Background
Salmonella are highly prevalent in swine where they affect about one third of all production holdings in the European Union [
1].
Salmonella negatively impact pig health and the productivity of livestock. Transmission to humans occurs via the food-chain, leading to severe infections. Therefore,
Salmonella control must be initiated at the farm level. Since antibiotics for growth promotion have been banned, alternative strategies to improve gut health are necessary to maintain productivity. Gut microbial composition and activity can be directly influenced via the diet [
2]. This in turn impacts the colonization ability of enteric pathogens, such as
Salmonella, through competitive exclusion mechanisms [
3]. Probiotics and prebiotics, known for their potential to modulate gut microbial composition and activity, are amongst the promising alternative strategies [
4].
Probiotics are defined as “live microorganisms which, when administered in adequate amounts, confer a health benefit on the host” [
5]. Beneficial effects attributed to probiotics in pig feed include reduced incidence and severity of infections and decreased shedding of pathogens [
6-
8]. For example, weaned pigs treated with a five strain probiotic mixture (four
Lactobacillus strains and one
Pediococcus strain) showed significantly reduced (>2 log
10 cfu/g feces)
Salmonella numbers at 15 days post-infection [
7]. Other authors report a lower incidence of diarrhea and fecal coliform numbers when feeding
Lactobacillus rhamnosus GG [
9], reduced carriage of
Escherichia coli with
Bifidobacterium lactis HN019 [
10], or decreased
Salmonella counts in feces and tissues after feeding pigs a combination of
Lactobacillus acidophilus and
Lactobacillus reuteri[
8].
Prebiotics are non-digestible food-ingredients that are readily fermentable in the colon and stimulate potentially health-promoting bacteria, mainly bifidobacteria and/or lactobacilli, thereby beneficially shifting the microbial equilibrium of the host gut [
11]. For example, Patterson et al. [
12] reported stimulation of
Bifidobacterium spp. and
Lactobacillus spp. with a concomitant suppression of
Clostridium spp. and members of
Enterobacteriaceae spp. upon feeding of inulin to pigs. Prebiotics can stimulate short chain fatty acid (SCFA) production, known to play a key role in intestinal host health. For example, butyrate, the main energy source for colonocytes, has anti-inflammatory and anti-carcinogenic properties (reviewed by Russell et al. [
13]) and down-regulates the expression of genes associated with
Salmonella invasion [
14]. However, conflicting results have been reported for the effects of prebiotic feeding in pigs. Tzortzis et al. [
15] reported higher acetate concentrations and increased bifidobacteria numbers after feeding GOS to pigs, while Mikkelsen and Jensen [
16] showed increased butyrate production after feeding FOS to piglets. In contrast, no effect was observed with FOS on bifidobacterial populations [
17] and on fecal SCFA concentrations [
18]. Prebiotics are increasingly combined with probiotics (synbiotics) to enhance probiotic survival and growth. Synbiotic formulations tested in pigs decreased the level of
Enterobacteriaceae in pig fecal samples [
19], and reduced adherence of
Escherichia coli O8:K88 to the jejunal and colonic mucosa [
20]. However, synbiotic formulations have been much less studied for pathogen inhibition. Yet, they have a promising potential considering the competitive advantage of the probiotic through simultaneous application of a prebiotic with high specificity [
21,
22].
The species
B. thermophilum belongs to the commensals of the pig gut microbiota [
23].
Bifidobacterium thermophilum RBL67 (RBL67) previously isolated from baby feces was shown to produce a bacteriocin-like substance (BLIS) with
in vitro activity against
Listeria and
Salmonella[
24-
26]
. Furthermore, we recently showed that RBL67 has antagonistic effects on
Salmonella infection in an
in vitro continuous intestinal fermentation model simulating the child proximal colon [
27]. This strain was reported to adhere to human intestinal cell lines [
28] and to exert protective effects on epithelial HT29-MTX cell culture integrity upon
Salmonella challenge in combined cellular and colonic fermentation models [
29]. Inulin supplemented in a three-stage continuous intestinal fermentation model of the child induced an increase of
B. thermophilum numbers in the proximal, transverse and distal colon sections while SCFA production was shifted towards higher butyrate concentrations [
30]. However, inulin in the proximal colon environment of the model was also shown to promote
Salmonella growth [
30], and to increase the efficiency of HT29-MTX cell invasion [
29]. Finally, RBL67 has technological features of interest for application, such as being moderately oxygen-tolerant, growing at high cell density, low pH and high temperatures of up to 47°C [
31].
