Strategies to improve the regulatory assessment of developmental neurotoxicity (DNT) using in vitro methods

Abstract

Currently, the identification of chemicals that have the potential to induce developmental neurotoxicity (DNT) is based on animal testing. Since at the regulatory level, systematic testing of DNT is not a standard requirement within the EU or USA chemical legislation safety assessment, DNT testing is only performed in higher tiered testing triggered based on chemical structure activity relationships or evidence of neurotoxicity in systemic acute or repeated dose toxicity studies. However, these triggers are rarely used and, in addition, do not always serve as reliable indicators of DNT, as they are generally based on observations in adult rodents. Therefore, there is a pressing need for developing alternative methodologies that can reliably support identification of DNT triggers, and more rapidly and cost-effectively support the identification and characterization of chemicals with DNT potential. We propose to incorporate mechanistic knowledge and data derived from in vitro studies to support various regulatory applications including: (a) the identification of potential DNT triggers, (b) initial chemical screening and prioritization, (c) hazard identification and characterization, (d) chemical biological grouping, and (e) assessment of exposure to chemical mixtures. Ideally, currently available cellular neuronal/glial models derived from human induced pluripotent stem cells (hiPSCs) should be used as they allow evaluation of chemical impacts on key neurodevelopmental processes, by reproducing different windows of exposure during human brain development. A battery of DNT in vitro test methods derived from hiPSCs could generate valuable mechanistic data, speeding up the evaluation of thousands of compounds present in industrial, agricultural and consumer products that lack safety data on DNT potential.

Keywords: Adverse outcome pathways; Developmental neurotoxicity; Human in vitro test systems; Integrated Approaches to Testing and Assessment; Regulatory purposes.

Fig. 1

Battery of in vitro assays anchored to key neurodevelopmental processes, non-mammalian models and in silico approaches suitable for evaluation of DNT effects. Human induced pluripotent stem cells (hiPSCs) (A) can be used to form rosettes (neuroectodermal cells, resembling neural tube formation in vitro) (B); neural progenitor cells (NPCs) (C) can be derived from rosettes, their migration can be measured (D), and NPCs can be further differentiated into various neuronal and glial sub-types (E–L). Apart from the image showing myelination (K) (modified from https://www.mpg.de/11583034/original-1508156154.jpg), the displayed images are representative pictures of IMR90-hiPSCs differentiated in house as detailed in (Pistollato et al., 2017b). These key neurodevelopmental processes can be measured by gene and protein analysis of markers specific for PSCs, neuroectoderm, and NPCs, and sequential neurodevelopmental processes as shown in the figure. Such analysis can be combined with functional in vitro assays (e.g. MEA measurments) and non-mammalian behavioral studies, if required, (e.g., ZF embryos) and/or in silico models (e.g., QSAR, read-across, IVIVE, etc.) in a battery of tests to support DNT testing. Further efforts are still needed to optimize the assays for evaluation of chemical impact on hiPSCs and NPCs differentiation into mature oligodendrocytes (able to form myelin), microglia, and fully mature neurons (indicated by a red hand symbol). Images show staining for: nestin (green)/β-III-tubulin (red) (B), nestin (red) (C), GFAP (green)/β-III-tubulin (red) (E), synapsin-1 (green)/β-III-tubulin (red) (G), β-III-tubulin (red) (H), synaptophysin (green)/PSD95 (red) (I), GFAP (green) (J), and Iba1 (red) (L). Other abbreviations: Oct4, octamer-binding transcription factor 4; Sox1 (and Sox2), Sex Determining Region Y-Box 1 (and Box 2); Pax6, paired Box 6; HCI, high content imaging; PARP, poly (ADP-ribose) polymerase; MAP2, microtubule-associated protein 2; NF68, neurofilament 68 kDa; NF200, neurofilament 200 kDa; GABA, gamma-aminobutyric acid; VGlut1, vesicular glutamate transporter 1; TH, tyrosine hydroxylase; GFAP, glial fibrillary acidic protein; S100B, S100 calcium-binding protein B; O4, oligodendrocyte marker 4; GalC, galactocerebroside; MBP, myelin basic protein; CNPase, 2′,3′-cyclic-nucleotide 3′-phosphodiesterase; Iba1, ionized calcium binding adaptor molecule 1; CD68, cluster of differentiation 68; TMEM119, transmembrane protein 119; SYP, synaptophysin; SYN1, synapsin 1; PSD95, postsynaptic density protein 95; EM, electro-microscopy; MFR, mean firing rate; MEA, multi-electrode array; IVIVE, in vitro to in vivo extrapolation; ZF, Zebrafish; TUNEL, Terminal deoxynucleotidyl transferase dUTP nick end labelling. (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article.)

Fig. 2

General outline of an Integrated Approach to Testing and Assessment (IATA) which integrates all available sources of existing information (human data, in vivo, in vitro and non-testing data) (modified from (OECD, 2016). Such an IATA can guide the targeted generation of new data based on in vitro DNT assays and, if required, can be combined with in silico approaches. Other abbreviations: WoE, weight of evidence; QSAR, quantitative structure–activity relationship; MIE, molecular initiating event; KE, key event; DNT, developmental neurotoxicity; AOP, adverse outcome pathway.