Developmental toxicity testing in the 21st century: the sword of Damocles shattered by embryonic stem cell assays?

High Accuracy Classification of Developmental Toxicants by In Vitro Tests of Human Neuroepithelial and Cardiomyoblast Differentiation

Human-relevant tests to predict developmental toxicity are urgently needed. A currently intensively studied approach makes use of differentiating human stem cells to measure chemically-induced deviations of the normal developmental program, as in a recent study based on cardiac differentiation (UKK2). Here, we (i) tested the performance of an assay modeling neuroepithelial differentiation (UKN1), and (ii) explored the benefit of combining assays (UKN1 and UKK2) that model different germ layers. Substance-induced cytotoxicity and genome-wide expression profiles of 23 teratogens and 16 non-teratogens at human-relevant concentrations were generated and used for statistical classification, resulting in accuracies of the UKN1 assay of 87-90%. A comparison to the UKK2 assay (accuracies of 90-92%) showed, in general, a high congruence in compound classification that may be explained by the fact that there was a high overlap of signaling pathways. Finally, the combination of both assays improved the prediction compared to each test alone, and reached accuracies of 92-95%. Although some compounds were misclassified by the individual tests, we conclude that UKN1 and UKK2 can be used for a reliable detection of teratogens in vitro, and that a combined analysis of tests that differentiate hiPSCs into different germ layers and cell types can even further improve the prediction of developmental toxicants.

Neuronal and cardiac toxicity of pharmacological compounds identified through transcriptomic analysis of human pluripotent stem cell-derived embryoid bodies

Concurrent with the '3R' principle, the embryonic stem cell test (EST) using mouse embryonic stem cells, developed in 2000, remains the solely accepted in vitro method for embryotoxicity testing. However, the scope and implementation of EST for embryotoxicity screening, compliant with regulatory requirements, are limited. This is due to its technical complexity, long testing period, labor-intensive methodology, and limited endpoint data, leading to misclassification of embryotoxic potential. In this study, we used human induced pluripotent stem cell (hiPSC)-derived embryoid bodies (EB) as an in vitro model to investigate the embryotoxic effects of a carefully selected set of pharmacological compounds. Morphology, viability, and differentiation potential were investigated after exposing EBs to folic acid, all-trans-retinoic acid, dexamethasone, and valproic acid for 15 days. The results showed that the compounds differentially repressed cell growth, compromised morphology, and triggered apoptosis in the EBs. Further, transcriptomics was employed to compare subtle temporal changes between treated and untreated cultures. Gene ontology and pathway analysis revealed that dysregulation of a large number of genes strongly correlated with impaired neuroectoderm and cardiac mesoderm formation. This aberrant gene expression pattern was associated with several disorders of the brain like mental retardation, multiple sclerosis, stroke and of the heart like dilated cardiomyopathy, ventricular tachycardia, and ventricular arrhythmia. Lastly, these in vitro findings were validated using in ovo chick embryo model. Taken together, pharmacological compound or drug-induced defective EB development from hiPSCs could potentially be used as a suitable in vitro platform for embryotoxicity screening.

Creating a human-induced pluripotent stem cell-based NKX2.5 reporter gene assay for developmental toxicity testing

To test large numbers of chemicals for developmental toxicity, rapid in vitro tests with standardized readouts for automated data acquisition are needed. However, the most widely used assay, the embryonic stem cell test, relies on the counting of beating embryoid bodies by visual inspection, which is laborious and time consuming. We previously developed the PluriBeat assay based on differentiation of human induced pluripotent stem cells (hiPSC) that we demonstrated to be predictive for known teratogens at relevant concentrations using the readout of beating cardiomyocytes. Here, we report the development of a novel assay, which we term the PluriLum assay, where we have introduced a luciferase reporter gene into the locus of NKX2.5 of our hiPSC line. This enabled us to measure luminescence intensities instead of counting beating cardiomyocytes, which is less labor intensive. We established two NKX2.5 reporter cell lines and validated their pluripotency and genetic stability. Moreover, we confirmed that the genetically engineered NKX2.5 reporter cell line differentiated into cardiomyocytes with the same efficiency as the original wild-type line. We then exposed the cells to valproic acid (25–300 μM) and thalidomide (0.1–36 µM) and compared the PluriBeat readout of the cardiomyocytes with the luminescence intensity of the PluriLum assay. The results showed that thalidomide decreased luminescence intensity significantly with a higher potency and efficacy compared to the beating readout. With this, we have developed a novel hiPSC-based assay with a standardized readout that may have the potential for higher throughput screening for developmental toxicity.

