Irritable bowel syndrome and the gut microbiota

ABSTRACT

Irritable Bowel Syndrome (IBS) is a common, chronic and distressing functional gastrointestinal disorder (FGID) which is currently considered a dysregulation of the brain-gut axis (BGA). Multiple biological systems have been implicated in the pathophysiology of IBS, however there is currently no reliable biomarkers for this disorder and no cure. Observations that the gut microbiota of IBS patients differs from healthy controls have been made for decades as well as the role of the gut microbiota in the pathophysiology of multiple disease. The use of probiotics, antibiotics and elimination diets have been proven to be efficacious in the treatment of IBS symptoms. A subset of IBS patients develop chronic IBS symptoms after contraction of infectious gastroenteritis or after antibiotic use. Furthermore, the gut microbiota are known to modulate the BGA. We review here the history and current state of knowledge of the role that the gut microbiota plays in the pathophysiology of IBS.

KEYWORDS:

• Irritable bowel syndrome
• microbiota
• brain gut axis
• diet
• antibiotics
• probiotics

Introduction

IBS is a common, chronic, and distressing FGID characterised by pain and a change in bowel habit in the absence of an obvious physical cause (Tack and Drossman 2017). Diagnosis is based on the presentation of these symptoms according to the validated Rome IV criteria (Drossman 2016). Based on patients’ predominant bowel habit, IBS is subtyped into diarrhoea (IBS-D), constipation (IBS-C), or a mixed phenotype (IBS-M) (Tack and Drossman 2017). Post infectious (Pi-) IBS, in which patients develop ongoing IBS symptoms after contracting an infectious gastroenteritis is also considered an IBS subtype (Spiller and Lam 2012). The global prevalence of IBS is estimated to be around 11% (Lovell and Ford 2012; Sperber et al. 2017) and constitutes a major health problem in the western world due to significant direct and indirect health care costs and a decreased quality of life (Buono et al. 2017). However, due to its multifactorial pathophysiology there is neither consensus concerning the aetiology nor reliable biological markers for IBS diagnosis. Patients’ responses to treatment are also variable (Mujagic et al. 2017). There is evidence of multiple biological system dysfunction in IBS patients compared to healthy controls, but current understanding of this disorder is that symptoms arise from dysregulation of the BGA (Quigley 2018). The BGA describes the bidirectional communication between the cognitive and emotional areas of the central nervous system (CNS) and the peripheral functions of the gastrointestinal tract (GIT). This is mediated by neurons, hormones, neurotransmitters, and the immune system (Bonaz et al. 2018; Florens et al. 2019; Parker et al. 2019) as well as the gut microbiota (Pusceddu et al. 2018). The GIT contains more than 100 different bacterial species which perform different functions in the gut, including developing and maintaining the gut epithelial barrier, protecting the host from pathogen colonisation, modulation of the immune and nervous systems, and digestion of dietary components (Sánchez et al. 2017). There is significant inter-individual variation of the gut microbiota, which is governed by the environment and host genetics (Rothschild et al. 2018; Johnson et al. 2019), as well as differences between healthy individuals and those with disease (Zhuang et al. 2017). Recently, the functions of the microbiota in healthy individuals have been shown to be compromised in IBS patients (Sundin et al. 2017).

IBS and the brain gut axis

Discovering IBS

After the First and Second World Wars, there were numerous reports of soldiers who had contracted infectious dysentery and then experienced ongoing gastrointestinal (GI) symptoms which could remit and recur over weeks and months. These symptoms were not due to strictures, or other organic causes, and there was no evidence of ongoing infection. Today, these soldiers would be described as likely experiencing Pi-IBS. Interestingly, the physicians of the day also recorded the personalities and mental states of their patients who exhibited these symptoms. It was noted that the majority of IBS patients were anxious and depressed (Osler and McCrae 1892; Ryle 1928; Jordan and Kieder 1929; White and Jones 1940; Hurst and Barber 1943; Alvarez 1949; Bargen 1950; Stewart 1950; Whorton 2000; Drossman 2016) and it was theorised at the time that symptoms in patients may be a physical reaction to these emotions. This was subsequently reinforced by a prospective study in 1996 that found patients hospitalised for enteric infection who scored higher on psychometric tests for anxiety, depression, somatisation and neuroses, were more likely to go on to develop Pi-IBS (Gwee et al. 1996). Today anxiety and depression are the most commonly reported non-GI co-morbidities of IBS (White et al. 2010), and their association with worse GI and non-GI symptoms in IBS as well as worse quality of life (Midenfjord et al. 2019) may, in part, reflect altered processing within the CNS of IBS symptoms (Elsenbruch et al. 2010).

