The ingestion of T-2 toxin contaminated feed by pigs resulted in a significant reduction of weight gain, after only 18 days, compared to control pigs that received blank feed (Table 1). To our knowledge, this is the first time such an effect has been reported due to a low concentration of T-2 toxin. Since contamination of human foodstuff with T-2 toxin is an emerging issue and concentrations up to 1810 μg T-2 toxin per kg wheat have been reported in Germany , it is feasible that T-2 toxin may also affect human metabolism. Different studies describe that, at high doses, T-2 toxin affects the intestinal absorption of nutrients and reduces the daily feed intake, resulting in a reduced body weight gain [44–46]. However, due to the housing conditions of the animals, we were not able to record the daily feed intake of the animals. Therefore, we cannot conclude whether the reduced weight gain of the pigs was the result of a decreased daily feed intake.
iTRAQ analysis showed that even an extreme low concentration of 5 ng/mL T-2 toxin affects protein expression in differentiated IPEC-J2 cells compared to untreated cells (Table 3). The main mechanism by which T-2 toxin causes its toxic effects is through inhibition of protein synthesis, leading to a ribotoxic stress response. This activates c-Jun N-terminal kinase (JNK)/p38 MAPKs and as a consequence modulates numerous physiological processes including cellular homeostasis, cell growth, differentiation and apoptosis . Proteomic analysis showed an upregulation of proteins involved in ribosome biogenesis, protein synthesis, protein folding and c-Jun N-Terminal kinase signalling. The increased expression of these proteins could be a rescue mechanism, highlighting that even a low concentration of 5 ng/mL T-2 toxin leads to a ribotoxic stress response in differentiated IPEC-J2 cells. The toxic character of T-2 toxin was also shown by the upregulation of heat shock proteins, pre-mRNA splicing factor heterogeneous nuclear ribonucleoprotein F, which could be a mechanism to increase mRNA stability , and 14-3-3 sigma. The protein 14-3-3 sigma is a p53-regulated inhibitor of G2/M progression  and its upregulation might emphasize the DNA damage caused by T-2 toxin . Overall, these iTRAQ data may indicate that T-2 toxin damages the porcine enterocyte, and by doing so, harms the absorption of nutrients with a reduced weight gain as result.
In contrast to other Fusarium mycotoxins, there is no guidance value set by the European Commission for the amount of T-2 toxin in complementary and complete feed for pigs. As shown by a neutral red assay, T-2 toxin affects cell viability at very low concentrations (Figure 3). The in vitro viability of porcine macrophages, undifferentiated and differentiated porcine intestinal epithelial cells was significantly decreased at concentrations ≥ 1 ng/mL, ≥ 2.5 ng/mL and ≥ 15 ng/mL, respectively. Taking into account that such low concentrations negatively affect cell viability in vitro, and that these concentrations are relevant in practice [9, 23], it is of utmost importance that maximum levels are set for this mycotoxin as well.
Ingestion of low and relevant concentrations of T-2 toxin results in reduced numbers of Salmonella Typhimurium bacteria in the cecum contents of pigs, and a tendency to a reduced colonization of the jejunum, ileum, cecum, colon and colon contents (Figure 1). With T-2 toxin and Salmonella Typhimurium being major problems in swine industry and with salmonellosis being one of the most important zoonotic bacterial diseases in both developed and developing countries, we aimed at evaluating the effect of T-2 toxin on the pathogenesis of a Salmonella Typhimurium infection in pigs. Until now, conflicting results have been published concerning the effect of mycotoxins on the susceptibility to intestinal infections and still little is known about the effects of low concentrations of these mycotoxins [16–18, 51–53]. According to Ziprin and McMurray, T-2 toxin did not affect the course of salmonellosis in mice . In the present study, we provide evidence that these data cannot be extrapolated to a pig host. Since the porcine intestine shows physiological, anatomical and pathological similarities to the human gut , it is not unlikely that T-2 toxin similarly affects the pathogenesis of a Salmonella infection in the human host as in the pig host.
The ingestion of 15 μg T-2 toxin per kg feed resulted in a significant decreased expression of IL-1β (Figure 2). Once Salmonella has invaded the intestinal epithelium, the innate immune system is triggered and the porcine gut will react with the production of several cytokines . Both Salmonella and mycotoxins affect the innate immune system. Zhou et al. found that deoxynivalenol (DON) increases the expression of TGF-β and IFN-γ in the small intestine of mice . Recently, Kruber et al. established that T-2 toxin strongly induces IL-8 production in a Caco-2 intestinal epithelial cell line . According to Vandenbroucke et al., DON and Salmonella Typhimurium synergistically potentiate intestinal inflammation in an ileal loop model of pigs . As our control group is Salmonella positive, we cannot conclude whether the decreased expression of IL-1β in the T-2 toxin treated pigs is caused by the effects of T-2 toxin on the innate immune system, the reduced numbers of Salmonella Typhimurium in the gut, or a combination of both. Furthermore, by the use of ELISA analysis, Maresca et al. showed that DON caused a biphasic effect on IL-8 secretion by Caco-2 cells . They also pointed out that this biphasic effect was not observed at mRNA level, where a dose-dependent increase in IL-8 mRNA was noticed . These data implicate that in order to obtain results about the secretion of IL-1β, ELISA analysis on the ileum should be performed.
