The microbial ecosystem residing in the human gut consists of over 100-fold more genes than the human genome( Reference Ley, Turnbaugh and Klein 1 – Reference Backhed, Ding and Wang 3 ) and is tantamount to a virtual organ. To a large extent, intestinal ecological conditions are set by the host and resident commensals must adapt to this environment. Host–microbe, environment–microbe and microbe–microbe interactions may also dictate the composition of this microbial community. A symbiotic relationship exists between the gut microbiota and host such that both partners benefit; the host provides protection and nutrients for the micro-organisms to flourish within( Reference Maynard, Elson and Hatton 4 ), whereas the microbiota contribute to food digestion, inhibit the growth of potential invading pathogens, convert harmful compounds into less toxic substances and produce bioactive molecules, which play a role in host physiology( Reference Marques, Wall and Ross 5 ). Disruptions to this symbiotic relationship can occur, for example, during certain disease states and during adoptive pathogenesis by certain commensal gut microbes causing small intestinal bacterial overgrowth( Reference Bouhnik, Alain and Attar 6 , Reference Riordan, McIver and Wakefield 7 ) and/or translocation to other tissues and organs( Reference Maes, Kubera and Leunis 8 , Reference Teltschik, Wiest and Beisner 9 ). However, it remains unclear whether disease development is causal or consequential of an altered intestinal microbiota, with an increasing body of evidence describing a link between the two.
Microbial colonisation of the infant intestine begins at birth( Reference Koenig, Spor and Scalfone 10 ). Extrinsic factors contribute to the initial colonisation of the infant gut( Reference Scholtens, Oozeer and Martin 11 , Reference Butel, Suau and Campeotto 12 ), including mode of delivery( Reference Dominguez-Bello, Costello and Contreras 13 ), feeding regime( Reference Bezirtzoglou, Tsiotsias and Welling 14 , Reference Fallani, Young and Scott 15 ), gestational age at birth( Reference Hallab, Leach and Zhang 16 ) and antibiotic therapy( Reference Fouhy, Guinane and Hussey 17 , Reference Hussey, Wall and Gruffman 18 ). Intestinal establishment of a healthy microbiota is believed to have a profound impact on the development and maturation of the immune system( Reference Palmer, Bik and DiGiulio 19 ). Vaginally born infants are initially colonised by faecal and vaginal bacteria from the mother, whereas infants delivered by Caesarean-section render the gut susceptible to colonisation by maternal skin microbiota and bacteria from the hospital environment( Reference Dominguez-Bello, Costello and Contreras 13 , Reference Penders, Thijs and Vink 20 , Reference Huurre, Kalliomaki and Rautava 21 ). It has been shown that vaginally born babies have higher numbers of Lactobacillus and Bifidobacterium, compared with infants delivered by Caesarean-section( Reference Dominguez-Bello, Costello and Contreras 13 ). The weaning process determines the transition of an unstable infant microbiota to a more complex adult-like microbial ecosystem( Reference Koenig, Spor and Scalfone 10 , Reference Fallani, Amarri and Uusijarvi 22 ). The development and diversification of the gut microbiota continues into adulthood and is further influenced by several factors, including diet and environment( Reference Yatsunenko, Rey and Manary 23 ). To a large extent, the gut microbiota remains relatively stable throughout adulthood, unless perturbed by extrinsic or host factors, including antibiotic treatment and inflammation, respectively.
The tools used for studying the link between gut microbial diversity and health status have improved our knowledge of the host–microbe relationship significantly. Culture-independent analysis of the composition and functional capacity of the gut microbiome targets the 16S rRNA gene, due to its presence in all prokaryotes with the existence of variable domains that allow different taxa to be identified. Although compositional studies generate a large volume of data, they fail to provide direct information regarding the microbial viability or the functional potential of the populations present and so the knowledge generated is somewhat limited in these aspects. Metagenomic studies go beyond the 16S rRNA gene to sequence small fragments of metagenomic DNA at random to characterise the full genetic content and functional potential of the microbial community( Reference Qin, Li and Raes 2 , Reference Kurokawa, Itoh and Kuwahara 24 , Reference Turnbaugh, Ridaura and Faith 25 ). The development of methods used to analyse gene expression (metatranscriptomics), protein products (metaproteomics) and metabolic profiles (metabolomics) of the gut microbiota has further enabled such studies to identify the microbial activity and to link this with compositional analysis, to determine host–microbe interactions( Reference Guinane and Cotter 26 ).