Studying the complex interplay of pro- and prebiotics with the gut microbiota and pathogens is hindered by the inaccessibility of the gastrointestinal tract. Studies are further challenged by ethical limits to conduct
in vivo animal infection trials. In this context,
in vitro models represent a cost-effective and ethically less constraint strategy [
32]. We recently reported and validated a novel two-stage
in vitro continuous fermentation model (PolyFermS) inoculated with immobilized fecal microbiota simulating the swine proximal colon. This model allows the parallel operation of five self-contained independent fermentations to simultaneously test different nutritional factors with the same microbiota [
33]. In this study, we used this PolyFermS model of the swine proximal colon to investigate the effects of
B. thermophilum RBL67 and prebiotics (FOS, GOS and MOS) on the gut microbiota composition and activity and on the colonization of the enteric pathogen
Salmonella enterica subsp.
enterica serovar Typhimurium N-15 (N-15).
Discussion
We recently described and validated a novel
in vitro continuous fermentation model (PolyFermS) simulating conditions of the swine proximal colon. The model consists of parallel reactors inoculated with the same microbiota [
33]. In this study, we report the first time application of this swine PolyFermS model to investigate the effects of a probiotic strain,
B. thermophilum RBL67, prebiotics (FOS, GOS, MOS) and combinations thereof, on
S. Typhimurium N-15 colonization in the presence of a diverse gut microbiota.
In a first test, RBL67 and N-15 were shown to colonize the system after one single inoculation. They reached stable and similar numbers after 1 to 2 days. Our
in vitro model data suggest competitive and adaptive traits of RBL67 and N-15 in co-culture with the modeled porcine microbiota. These results are in agreement with previous studies done with one- and three-stage chemostat models of the child colon [
27,
34]. The increasing capacity of N-15 to colonize the model observed from periods 2 to 4, underlines the robustness and/or adaptation of
Salmonella in simulated colonic conditions of the swine colon. This suggests that the PolyFermS model is suitable to mimic a
Salmonella carrier state of pigs with continuous shedding of
Salmonella[
35]. Moreover, an incomplete removal of N-15 during washing periods of reactors may partly explain the enhanced competition of N-15 over time, because viable cells of
Salmonella were detected in the effluents by plating after careful washing with 10% chlorine for 1 h and prior to N-15 challenge in periods 3 and 4 (data not shown). This persistence of
Salmonella could be due to the formation of biofilms in the reactor, which is known to increase sterilization resistance [
36]. This effect may be avoided in the future by replacing the test reactors with sterile units before each new treatment period. We also reported an increase of the family
Succinivibrionaceae during the course of the fermentation for the first-stage immobilized cell and all second-stage reactors for the same fermentation test [
33].
Salmonella and
Succinivibrionaceae belong to the ?-subclass of the phylum Proteobacteria [
37]. Increased numbers of
Succinivibrionaceae correlated with the increased capacity of N-15 to grow in the system, suggesting that this group potentially supported N-15 persistence and growth in periods 3 and 4 after washing. Such co-occurrence of related bacteria has been previously reported for
Salmonella invasion in a mouse infection model in the presence of high titers of
E. coli[
38].
Colonization of N-15 in the porcine PolyFermS was strongly inhibited by the addition of FOS or GOS. This correlated with an increase of SCFA production, especially acetate and propionate. A 5 mM undissociated acetic acid solution was reported to inhibit
Salmonella growth [
39-
41]. In our study, concentrations of undissociated acetic acids were calculated to be >6 mM (pH?=?6.0) for treatments with FOS and GOS, compared to levels ?5 mM in the reactor spiked with N-15 alone. RBL67 combined with FOS or GOS showed an enhanced inhibition of N-15 compared to single treatments with pro- or prebiotics. We chose strain RBL67, because it produces BLIS (thermophilicin B67), which exhibits an antagonistic effect against
Salmonella and
Listeria[
24-
26]. The production of acetate was decreased for R-FOS and R-GOS compared to prebiotics alone (Table
2). This suggests that BLIS contributed to N-15 inhibition in combination with organic acids produced by FOS and GOS. The lower dosage of the prebiotic in R-MOS compared to the other combinations and the stimulation of propionate rather than acetate production, may explain the less pronounced effect on N-15 colonization. However, MOS has previously been shown to block enteropathogen adhesion to the mannose-rich surface glycoproteins of epithelial villi via binding of its ?-D-Mannan to Type 1 fimbriae of enteropathogens and thus may reduce the risk of infection by this mechanism [
42].
