There is an increasing need for alternatives strategies to treat livestock bacterial diseases without the use of antimicrobials. In this study, we used in vitro porcine colon culture to evaluate the efficacy of non-antimicrobial compounds in preventing tissue damage following exposure to B. hyodysenteriae, L. intracellularis or S. Typhimurium.
Compound P treatment, a blend of MCFA and SCFA, improved explant epithelial coverage, decreased the accumulation of mucus, and the expression of TNF-α mRNA following challenge with B. hyodysenteriae (Fig. 1A). A trend towards downregulation of IFN-γ mRNA expression following challenge was also observed (Fig. 1C). Compound S, a blend of OA decreased the accumulation of mucus, and reduced the expression of TNF-α and iNOS mRNA following challenge with B. hyodysenteriae. TNF-α and IFN-γ have a recognized role in tight junction regulation (35,36). Tight junction proteins, such as occludins, claudins and zonulae occludentes (ZO), are crucial for the maintenance of epithelial barrier integrity and to regulate the paracellular movement of ions and water (37,38). Fatty acids appear to modulate tight junction permeability and have an anti-inflammatory effect in the colon (39–41). Increased TNF-α and IFN-γ levels lead to the rearrangement of myosin molecules associated with tight-junction proteins, consequently increasing paracellular permeability (42–44). In our study, TNF-α and IFN-γ mRNA expression was down regulated when explants were treated with a blend of MCFA and SCFA, including butyrates (compound P), while an upregulation was observed in explants being treated with compound S, that does not contain butyrates. Similar responses were identified in weaned pigs supplemented with butyrate, and when culturing human colonic biopsies, human colonic cell lines and isolated lamina propria cells with butyrate (45–48). Intestinal epithelial cells exposed to TNF-α and IFN-γ have reduced cystic fibrosis transmembrane conductance regulator (CFTR) expression and chloride (Cl−) secretion (49–51). This impairment of anion secretion affects the mucus layer integrity. Mucins require the interaction of bicarbonate (HCO3−) and Cl- with calcium (Ca2+) for proper release and expansion from goblet cells (52,53). A recent study indicated that host cytokines are not responsible for the impairment of anion channels, and that B. hyodysenteriae may directly cause the decrease in Cl− secretion and which may lead to mucin aggregation and accumulation (54). In contrast, our findings suggest a relationship between reduced gene expression of TNF-α and IFN-γ and a reduction in mucus secretion following infection with B. hyodysenteriae and treatment with compound P. This link between host cytokines and mucus secretory response in SD remains to be clarified. In addition, it is important to highlight that the fold changes observed in this study were quantitatively small, when compared to previously published data. This could be an effect of the model used and the biological significance remains to be explored.
Explants treated with compound F (prebiotic based on Agaricus subrufescens fermented rye) had higher epithelial coverage when challenged with L. intracellularis (Fig. 2A) than those untreated. Riboglucans, β-glucans and glucomannans are examples of bioactive polysaccharides isolated from A. subrufescens (55). These molecules can act as a substrate for bacterial adherence, as they mimic the host glycocalyx (56). D-mannose, a prebiotic, reduced the adhesion of Escherichia coli, Vibrio cholerae, Campylobacter jejuni, and S. Typhimurium to HT-29 cells as per the concept described above (57). This effect was also observed in animal studies, when weaned piglets feed was supplemented with Lentinus edodes mycelium extracts, leading to reduced viable counts of E. coli and Streptococci in the digesta (stomach, jejunum) and mucosal scrapings of the small intestine (58). In vitro studies with ingredients in compound F have also proven binding affinity to S. Typhimurium and S. Enteritidis, and in vivo reducing peak and average shedding of these bacteria (59,60). However, our data revealed no significant effect of compound F in epithelial coverage or cytokine expression following S. Typhimurium challenge. To the best of our knowledge, this is the first report which evaluate the effectiveness of A. subrufescens rye fermentation against L. intracellularis.
