The gastrointestinal tract can be exposed to high concentrations of mycotoxins after ingestion of mycotoxin contaminated food or feed. Since swine are the most sensitive domestic animal species to mycotoxins , this study aimed to investigate the toxic effect of four common Fusarium toxins, T-2, DON, ZEA and FB1 on IPEC-J2 cells. Studies concerning the effects of mycotoxins on IPEC-J2 cells are limited. The concentrations of the mycotoxins tested in this study are the result of several optimization experiments performed to refine the concentration range to be used for these in vitro experiments. According to Sergent et al. DON concentrations in human intestine are estimated between 0.16 μg/ml and 2 μg/ml depending on the contamination level of the feed, including thus our tested range. Mycotoxin concentrations in the test solutions are not based on levels known to cause problems in pigs, but are based on the cytotoxicity results of the flow cytometer experiments, the solubility of the mycotoxin in aqueous solvents, and cost of the mycotoxin.
In vitro cell viability assays have a central role in predictive toxicology. The capability of commonly used colorimetric assays, such as the WST-1, MTT (3-(4,5-dimethyltiazol-2-yl)-2,5-diphenyltetrazolium bromide) and MTS assay (3-(4,5-dimethylthiazol-2-yl)-5-(3-carboxymethoxyphenyl)-2-(4-sulfophenyl)-2H-tetrazolium) or the lactate dehydrogenase (LDH) bioassay to detect cytotoxic effects is, however, limited. Cells that are apoptotic remain undetected with these assays. Our study used a flow cytometric technique, which allows to distinguish Annexin-V-FITC and PI negative (viable) cells from Annexin-V-FITC positive and PI negative (apoptotic) cells and Annexin-V-FITC and PI positive (necrotic) cells.
After 72 h of exposure with T-2, DON, FB1 or ZEA, T-2 clearly was the most cytotoxic with a toxicity in the ng/ml range, while the other mycotoxins are only toxic at concentrations in the μg/ml range. This finding is in accordance with the study of Calvert et al. , who demonstrated that T-2 was most cytotoxic against three human cell lines tested as determined by the MTT assay.
When comparing the percentage Annexin-V-FITC positive and PI negative cells and Annexin-V-FITC and PI positive cells, it is striking that more Annexin-V-FITC and PI positive cells than Annexin-V-FITC positive and PI negative cells were found after treatment of the cells with T-2. This corresponds, however, with findings in practice where T-2, present in feed, causes skin necrosis [25, 26]. Also necrosis of the lymphoid cells of the intestinal mucosa was seen after intake of T-2 contaminated feed [25, 27]. DON, on the other hand, causes more Annexin-V-FITC positive and PI negative cells than Annexin-V-FITC and PI positive cells. The discrepancy between type A (T-2) and type B (DON) trichothecenes was also demonstrated by Nasri et al.  who showed that DON causes rather apoptosis and T-2 rather necrosis in Jurkat T-lymphocytes. As it is known that type B trichothecenes activate the MAPK signaling pathway resulting in apoptosis, while type A trichothecenes are not able to activate the MAPK families [29–31], the authors suggested this as an explanation for the results. For ZEA and FB1, although FB1 was not cytotoxic at the tested concentrations, an approximately equal percentage of apoptotic and necrotic cells were seen after treatment of IPEC-J2 cells for 72 h with those toxins. Our results suggest that different toxins may induce different types of cell death in IPEC-J2 cells. To our knowledge, this is the first report of DON and T-2 induced different mechanisms of cell death in IPEC-J2 cells. Since it was not the principal aim of this study to elaborate further the differences in the mode of cell death induced by different mycotoxins, no further experiments were conducted. Nevertheless, it could be interesting to investigate these differences in more detail, including the investigation of associated pathway molecules, the analysis of potential protective effects of caspase inhibitors and/or necrose inhibitors, etc. Also other types of cell death can be taken into account since there are more than two ways for a cell to die. Cell death can be classified according to its morphological appearance (apoptotic, necrotic, autophagic or associated with mitosis), enzymological criteria (with and without the involvement of nucleases or distinct classes of proteases, such as caspases, calpains, cathepsins and transglutaminases), functional aspects (programmed or accidental, physiological or pathological) or immunological characteristics (immunogenic or non-immunogenic) . It could be interesting exploring these differences in relation with mycotoxins.
Correlatively with direct cytotoxicity, the consequences on barrier integrity (TEER and passage of drugs) were also assessed. The effect on TEER and passage was only studied for DON and T-2 because, of the four mycotoxins tested in this study, these are the most toxic ones and their ratio of Annexin-V-FITC positive and PI negative cells and Annexin-V-FITC and PI positive cells is antithetic. Notwithstanding the assertion of Geens and Niewold  that IPEC-J2 cells cultured on Transwell®-collagen coated inserts with a surface area of 0.33 cm2 are not appropriate to measure TEER values, our IPEC-J2 monolayer showed TEER values above 10 kΩ cm2 after 3 weeks of culturing. According to Fromter and Diamond , IPEC-J2 cell lines with TEER values of approximately 2 kΩ cm2 can be considered to be tight epithelia. When IPEC-J2 cells were incubated with cytotoxic concentrations of DON (5 and 10 μg/ml) and T-2 (10 and 100 ng/ml) TEER values dropped to nearly 0 kΩ cm2 after 24 h suggesting a complete disruption of the epithelial monolayer within 24 h. In contrast, TEER values of IPEC-J2 cells incubated with non-toxic concentrations (0.5 μg/ml DON and 0.1 and 1 ng/ml T-2) were comparable to the TEER of control cells. Treatment with 1 μg/ml DON on the other hand shows a time-dependent reduction of the TEER. This was also reported by Pinton et al.  who showed a time- and dose-dependent reduction of the TEER after exposure of IPEC-1 and Caco-2 cells to different concentrations (0, 3, 6, 15, 30 μg/ml) of DON. A decreased barrier function in vitro is reflected in both a decreased TEER and an increased tracer flux [17, 35]. In our study, the decrease in TEER is related to the passage of the used tracers, doxycycline (molecular weight (mw) 513 Da) and paromomycin (mw 714 Da), two commonly used antibiotics in pig industry. The higher the toxin concentration, the higher the marker passage. Most studies use tracers such as FITC-dextran, while in this study and for the first time the antibiotics doxycycline and paromomycin were used as markers. We were, however, not capable of demonstrating that induction of more apoptosis or more necrosis results in differences in transepithelial passage of antibiotics. Both doxycycline and paromomycin pass the epithelial monolayer, regardless the induction of a different cell death mechanism.
It is very interesting to study the passage of these therapeutic compounds during in vivo trials, as veterinary drugs are often used in pig mass medication and frequently administered by the oral route, i.e. mixed in the feed or drinking water.