Positioning the “Worm-on-Chip” technology as part of the toolbox for NAMs.

Laurent Mouchiroud, Elena Katsyuba, David Liaskos, David Rouquié, Ainhoa Letamendia, Matteo Cornaglia

2022 Society of Toxicology Annual Meeting


Currently existing toxicology testing still implies an extensive experimentation in mammalian models, which is expensive and associated with important ethical concerns. The alternative methods to animal testing are typically based on cellular models. The main limitation of these in vitro approaches is that they often cannot predict complex responses at the level of an organism, usually involving a multi-organ crosstalk and metabolic processing of the test molecules.

Nowadays nematode Caenorhabditis elegans (C. elegans) starts to get recognition as a valuable alternative model in predictive toxicology studies, that can complement in vitro models to better predict the outcomes in mammals. However, experimentation in C. elegans is still mainly based on manual handling techniques and direct observation by the operator, hence largely limiting the potential of the worms for high-throughput and high-content screenings required for toxicology studies.

We developed a microfluidic-based robotic platform that can perform automated high-content phenotypic analysis of C. elegans and execute different types of toxicology assays, including the assessments of fertility, embryotoxicity and acute toxicity.

As an illustration, we present here the results of a study characterizing the effects on reproduction of fourteen benchmark chemicals, including 5-fluorouracil, paraquat, sodium chloride, methoxyacetate, lithium chloride, penicilin G, busulfan, dexamethasone, thalidomide, triadimenol, bisphenol A, diphenylhydantoin, benzalkonium chloride and methotrexate. Synchronized populations of worms were chronically exposed to 5 doses of these compounds starting from the last larval stage (L4) until day 3 of adulthood. The images of each worm were recorded every hour and time-resolved phenotypic read-outs were then extracted from the collected images, including growth dynamics, sexual maturity, fertility, embryonic viability and progeny accumulation rate. The phenotypic outcomes were compared to those of positive (5-fluorouracil) and negative (1% DMSO) controls. Out of the tested compounds methotrexate showed the most pronounced adverse effects on embryonic viability, while bisphenol A strongly affected the development of the mothers.

In conclusion, we propose an innovative solution for rapid identification of toxic compounds and their potential mechanism of toxicity, using a biological model that perfectly bridges the gap between in vitro and in vivo assays.