Dietary interventions with probiotics and prebiotics, due to the dynamic nature of the gut microbial ecosystem, have become an attractive means of self-manipulating the microbiota to improve health status, and have been extensively reviewed( Reference Vyas and Ranganathan 27 , Reference Sanders, Guarner and Guerrant 28 ). Probiotics are defined as ‘live micro-organisms which when administered in adequate amounts confer a health benefit on the host’( 29 ) and have been shown to improve intestinal barrier function, modulate the immune system and enhance the host defence system by competing against pathogens for nutrients and binding sites. In addition, numerous probiotic intervention studies have revealed their functional capacity to improve certain gastrointestinal disorders, for example, irritable bowel syndrome, inflammatory bowel disease (IBD)( Reference Cappello, Tremolaterra and Pascariello 30 , Reference Whelan and Quigley 31 ) and antibiotic-associated diarrhoea( Reference Hempel, Newberry and Maher 32 , Reference Hickson 33 ). Bifidobacterium and Lactobacillus are the main genera of micro-organisms used as probiotics as many of them can survive gastrointestinal transit, have the capability to adhere to intestinal epithelial cells and are regarded as safe( Reference Marques, Cryan and Shanahan 34 ). Prebiotics are non-digestible food ingredients that selectively stimulate the growth of beneficial indigenous microbes already established within the gut such as bifidobacteria and lactobacilli( Reference Saulnier, Spinler and Gibson 35 ). Typically, prebiotics must reach the large intestine unaltered, resisting host digestion, absorption and adsorption to be fermented by the gut microbiota. Commonly used prebiotics include inulin, fructo-oligosaccharides and galacto-oligosaccharides( Reference Preidis and Versalovic 36 ). The fermentation of prebiotics by the gut microbiota generates SCFA, such as butyric and acetic acids, which are linked with numerous health benefits in vivo ( Reference Saulnier, Spinler and Gibson 35 , Reference Preidis and Versalovic 36 ). This review will focus on the importance of the microbiota to host health and in establishing a healthy immune system and describes some of the known beneficial bioactive metabolites produced by the microbiota that impact on health.
Role of the microbiota in establishing a healthy immune system
During the first year of life, the immature developing gut microbial ecosystem rapidly shapes the maturation of the infant immune system, while the immune system also influences the gut microbiota( Reference Hansen, Nielsen and Kverka 37 ). From birth, breast milk provides passive transfer of maternal antibodies to the infant which shapes both the immature immune system and gut microbiota( Reference Maynard, Elson and Hatton 4 ). Much of the information regarding the influence of gut microbes on the host immune system is generated from studies using germ-free animals, i.e. those born and reared without exposure to micro-organisms such that the immune responses have not been influenced by interactions with molecules of commensal and pathogenic micro-organisms. Consequently, germ-free animals show defects in both the development of the immune system and in immune responses. One of the first immunological defects observed in these animals was a marked reduction in antibodies produced within the intestine( Reference Moreau, Ducluzeau and Guy-Grand 38 ). Furthermore, germ-free animals show extensive defects in the development of gut-associated lymphoid tissue and have fewer and smaller Peyer's patches and mesenteric lymph nodes, compared with animals housed under specific pathogen-free conditions( Reference Macpherson and Harris 39 – Reference Hoshi, Aijima and Horie 42 ). Intestinal epithelial cells have many immunological functions; they secrete and respond to various cytokines and express molecules that directly interact with lymphocytes and line the gut to form a physical barrier between the luminal contents (including the microbiota) and the underlying cells of the immune system( Reference Round and Mazmanian 43 ). Germ-free mice demonstrate a reduced number of these cells, whereby their function is compromised( Reference Imaoka, Matsumoto and Setoyama 44 , Reference Umesaki, Setoyama and Matsumoto 45 ) and demonstrate decreased cell turnover rates of these cells( Reference Abrams, Bauer and Sprinz 46 ). Furthermore, animals lacking a gut microbiota are more susceptible to infection due to a poorly developed immune system; e.g. germ-free guinea-pigs challenged with the enteric pathogen Shigella flexneri demonstrated a decrease in the immune resistance to infection coupled with an increase in mortality( Reference Sprinz, Kundel and Dammin 47 ), while infection with the intracellular pathogen Listeria monocytogenes in germ-free mice resulted in decreased pathogen clearance, compared with conventionalised animals( Reference Zachar and Savage 48 ). Deliberate colonisation of the sterile gut of these animals either with a single microbial species or a defined species mixture, termed ‘gnotobiotics’ is a powerful technological tool for determining which host immune functions are genetically encoded and which require interactions with microbes( Reference Hooper and Macpherson 49 ). For example, colonisation of germ-free animals with a single bacterium, Bacteroides fragilis, has been shown to protect against inflammation in an animal model of experimental colitis( Reference Mazmanian, Round and Kasper 50 , Reference Mazmanian, Liu and Tzianabos 51 ). Collectively, these observations suggest that developmental defects through the absence of a gut microbial ecosystem compromise immune function of the host at the tissue, cellular and molecular levels and highlight a role of the microbiota in the establishment of a functional immune system.
