A multiparametric 3D cortical neurosphere NAM for developmental neurotoxicity: Chlorpyrifos and a...

Developmental neurotoxicity (DNT) arises from disruption of key neurodevelopmental processes-including neural stem cell proliferation, neuronal and glial differentiation, radial migration, and synaptogenesis-that collectively shape corticogenesis. Traditional in vivo guideline studies are costly, low-throughput, and provide limited mechanistic insight, prompting OECD and EFSA to promote new approach methodologies (NAMs) such as the neurosphere assay (NSA).

We refined and characterized a mouse cortical NSA over a three-week differentiation period using a multiparametric endpoint battery encompassing proliferation, neuronal and astrocytic differentiation, radial migration, synaptogenesis, and astrocytic maturation. Baseline differentiation was defined by flow cytometry, confocal immunofluorescence, and qPCR. Two chronic exposure scenarios were implemented: (i) during the 7-day proliferation phase, and (ii) from the onset of differentiation and migration throughout the three-week maturation period. Chlorpyrifos (CPF) was used as a DNT-positive reference compound, and a biomonitoring-informed PFAS mixture of PFOS, PFOA and PFUnDA was designed using French Esteban data to reflect low-nM, environmentally relevant exposure levels; due to availability constraints, PFHxS was not included and 4:2 fluorotelomer sulfonic acid (4:2 FTSA) was used as the short-chain sulfonate component.

Baseline analyses showed progressive acquisition of neuronal (TUBB3, MAP2, SATB2) and astrocytic (GFAP) phenotypes, emergence of SYP⁺/PSD95⁺ synaptic structures, and dependence on mitogenic signaling, consistent with key features of mid-gestational corticogenesis. CPF exerted biphasic effects, with early neurosphere enlargement followed by growth arrest and impaired radial migration; at 250 µM, CPF induced overt cytotoxicity and was associated with reduced GFAP expression. In contrast, PFAS mixture produced only modest effects on bulk viability yet consistently reduced radial migration and significantly downregulated Gfap and Syp at low-nM concentrations, in line with epidemiological and experimental evidence implicating PFAS in neurodevelopmental disorders. Within this proof-of-concept dataset, radial migration emerged as a more sensitive endpoint than bulk viability, revealing functional impairments below overt toxicity thresholds.

The cortical NSA captures key cellular and functional features of mid-gestational corticogenesis and discriminates compound-specific DNT liabilities, with CPF linked to reduced Gfap expression under overtly toxic conditions and PFAS mixture disrupting astrocytic and synaptic programs at biomonitoring-relevant levels. By integrating complementary endpoints under chronic, developmentally targeted exposures, this work advances the NSA as a mechanistic, regulatory-relevant NAM and a strong candidate for inclusion in the OECD DNT in vitro battery.