Stress impacts greatly on virtually all aspects of gut physiology relevant to IBS including motility, visceral perception, gastrointestinal secretion and intestinal permeability while also having negative effects on the intestinal microbiota[17,143,144]. A maladaptive stress response may thus be fundamental to the initiation, persistence and severity of symptoms in IBS as well as the stress-related psychiatric comorbidities[145]. Although the findings pertaining to HPA axis irregularities in IBS are far from consistent[8,137], the well validated Trier Social Stress Test (TSST)[146] has recently been used to demonstrate a sustained HPA axis response to an acute stress in IBS, possibly indicating an inability to appropriately shutdown the stress response[147].

Accumulating evidence suggests aberrant stress responses could be mediated via the gut microbiota. A landmark study by Sudo et al[148] neatly validated this possibility by demonstrating the absence of a gut microbiota impaired control of the stress response, at least in terms of the exaggerated corticosterone production following acute stress in germ-free mice[148]. Subsequently independently replicated[105], the ability of the microbiota to modulate the stress response is also evident following probiotic administration[115], C. rodentium infection[131] and indeed following colonisation of germ-free mice[148]. Many now view the gut microbiota as an endocrine organ and as a key regulator of the stress response[18,19]. It must also be acknowledged that whilst the microbiota can modulate the stress response, stress can also affect the composition of the gut microbiota[17]. Thus, stress induced changes in the microbiota may precede any subsequent GI and central disturbances in IBS.

Mechanisms

When considering the preclinical evidence reviewed above, and preliminary evidence from healthy humans[142] it appears that the perturbations in composition of the gut microbiota may be considered as a primary factor in driving changes in central function in IBS. However, as IBS is a stress related disorder, the preclinical evidence indicating that chronic stress can alter the gut microbiota must also be borne in mind. As noted, stress and the gut microbiota have been shown to interact in a complex manner to influence brain function, at least in rodents[131] and it will be important to delineate this interaction in IBS. Nevertheless, when considering the gut microbiota as the primary factor in driving changes in central functioning, a number of potential mechanisms have been considered, with varying degrees of evidence supporting both humoral and neural lines of communication to the level of the CNS as well as more localised effects from compositional alterations.

Tryptophan, an essential amino acid and precursor for the neurotransmitter serotonin (5-HT), in particular has received much attention (Figure ). 5-HT is a key signalling molecule in the brain-gut axis, both in the enteric nervous system[149] and the CNS[150]. The information gleaned from studies in germ-free animals suggests that the peripheral availability of tryptophan, which is critical for CNS 5-HT synthesis, is coordinated by the gut microbiota[105]. Plasma tryptophan concentrations can be normalised following colonisation of germ-free animals[105] and can also be augmented following administration of the probiotic B. infantis[116]. How the bacteria in our gut regulate circulating tryptophan concentrations is unclear but may involve controlling the degradation of tryptophan along an alternative and physiologically dominant metabolic route, the kynurenine pathway[151,152]. The enzymes responsible for the initial metabolic step in this pathway, indoleamine-2,3-dioxygenase (IDO) and tryptophan-2,3-dioxygenase (TDO), are immune and glucocorticoid responsive respectively and the decreased ratio of kynurenine to tryptophan (an index of IDO/TDO activity) in germ-free animals implicates this pathway in the reported alterations (Figure )[105]. Moreover, an increased ratio is observed following infection with Trichuris Muris, likely due to increased IDO activity following the associated chronic gastrointestinal inflammation[153].

The relevance of these preclinical findings to IBS is well reflected in the clinical literature which has demonstrated increased IDO activity in both male and female IBS populations[154-156]. Interestingly, TLR receptors, which have altered expression and activity in both clinical IBS populations[157,158] and animal models of the disorder[159], might drive the low grade inflammation in IBS and mediate the immune consequences of the misfiring engagement between the microbiota and the host in IBS. In this context, it is interesting to note that once TLR receptors are engaged by their cognate ligands, degradation of tryptophan can ensue in general[155,160,161] and there appears to be a differential TLR-specific pattern of kynurenine production in IBS[155].

There are also other potential explanations for the alterations in tryptophan supply due to microbiota alterations and in addition to the growth requirements for bacteria[162], a bacteria-specific tryptophanase enzyme also recruits tryptophan for indole production[163,164]. One such bacteria, Bacteroides fragilis, harbours this enzyme and has recently been linked to gastrointestinal abnormalities in autism spectrum disorders[165]. Of further interest and adding to the complexity of the narrative is that, in contrast to eukaryotes, bacteria retain a capacity for tryptophan biosynthesis via enzymes such as tryptophan synthase[166,167]. It seems a curious quirk of the evolutionary process that we have lost the capacity for endogenous tryptophan synthesis, given the pivotal nature of this amino acid not alone as a precursor to serotonin, which itself has an expansive physiological repertoire[168], but also the other metabolic pathways it serves[150,151].

The production of serotonin from tryptophan, at least in-vitro, is also possible in some bacterial strains[169-171]. Harnessing this knowledge to specifically target the 5-HT receptors and receptor subtypes expressed in the gut of most relevance to IBS[172-175] or indeed alternative receptors activated by kynurenine pathway metabolites that interact with gastrointestinal functions[176] presents an interesting challenge. Similarly, whether we can accurately “titer” the gut microbiota to deliver precise circulating or regional tryptophan concentrations is an intriguing possibility but one beyond our current capabilities.