In order to elucidate how T-2 toxin causes reduced numbers of Salmonella Typhimurium bacteria in the cecum contents of pigs, and a tendency to a reduced colonization of the jejunum, ileum, cecum, colon and colon contents, we investigated the effects of T-2 toxin on the interactions of Salmonella Typhimurium with porcine macrophages and intestinal epithelial cells, two cell types that play an important role in the pathogenesis of a Salmonella infection. In vitro treatment of the host cells with T-2 toxin rendered them more susceptible to invasion, in a SPI-1 independent manner, and increased the transepithelial passage of the bacterium (Figures 4 and 6). This is in accordance with Vandenbroucke et al. who showed that DON promotes the invasion and translocation of Salmonella Typhimurium over porcine host cells, by a mechanism that is not SPI-1 dependent [24, 59]. The results obtained by Maresca et al. also confirm our results since they pointed out that DON concentrations that do not compromise the barrier function, significantly increase the passage of non-invasive Escherichia coli bacteria through Caco-2 inserts . As reviewed by Maresca and Fantini  such increase in bacterial passage through intestinal epithelial cells could be involved in inducing inflammatory bowel diseases. Extrapolating these results to the in vivo situation, one would expect an increased colonization by Salmonella in pigs. However, we showed that ingestion of low and relevant concentrations of T-2 toxin resulted in a significantly decreased amount of Salmonella Typhimurium bacteria in the cecum contents and in a tendency to reduced colonization of the jejunum, ileum, cecum, colon and colon contents. In vitro, T-2 toxin decreased the intracellular survival of Salmonella Typhimurium in PAM, undifferentiated IPEC-J2 cells and differentiated IPEC-J2 cells (Figure 4) at concentrations which significantly reduced the cell viability (Figure 3). Possibly this reduced survival plays an important role in the in vivo outcome. However, whether this reduced survival is due to a decrease in viable cells, a diminished replication capacity of the bacterium or a combination of both, is unknown.
Invasion of Salmonella in nonphagocytic cells involves a series of cytoskeletal changes, characterized by actin polymerization and the formation of membrane ruffles. These cytoskeletal changes are important for the uptake and the cytoplasmic transport of the bacterium, as well as for the establishment and the stability of the bacterial replicative niche, also called Salmonella containing vacuole (SCV) . By the use of iTRAQ analysis, we demonstrated that 5 ng/mL T-2 toxin induces alterations in the expression of proteins that are involved in the cytoskeleton formation of differentiated IPEC-J2 cells. T-2 toxin causes a decreased expression of cytokeratin 18, a member of the intermediate filament network that provides support and integrity to the cytoskeleton , of myristoylated alanine-rich C-kinase substrate, a filamentous actin crosslinking protein  and of putative beta-actin, which is a major component of the cytoskeleton. Furthermore, T-2 toxin causes an increased expression of thymosin beta-10, an actin-sequestering protein involved in cytoskeleton organization and biogenesis , of cysteine and glycine-rich protein 1 isoform 1, a regulator for actin filament bundling  and of profilin, an actin-binding protein that can sequester G-actin or actively participate in filament growth . According to Vandenbroucke et al., low concentrations of DON can modulate the cytoskeleton of macrophages resulting in an enhanced uptake of Salmonella Typhimurium in porcine macrophages . The observed changes in protein expression are not sufficient to induce morphological changes, as assessed with TEM (Figure 7). However, the T-2 toxin induced altered expression of cytoskeleton associated proteins could influence the interactions between IPEC-J2 cells and Salmonella. Thus T-2 toxin and Salmonella Typhimurium appear to act synergistically, inducing cytoskeleton reorganizations which increase the invasion of the bacterium.
We also examined the effects of T-2 toxin on Salmonella Typhimurium gene expression. Microarray analysis revealed that T-2 toxin caused a general downregulation of Salmonella Typhimurium metabolism (Additional file 5) and notably of ribosome synthesis. To our knowledge, this is the first time it has been shown that T-2 toxin affects ribosomal gene expression in both eukaryotic  and prokaryotic cells. Microarray analysis also showed that T-2 toxin causes a downregulation of flagella gene expression (Additional file 5) and consequently resulted in decreased motility of Salmonella Typhimurium (Figure 8). Motility of Salmonella increases the probability that the bacterium will reach suitable sites for invasion and successful infections . Transcriptomic analysis furthermore demonstrated that exposure to T-2 toxin results in reduced expression of many SPI-1 genes. According to Boyen et al., SPI-1 plays a crucial role in the invasion and colonization of the porcine gut and in the induction of influx of neutrophils . Shah et al. indicated that the pathogenicity of Salmonella Enteritidis isolates is associated with both motility and secretion of the type III secretion system (TTSS) effector proteins . Isolates with low invasiveness had impaired motility and impaired secretion of FlgK, FljB and FlfL or TTSS secreted SipA and SipD. Therefore, a T-2 toxin induced downregulation of SPI-1 and motility genes and a reduced motility may lead to a reduced colonization by the bacterium in pigs.
In conclusion, we showed that the presence of low and in practice relevant concentrations of T-2 toxin in the feed causes a decrease in the amount of Salmonella Typhimurium bacteria present in the cecum contents of pigs, and a tendency to a reduced colonization of the jejunum, ileum, cecum, colon and colon contents. In vitro, T-2 toxin causes an increased invasion and transepithelial passage of the bacterium in and through T-2 toxin treated porcine cells, in a SPI-1 independent manner. However, T-2 toxin significantly reduces the SPI-1 gene expression, invasiveness and motility of the bacterium. Therefore, in vivo, the effect of T-2 toxin on the bacterium is probably more pronounced than the host cell-mediated effect.