Dysbiosis or dysbacteriosis

Next to the above-mentioned beneficial microbes, the intestinal tract also harbours (facultative) pathogenic bacteria. These two populations interact with one another. In a perfect world, the beneficial microbiota competes with pathogenic invaders and thus largely wards off the latter from colonising the intestinal tract (Hooper and Macpherson, Reference Hooper and Macpherson2010). Dysbiosis (or dysbacteriosis) is defined as an undesirable alteration of the microbiota resulting in an imbalance between protective and harmful bacteria (Chan et al., Reference Chan, Estaki and Gibson2013).

The distinction between beneficial and harmful bacteria is, however, not clear cut. Indeed, a large array of intestinal microorganisms, which normally are commensals, can at some point become a potential threat to the host. The outgrowth of (facultative) pathogenic bacteria in the intestinal lumen and/or invasion across the intestinal barrier needs to be stopped by an aggressive immune response (Biswas and Kobayashi, Reference Biswas and Kobayashi2013). Consequently, the challenge to the intestinal immune system is to protect the host tissues from microbial invasion, while at the same time maintaining the symbiotic benefits from the microbial presence (Hooper et al., Reference Hooper, Littman and Macpherson2012). Tolerance towards the microbiota in the intestinal lumen is achieved through antigen-presenting cells continuously sampling the intestinal lumen and presenting these samples to the immune cells in the cortex of the GALT in mammals (in the spleen in birds). It has been established recently that Clostridiaceae, belonging to the phylum Firmicutes, play an essential role in establishing the tolerance response by the induction of regulatory T-lymphocytes (Nagano et al., Reference Nagano, Itoh and Honda2012). Invading aggressors, on the other hand, are sensed through the TLRs expressed on the baso-lateral plasma membrane (under the tight junctions) and through the NLRs expressed in the cytoplasm of the epithelial cells (Biswas and Kobayashi, Reference Biswas and Kobayashi2013).

Dysbiosis has been associated with many disease conditions in humans, including not only local gastrointestinal diseases but also systemic diseases (Chan et al., Reference Chan, Estaki and Gibson2013). However, it is not always clear whether dysbiosis is causing the disease, or the other way round, or whether both are mere concomitant phenomena. However, it is well established that, in man, dietary factors can trigger dysbiosis. The dysbiosis in turn affects the host–microbe interactions, leading essentially to inflammation through the triggering of the TLR and NLR (Devkota and Chang, Reference Devkota and Chang2013). Clinical studies have identified dysbiosis in human patients with inflammatory bowel diseases, including both Crohn’s disease and ulcerative colitis (Chan et al., Reference Chan, Estaki and Gibson2013).

In domestic animals, dysbiosis was practically unknown until the ban on antimicrobial growth promoters (AGP) in animal feed in Europe on 1 January 2006 (Teirlynck et al., Reference Teirlynck, De Gussem, Marien, Vancraeynest, Haesebrouck, Ducatelle and Van Immerseel2011). Since then, it is perceived in the field as one of the most challenging problems, especially in broilers and in newly weaned pigs. We reported that dysbiosis in broilers is associated with intestinal inflammation and shortening of small intestinal villi (Teirlynck et al., Reference Teirlynck, De Gussem, Marien, Vancraeynest, Haesebrouck, Ducatelle and Van Immerseel2011). Under experimental conditions in broilers, dysbiosis, characterised by typical inflammation and villus shortening, can be induced by feeding a diet containing high amounts of non-starch polysaccharides (NSP) and no NSP-degrading enzymes (Teirlynck et al., Reference Teirlynck, Bjerrum, Eeckhaut, Huyghebaert, Pasmans, Haesebrouck, Dewulf, Ducatelle and Van Immerseel2009a). Moreover, such a diet makes the birds more susceptible to Salmonella enteritidis colonisation, an observation that further documents the imbalance between beneficial and pathogenic microorganisms (Teirlynck et al., Reference Teirlynck, Haesebrouck, Pasmans, Dewulf, Ducatelle and Van Immerseel2009b). This shows that, in broilers, as in humans, dysbiosis can be diet induced. It also suggests that the AGPs that were included in animal feed before the ban prevented the detrimental microbiota shift that is called dysbiosis. The AGP Zn-bacitracin indeed was able to reverse the inflammatory changes and villus shortening induced by the high NSP-containing diet in broilers, in association with a shift in the microbiota (Teirlynck et al., Reference Teirlynck, Bjerrum, Eeckhaut, Huyghebaert, Pasmans, Haesebrouck, Dewulf, Ducatelle and Van Immerseel2009a). It thus appears that the use of AGP in the past has masked the dysbiosis-inducing effects of many feed formulae used in broilers (and possibly in pigs and other farm animals). Nevertheless, other modes of action of AGP, not implicating a role for the microbiota, have been suggested (Niewold, Reference Niewold2007).