A novel human pluripotent stem cell-based assay to predict developmental toxicity

There is a great need for novel in vitro methods to predict human developmental toxicity to comply with the 3R principles and to improve human safety. Human-induced pluripotent stem cells (hiPSC) are ideal for the development of such methods, because they are easy to retrieve by conversion of adult somatic cells and can differentiate into most cell types of the body. Advanced three-dimensional (3D) cultures of these cells, so-called embryoid bodies (EBs), moreover mimic the early developing embryo. We took advantage of this to develop a novel human toxicity assay to predict chemically induced developmental toxicity, which we termed the PluriBeat assay. We employed three different hiPSC lines from male and female donors and a robust microtiter plate-based method to produce EBs. We differentiated the cells into cardiomyocytes and introduced a scoring system for a quantitative readout of the assay-cardiomyocyte contractions in the EBs observed on day 7. Finally, we tested the three compounds thalidomide (2.3-36 µM), valproic acid (25-300 µM), and epoxiconazole (1.3-20 µM) on beating and size of the EBs. We were able to detect the human-specific teratogenicity of thalidomide and found the rodent toxicant epoxiconazole as more potent than thalidomide in our assay. We conclude that the PluriBeat assay is a novel method for predicting chemicals' adverse effects on embryonic development.

Combining mouse embryonic stem cells and zebrafish embryos to evaluate developmental toxicity of chemical exposure

The assays in this study utilize mouse embryonic stem cells (mESCs) and zebrafish embryos to evaluate the potential developmental toxicity of industrial and pharmaceutical chemicals. A set of eleven chemicals of known mammalian in vivo teratogenicity were tested in the assays and correlations to mammalian data. Using mESCs, proliferation, differentiation, and cytotoxicity of the chemicals were measured. In zebrafish embryos, lethality and the lowest effect level concentrations for morphological malformations were determined. Clustering of the assays based on frequency of affected assays resulted in a ranking of the test compounds that correlated to in vivo rodent data (R = 0.88, P < 0.001). We conclude that the combination of ESC- and zebrafish-based assays provides a valuable platform for the prioritization of pharmaceutical and industrial chemicals for further testing of developmental toxicity in rodents.

In Vitro Effects of Chronic Spirolide Treatment on Human Neuronal Stem Cell Differentiation and Cholinergic System Development

Spirolides (SPX) are marine toxins, produced by dinoflagellates that act as potent antagonists of nicotinic acetylcholine receptors. These compounds are not toxic for humans, and since there are no reports of human intoxications caused by this group of toxins they are not yet currently regulated in Europe. Currently 13-desmethyl spirolide C, 13,19-didesmethyl spirolide C, and 20-methyl spirolide G are commercially available as reference materials. Previous work in our laboratory has demonstrated that after 4 days of treatment of primary mice cortical neurons with 13-desmethyl spirolide C, the compound ameliorated the glutamate induced toxicity and increased acetylcholine levels and the expression of the acetylcholine synthesizing enzyme being useful both in vitro and in vivo to decrease the brain pathology associated with Alzheimer's disease. In this work, we aimed to extend the study of the neuronal effects of spirolides in human neuronal cells. To this end, human neuronal progenitor cells CTX0E16 were employed to evaluate the in vitro effect of spirolides on neuronal development. The results presented here indicate that long-term exposure (30 days) of human neuronal stem cells to SPX compounds, at concentrations up to 50 nM, ameliorated the MPP⁺-induced neurotoxicity and increased the expression of neuritic and dendritic markers, the levels of the choline acetyltransferase enzyme and the protein levels of the α7 subunit of nicotinic acetylcholine receptors. These effects are presumably due to the previously described interaction of these compounds with nicotinic receptors containing both α7 and α4 subunits. All together, these data emphasize the idea that SPX could be attractive lead molecules against neurodegenerative disorders.