Of course, immune system mediators and glucocorticoids can impact both locally in the gut and at the level of the CNS independently of their effects on tryptophan metabolism and represent viable alternative routes through which the gut microbiota can modulate gut-brain axis signalling and influence IBS symptoms[14,19,104,177,178]. In addition, the more general concept of a “leaky gut” has been proposed to explain the common feature of a low-grade circulating inflammation in both IBS itself and depression, which, as outlined above, is a prominent psychiatric comorbidity in IBS[179-182]. This model relies on the presence of increased intestinal permeability in IBS which allows the gut microbiota to drive the reported proinflammatory state and influence the CNS via the ensuing elevations in circulating cytokines[104] as well as visceral hypersensitivity via local gut mechanisms[183]. There is certainly accumulating evidence to support the hypothesis of altered intestinal permeability, a compromised integrity of the intestinal epithelial barrier and related tight junction disturbances in IBS, if not in depression[183-187].

Defects of the intestinal epithelial barrier may also play a significant role in cognitive dysfunction in IBS. The maternal immune activation (MIA) mouse model produces epithelial barrier defects, changes in the gut microbiota, and associated cognitive and behavioural features of neurodevelopmental disorders in rodents[188]. A recent study has provided strong evidence that maternal infection in the MIA model drives changes in the gut microbiota in the offspring, which subsequently leads to the cognitive and behavioural alterations in this model. Treatment with B. fragillis in MIA offspring restored gut barrier integrity and alleviated some of the cognitive and behavioural defects displayed by these animals[189]. Importantly, restoration of gut barrier integrity in MIA offspring appeared to stop a number of neuroactive metabolites being released systemically to reach the CNS and affect behavioural and cognitive function[189]. Thus, when extrapolated to IBS, epithelial barrier dysfunction may lead to the release of numerous metabolites that could impact centrally and impair cognitive performance. Of note, some probiotic strains have shown efficacy in repairing epithelial barrier function[190] in preclinical models which may also explain the efficacy in treating some GI symptoms in IBS[57]. If probiotics also prove beneficial in alleviating central disturbances in IBS, this may potentially be via restoration of epithelial barrier integrity leading to the reduction of harmful neuroactive metabolites being released from the gut and impacting centrally.

The gut microbiome can also be considered a metabolic organ[191,192] and the array of microbial metabolites produced can impact greatly on GI health and the gut-brain axis scaffolding. Interestingly, dietary restriction of fermentable carbohydrates (fermentable oligosaccharides, disaccharides, monosaccharides and polyols: the low FODMAP diet) has received much attention for the management of symptoms in IBS[193,194]. Although microbial metabolism of carbohydrates, proteins and amino acids by human gut bacteria generates a variety of compounds[195], short chain fatty acids (SCFAs) may be of particular importance in the context of microbiome-gut-brain axis signalling. For example, these organic acids are altered in IBS and may be related to symptoms[52,196]. Preclinically, administration of sodium butyrate increases visceral sensitivity in rats[197]. Interestingly, it has recently been demonstrated that butyrate can regulate intestinal macrophage function via histone deacetylase inhibition[198] which is in line with the proposed epigenetic mechanism of gut-brain axis dysfunction[199,200]. Butyrate can also mediate its immunomodulatory effects via G-protein coupled receptors[201] or indirectly via TLRs[202].

Receptors and transporters for SCFAs are expressed in the gastrointestinal tract and appear to be of relevance to gastrointestinal function[203-208]. For example, SCFAs may modulate both 5-HT secretion[18] and peptide YY release, an important neuropeptide at multiple levels of the gut-brain axis[209]. Thus, there is patently a role for these microbial metabolites beyond the regulation of energy homeostasis[210]. Interestingly, intraventricular administration of propionic acid in rats induces a variety of behavioural alterations although it is unclear if this occurs via similar mechanisms to the periphery[211]. It is worth noting that G protein-coupled receptor (GPR) 41, a receptor activated by propionic acid, is highly expressed in rat brain tissue[212]. Although we know that fibre metabolized by the gut microbiota can increase the concentration of circulating SCFAs[213], it remains to be established if this is reflected at physiologically relevant concentrations in the CNS.

The gut microbiota can also engage neural mechanisms to influence brain-gut axis signalling. In particular, many of the behavioural effects of specific probiotic strains are abolished in vagotomized animals[113,115]. Germ-free studies have confirmed that the presence of intestinal bacteria is also essential for normal postnatal development of the ENS[214] and for normal gut intrinsic primary afferent neuron excitability in the mouse[215]. Thus, there is direct evidence of bacterial communication to the enteric nervous system while as indicated above, the microbiota is also a potential source of relevant ENS neurotransmitters including serotonin and GABA[216-218]. Interestingly, colorectal distension induces specific of patterns of prefrontal cortex activation in the viscerally hypersensitive maternal separation model of IBS, in which microbiota alterations are also manifested[219]. Taken together, it seems likely that the gut microbiota can modulate both the physiological information flow to the CNS via vagal afferents and the noxious information that is encoded by spinal afferents[10,220].