Probiotics

Probiotics are defined as live microorganisms that, when consumed in adequate amounts, confer a health effect on the host (Anonymous, 2002). The concept originally comes from Elie Metchnikoff at the beginning of the 20th century who correlated the longevity of Bulgarians to their high consumption of fermented milk. Following on these ancient observations, the most frequently used bacterial probiotic genera still are lactic acid bacteria, mainly the Lactobacillus and Bifidobacterium genera. Other genera are also used, including Enterococcus and Streptococcus. A wide array of different health benefits have been attributed to the consumption of these probiotics in human medicine (Nagpal et al., Reference Nagpal, Kumar, Kumar, Behare, Jain and Yadav2012). These health benefits are not limited to the intestinal tract but also include systemic metabolic diseases such as obesity and type 2 diabetes (Nakamura and Omaye, Reference Nakamura and Omaye2012).

In the original definition as launched by Fuller (Reference Fuller1989), it was mentioned that probiotics exerted their effect by improving the intestinal microbial balance of the host. This aspect is no longer retained in the current definition (Anonymous, 2002), thus clearly underscoring the uncertainty regarding the mode of action of probiotics. Indeed, notwithstanding the long tradition of using probiotics in human nutrition, up until the present day, the discovery and development of new probiotics still is an empirical process. In animal nutrition, where the move towards the application of live microorganisms in feed was much more recent, these things were starting off by copying from human nutrition. Nevertheless, in the Register of Feed Additives of the European Union, more than 100 microorganisms and more than 30 so-called gut flora stabilisers are listed. This illustrates the widespread current interest in developing probiotics for use in animal feed. The interest was definitely boosted as a consequence of the ban on the use of growth-promoting antibiotics in animal feed.

It has been claimed indeed that probiotics can serve as alternatives to antibiotics for increasing performance in poultry (Patterson and Burkholder, Reference Patterson and Burkholder2003). In a number of trials, certain formulations of certain probiotics yielded performance levels comparable to those achieved with growth-promoting antibiotics, provided the microorganisms could survive the passage through the upper alimentary tract – for example, by microencapsulation (Han et al., Reference Han, Zhang, Wang, Li, Liu and Zhao2013). However, taking together the results of many different feeding trials, there is not always an unequivocal beneficial effect of the probiotics (for a review in broilers, see Huyghebaert et al., Reference Huyghebaert, Ducatelle and Van Immerseel2011; for a review in weaned pigs, see Heo et al., Reference Heo, Opapeju, Pluske, Kim, Hampson and Nyachoti2012). This may have various reasons, not in the least because probiotic effects are strain dependent (Butel, Reference Butel2014). Thus, studies carried out with different strains of the same probiotic species may yield different results. However, one striking consistency in the literature is the improvement of performance under challenge conditions. Regardless of whether the challenge is coccidiosis, Clostridium perfringens (Rahimi et al., Reference Rahimi, Kathariou, Grimes and Siletzky2011; Jerzele et al., Reference Jerzele, Szeker, Csizinszky, Gere, Jakab, Mallo and Galfi2012; Jayaraman et al., Reference Jayaraman, Thangavel, Kurian, Mani, Mukkalil and Chirakkal2013) or Salmonella (Wolfenden et al., Reference Wolfenden, Pumford, Morgan, Shivaramaiah, Wolfenden, Pixley, Green, Tellez and Hargis2011; Tellez et al., Reference Tellez, Pixley, Wolfenden, Layton and Hargis2012), or simply LPS injection (Jiang et al., Reference Jiang, Schatzmayr, Mohnl and Applegate2010), a beneficial effect is observed following the application of a probiotic. Similar studies have, to our knowledge, not yet been carried out in dysbiosis models. Nevertheless, dysbiosis definitely is a challenge condition, although not triggered by a single well-defined pathogen. Being a challenge condition, it is to be expected that probiotics may help protect against dysbiosis.