The BGA: symptoms common to IBS patients

Visceral hypersensitivity

Visceral hypersensitivity, or an abnormal pain response to a normal stimulus, was elegantly demonstrated in IBS patients in 1973, when James Ritchie established that inflating a balloon in the colon resulted in pain in a disproportionate number of IBS patients when compared to healthy controls (Ritchie 1973). This finding is suggestive of a significantly lower pain threshold in this group (Whitehead et al. 1990). In the absence of a detectable biological cause it was postulated that the altered bowel function, disturbed motility, secretion, and absorption in IBS patients (Baird 1946, p. 503) could all potentially be explained by changes in the autonomic nervous system (ANS) brought on by an over-exaggerated stress response mediated through the HPA axis (Selye 1950). However, whereas IBS patients were shown to exhibit abnormal GI motility in response to stressful situations (Almy 1951), animal studies found no association between the adrenal and/or the pituitary gland, and visceral hypersensitivity in response to stress (Williams et al. 1988). Collectively, these findings suggested that while the HPA axis is not involved, abnormal communication between the brain and periphery in IBS patients was likely to play a role in triggering of IBS symptoms.

Current research suggests that decreased pain thresholds in IBS patients likely contributes to visceral hypersensitivity but the underlying mechanism within the CNS remains unknown. Intriguingly, functional magnetic resonance imaging (fMRI) and positron emission tomography (PET) brain scans of IBS patients identify functional alterations in various regions of the CNS that include the thalamus, hippocampus, and prefrontal cortex (Silverman et al. 1997; Mertz et al. 2000). These are areas that are important in the perception of sensory and motor signals, the relay of these signals between the CNS and the periphery, as well as memory, emotion and behaviour (Mertz et al. 2000; Naliboff et al. 2001; Ma et al. 2015). Functional changes to the way that these areas of the brain communicate are thought to be important in the pathophysiology of IBS. These functional changes potentially include an inability to effectively modulate sensory information from being brought to the brain or increased ‘attention’ being paid to ‘normal’ sensations in the GIT. Both of these changes could be the root of visceral hypersensitivity, where ‘normal’ sensations are relayed to the brain as painful sensations (Icenhour et al. 2017).

Corticotrophin Releasing Hormone (CRH), a vital mediator of the HPA axis and the stress response, is also thought to have a role in mediating IBS symptoms in the gut via modification of the inflammasome. The inflammasome is a component of the immune system that responds to the presence of pathogens via inflammatory responses, resulting in visceral hypersensitivity and increased GI barrier permeability (Ren et al. 2017). An increase in barrier permeability that exposes the enteric immune system (ENS) to luminal antigens then has the potential to perpetuate a positive feedback loop of inflammation, visceral pain, and altered colonic motility.

Low-grade inflammation potentially underpins visceral hypersensitivity in IBS patients via the increased number of histamine-producing mast cells (MCs) in the myenteric plexus of patients with IBS (Figure 1). There is evidence to suggest that visceral pain positively correlates with the physical closeness of MCs to enteric neurons (Törnblom et al. 2002; Wouters et al. 2016). However, while inflammation potentially has a role in triggering visceral hypersensitivity, it is not necessarily involved in its maintenance (Balemans et al. 2017). Low-grade inflammation in the gut is likely to be linked to the gut microbiota and this is emphasised by depolarisation of enteric neurons following the application of bacterial supernatants from enteric mucosal biopsies (Balemans et al. 2017). Colonic biopsy supernatant from patients with IBS-D (but not IBS-C) are also shown to cause neuronal excitement via the protease signalling pathway, PAR2 (Valdez-Morales et al. 2013). This demonstrates that while both IBS-D and IBS-C patients experience abdominal pain by Rome IV definition, the mechanisms behind visceral hypersensitivity are likely heterogeneous in nature, despite being part of the communication between the periphery of the gut and the pain centres of the brain. More recently a genetic predisposition for visceral hypersensitivity has been demonstrated in Pi-IBS patients due to a polymorphism which produces increased expression of ion channels known to be involved in the transmission of nociceptive signals to the CNS (Lashermes et al. 2018). Intriguingly these are the same channels that become sensitised during inflammation (Burnstock 2016).

Figure 1. Anatomy of the gastrointestinal tract; the myenteric plexus (6), a part of the enteric nervous system, sits between the two layers of smooth muscle that form the muscularis (5). The gut microbiota live in and on the mucous layer (2) that covers the mucosa (3) as well as within the lumen of the gastrointestinal tract (1). The gut microbiota affect and are effected by the enteric, autonomic and central nervous systems as well as being effected by the life style of the host through a variety of mechanisms. (A) The position of the small and large intestines in the human body; (B) an enlarged cross section of the small intestine; (C) a magnification of B. (1) Gut microbiota in the GI lumen; (2) mucous layer; (3) mucosa; (4) submucosa; (5) muscularis; (6) myenteric plexus.