Implications of a perturbed gut microbiota on immune function and health
Antibiotic therapy soon after birth has been shown to impact gut microbiota composition up to 8 weeks after treatment( Reference Fouhy, Guinane and Hussey 17 , Reference Hussey, Wall and Gruffman 18 ). This could have a negative impact on the development of the immune system, predisposing the infant to development of asthma, obesity and allergies( Reference Fouhy, Guinane and Hussey 17 , Reference Hussey, Wall and Gruffman 18 ). Thus, disruptions in the host–microbe relationship can undoubtedly predispose to disease, from infancy to adulthood. There has been a rapid increase in the development of disorders such as IBD, asthma, rheumatoid arthritis, diabetes and obesity, particularly within developed, Western populations. Indeed, the role of a perturbed gut microbial ecosystem in such diseases is becoming more evident, although future work is needed to decipher the mechanisms involved.
Inflammatory bowel disease
IBD comprises a group of disorders characterised by severe intestinal inflammation and is characterised as either Crohn's disease or ulcerative colitis, based on the location of the gastrointestinal tract affected. Although the exact causes of IBD remain unclear, the onset of both conditions is generally thought to be due to an overall disruption in the host–microbe relationship, and not by a single causal organism( Reference Lepage, Hasler and Spehlmann 52 , Reference Martinez, Antolin and Santos 53 ). Recently reviewed, numerous studies which indicate a role of the gut microbiota in the manifestation of IBD generally conclude that the gut microbiota are involved in the development of mucosal lesions causing intestinal inflammation( Reference Manichanh, Borruel and Casellas 54 ). Inflammatory damage in IBD has been linked to alterations in the relative abundances of Enterobacteriaceae, Ruminococcaceae and Leuconostocaceae ( Reference Morgan, Tickle and Sokol 55 ) and an overall decrease in bacterial diversity( Reference Manichanh, Rigottier-Gois and Bonnaud 56 , Reference Sokol, Seksik and Rigottier-Gois 57 ). The incidence of Clostridium difficile carriage, an opportunistic pathogen frequently linked with antibiotic-associated diarrhoea, has been reported to be over 8-fold higher in patients suffering from IBD, compared with healthy controls( Reference Clayton, Rea and Shanahan 58 ).
Type-1 and type-2 diabetes
The incidence of both type-1- (T1D) and type-2- (T2D) diabetes have increased dramatically in recent decades. Although genetic factors play a role in disease onset, particularly in predisposing individuals to T1D, T2D is principally linked to obesity associated insulin resistance. Recent studies demonstrate disruptions to the host–microbe relationship and gut microbial composition and diversity associated with both T1D and T2D. Compositional sequencing studies have revealed a reduction in the relative proportions of Firmicutes, while Bacteroidetes were enriched in T2D subjects, compared with healthy controls( Reference Larsen, Vogensen and van den Berg 59 ). Identification of gut microbial markers associated with the moderate degree of microbial disruption in patients with T2D could be useful in the future management of this disease( Reference Qin, Li and Cai 60 ). Furthermore, increases in opportunistic pathogens such as Clostridium were identified as contributing to disruption of host–microbe interactions associated with T2D( Reference Qin, Li and Cai 60 ). Creating a link between the gut microbiota and T1D is more difficult, since genetic factors play a more significant role in this disease. However, evidence indicates that alterations in the intestinal microbiota are associated with T1D and subsequent insulin dependence in various models of the disease. While one study demonstrated that the stool of bio-breeding diabetes-resistant rats contained higher relative abundances of Lactobacillus and Bifidobacterium, compared with bio-breeding diabetes-prone rats( Reference Roesch, Lorca and Casella 61 ), others have reported that lactate-producing species such as Lactobacillus, Lactococcus and Bifidobacterium were increased in the stool of children who tested positive for T1D-associated autoimmunity( Reference Brown, Davis-Richardson and Giongo 62 , Reference Giongo, Gano and Crabb 63 ). Furthermore, low relative abundances of two of the most common Bifidobacterium species, B. adolescentis and B. pseudocatenulatum have been associated with autoimmunity in children who tested positive for at least two T1D-associated autoantibodies( Reference de Goffau, Luopajarvi and Knip 64 ).
Obesity
Excessive energy intake over expenditure is the main cause of obesity, since host-negative feedback signals are insufficient to maintain normal weight in circumstances of plentiful food/energy supply. Although lifestyle, genetic factors, diet and exercise undoubtedly contribute largely to this modern epidemic, an increasing body of evidence suggests that disruptions to the host–microbe relationship also contribute( Reference Ley 65 – Reference Turnbaugh, Ley and Mahowald 68 ). Identifying specific populations which may be associated with weight gain has been the subject of much debate, often differing among various models of obesity in both rodent and human subjects. Genetically (ob/ob) and diet-induced obese mice have been shown to harbour an increased Firmicutes:Bacteroidetes ratio, compared with their lean counterparts( Reference Ley, Backhed and Turnbaugh 66 ). Furthermore, weight loss in human subjects has been linked with decreased Firmicutes:Bacteroidetes ratio( Reference Ley, Turnbaugh and Klein 1 ), but yet the relevance of the Firmicutes:Bacteroidetes ratio in obesity remains unclear( Reference Schwiertz, Taras and Schafer 69 ). The gut microbiota also increase the dietary energy-harvesting capacity of the host( Reference Murphy, Cotter and Healy 70 ) and conventionally raised mice have been shown to contain 40 % more body fat than their germ-free counterparts, while colonisation with a conventional gut microbiota induced hepatic lipogenesis and increased lipid storage in adipocytes( Reference Backhed, Ding and Wang 3 ).