Several problems still limit the use of probiotics in animal feed for the control of dysbiosis. First of all, there is the aspect of safety. As mentioned above, there is not always a clear-cut distinction between beneficial microorganisms on the one hand and facultative pathogens and opportunistic pathogens on the other. Indeed, many different microorganisms residing in the intestinal tract may, under certain conditions, translocate from the intestinal lumen into the systemic circulation and cause disease. Predisposing factors for bacterial translocation may include the immune status of the host, the type of bacteria, level of inoculum or pregnancy. The latter was suggested in a recent paper on Enterococcus faecalis in humans and in a mouse model (Tan et al., Reference Tan, Xu, Xu, Aguilar, Yang, Dong, Chen and Wei2013). It is unclear whether the inflammatory condition, which characterises dysbiosis, predisposes for translocation of microorganisms. Conversely, there are indications that some probiotic strains may have a protective effect on the translocation of opportunistic pathogens. Indeed, in a recent study using a wire flooring model for inducing lameness in broilers, Wideman et al. (Reference Wideman, Hamal, Stark, Blankenship, Lester, Mitchell, Lorenzi and Pevzner2012) showed that a commercial blend of probiotic microorganisms significantly reduced the numbers of lame birds.

Another problem limiting the use of probiotics is a mere practical one. As most poultry feed is pelleted, bacteria added during mixing may be killed by the heat generated during the pelleting process. Lactobacilli, bifidobacteria and many other vegetative bacteria are prone to heat inactivation. This is, however, not a problem for Bacillus species, which are usually incorporated in feed as heat-resistant spores (Shivaramaiah et al., Reference Shivaramaiah, Pumford, Morgan, Wolfenden, Wolfenden, Torres-Rodriguez, Hargis and Tellez2011). To the authors’ knowledge, there is very little information on the survival of freeze-dried vegetative bacteria during the pelleting of animal feed; however, unpublished data from our lab show that mixing a freeze-dried probiotic strain in mash feed certainly is a viable option.

Prebiotics

A prebiotic is defined as a selectively fermented dietary ingredient that results in specific changes in the composition and/or activity of the gastrointestinal microbiota, thus conferring benefit(s) upon host health (for review see Roberfroid et al., Reference Roberfroid, Gibson, Hoyles, McCartney, Rastall, Rowland, Wolvers, Watzl, Szajevska, Stahl, Guarner, Respondek, Whelan, Coxam, Davicco, Leotoing, Wittrant, Delzenne, Cani, Neyrinck and Meheust2010). Most prebiotics belong to the categories of so-called NSP or non-starch oligosaccharides. The latter include a.o. fructooligosaccharides (FOS), galactooligosaccharides, arabinoxylan oligosaccharides, raffinose family oligosaccharides (RFOs), etc. Some authors do not consider mannanoligosaccharides as prebiotics sensu stricto, because they apparently do not alter the intestinal microbiota, even though they can improve performance in a variety of animal species, including weaned piglets and rabbits (Bovera et al., Reference Bovera, Lestingi, Marono, Iannacone, Nizza, Mallardo, deMartino and Tateo2012; Zhao et al., Reference Zhao, Jung and Kim2012). Conversely, some prebiotics have beneficial effects in chickens after in ovo injection, thus in the absence of a developed microbiota, as documented for RFOs (Bednarczyk et al., Reference Bednarczyk, Urbanowski, Gulewicz, Kasperczyk, Maiorano and Szwaczkowski2011).