Application of Mouse Embryonic Stem Cell Test to Detect Gender-Specific Effect of Chemicals: A Supplementary Tool for Embryotoxicity Prediction

Gender effect is an inherent property of chemicals, characterized by variations caused by the chemical-biology interaction. It has widely existed, but the shortage of an appropriate model restricts the study on gender-specific effect. The embryonic stem cell test (EST) has been utilized as an alternative test for developmental toxicity. Despite its numerous improvements, mouse embryonic stem cells with an XX karyotype have not been used in the EST, which restricts the ability of the EST to identify gender-specific effects during high-throughput-screening (HTS) of chemicals to date. To address this, the embryonic stem cell (ESC) SP3 line with an XX karyotype was used to establish a "female" model as a complement to EST. Here, we proposed a "double-objects in unison" (DOU)-EST, which consisted of male ESC and female ESC; a seven-day EST protocol was utilized, and the gender-specific effect of chemicals was determined and discriminated; the replacement of myosin heavy chain (MHC) with myosin light chain (MLC) provided a suitable molecular biomarker in the DOU-EST. New linear discriminant functions were given in the purpose of distinguishing chemicals into three classes, namely, no gender-specific effect, male-susceptive, and female-susceptive. For 15 chemicals in the training set, the concordances of prediction result as no gender effect, male susceptive, and female susceptive were 86.67%, 86.67%, and 93.33%, respectively, the sensitivities were 66.67%, 83.33%, and 83.33%, respectively, and the specificities were 91.67%, 88.89%, and 100%, respectively; the total accuracy of DOU-EST was 86.67%. For three chemicals in the test set, one was incorrectively predicted. The possible reason for misclassification may due to the absence of hormone environment in vitro. Leave-one-out cross-validation (LOOCV) indicated a mean error rate of 18.34%. Taken together, these data suggested a good performance of the proposed DOU-EST. Emerging chemicals with undiscovered gender-specific effects are anticipated to be screened with the DOU-EST.

High-throughput multi-parameter profiling of electrophysiological drug effects in human embryonic stem cell derived cardiomyocytes using multi-electrode arrays

Human stem cell derived cardiomyocytes (hESC-CM) provide a potential model for development of improved assays for pre-clinical predictive drug safety screening. We have used multi-electrode array (MEA) analysis of hESC-CM to generate multi-parameter data to profile drug impact on cardiomyocyte electrophysiology using a panel of 21 compounds active against key cardiac ion channels. Our study is the first to apply multi-parameter phenotypic profiling and clustering techniques commonly used for high-content imaging and microarray data to the analysis of electrophysiology data obtained by MEA analysis. Our data show good correlations with previous studies in stem cell derived cardiomyocytes and demonstrate improved specificity in compound risk assignment over convention single-parametric approaches. These analyses indicate great potential for multi-parameter MEA data acquired from hESC-CM to enable drug electrophysiological liabilities to be assessed in pre-clinical cardiotoxicity assays, facilitating informed decision making and liability management at the optimum point in drug development.

