

Alternative assays for developmental toxicity testing are needed to reduce animal use in regulatory toxicology. The in vitro murine neural embryonic stem cell test (ESTn) was designed as an alternative for neurodevelopmental toxicity testing. The integration of toxicogenomic-based approaches may further increase predictivity as well as provide insight into underlying mechanisms of developmental toxicity. In the present study, we investigated concentration-dependent effects of six mechanistically diverse compounds, acetaldehyde (ACE), carbamazepine (CBZ), flusilazole (FLU), monoethylhexyl phthalate (MEHP), penicillin G (PENG) and phenytoin (PHE), on the transcriptome and neural differentiation in the ESTn. All compounds with the exception of PENG altered ESTn morphology (cytotoxicity and neural differentiation) in a concentration-dependent manner. Compound induced gene expression changes and corresponding enriched gene ontology biological processes (GO-BP) were identified after 24h exposure at equipotent differentiation-inhibiting concentrations of the compounds. Both compound-specific and common gene expression changes were observed between subsets of tested compounds, in terms of significance, magnitude of regulation and functionality. For example, ACE, CBZ and FLU induced robust changes in number of significantly altered genes (≥ 687 genes) as well as a variety of GO-BP, as compared to MEHP, PHE and PENG (≤ 55 genes with no significant changes in GO-BP observed). Genes associated with developmentally related processes (embryonic morphogenesis, neuron differentiation, and Wnt signaling) showed diverse regulation after exposure to ACE, CBZ and FLU. In addition, gene expression and GO-BP enrichment showed concentration dependence, allowing discrimination of non-toxic versus toxic concentrations on the basis of transcriptomics. This information may be used to define adaptive versus toxic responses at the transcriptome level.

An optimized gene set for transcriptomics based neurodevelopmental toxicity prediction in the neural embryonic stem cell test

The murine neural embryonic stem cell test (ESTn) is an in vitro model for neurodevelopmental toxicity testing. Recent studies have shown that application of transcriptomics analyses in the ESTn is useful for obtaining more accurate predictions as well as mechanistic insights. Gene expression responses due to stem cell neural differentiation versus toxicant exposure could be distinguished using the Principal Component Analysis based differentiation track algorithm. In this study, we performed a de novo analysis on combined raw data (10 compounds, 19 exposures) from three previous transcriptomics studies to identify an optimized gene set for neurodevelopmental toxicity prediction in the ESTn. By evaluating predictions of 200,000 randomly selected gene sets, we identified genes which significantly contributed to the prediction reliability. A set of 100 genes was obtained, predominantly involved in (neural) development. Further stringency restrictions resulted in a set of 29 genes that allowed for 84% prediction accuracy (area under the curve 94%). We anticipate these gene sets will contribute to further improve ESTn transcriptomics studies aimed at compound risk assessment.

Innovative approaches in the embryonic stem cell test (EST).

The embryonic stem cell test (EST) is a high-throughput in vitro screening assay for developmental toxicity free of animal use. The EST uses the ability of murine embryonic stem cells to differentiate into the mesodermal cardiac lineage in combination with two cytotoxicity test systems. Validation of the EST showed that the test system is very promising as an alternative method to animal testing, however to optimize predictability and increase knowledge on the applicability domain of the EST, improvements to the method were proposed and studied. In this review we discuss the first definition of the EST followed by the innovative approaches which have been proposed to increase the predictivity of the EST, including implementation of molecular endpoints in the EST, such as omics technologies and the addition of alternative differentiation models to the testing paradigm, such as neural and osteoblast differentiation and the use of human stem cells. These efforts to improve the EST increase the value of embryonic stem cells used as in vitro systems to predict developmental toxicity.

Comparing rat and rabbit embryo-fetal developmental toxicity data for 379 pharmaceuticals: on systemic dose and developmental effects.

Abstract A database of embryo-fetal developmental toxicity (EFDT) studies of 379 pharmaceutical compounds in rat and rabbit was analyzed for species differences based on toxicokinetic parameters of area under the curve (AUC) and maximum concentration (Cmax) at the developmental lowest adverse effect level (dLOAEL). For the vast majority of cases (83% based on AUC of n = 283), dLOAELs in rats and rabbits were within the same order of magnitude (less than 10-fold different) when compared based on available data on AUC and Cmax exposures. For 13.5% of the compounds the rabbit was more sensitive and for 3.5% of compounds the rat was more sensitive when compared based on AUC exposures. For 12% of the compounds the rabbit was more sensitive and for 1.3% of compounds the rat was more sensitive based on Cmax exposures. When evaluated based on human equivalent dose (HED) conversion using standard factors, the rat and rabbit were equally sensitive. The relative extent of embryo-fetal toxicity in the presence of maternal toxicity was not different between species. Overall effect severity incidences were distributed similarly in rat and rabbit studies. Individual rat and rabbit strains did not show a different general distribution of systemic exposure LOAELs as compared to all strains combined for each species. There were no apparent species differences in the occurrence of embryo-fetal variations. Based on power of detection and given differences in the nature of developmental effects between rat and rabbit study outcomes for individual compounds, EFDT studies in two species have added value over single studies.

Valproic acid-induced gene expression responses in rat whole embryo culture and comparison across in vitro developmental and non-developmental models.

Transcriptomic evaluations may improve toxicity prediction of in vitro-based developmental models. In this study, transcriptomics was used to identify VPA-induced gene expression changes in rat whole embryo culture (WEC). Furthermore, VPA-induced responses were compared across in vitro-based developmental models, such as the cardiac and neural embryonic stem cells (ESTc and ESTn, respectively) and the zebrafish embryotoxicity model. VPA-induced gene regulation in WEC corresponded with observed morphological effects and previously suggested mechanisms of toxicity. Gene Ontology term-directed analysis showed conservation of VPA-induced gene expression changes across in vitro-based developmental models, with ESTc and ESTn exhibiting complementary responses. Furthermore, comparison of in vitro-based developmental and non-developmental models revealed that more generalized VPA-induced effects can be detected using non-developmental models whereas developmental models provide added value when assessing developmental-specific effects. These analyses can be used to optimize test batteries for the detection of developmental toxicants in vitro.