In a number of feeding experiments in broilers, some prebiotics, including 0.25% or 0.5% FOS, enhanced performance to a level similar to that of AGP (Kim et al., Reference Kim, Seo, Kim and Paik2011; Khodambashi Emami et al., Reference Khodambashi Emami, Samie, Rahmani and Ruiz-Feria2012). Nevertheless, not all studies report unequivocal positive results of in-feed prebiotics, even under stress conditions such as Eimeria challenge (Faber et al., Reference Faber, Dilger, Hopkins, Price and Fahey2012). Grmanova et al. (Reference Grmanova, Rada, Sirotek and Vlkova2010) stated that chicken feed mixtures naturally contain prebiotic oligosaccharides in the form of RFOs in higher concentrations than what is commonly recommended as artificial supplementation. Thus, inconsistency of results obtained in different trials with different prebiotic supplements may depend on the combination of other ingredients in the feed formula. In addition, there are a lot of other factors that may influence the success of prebiotic supplements. As a typical example, it has been clearly shown for arabinoxylans from wheat and other cereals that differences in pretreatment may determine the expected effect in different animal species and in humans (Broekaert et al., Reference Broekaert, Courtin, Verbeke, Van De Wiele, Verstraete and Delcour2011), to the point that different degrees of polymerisation and substitution of arabinoxylan–oligosaccharides may exert different effects (Damen et al., Reference Damen, Verspreet, Pollet, Broekaert, Delcour and Courtin2011; Geraylou et al., Reference Geraylou, Souffreau, Rurangwa, D’hondt, Callewaert, Courtin, Delcour, Buyse and Ollevier2012). These compounds do induce a shift in the microbiota composition and can suppress Salmonella colonisation (Eeckhaut et al., Reference Eeckhaut, Van Immerseel, Dewulf, Pasmans, Haesebrouck, Ducatelle, Courtin, Declour and Broekaert2008). These results suggest that certain prebiotics may help to control dysbiosis in animals, as has been suggested also for human dysbiosis (Di Bartolomeo et al., Reference Di Bartolomeo, Startek and Van den Ende2013). Nevertheless, further experimental and field studies are required to support this suggestion.

Future perspectives

Numerous feeding trials have been carried out already using a variety of probiotics and prebiotics, especially in broilers. However, the endpoints were mostly feed conversion and daily weight gain, and the approach was mostly empirical. A protective effect of a probiotic against experimental challenge with a specific pathogen has been documented a.o. in Clostridium perfringens-associated necrotic enteritis in broilers (Jayaraman et al., Reference Jayaraman, Thangavel, Kurian, Mani, Mukkalil and Chirakkal2013). Further studies on the health-protecting effects of pre- and probiotics are underway. Further research also on new prebiotics and probiotics is ongoing, as legislative pressure on the use of growth-promoting antibiotics and other traditional control measures for dysbiosis (e.g. zinc salts) is getting tighter.

However, a major breakthrough in this field will require a better understanding of the intestinal ecosystem. It is clear that this ecosystem is extremely complex, with numerous intermicrobial interactions as well as microbiota–host interactions. Numerous attempts are currently being undertaken to try and unravel these complex interactions. One approach is based on the keystone–pathogen hypothesis, stating that certain low-abundance microbial pathogens can orchestrate inflammatory disease by remodelling a normally benign microbiota into a dysbiotic one (Hajishengallis et al., Reference Hajishengallis, Darveau and Curtis2012). In poultry this ‘alpha bug’ (overruling all the others and multiplying at an extremely high pace) could be Clostridium perfringens, leading to dysbiosis in subclinical cases, and to necrotic enteritis in severe clinical cases. Clostridium perfringens indeed possesses an impressive armoury of virulence factors (Timbermont et al., Reference Timbermont, Haesebrouck, Ducatelle and Van Immerseel2011) that could disrupt the normal equilibrium of a healthy microbiota (Stanley et al., Reference Stanley, Keyburn, Denman and Moore2012).