Obesity is also associated with low-grade inflammation, which may be linked to host–microbe interactions. Data from several studies have revealed that the lipopolysaccharide (LPS) endotoxin derived from certain components of the gut microbiota contributes towards obesity-associated inflammation. Endogenous LPS is continuously produced in the gut as a consequence of inactivation of Gram-negative bacteria, since LPS is a component of the Gram-negative bacterial cell wall and acts through the Toll-like receptor 4/MyD88/NF-κB-signalling pathway. LPS-induced inflammation could also be an early factor which triggers high-fat diet-induced metabolic diseases, otherwise known as metabolic endotoxemia( Reference Cani, Amar and Iglesias 71 ). It has been shown that high-fat feeding increased plasma LPS levels throughout the day, compared with controls, resulting in significant increases in fasting blood glucose, insulin, liver TAG content, body weight and proinflammatory cytokine mRNA expression, similar to mice that were infused with LPS( Reference Cani, Amar and Iglesias 71 ). Further studies examined the effect of changes in the gut microbiota leading to LPS-induced metabolic endotoxemia( Reference Cani, Bibiloni and Knauf 72 ). It was revealed that while plasma LPS levels were increased following high-fat feeding relative to controls, this result was overturned in high-fat diet-fed mice following antibiotic treatment( Reference Cani, Bibiloni and Knauf 72 ). Such studies reveal that disruptions to host–microbe interactions within the gut following obesity and high-fat diet may generate increased gastrointestinal levels of microbial-derived LPS endotoxin, associated with metabolic endotoxemia.
Microbial metabolism of choline and CVD
Choline is a water-soluble essential nutrient, an important component of cell membranes and mediates lipid metabolism and VLDL synthesis in the liver( Reference Vance 73 ). In addition, choline is a precursor to the neurotransmitter acetylcholine, which has important functions in cognition, as discussed later. While small quantities of choline are continuously synthesised by the host, it is mostly obtained from foods such as red meats and eggs. From infancy, breast milk is an important source of choline and the US Food and Drug Administration requires that infant formula not made from cow's milk be supplemented with choline. Much like a choline-deficient diet, microbial metabolism of choline decreases the bioavailable levels of this essential nutrient and triggers non-alcoholic fatty liver disease( Reference Dumas, Barton and Toye 74 ). The gut microbiota play a role in the transformation of dietary choline to trimethylamine with subsequent metabolism in the liver to the toxic methylamine, trimethylamine-N-oxide( Reference Dumas, Barton and Toye 74 , Reference Prentiss, Rosen and Brown 75 ). Excess plasma levels of the pro-atherosclerotic metabolite trimethylamine-N-oxide and its metabolites are associated with CVD( Reference Wang, Klipfell and Bennett 76 ). One recent study highlighted a direct link between increased plasma trimethylamine-N-oxide levels, increased risk of major adverse cardiovascular events and the gut microbiota( Reference Tang, Wang and Levison 77 ). Plasma trimethylamine-N-oxide levels were suppressed following antibiotic treatment, but reappeared following antibiotic withdrawal( Reference Tang, Wang and Levison 77 ).
Host–microbe interactions, generation of long-chain PUFA and microbial metabolite production with health effects
The products of human enteric microbial metabolism often act as signalling molecules, developing ‘intelligent communication systems’ in the body. These pharmabiotics can exert beneficial health effects, which directly impact host intestinal function but may also affect the liver and brain( Reference Shanahan 78 ). Host–microbe interactions can together co-metabolise dietary components to produce a large array of molecules with beneficial impacts on health. Commensal bacteria have been shown to synthesise essential vitamins such as vitamin K2 and B vitamins( Reference Said 79 ), can alter n-3 PUFA metabolism to generate increased levels of long-chain PUFA metabolites such as EPA and DHA( Reference Wall, Ross and Shanahan 80 , Reference Wall, Marques and O'Sullivan 81 ), can produce conjugated fatty acid derivatives of PUFA such as conjugated linoleic acid (CLA) and conjugated α-linolenic acid( Reference Barrett, Fitzgerald and Dinan 82 , Reference Hennessy, Barrett and Ross 83 ) and can increase production of SCFA( Reference Wall, Marques and O'Sullivan 81 ). The beneficial impacts of some of these bioactive compounds on host health are reviewed later and summarised in Table 1.
Table 1. Effects of probiotic metabolite production on host metabolic and psychiatric health
CLA, conjugated linoleic acid; EPS, exopolysaccharides; GABA, γ-aminobutyric acid.
