Changes of physico-chemical properties of nano-biomaterials by digestion fluids affect the physiological properties of epithelial intestinal cells and barrier models

International audienceBackground: The widespread use of nano-biomaterials (NBMs) has increased the chance of human exposure. Although ingestion is one of the major routes of exposure to NBMs, it is not thoroughly studied to date. NBMs are expected to be dramatically modified following the transit into the oral-gastric-intestinal (OGI) tract. How these transformations affect their interaction with intestinal cells is still poorly understood. NBMs of different chemical nature—lipid-surfactant nanoparticles (LSNPs), carbon nanoparticles (CNPs), surface modified Fe3O4 nanoparticles (FNPs) and hydroxyapatite nanoparticles (HNPs)—were treated in a simulated human digestive system (SHDS) and then characterised. The biological effects of SHDS-treated and untreated NBMs were evaluated on primary (HCoEpiC) and immortalised (Caco-2, HCT116) epithelial intestinal cells and on an intestinal barrier model.Results: The application of the in vitro SDHS modified the biocompatibility of NBMs on gastrointestinal cells. The differences between SHDS-treated and untreated NBMs could be attributed to the irreversible modification of the NBMs in the SHDS. Aggregation was detected for all NBMs regardless of their chemical nature, while pH- or enzyme-mediated partial degradation was detected for hydroxyapatite or polymer-coated iron oxide nanoparticles and lipid nanoparticles, respectively. The formation of a bio-corona, which contains proteases, was also demonstrated on all the analysed NBMs. In viability assays, undifferentiated primary cells were more sensitive than immortalised cells to digested NBMs, but neither pristine nor treated NBMs affected the intestinal barrier viability and permeability. SHDS-treated NBMs up-regulated the tight junction genes (claudin 3 and 5, occludin, zonula occludens 1) in intestinal barrier, with different patterns between each NBM, and increase the expression of both pro- and anti-inflammatory cytokines (IL-1β, TNF-α, IL-22, IL-10). Notably, none of these NBMs showed any significant genotoxic effect.Conclusions: Overall, the results add a piece of evidence on the importance of applying validated in vitro SHDS models for the assessment of NBM intestinal toxicity/biocompatibility. We propose the association of chemical and microscopic characterization, SHDS and in vitro tests on both immortalised and primary cells as a robust screening pipeline useful to monitor the changes in the physico-chemical properties of ingested NBMs and their effects on intestinal cells