Another approach to understanding the intestinal ecosystem is based on metagenomic studies comparing the whole microbiota composition of healthy individuals v. individuals suffering from dysbiosis, taking advantage of currently available and affordable mass sequencing and mathematical analysis tools. This approach, however, still suffers from the enormous variation in microbiota composition between healthy individuals and the fact that even with currently available technology only a small fraction of the total 16S ribosomal-RNA-encoding gene pool of a given individual is being sequenced. Nevertheless, the first successes are there, as different enterotypes of the gut microbiome have been identified in the human population (Arumugam et al., Reference Arumugam, Raes, Pelletier, Le Paslier, Yamada, Mende, Fernandes, Tap, Bruls, Batto, Bertalan, Borruel, Casellas, Fernandez, Gautier, Hansen, Hattori, Hayashi, Kleerebezem, Kurokawa, Leclerc, Levenez, Manichanh, Nielsen, Nielsen, Pons, Poulain, Qin, Sicheritz-Ponten, Tims, Torrents, Ugarte, Zoetendal, Wang, Guarner, Pedersen, de Vos, Brunak, Doré, Consortium, Weissenbach, Dusko Ehrlich and Bork2011). In broilers, attempts have been made to identify potential performance-related beneficial gut microbiotas across feeding trials (Torok et al., Reference Torok, Hughes, Mikkelsen, Perez-Maldonado, Balding, Macalpine, Percy and Ophel-Keller2011; Stanley et al., Reference Stanley, Geier, Denman, Hazring, Crowley, Hughes and Moore2013). Thus, the available tools could be applied to investigate dysbiosis.

The third road to understanding what is going on in a healthy v. diseased intestinal tract is the metabolomics approach. Considering the vast numbers of microbes present in the lower intestinal tract, huge amounts of metabolites must be produced continuously. These metabolites impact not only on the microbiota itself but also on the intestinal mucosa and even on the general health of the host. Thus, metabolite analysis may contribute to our understanding of the intestinal ecosystem (Nyangale et al., Reference Nyangale, Mottram and Gibson2012). Metabolites produced by the different microorganisms may be toxic for other microorganisms. This is the case, for example, with metabolites produced by certain strains of Clostridium perfringens killing other closely related strains, even belonging to the same species (Timbermont et al., Reference Timbermont, Lanckriet, Pasmans, Haesebrouck, Ducatelle and Van Immerseel2009). However, metabolites produced by one group of microorganisms may also serve as the nutrient source for other microorganisms. This phenomenon of cross-feeding is probably equally important in shaping the microbial population balance in the intestinal tract. Indeed, the production of excess amounts or inadequate amounts of certain metabolites as such may be sufficient to cause dysbiosis. Understanding these metabolic cascades may help in the rational design of next-generation prebiotics.

Finally, there is definitely still a role to play for good old bacterial culturing as a tool to foster our understanding of the intestinal ecosystem. Considering that only a small minority of the microbial species residing in the intestinal tract have ever been cultured, it is clear that isolation and characterisation of new bacterial species from the gut can yield useful information. In the past, isolation attempts have focused mostly on pathogenic microorganisms. Future efforts should focus more on the beneficial microbes. However, this appears to be much more challenging. We recently isolated a collection of butyrate-producing strict anaerobes from the caeca of chickens (Eeckhaut et al., Reference Eeckhaut, Van Immerseel, Croubels, De Baere, Haesebrouck, Ducatelle, Louis and Vandamme2011). One of these strains turned out to be highly protective in a laboratory animal model for inflammatory bowel disease, through suppression of the inflammatory cascade (Eeckhaut et al., Reference Eeckhaut, Machiels, Perrier, Romero, Maes, Flahou, Steppe, Haesebrouck, Sas, Ducatelle, Vermeire and Van Immerseel2013).

In conclusion, for the time being, prebiotics and probiotics should not be considered as simple alternatives to growth-promoting antibiotics. Further fundamental research is needed to understand what goes on in the intestinal tract in health and disease, taking into account the complexity introduced by all three players of the intestinal ecosystem – namely the host, the feed and the microbes.