Introduction
Animals used for food production have been genetically selected for feed intake and muscle development (or milk and egg production) and are therefore divergent from their ancestors. In addition, these animals are reared in conditions that favor fast spread of pathogens. The high uptake of feed and the fast growth make these animals prone to intestinal disorders, what has been neglected in the past because of the use of low doses of specific antimicrobial compounds, called antimicrobial growth promoters (AGPs). These were used worldwide to maintain the profitability of the broiler and pig industry. These antibiotic substances, added at sub-therapeutic level as feed additives, increased animal performance. A ban on the use of AGPs, mainly driven by consumer concerns about increases in antimicrobial resistance, was installed in the EU in 2006, followed by a global concern that led to a decreased use or ban, depending on the region. In 2012 the Center for Veterinary Medicine of the US Food and Drug Administration wrote a ‘Guidance for Industry’ document mentioning that antibiotics should only be used in case of specific diseases and not for growth promotion. The mode of action of the AGPs is still under debate, but a variety of mechanisms have been proposed, including a reduction in total bacterial counts in the gut (and consequently less competition for nutrients), a reduction of specific pathogens (e.g. Clostridium perfringens), a decreased abundance of specific harmful bacterial properties (e.g. bile salt hydrolase activity and thus poor fat digestion), a reduced inflammatory reaction because of the decreased pathogen load, amongst others (Butaye et al., 2003; Knarreborg et al., 2004). Also direct immune-modulatory effects by AGPs have been suggested. Whatever the mechanism of action is, it is evident that host-microbiota interactions are involved. The gut-microbiota interactions are very complex since the gut is an organ that contains multiple cell types that fulfill many functions, and is hosting a diverse microbiota that carries out many functions as well, including breakdown of dietary molecules and consequently production of absorbable end products, and maturation and development of the (mucosal) immune system. The term ‘gut ecosystem’ is used to describe that the gut and the gut microbiota are forming one organ, with specific functions that are derived from both the gut microbiota’s genetic potential (the microbiome), and the functions of the host gut wall. Novel technologies (-omics technologies) have been used recently to get a better understanding of the host-microbiota interactions. More specifically, various studies using 16S rDNA sequencing led to the identification of microbial taxa that are associated with beneficial or harmful host responses, and metabolomics has been used to identify microbial metabolites that trigger these effects. The production of microbial metabolites can be steered using nutritional factors, creating an excellent opportunity to make animals more resilient against non-infectious and infectious challenges.
The host side: epithelial cells as major signal sensors
There are obvious anatomical and morphological differences between the gastro-intestinal tracts of poultry, pigs and cattle, with the latter being distinct because of the presence of forestomaches. While the rumen is a key fermentation chamber in adult cattle, calves fed on liquid diets (veal calves) do not develop forestomaches and can be considered as monogastrics, similar to chickens and pigs. Despite differences in anatomy and morphology, the cell types present in the gut wall are similar in these animal species. The luminal side of the intestinal wall is lined with absorptive epithelial cells, whose major task is water and nutrient uptake, and secretion of enzymes. They form a semi-permeable barrier between the outside world (the gut lumen) and the internal host tissues. The semi-permeable barrier is not only formed by the cell membranes of the epithelial cells, but also by tight junctions that connect neighboring epithelial cells (Piche, 2014). These connections are regulated at different levels (e.g. by cytokines). The permeability of the intestinal epithelial cell layer can be affected by epithelial cell death but also by luminal signals that increase the epithelial layer permeability by affecting the tight junctions or inducing cell death, and thus causing loss of integrity of an important barrier between the ‘inside’ and the ‘outside’ of the gut (Hooper, 2015). Loss of intestinal epithelial integrity can cause losses of host proteins (‘leaky gut’) into the lumen and can allow luminal molecules (including toxins) and micro-organisms to reach the gut submucosa under the epithelial layer. If these components have pro-inflammatory properties, this can yield massive infiltration of immune cells, which is energy-demanding for the host. Inflammation is mediated by binding of pathogen associated molecular patterns (e.g. LPS, peptidoglycan lipoproteins, flagellin) to receptors (e.g. Toll like receptors (TLRs)) that transmit signals in a cascade ultimately leading to inflammatory cell infiltration in the mucosa (Brown et al., 2011). Although this is a protective response, this inflammatory cascade