Due to several gaps remaining in the toxicological evaluation of nanomaterials (NMs), consumers and public health agencies have shown increasing concern for human health protection. In addition to aluminum (Al) microparticles, Al-containing nanomaterials (Al NMs) have been applied by food industry as additives and contact materials. Due to the limited amount of literature on the toxicity of Al NMs, this study aimed to evaluate the in vivo genotoxic potential of Al⁰ and Al₂O₃ NMs after acute oral exposure. Male Sprague-Dawley rats were administered three successive gavages at 6, 12.5 and 25 mg/kg bw. A comparison with AlCl₃ was done in order to assess the potential effect of dissolution into Al ions. Both DNA strand breaks and oxidative DNA damage were investigated in six organs/tissues (duodenum, liver, kidney, spleen, blood and bone marrow) with the alkaline and the Fpg-modified comet assays. Concomitantly, chromosomal damage was investigated in bone marrow and colon with the micronucleus assay. The comet assay only showed DNA damage with Al₂O₃ NMs in bone marrow (BM), while AlCl₃ induced slight but non-significant oxidative DNA damage in blood. No increase of chromosomal mutations was observed after treatment with the two Al MNs either in the BM or in the colons of rats.

Bacillus cereus (Bc) is a wide group of Gram-positive and spore-forming bacteria, known to be the etiological agents of various human infections, primarily food poisoning. The Bc group includes enteropathogenic strains able to germinate in the digestive tract and to produce enterotoxins such as Nhe, Hbl, and CytK. One species of the group, Bacillus thuringiensis (Bt), has the unique feature of producing insecticidal crystals during sporulation, making it an important alternative to chemical pesticides to protect crops from insect pest larvae. Nevertheless, several studies have suggested a link between the ingestion of pesticide strains and human cases of food poisoning, calling their safety into question. Consequently, reliable tools for virulence assessment are worth developing to aid decision making in pesticide regulation. Here, we propose complementary approaches based on two biological models, the human intestinal Caco-2 cell line and the insect Drosophila melanogaster, to assess and rank the enteric virulence potency of Bt strains in comparison with other Bc group members. Using a dataset of 48 Bacillus spp. strains, we showed that some Bc group strains, including Bt, were able to induce cytotoxicity in Caco-2 cells with concomitant release of IL-8 cytokine, a landmark of pro-inflammatory response. In the D. melanogaster model, we were able to sort a panel of 39 strains into four different classes of virulence, ranging from no virulence to strong virulence. Importantly, for the most virulent strains, mortality was associated with a loss of intestinal barrier integrity. Interestingly, although strains can share a common toxinotype, they display different degrees of virulence, suggesting the existence of specific mechanisms of virulence expression in vivo in the intestine.

Microcystins (MCs) are toxins produced by several cyanobacterial species found worldwide. While MCs have a common structure, the variation of two amino acids in their structure affects their toxicity. As toxicodynamics are very similar between the MC variants, their differential toxicity could rather be explained by toxicokinetic parameters. Microcystin-RR (MC-RR) is the second most abundant congener and induces toxicity through oral exposure. As intestinal permeability is a key parameter of oral toxicokinetics, the apparent permeability of MC-RR across a differentiated intestinal Caco-2 cell monolayer was investigated. We observed a rapid and large decrease of MC-RR levels in the donor compartment. However, irrespective of the loaded concentration and exposure time, the permeabilities were very low from apical to basolateral compartments (from 4 to 15 × 10⁻⁸ cm·s⁻¹) and from basolateral to apical compartments (from 2 to 37 × 10⁻⁸ cm·s⁻¹). Our results suggested that MC-RR would be poorly absorbed orally. As similar low permeability was reported for the most abundant congener microcystin-LR, and this variant presented a greater acute oral toxicity than MC-RR, we concluded that the intestinal permeability was probably not involved in the differential toxicity between them, in contrast to the hepatic uptake and metabolism.

The pinnatoxins (PnTXs) and portimines, produced by Vulcanodinium rugosum, have been detected in several countries, raising concerns for human health. Although no human poisoning from these toxins has been reported so far, they have been shown to distribute throughout the rodent body after oral administration. Therefore, we investigated the impact of PnTX analogs (PnTX-A, -E, -F, -G, and -H) and portimine (8, 16, and 32 ng/mL) on intestinal barrier integrity and their oral bioavailability using human Caco-2 cell monolayers treated for 2, 6, and 24 h. Our results demonstrated that all of the toxins could impair barrier integrity after 24 h, with differences observed for PnTX-A, -E, and -F, as well as portimine, the most potent of all. While PnTX-A and -E exhibited poor permeability, the other PnTXs were more penetrative, with a Papp > 1.5 × 10⁻⁶ cm·s⁻¹. Portimine was the only toxin displaying both a time- and concentration-dependent passage, likely involving a passive diffusion process. The experimental results were compared to predictions obtained by QSAR tools. Although only qualitative, our results suggest that some of these compounds may be more likely to be distributed throughout the body. Further in vivo studies are required to estimate oral bioavailability and potential public health concerns.