Vitamin synthesis
Some commensals of the human gut microbiota possess the ability to synthesise menaquinone (vitamin K2), as well as many of the water-soluble B vitamins such as biotin, cobalamin, folate, nicotinic acid, panthotenic acid, pyridoxine, riboflavin and thiamine( Reference LeBlanc, Milani and de Giori 92 ). In particular, many Bifidobacterium strains have been shown to exhibit vitamin production capabilities( Reference Deguchi, Morishita and Mutai 93 – Reference Pompei, Cordisco and Amaretti 95 ). Vitamin K is a lipophilic vitamin which acts as a co-factor for the enzyme γ-carboxylase, which converts specific glutamyl residues in a limited number of proteins to γ-carboxyglutamyl (Gla) residues, responsible for high-affinity binding of calcium ions( Reference LeBlanc, Milani and de Giori 92 ). The daily requirement for vitamin K is fulfilled by dietary phylloquinone, present in plants and to an undetermined extent, by bacterially produced vitamin K2 ( Reference LeBlanc, Milani and de Giori 92 ). Vitamin K is important for blood clotting, bone and vascular health and deficiencies have been associated with low bone mineral density( Reference Szulc, Arlot and Chapuy 96 , Reference Knapen, Nieuwenhuijzen Kruseman and Wouters 97 ), increased risk of fracture( Reference Szulc, Chapuy and Meunier 98 , Reference Luukinen, Kakonen and Pettersson 99 ) and CVD( Reference Geleijnse, Vermeer and Grobbee 100 ).
Vitamin B12 is a type of cobalt corrinoid, particularly of the cobalamin group and is solely synthesised by some bacteria and archea. Vitamin B12 biosynthesis was first identified in Propionibacteria freudenreichii, now used in the commercial production of the vitamin( Reference Roessner, Huang and Warren 101 ). Furthermore, Lactobacillus reuteri CRL1098 was shown to be the first lactic acid producing bacterial strain capable of producing a cobalamin-like compound with an absorption spectrum resembling that of standard cobalamin( Reference Taranto, Vera and Hugenholtz 102 ). The genetic pathway responsible for the de novo synthesis of vitamin B12 by L. reuteri has previously been described for two L. reuteri strains( Reference Saulnier, Santos and Roos 103 ). More recently, the presence of bifidobacterial genes, predicted to be involved in the biosynthesis of several B vitamins has been identified in bifidobacteria residing in faecal samples of adult subjects( Reference Klaassens, Boesten and Haarman 104 – Reference Gosalbes, Durban and Pignatelli 106 ).
PUFA and conjugated fatty acid synthesis
PUFA contain two or more double bonds and are classified as either n-3, n-6 or n-9, based on the location of the last double bond relative to the terminal methyl end of the molecule. Linoleic acid (18:2n-6; precursor to the n-6 series of fatty acids) and α-linolenic acid (18:3n-3; precursor to the n-3 series of fatty acids) are the simplest members of each family of PUFA and are essential fatty acids. PUFA regulate a wide variety of biological functions, ranging from blood pressure and blood clotting, to the development and functioning of the brain and nervous system. It has been shown that the gut microbiota not only affects fat quantity( Reference Backhed, Ding and Wang 3 ), but also affects fat quality( Reference Wall, Ross and Shanahan 80 , Reference Wall, Ross and Shanahan 86 ) in various animal models. CLA refers to a family of positional and geometric isomers of linoleic acid, which have been associated with several health benefits. The CLA isomers cis-9, trans-11 and trans-10, cis-12 are most often studied for their beneficial in vitro and in some cases in vivo health effects associated with various types of cancer, atherosclerosis, obesity, diabetes, as well as an ability to improve immune function, body composition and bone formation( Reference Belury 107 – Reference Watras, Buchholz and Close 119 ). It has recently been shown that plasma CLA metabolite concentrations in human subjects following dietary CLA supplementation were comparable with those previously observed in experimental animal models and sufficient enough to exert health benefits( Reference Mele, Cannelli and Carta 120 ). Considerable species variations among bifidobacteria have been observed for PUFA and CLA productions. Although Bifidobacterium breve has been reported as one of the most efficient CLA producers among various strains tested( Reference Coakley, Ross and Nordgren 121 , Reference Rosberg-Cody, Ross and Hussey 122 ), Bifidobacterium bifidum ( Reference Rosberg-Cody, Ross and Hussey 122 ) and Bifidobacterium dentium ( Reference Coakley, Ross and Nordgren 121 ) have also demonstrated good conversion rates in vitro. It has been shown that administered CLA-producing strains of bifidobacteria are metabolically active in the gastrointestinal tract of mice and pigs( Reference Wall, Ross and Shanahan 80 ). Furthermore, administration of B. breve NCIMB 702258 in combination with linoleic acid resulted in modulation of tissue fatty acid composition, significantly increasing levels of cis-9, trans-11 CLA in the liver of both mice and pigs( Reference Wall, Ross and Shanahan 80 ). Increased tissue concentrations of n-3 long-chain PUFA, EPA and DHA were also found in the adipose tissue of both mice and pigs( Reference Wall, Ross and Shanahan 80 ). Furthermore, the ratio of arachidonic acid to EPA in the liver and adipose tissue was reduced following B. breve supplementation, coupled with an anti-inflammatory cytokine profile in the host( Reference Wall, Ross and Shanahan 80 ). Both EPA and DHA have previously been shown to exert anti-inflammatory properties( Reference Nobre, Correia and Borges Mde 123 ). In a related study, it was found that administration of B. breve NCIMB 702258 in combination with α-linolenic acid was associated with alterations in the fatty acid composition of the brain, with elevated levels of EPA and DHA( Reference Wall, Marques and O'Sullivan 81 ). Such studies have demonstrated that manipulation of the gut microbiota with metabolically active strains may represent a therapeutic strategy for various disorders related to inflammation in the host, through the production of long-chain PUFA and PUFA-derived conjugated fatty acids.