should be brought back to normal conditions when the trigger is eliminated. Also intracellular receptors (NOD-like receptors) can sense bacterial compounds and can even induce tolerance (eg. peptidoglycan-derived muramyldipeptides (MDPs)). Apart from absorptive epithelial cells, also other epithelial cell types are present in the lining of the gut wall. These include mucin-producing goblet cells and antimicrobial peptide producing Paneth cells (in the crypts, not present in all animal species), important in innate defenses (Muniz et al., 2012). Entero-endocrine cells can secrete peptide hormones at the basal side of the cells that can reach the bloodstream. These peptide hormones have a variety of functions, including effects on epithelial cell proliferation, inflammation, and consequently intestinal integrity, even at distant segments of the intestine. One of the key hormones is glucagon-like peptide 2 (GLP-2), a hormone that is important in maintaining epithelial integrity (Baldassano and Amato, 2014). Entero-endocrine cells are responsive to various luminal signals, including microbial signals. Below the epithelial lining, many other cell types are present that form the lamina propria of the intestinal mucosa. These are immune cells, fibroblasts, nerve cells and muscle cells, amongst others. Intestinal integrity, inflammation and gut function all are influenced by luminal signals, of which many are produced by the microbiota (Havenaar, 2011). The above mentioned cells sense microbial signals and transmit these signals to other cell types and to other parts in the body of the animal. The microbiota composition and the metabolites produced by the bacteria are thus crucial for health and productivity.
The microbial side: the microbiota as signal producers
The microbiota composition in the gut varies with age and with the gastrointestinal segment (Stanley et al., 2014; Song et al., 2017; Sun et al., 2019; Yang et al., 2019). In general, the diversity of the microbiota increases with age. In industrial animal production, often the birth or hatching environment is as sterile as possible. This situation can be considered as unfavorable because the establishment of a protective microbiota is delayed, and the young animals are more prone to colonization by pathogens. The microbiota composition varies between the different segments of the gut. In general low numbers of bacteria are found in the proximal parts of the gut while the numbers increase towards the distal ileum, caecum or colon. The rumen of cattle is an exception, as this part is the main fermentation chamber in this animal species. The diversity generally increases significantly towards the distal gut (exception is rumen), and while in the small intestine a limited variability is found, with lactobacilli often dominant, the distal intestinal tract harbors a huge number of different bacterial groups. The distal intestinal tract of healthy subjects is mostly dominated by bacteria from the phyla Bacteroidetes and Firmicutes (together comprising more than 80% of the microbiota), the former containing many polysaccharide degrading bacterial species, while the latter contains a variety of bacterial families, including Ruminococcaceae and Lachnospiraceae families, that are considered important health-promoting populations, due to butyrate production. Also members of the phylum Proteobacteria are mostly present, although in lower numbers. These include Enterobacteriaceae, such as Escherichiacoli, thus Gram-negative bacteria that contain opportunistic pathogens and often are associated with harmful inflammatory effects. The bacterial community has the genetic potential to carry out an enormous number of physiological functions. The number of microbial genes in the gut, the microbiome, exceeds the number of animal genes, and together they form a ‘hologenome’ (Rosenberg and Zilber-Rosenberg, 2011). The variety of bacterial functions includes degradation of complex substrates (polysaccharides, proteins, fat), fermentation of substrates to yield acidic compounds, immunomodulation, communication with other bacteria, and many more. The metabolites produced by the bacterial community are of vital importance for maintaining gut health and controlling pathogen colonization.
Polysaccharide breakdown is performed by the microbiota in a cascade in which different bacterial members take care of specific catalytic steps in degrading the substrates (Flint et al., 2012). Complex substrates (such as polysaccharides, including arabinoxylans, pectins, and cellulose) are converted to oligosaccharides by specific bacterial populations (e.g. lactobacilli, some Bacteroides species, and others), and these oligosaccharides (e.g. arabinoxylanoligosaccharides (AXOS)) are further used by other bacterial groups to produce short-chain fatty acids (SCFAs, i.e. acetic, propionic and butyric acid), lactate and gases. Some polysaccharides can be broken down to glucose by single species. An example is cellulose breakdown by Ruminococcus spp. due to the production of extracellular enzymatically active protein complexes, called cellulosomes (La Reau and Suen, 2018). Depending on the bacterial network or the bacterial strains that carry out these conversions, other end products can be produced.