The DNT-EST: a predictive embryonic stem cell-based assay for developmental neurotoxicity testing in vitro

As the developing brain is exquisitely vulnerable to chemical disturbances, testing for developmental neurotoxicity of a substance is an important aspect of characterizing its tissue specific toxicity. Mouse embryonic stem cells (mESCs) can be differentiated toward a neural phenotype, and this can be used as a model for early brain development. We developed a new in vitro assay using mESCs to predict adverse effects of chemicals and other compounds on neural development - the so-called DNT-EST. After treatment of differentiating stem cells for 48h or 72h, at two key developmental stages endpoint for neural differentiation, viability, and proliferation were assessed. As a reference, we similarly treated undifferentiated stem cells 2 days after plating for 48h or 72h in parallel to the differentiating stem cells. Here, we show that chemical testing of a training set comprising nine substances (six substances of known developmental toxicity and three without specific developmental neurotoxicity) enabled a mathematical prediction model to be formulated that provided 100% predictivity and accuracy for the given substances, including in leave-one-out cross-validation. The described test method can be performed within two weeks, including data analysis, and provides a prediction of the developmental neurotoxicity potency of a substance.

Human embryonic stem cell-derived test systems for developmental neurotoxicity: a transcriptomics approach

Developmental neurotoxicity (DNT) and many forms of reproductive toxicity (RT) often manifest themselves in functional deficits that are not necessarily based on cell death, but rather on minor changes relating to cell differentiation or communication. The fields of DNT/RT would greatly benefit from in vitro tests that allow the identification of toxicant-induced changes of the cellular proteostasis, or of its underlying transcriptome network. Therefore, the ‘human embryonic stem cell (hESC)-derived novel alternative test systems (ESNATS)’ European commission research project established RT tests based on defined differentiation protocols of hESC and their progeny. Valproic acid (VPA) and methylmercury (MeHg) were used as positive control compounds to address the following fundamental questions: (1) Does transcriptome analysis allow discrimination of the two compounds? (2) How does analysis of enriched transcription factor binding sites (TFBS) and of individual probe sets (PS) distinguish between test systems? (3) Can batch effects be controlled? (4) How many DNA microarrays are needed? (5) Is the highest non-cytotoxic concentration optimal and relevant for the study of transcriptome changes? VPA triggered vast transcriptional changes, whereas MeHg altered fewer transcripts. To attenuate batch effects, analysis has been focused on the 500 PS with highest variability. The test systems differed significantly in their responses (<20 % overlap). Moreover, within one test system, little overlap between the PS changed by the two compounds has been observed. However, using TFBS enrichment, a relatively large ‘common response’ to VPA and MeHg could be distinguished from ‘compound-specific’ responses. In conclusion, the ESNATS assay battery allows classification of human DNT/RT toxicants on the basis of their transcriptome profiles.

Don't judge a neuron only by its cover: neuronal function in in vitro developmental neurotoxicity testing

Classical cases of developmental neurotoxicity (DNT) in humans and advances in risk assessment methods did not prevent the emergence of new chemicals with (suspected) DNT potential. Exposure to these chemicals may be related to the increased worldwide incidence of learning and neurodevelopmental disorders in children. DNT is often investigated in a traditional manner (in vivo using large numbers of experimental animals), whereas development of in vitro methods for DNT reduces animal use and increases insight into cellular and molecular mechanisms of DNT. Several essential neurodevelopmental processes, including proliferation, migration, differentiation, formation of axons and dendrites, synaptogenesis, and apoptosis, are already being evaluated in vitro using biochemical and morphological endpoints. Yet, investigation of chemical-induced effects on the development of functional neuronal networks, including network formation, inter- and intracellular signaling and neuronal network function, is underrepresented in DNT testing. This view therefore focuses on in vitro models and innovative experimental approaches for functional DNT testing, ranging from optical and electrophysiological measurements of intra- and intercellular signaling in neural stem/progenitor cells to measurements of network activity in neuronal networks using multielectrode arrays. The development of functional DNT assays will strongly support the decision-making process for measures to prevent potential chemical-induced adverse effects on neurodevelopment and cognition in humans. We therefore argue that for risk assessment, biochemical and morphological approaches should be complemented with investigations of neuronal (network) functionality.