Production of SCFA
SCFA are the end products of anaerobic gut microbial fermentation of undigested dietary fibres and have important functions in host energy metabolism. Indeed, SCFA play a key role in the prevention and treatment of metabolic and bowel disorders and certain types of cancer( Reference Hu, Chen and Xu 124 – Reference Tang, Chen and Jiang 128 ). The positive influence of SCFA treatment on ulcerative colitis and Crohn's disease have been demonstrated in various clinical studies( Reference Harig, Soergel and Komorowski 129 – Reference Di Sabatino, Morera and Ciccocioppo 133 ). Butyrate is the primary energy source for cellular metabolism in the colonic epithelium( Reference den Besten, van Eunen and Groen 134 ). The colonic epithelial cells of germ-free mice are severely energy-deprived and are characterised by increased activation of AMP-activated protein kinase, which senses cellular energy status( Reference Donohoe, Garge and Zhang 135 ). SCFA also regulate gene expression in the host by binding to the G-protein-coupled receptors, GPR41 and GPR43 to impact on several different cellular functions in the host, depending on the cell type( Reference Tremaroli and Backhed 136 ). For example, SCFA suppress inflammation through GPR43 signalling in immune cells( Reference Maslowski, Vieira and Ng 137 , Reference Sina, Gavrilova and Forster 138 ) and modulate secretion of the insulin secreting and antidiabetic hormone glucagon-like peptide-1 in the distal small intestine and colon( Reference Tolhurst, Heffron and Lam 139 ).
Microbial production of exopolysaccharides
Many organisms including some resident microbes of the gut microbial ecosystem have the ability to synthesise exopolysaccharides (EPS) with a large variation in composition, charge and molecular structure( Reference Stack, Kearney and Stanton 140 ). EPS-producing strains are responsible for a ‘ropy’ phenotype and are beneficial in the food and health industries. Health benefits associated with EPS include immunostimulatory effects( Reference Vinderola, Perdigon and Duarte 88 , Reference Kitazawa, Harata and Uemura 141 ), blood cholesterol-lowering effects( Reference Nakajima, Suzuki and Kaizu 87 , Reference Maeda, Zhu and Omura 142 ) and prebiotic effects( Reference Korakli, Ganzle and Vogel 143 , Reference O'Connor, Barrett and Fitzgerald 144 ). β-Glucan is a water-soluble fibre found in cereals, as well as in yeast, bacteria, algae and mushrooms( Reference Theuwissen and Mensink 145 ). The EPS β-glucan has been reported to have many health promoting properties, including immunomodulatory effects( Reference Akramiene, Kondrotas and Didziapetriene 146 – Reference Volman, Ramakers and Plat 149 ), lowering serum cholesterol levels( Reference Theuwissen and Mensink 145 , Reference Wilson, Nicolosi and Delaney 150 ), antiosteoporotic( Reference Shin, Yang and Park 151 ), antitumourigenic, anticytotoxic and antimutagenic effects( Reference Gu, Fujimiya and Itokawa 152 , Reference Mantovani, Bellini and Angeli 153 ). Furthermore, oat β-glucan has also been associated with the ability to modulate satiety, thus controlling appetite( Reference Beck, Tapsell and Batterham 154 , Reference Beck, Tosh and Batterham 155 ). Heterologous expression of the pediococcal glycotransferase (gtf) gene responsible for the synthesis and secretion of the two substituted (1,3) β-d-glucan in Lactobacillus paracasei NFBC 338 increased the stress tolerance of the probiotic, due to EPS production( Reference Stack, Kearney and Stanton 140 ). Furthermore, B. breve UCC2003 has been shown to produce two EPS which have been associated with an increased resilience of this strain to tolerate acid and bile while reducing the intestinal colonisation levels of pathogenic Citrobacter rodentium ( Reference Fanning, Hall and van Sinderen 156 ). Thus, EPS production is thought to be important not only in host interactions, but also for protection against pathogenic infection.