Short-chain fatty acids are the best known end products of polysaccharide fermentation. Apart from the typical acids, acetic, propionic and butyric acid, also succinate and lactate can be produced. The most important butyric acid producing bacteria belong to the Ruminococcaceae (Clostridial cluster IV) and Lachnospiraceae (Clostridial cluster XIVa) families (Pryde et al., 2002). These families contain strictly anaerobic bacteria that are highly abundant in the distal gut. Some of the Lachnospiraceaeconsume lactic acid to produce butyric acid (Duncan et al., 2004). Butyric acid has a variety of beneficial properties, including pathogen control, anti-inflammatory effects, increased mucin and antimicrobial peptide production, strengthening of the epithelial barrier, etc. (Guilloteau et al., 2010). Fermentation to butyrate in the distal gut can affect small intestinal function by stimulating GLP-2 secretion by entero-endocrine cells in the blood stream (Tappenden et al., 2003). This GLP-2 can have effects on various cell types in the small intestine, leading to anti-inflammatory effects, effects on the integrity of the epithelial barrier and increased cell proliferation (Rowland and Brubaker, 2011). Also proteins can be enzymatically broken down by the microbiota when they arrive in the distal compartment of the intestine (only possible when protein digestion by the host through pepsin, trypsin and peptidases did not completely degrade protein into amino acids for absorption. This can yield branched-chain fatty acids and amines, and the nature of these components depends on the amino acids being used by the bacteria. The effects of these specific metabolites (eg. polyamines spermidine, cadaverine, but also indoles) on host cells are not always clear, but some of these seem to have effects on transepithelial resistance as well. If an unfavorable shift in the microbiota composition occurs (= dysbiosis) certain crucial steps in certain pathways may be deficient, yielding shifts in bacterial metabolites in the gut. In general, there is a complex interaction between different bacterial populations for specific substrates, and the outcome of this competition can drive the microbiota to one that produces beneficial metabolites that promote gut health or toxic metabolites. When epithelial cells are killed or when the tight junctions between epithelial cells are damaged (e.g. by hydrogen sulphide, toxins, parasites), some opportunistic pathogens can take benefit by gaining access to the basolateral side of the epithelial cells and induce inflammation. Nutrient leakage and inflammation will cost energy for the animal, and will cause villus shortening or blunting, decreasing performance. When butyrate-producing bacteria are present in high amounts, the epithelial barrier integrity will in general be strong, the epithelial proliferation and thus the villus length optimal and inflammatory reactions will be reduced, while the stimulation of regulatory T-lymphocytes will yield a state of tolerance to non-harmful bacteria. It is believed that the mucosa-associated microbiota is very important, as the bacteria and their metabolites are in intimate contact with host cells.
Interfering with bacterial signal production and host sensing, by nutritional interventions
A variety of feed additives are used nowadays as either antimicrobial growth promotor alternatives or gut health stabilizers. Some might inhibit certain bacterial groups but most are supposed to steer the microbiota composition to a more favorable one, and have important host effects, either direct or indirect, the latter through the microbiota. As discussed above, feed formulas or feed additives should improve intestinal epithelial integrity, stimulate tolerance responses towards non-harmful bacteria, avoid an excess inflammation, stimulate host antibacterial responses (mucin and antimicrobial peptide production) and bring the host in a steady state of mutualism with its microbiota. This means that these feed additives or formulae should favor beneficial microbes and inhibit the microbes that produce harmful metabolites, or reduce pathogen colonization. Below a short overview is given on dietary additives that affect gut health.
a) Feed composition and enzymes
Gut inflammation and villus shortening can be induced by feeding a diet containing high amounts of non-starch polysaccharides (NSP) without NSP-degrading enzymes (Teirlynck et al., 2009). AGPs are able to reverse the inflammatory changes and villus shortening induced by the high NSP containing diet, in association with a shift in the microbiota (Teirlynck et al., 2009). It appears that the use of AGPs in the past has masked the dysbiosis-inducing effects of many feed formulas used in monogastrics. Also the feed structure, protein source and the choice of ingredients can affect gut health. Enzymes such as xylanases convert large polysaccharides to shorter oligo-saccharides and thus perform one of the initial steps in the breakdown of these substrates, as is done in the gut by bacterial species in cross-feeding pathways. This also reduces viscosity and bacterial overgrowth in the small intestine. More info on the effect of feed constituents and gut health can be read in a review paper by Choct (2009), and in the paper of Kiarie in this proceedings book.