The antagonistic effect of RBL67 was less pronounced in this study compared to a previous report [
27]. A strong inhibition of
Salmonella and a rapid metabolic rebalancing of the gut microbiota after antibiotic treatments were observed when RBL67 was added before or after infection in an
in vitro intestinal fermentation model inoculated with child microbiota [
27]. In contrast, Zihler et al. [
30] did not detect an anti-
Salmonella effect of RBL67. This may be explained by different host microbiota, model set-up and probiotic:pathogen ratios used for all these studies, i.e. 16:1 (this study), 3050:1 [
27] and 2:1 [
30].
FOS has been reported to stimulate butyrate production in some studies with piglets [
16,
43]. In our study, we observed an increased butyrate production with the combination of FOS and RBL67. Because bifidobacteria do not produce butyrate [
44], we presume that FOS was first degraded e.g. by RBL67, followed by cross-feeding reactions with butyrate-producing bacteria (e.g.
Roseburia spp. or
Megasphaera; [
45]). Interestingly, while butyrate has been linked to a series of health-related properties (reviewed by Russell et al. [
13]), it was also shown to repress invasion gene expression of
Salmonella[
14].
The microbiota composition from CR to TR effluents only changed marginally after RBL67 and prebiotic treatments. In particular, we did not observe a growth stimulation of bifidobacteria or lactobacilli in the FOS and GOS treatments, as it was previously shown
in vitro with human gut microbiota treated with FOS and inulin [
30,
46] or pig microbiota treated with GOS [
15,
47]. Divergent results have been reported concerning the effect of FOS and GOS
in vivo. Patterson
et al.[
12] reported increased numbers of bifidobacteria and lactobacilli in young pigs fed with inulin. In contrast, Mountzouris et al. [
17] and Mikkelsen and Jensen [
16] did not observe a significant stimulation of bifidobacteria and lactobacilli in pigs fed with FOS and transgalactooligosaccharides. These discrepancies may be explained by different prebiotic structures, dosage and methodology [
4,
48], complicating a direct comparison between the studies. Furthermore, other bacteria of the gut microbiota, including
Salmonella and members of
Roseburia and
Bacteroides, can efficiently utilize FOS and GOS as growth substrates [
49-
51] and can directly compete for these nutrients with bifidobacteria and lactobacilli.
Using 454 pyrosequencing, we detected a consistent increase in the relative abundance of the genus
Sharpea upon addition of prebiotics. This suggests that
Sharpea spp
. play a role for prebiotic degradation. They belong to the family
Erysipelotrichaceae within the
Clostridium Cluster XVII. Members of this genus are heterofermentative and produce lactic acid and CO
2 from glucose. They were first isolated from horse feces and are closely related to
Eggerthia catenaformis[
52,
53]. Higher net substrate availability upon prebiotic addition may be responsible for a higher abundance of
Sharpea spp
. Erysipelotrichaceae were also more abundant in pigs with increased feed consumption [
54,
55], and accounted for a sevenfold higher proportion in mice fed a high energy diet [
56]. Yet, the exact role of the genus
Sharpea remains unclear and further insights into prebiotic degradation or its involvement in possible cross-feeding reactions should be elucidated in future research.
Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made.
The images or other third party material in this article are included in the article’s Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder.
Competing interests
The authors declare that they have no competing interests.
Authors contributions
SAT designed and performed experiments, carried out data analysis and wrote the manuscript. CC and CL participated in the design of the experiments, analysis and interpretation of data and writing of the manuscript. AZB designed experiments and analyzed the data. All authors read and approved the final manuscript.