The gut–brain axis: microbial metabolite production with implications on host psychiatric health
The gut–brain axis is a bidirectional communication system between the brain and the gut, including the metabolically complex gut microbiota which integrates neural, hormonal and immunological signalling between the gut and the brain( Reference Collins, Surette and Bercik 157 ). The gut microbiota and the metabolites they produce may also modulate the peripheral nervous system and central nervous system (CNS) to influence brain development and function( Reference Forsythe, Sudo and Dinan 158 ). To date, numerous studies have demonstrated the importance of the gut microbiota in the stress response( Reference Neufeld, Kang and Bienenstock 159 , Reference Sudo, Chida and Aiba 160 ) and neurodevelopmental disorders( Reference de Theije, Wu and da Silva 161 – Reference Finegold, Dowd and Gontcharova 163 ). Commensal microbiota have demonstrated the ability to interact with the serotonergic system in the host by regulating the development of the hypothalamus–pituitary–adrenal axis, a neuroendocrine system that controls reactions to stress( Reference Sudo, Chida and Aiba 160 ). Recent studies have demonstrated that germ-free mice display a reduction in anxiety-like behaviour( Reference Heijtza, Wang and Anuar 164 , Reference Clarke, Grenham and Scully 165 ), compared with conventionally colonised mice, possibly through an enhanced hypothalamus–pituitary–adrenal response. Another study using germ-free mice described how in the absence of gut microbiota, mice exhibited deficits in social motivation and preference for social novelty, behavioural characteristics indicative of disruptions in distinct normal social behaviours( Reference Desbonnet, Clarke and Shanahan 162 ). Probiotic intervention has proven successful for the treatment of psychiatric disorders such as anxiety( Reference Bravo, Forsythe and Chew 90 , Reference Messaoudi, Lalonde and Violle 166 ), depression( Reference Desbonnet, Garrett and Clarke 91 ) and autism( Reference Hsiao, McBride and Hsien 167 ). Bifidobacterium infantis 35 624, when administered to a maternal separation animal model of depression, exhibited antidepressant properties( Reference Desbonnet, Garrett and Clarke 91 ), Lactobacillus rhamnosus JB-1 has also demonstrated antianxiety and antidepressant properties through activation of the vagus nerve in mice, compared with broth-fed controls( Reference Bravo, Forsythe and Chew 90 ) and B. fragilis administration alleviated autistic-like behavioural impairments in communication, social behaviour, social abnormalities and restricted/repetitive behaviour in mice symptomatic of this disorder, compared with autistic, untreated controls( Reference Hsiao, McBride and Hsien 167 ). Furthermore, administration of B. breve NCIMB 702258 to mice had a significant impact on the fatty acid composition of the brain( Reference Wall, Marques and O'Sullivan 81 ). Mice that received the bacteria for 8 weeks exhibited higher concentrations of bioactive fatty acids, arachidonic acid and DHA, compared with unsupplemented controls( Reference Wall, Marques and O'Sullivan 81 ), whereby these bioactive fatty acids have a role in neurotransmission and protection against oxidative stress( Reference Henriksen, Haugholt and Lindgren 168 , Reference Yurko-Mauro, McCarthy and Rom 169 ).
A broad range of microbes, either probiotics or commensals can manufacture and secrete neurochemicals which can positively impact on mental health and thus, could be used for the treatment of CNS disorders, such as anxiety and depression. Recently defined, a psychobiotic is ‘a live micro-organism that, when ingested in adequate amounts, produces a health benefit in patients suffering from psychiatric illness’( Reference Dinan, Stanton and Cryan 170 ). Moreover, disruptions in the composition of the gut microbiota may lead to a deterioration of gastrointestinal, neuroendocrine and immune pathways, which could in turn lead to alterations in gut–brain interactions and consequently result in disease( Reference Cryan and O'Mahony 171 ). The gut microbiota produce a range of neurotransmitters and neuromodulators, bioactive metabolites which impact on host psychiatric health, only some of which have been demonstrated in vivo (Table 1).