b) Probiotics
Probiotics are defined as live micro-organisms that, when consumed in adequate amounts, confer a health effect on the host. The most widely used bacterial probiotics are Bacilli, as they are stable in formulation (spores) and produce antibacterial compounds, apart from beneficial metabolites that are under discovery. As an example, recently it was shown in our lab that specific Bacillus species produce high concentrations of niacin in vivo. Niacin is sensed by the receptor Gpr109a, that is also activated by butyrate, and activates anti-inflammatory responses (Singh et al., 2014). Niacin has been shown to reduce epithelial apoptosis in the rumen of cattle induced by excess butyrate (Luo et al., 2019). Apart from Bacillus species, also other single strain probiotics are marketed, including lactobacilli. Multi-strain products are on the market as well. Also competitive exclusion products, containing a freeze-dried mixture of gut content, are marketed. In the scientific literature, reports on the effect of probiotics on animal performance are published, and reports on the protection against pathogen colonization and disease are available. The question remains how many studies are not published because of inconsistent, no or negative effects observed. Data from our laboratory show that the efficacy of probiotics is highly depending on the model used and not all studies show clear reproducible beneficial results. Instead of empirically developing and marketing probiotics only because of their genus name, we should rethink the system and develop probiotics based on their mode of action. For example, based on the above described data, attempts could be made to evaluate strains that stimulate butyrate production by strains of Clostridial cluster IV and XIVa, or use these butyrate-producing strains as probiotics. These are, however, strict anaerobes and do not form spores consistently, while this is not a problem for Bacillus species, which are usually incorporated in feed as heat resistant spores (Shivaramaiah et al., 2011). For probiotic species that have consistent health effects, it would be good to identify mechanisms of action, and figure our which metabolites they produce that can exert direct or indirect effects on host health. This way new dietary additives can be developed.
c) Prebiotics
Prebiotics are defined as natural or processed functional foods which contain biologically active compounds that have documented benefits on health by altering the interactions between beneficial and pathogenic bacteria (Gibson and Roberfroid, 1995). Most prebiotics are oligosaccharides, such as fructooligosaccharides (FOS), galactooligosaccharides (GOS), arabinoxylan oligosaccharides (AXOS), xylan oligosaccharides (XOS) and raffinose family oligosaccharides (RFOs). Mannanoligosaccharides (MOS) are often not considered as prebiotics because they are not supposed to be fermented but have direct immunomodulatory effects. Prebiotics are complex molecules because of the chain length, the nature of the sugar bounds, and the nature of the side chains on the saccharides. All this can affect function. The scientific literature reports various studies in which prebiotics are having beneficial effects on broiler performance and pathogen control. As with probiotics, it is difficult to estimate the bias that is present using data derived from scientific papers, because only beneficial effects are mostly reported, and no or negative effects are seldom published. It is anyhow the case that the prebiotics need to be converted by the microbiota to metabolites. Because prebiotics are saccharides, the end products will be SCFAs, lactate and gases and thus the beneficial effect can theoretically be evaluated or predicted by measuring the ratio of beneficial versus harmful bacterial groups or metabolites. As such, prebiotics that increase colonization of butyrate-producing Clostridial cluster IV and XIVa bacteria are considered to be beneficial. Other parameters could include reductions in Enterobacteriacea. Also in the case of prebiotics we thus need to proceed in the future towards a science-driven development in which the mechanism of action plays a central role, instead of empirically developing prebiotics. For example, our group could show that XOS administration to a broiler diet increased the number of lactobacilli and Clostridial cluster XIVa strains in the distal gut, hereby stimulating cross-feeding of lactate to butyrate (De Maesschalck et al., 2015).
d) Synbiotics
Synbiotics are combinations of pre- and probiotics and thus offer the bacteria directly the substrate they can convert to beneficial metabolites (Roberfroid, 1998). In the above mentioned example, XOS and a Lactobacillus strain could for example be given to promote lactic acid production in the gut. This is a very simplified way of thinking because this depends on the fermentation of XOS by other bacteria at different sites in the gut, and the colonization site of the strain administered. There is thus a lot to learn on the nature of the substrates that are required to stimulate beneficial bacterial groups before these synbiotics can be developed with a high success rate.
