The aim of this novel research was to measure the metabolism of cells to determine the concentration at which a substance to which they are exposed becomes toxic. The study was initiated by Estelle Dubreil, research project leader in the Toxicology of Contaminants Unit at ANSES's Fougères Laboratory, and its findings were published in the Journal of Hazardous Materials in August 2024. "Ultimately, such methods could be used to develop toxicity reference values for substances that pose a risk to human health," explains Ludovic Le Hégarat, Deputy Head of the unit. These values are limits above which there is a health risk for the population as a whole.

An approach still in its early stages

"New approaches are being explored to assess the toxicity of molecules without recourse to animal experiments, but metabolomics [the study of metabolites] has not yet been widely applied to in vitro toxicology,” explains Estelle Dubreil. Metabolites, such as lipids or amino acids, are compounds produced by cells to help them function or defend themselves. The quantity of metabolites produced may increase or decrease in response to exposure to a toxic substance.

The study carried out by ANSES scientists used liver cells exposed to three pyrrolizidine alkaloids as an example. They belong to a large family of organic substances secreted by plants and known to be toxic in both humans and animals. Pyrrolizidine alkaloids can be found in beverages and foods, including herbal teas, honey and cereals.

Effects even at low concentrations

One of the advantages of the approach explored by ANSES is that we know precisely which metabolic function or pathway is affected at each concentration. The study revealed that the effects of substances are not always proportional to their concentration. Many of the responses are bell-shaped: when cells are exposed to low concentrations, the quantity of metabolites increases in step with the quantity of alkaloids, but then decreases at higher concentrations.

Among the metabolites produced by cells, lipids are the first to be affected: they could therefore be a marker of toxic effects on the liver at low concentrations. The researchers assume that an increase in the lipid content of cells is an initial response to compensate for the toxic effects of alkaloids, before other metabolic pathways (such as bile acids or amino acids) take over. "This study clearly highlights the need to focus on toxic effects at low concentrations, which are currently rarely investigated. Toxicology studies on animals continue to favour a ‘high dose’ approach, whereas studies at low doses enable us to more accurately reflect human exposure," comments Ludovic Le Hégarat.

Determining a single value from a multitude of data

The study authors tested a software program that uses the dose-response curves for each metabolite to determine an inflection point, i.e. a concentration at which the substance is considered to disrupt cellular metabolism. However, before it can be applied to human risk assessment, the scientific community will need to agree on the calculation method that most accurately predicts the toxicity of a substance. One of the questions to be settled, for example, is whether it is better to base decisions on the concentration at which the first effects are observed on certain marker metabolites, or whether a median value should be used for a group of affected metabolites.