Gamma-aminobutyric acid
γ-Aminobutyric acid (GABA) is a major inhibitory neurotransmitter of the vertebrate CNS and is the main inhibitory neurotransmitter in the brain. Dysfunctions of GABA have been linked with anxiety and depression( Reference Cryan and Kaupmann 172 , Reference Schousboe and Waagepetersen 173 ). Certain strains of Lactobacillus and Bifidobacterium secrete GABA via the same biosynthetic pathway as in neuronal tissue involving conversion of glutamate by the action of the enzyme glutamate decarboxylase and vitamin co-factor pyridoxal phosphate( Reference Komatsuzaki, Nakamura and Kimura 174 ). Furthermore, the GABA producing capability of some bacterial strains is thought to protect the organism from the acidic environment of the stomach( Reference Higuchi, Hayashi and Abe 175 ). Several human-derived lactobacilli and bifidobacteria were screened for their ability to produce GABA from monosodium glutamate, and it was found that five strains had this ability( Reference Barrett, Ross and O'Toole 176 ). Of these strains, Lactobacillus brevis and B. dentium were the most efficient GABA producers( Reference Barrett, Ross and O'Toole 176 ). Ko et al.( Reference Ko, Lin and Tsai 89 ), recently demonstrated GABA production in black soyabean milk by L. brevis FPA3709 and its administration to rats resulted in an antidepressant effect similar to that of fluoxetine, a common antidepressant drug, but without the side-effects such as appetite and weight loss( Reference Ko, Lin and Tsai 89 ). At the level of gene expression, ingestion of L. rhamnosus JB-1 altered the mRNA expression of both GABAA and GABAB, two GABA receptors which have been implicated in anxiety and depression( Reference Bravo, Forsythe and Chew 90 ).
Serotonin
Serotonin is a metabolite of the amino acid tryptophan and plays an important role in the regulation of a number of brain functions, including mood( Reference Dinan, Stanton and Cryan 170 ). The vast majority of antidepressant drugs work to increase serotonin levels in the brain and some studies have shown that bacteria can synthesise serotonin in vivo. For example, plasma serotonin levels were shown to be nearly 3-fold higher in conventional mice than in their germ-free counterparts( Reference Wikoff, Anfora and Liu 177 ). Oral ingestion of B. infantis 35 624 increased the plasma levels of tryptophan, precursor to serotonin, suggesting that commensal bacteria have the ability to influence tryptophan metabolism and could potentially act as antidepressants( Reference Desbonnet, Garrett and Clarke 91 ). This effect on tryptophan metabolism may be mediated by the impact of the microbiota on the expression of indoleamine-2,3-dioxygenase, a key enzyme in the physiologically dominant kynurenine pathway of tryptophan metabolism( Reference Forsythe, Sudo and Dinan 158 ). Early life stress induces changes in the gut microbiota and is a known risk factor for depression in adulthood( Reference O'Mahony, Marchesi and Scully 178 ). This phenomenon has been shown in rhesus monkeys, whereby prenatal stressors have been shown to alter the microbiome by reducing the overall numbers of bifidobacteria and lactobacilli( Reference Bailey and Coe 179 ).
Catecholamines and acetylcholine
Catecholamines such as dopamine and norepinephrine are the major neurotransmitters that mediate a variety of CNS functions such as motor control, cognition, memory processing, emotion and endocrine regulation. Tsavkelova et al.( Reference Tsavkelova, Botvinko and Kudrin 180 ), identified a wide range of bacteria, which produce mmol quantities of dopamine( Reference Tsavkelova, Botvinko and Kudrin 180 ) and which could be used for the treatment of Parkinson's disease, Alzheimer's disease and other major depressive disorders whereby dysfunctions in catecholamine neurotransmission are implicated. In addition, bacteria which constitute the normal gut microbiome in mice have been shown to be capable of the production of norepinephrine in vivo ( Reference Asano, Hiramoto and Nishino 181 ). Acetylcholine is a neurotransmitter found in the CNS and peripheral nervous system which plays a critical role in cognitive function, particularly in memory and learning. Previous studies have shown that acetylcholine is both a component of bacterial strains, including Lactobacillus plantarum and Bacillus subtilis ( Reference Girvin and Stevenson 182 – Reference Horiuchi, Kimura and Kato 184 ) and a microbial metabolite.
Conclusion
Pharmabiotics produced by the gut microbiota can undoubtedly influence a variety of physiological and metabolic systems/processes in the human body. At a local level they can induce changes in the gut epithelium and the enteric nervous system, while at a more systemic level processes as wide-ranging as immune function and CNS signalling may be affected( Reference Cryan and Dinan 185 ). Consequently it is not surprising that alternations in the microbial consortium are being found to be associated with a number of disease states such as IBD, diabetes, obesity, anxiety and depression. Disturbances to the delicate host–microbe relationship may disrupt development of the immune system, which may in turn result in disease. The gut microbiota have the ability to produce a variety of metabolites that exert beneficial effects on biological and neurological functions. Probiotics, prebiotics and dietary PUFA offer the potential to modulate the gut microbiota with knock-on health effects. Microbe manipulation to strengthen the host–microbe symbiotic relationship may be crucial for the future prevention of immune and psychiatric-related disorders.
Financial Support
E. P. was supported by funding from the Teagasc Walsh Fellowship Scheme (2010–2013) and the work was supported by the Science Foundation of Ireland – funded Centre for Science, Engineering and Technology, the Alimentary Pharmabiotic Centre, CSET grant 07/CE/B1368.
Conflicts of Interest
None.
Authorship
E. P. and C. S. wrote the manuscript; J. F. C., G. F. F., R. P. R. and T. G. D. made substantial contributions to the overall content of the manuscript and all authors had responsibility for the final content.