e) Essential oils, phytobiotics
Essential oils and botanical products are also well-known feed additives in the animal production industry. Biologically active constituents of plants are terpenoids (mono-and sesquiterpenes, steroids, etc.), phenolics (tannins), glycosides and alkaloids (present as alcohols, aldehydes, ketones, esters, ethers, lactones, etc.). Many of these, but not all, have antibacterial activity (Penalver et al., 2005; Barbosa et al., 2009). Effects on immune function and host responses have not yet been investigated thoroughly, and also for this class of compounds it is difficult to explain the mechanism of action of the products that have been published in the scientific literature as beneficial for gut health. According to Adams (1999) the antimicrobial activity is rather weak for ginger and pepper, medium for cumin (p-cymene), coriander (lialol), oregano (carvacrol), rosemary (cineol), sage (cineol) and thyme (thymol) and strong for clove (eugenol), mustard (allylisothiocyanate), cinnamon (cinnamaldehyde) and garlic (allicin). Also here the dosage, purity, extraction method from the plant (in case of mixtures, thus phytobiotics) or synthetic production method will determine the success of the products. It is clear that the antibacterial essential oils will affect the gut microbiota composition, and there is a need to clarify which ones are promoting beneficial bacterial species, using in vivo studies. Resin acids have recently been studied and seem to alter matrix metalloproteinase activity in the gut mucosa, that could be highly relevant in restoration of intestinal damage (Aguirre et al., 2019). Indeed, matrix metalloproteinase upregulation has been shown in gut inflammation models and are likely involved in extracellular matrix breakdown.
f) Short chain fatty (and other) acids
Drinking water and feed additives containing SCFAs, medium chain fatty acids and even aromatic acids (e.g. benzoic acid) are widely used in the animal production industry. While drinking water acidification is mainly for sanitation purposes, feed additives are used mainly for optimizing animal performance and for pathogen control (Van Immerseel et al., 2006). It is difficult to compare the relative efficacy of commercial products because they differ in the nature of the acids used (often combinations are used), the concentration and even more important the delivery method (pure, on a carrier, encapsulated …). The latter determines the site of release in the gut and can affect the outcome. While SCFAs are more considered as signaling molecules for the microbiota and the host, the medium-chain and aromatic acids are more antibacterial.
Final considerations
A huge number of experimental and field trials have been carried out using a variety of feed additives, in production animals. The most commonly measured outcome parameter is performance, either or not under challenge conditions. Some studies have been undertaken to determine the effect on pathogen colonization. The approach so far was mostly empirical and the products are thus mainly developed without a clear understanding of the reasons of the expected beneficial effects. Many feed additives that are meant to replace AGPs have variable activities. The only way to develop a product with an enhanced activity as compared to the already existing products will be based on a thorough understanding of the intestinal ecosystem, and the way the gut wall responds to the microbiota and their metabolites. Identifying the microbiota components that are crucial for gut health is ongoing and is essential for proper development of additives that affect gut health. This needs to be done by identifying both the beneficial ones and the harmful ones. In fact, current knowledge indicates that butyrate-producing bacteria need to be boosted or maintained while Enterobacteriaceae and specific pathogens such as C. perfringens need to be suppressed. These are easy to measure criteria and are well known to correlate with a good morphological structure of the gut. In fact, studies that are recently carried out and future studies using –omics technologies will be of value to identify potential performance-related beneficial gut microbiota components and metabolites (Torok et al., 2011). Although meta–omics tools (metagenomics, metabolomics, metaproteomics) can be very informative, there is a need to investigate the specific effect of the different bacteria and metabolites that are found to play a role in gut health. So bacterial culturing is a crucial tool to foster our understanding of the intestinal ecosystem and is essential to study the effect of a specific strain or species on gut health parameters (Walker et al., 2014). Only a small minority of the microbial species residing in the gut have been cultured so far, and thus isolation and characterization of new bacterial species from the gut will yield useful information. A recently established method to isolate, describe and genome sequence novel strains is called culturomics (Lagier et al., 2012). This method consists of plating gut or faecal samples on a variety (over 100) of media, followed by identification of the bacteria. Although this technique is able to discover new bacterial strains, a major breakthrough would be a method that permits to isolate strains that produce predefined metabolites, but this is not available yet. For the time being we have to rely on the isolation of potentially beneficial microbes and test their behavior, and more importantly study how we can promote their abundance in the animal gut. In the future many more health promoting and harmful bacterial groups and their respective metabolites will evidently be discovered and will steer our ways to optimize gut health and animal performance.