A recent study (70) showed the potential immunomodulatory effect of compound F when supplementing piglets post-weaning, with a reduction of pro-inflammatory cytokine production in jejunum, ileum and colon. In our study, no significant differences in cytokine mRNA levels were observed after L. intracellularis challenge (Fig. 2B). This observation may be due to the short period of in vitro incubation which may lead to a low level of bacteria infecting and propagating inside the epithelial cells. Previous authors reported that the pathogen may take up to 12 hours to invade cells after oral inoculation, or 6 hours when ligated intestinal loops were infected directly with vaccine inoculum (61,62). The ability of the attenuated vaccine strain to induce such changes is also questionable, but it has been shown to do so in vivo (63). However, the inoculum concentration used in current study for L. intracellularis challenge would not be considered to cause clinical disease and lesions in natural infections, and therefore can explain the lack of effect between the PCG and the CCG or for almost all TGs challenged with L. intracellularis. Thus, further studies investigating the immunomodulatory role of compound F following infection with a virulent L. intracellularis using longer incubation periods are strongly suggested.
Surprisingly, a lower degree of epithelial coverage was observed in explants exposed to compound F alone than explants exposed to B. hyodysenteriae (Fig. 1A and B). It is known that colon explants harbor a microbiota compositionally similar to the donor pig prior to euthanasia (64). Thus, we postulate that compound F may have served as a substrate for the microbiota already present in the explants, leading to bacterial overgrowth. The lack of colonic peristalsis, may have further contributed to our observations.
Explants infected with B. hyodysenteriae and treated with compound D (phytobiotic) had increased epithelial coverage and decreased levels of IL-1α, TNF-α and IFN-γ, when compared to infected, untreated explants (Fig. 1A and C). Thymol and carvacrol are present in the essential oils extracted from thyme (Thymus vulgaris), the active ingredients on compound D (65). Carvacrol was demonstrated to have a gastroprotective effect in a rodent model of gastritis (66,67). It was associated with reduced colonic lesions in colitis induced by 2,4,6-trinitrobenzenesulfonic (TNBS) in rats (68) and in acetic acid-induced colitis in mice (69). The protective effect of carvacrol was associated with its ability to regulate cyclooxygenase-2 (COX-2) expression (70,71). An in vitro T cell model also linked the reduction of IL-2 and IFN-γ expression to exposure to thymol and carvacrol (72). In contrast, IL-1β and TNF-α induce the expression of COX-2 (73). Mice treated with carvacrol had decreased TNF-α levels and milder lesions following acetic acid-induced colitis (69). Additionally to the effects of thyme, carob (Ceratonia siliqua, another ingredient in compound D) contains phenolic compounds such as flavonoids and gallotannins that also inhibit COX-2 (74). Thus, the effect of compound D was likely due to its anti-inflammatory effects associated with the inhibition COX-2 cascade.
In our study, no significant differences in cytokine mRNA levels were observed after L. intracellularis challenge. This observation may be due to the short period of in vitro incubation which may have led to a low level of bacteria infecting and propagating inside of the epithelial cells. Previous authors reported that the pathogen may take up to 12 hours to invade cells after oral inoculation, or 6 hours when ligated intestinal loops were infected directly with vaccine inoculum (71). The ability of the vaccine strain, at the same dose used in our study, to induce such changes is also questionable, but it has been shown to do so in vivo (73). Thus, further studies investigating the immunomodulatory role of compound F following infection with a virulent L. intracellularis during longer incubation periods are strongly suggested.
In conclusion, our findings suggest that the non-antimicrobial compounds studied may have a beneficial effect to the host based on the explant model data shown. Compound P supported epithelial survival and reduces mucus thickness when explants were exposed to B. hyodysenteriae. Compound D has an immune-modulating effect in explants challenged with B. hyodysenteriae. Compound F prevented epithelial death following L. intracellularis exposure. The authors warrant that further investigations are needed to verify compound effectiveness in vivo.