Altered colonic transit

The other defining symptom common to all IBS patients is altered colonic transit. Originally, it was thought that IBS patients exhibit abnormal colonic motility that, in turn, translated to both visceral pain and altered transit (Almy 1957), and early studies suggested that IBS was primarily a disorder of the motility of the gut (Snape Jr et al. 1977; Taylor et al. 1978). However, altered GI motility was also shown to correspond poorly with visceral pain (Mayer and Helen 1990). Current evidence, however, shows that while altered transit is observed in IBS patients (i.e. altered bowel habit), only a subset of IBS patients’ exhibit abnormal colonic motility, and that there is no overriding pattern that defines these patients (Simrén et al. 2000; Agrawal et al. 2009). While the exact mechanisms by which symptoms correlate to colonic motility remain unknown, patients who exhibit this abnormal colonic motility (particularly IBS-C patients), tend to respond well to pharmacological interventions that stimulate gut motility (Caldarella et al. 2002; Malagelada et al. 2017).

GI hormones and neurotransmitters

Abnormalities in the metabolism of gut hormones and neurotransmitters observed in IBS patients could be a common mechanism for both altered transit and visceral hypersensitivity. A significantly lower density of enteroendocrine cells (ECs) have been demonstrated in IBS patients (Gunawardene et al. 2011; El-Salhy 2012; El-Salhy and Gilja 2017). ECs produce and secrete GI hormones and neurotransmitters in response to the luminal environment as well as signals from the CNS, ANS and ENS (Mawe et al. 2006; Camilleri 2012; El-Salhy 2012; El-Salhy et al. 2012). ECs also mediate the communication between the ENS, ANS and CNS and are therefore involved in both the modulation of GI sensation and motility, this has been reviewed in detail elsewhere (Wade et al. 1996; May and Kaestner 2010; Gunawardene et al. 2011; El-Salhy 2012; El-Salhy et al. 2012; Gershon 2013). One of the predominant neurotransmitters produced by ECs is serotonin. Serotonin has a vital role in the regulation of colonic motility as well as secretion and sensation, and is a vital neurotransmitter within and between the CNS and the ENS (Sikander et al. 2009). Abnormal metabolism of serotonin and its precursor, 5-hydroxytryptophan, have been associated with IBS pathophysiology, as well as the different IBS subtypes, for a number of years (Pata et al. 2002; Dunlop et al. 2005). Interestingly a recent study conducted in Norway has found that modulating the gut microbiota of IBS patients through various different methods increased EC cell density towards the level of healthy controls, which was associated with improvements to symptoms and quality of life (Mazzawi et al. 2013, 2016).

All IBS patients, by Rome IV definition, exhibit altered transit and visceral hypersensitivity. Therefore pathophysiological heterogeneity makes IBS particularly hard to study. Even the most stringent study inclusion criteria, if based on clinical symptoms alone, may still result in a heterogeneous pool of IBS participants. Given this heterogeneity it is not surprising that current guidelines describe IBS pathophysiology as dysregulation of the BGA, an ‘umbrella’ term under which all of the symptoms and mechanisms of IBS can sit. It is interesting to note however that apart from its direct roles in the pathophysiology of IBS, the gut microbiota has the ability to regulate the BGA from the CNS to the gut epithelial barrier.

Gut microbiota and IBS

The gut microbiota is believed to be one of the main contributors to the aetiology of IBS. However, past and current research has been unable to determine exactly what components of the microbiota play a part in IBS development. An early study found that IBS patients had a greater proportion of facultative anaerobes cultured from stool samples compared to healthy controls (Bayliss et al. 1984). More recently, a systematic review and meta-analysis of IBS studies using culturing techniques found consistent reductions in Bifidobacterium and Lactobacillus in IBS patients compared to healthy controls (Zhuang et al. 2017). However, anaerobic culturing techniques are no longer considered appropriate for studying the complex ecology of the gut microbiota. Instead genetic sequencing techniques are now widely used, particularly 16S rRNA sequencing, which allows the investigation of complex microbial communities by amplifying the evolutionarily conserved 16S rRNA gene (Larsen et al. 1993). This sequencing technique is able to identify bacterial colonies that grow poorly in culture as well as detect species of bacteria previously uncultured. Subsequent analyses comparing culture and sequencing techniques found that 60%–80% of bacteria present in human stool are unculturable and only 24% of the sequencing results identify species that corresponded to previously described organisms (Suau et al. 1999). Despite these advances studies investigating differences in the gut microbiota in IBS populations are still contradictory.

A recent systematic review collated 16S rRNA studies and found that there are consistent detectable alterations in the gut microbiota of IBS patients, which may have a significant association with IBS pathophysiology. In particular the study identified a reduction in the gut microbiota α diversity and a shift in the Firmicutes:Bacteroidetes ratio. The Firmicutes and Bacteroidetes are the two anaerobic bacterial phyla in that dominate the human gut, with IBS patients demonstrating a higher abundance of Firmicutes and lower abundance of Bacteroidetes compared to healthy controls (Duan et al. 2019). While this finding has been consistent among different IBS populations, it is unknown how it is related to IBS pathophysiology and aetiology. It has been speculated that increases in Firmicutes could lead into increased gut barrier permeability and acid production (driven by Clostridia species) which could lead to low-grade inflammation and visceral hypersensitivity (Ringel-Kulka et al. 2015; Duan et al. 2019). In contrast, a lower abundance of Bacteroidetes could lead to increased inflammation through decreases in bacterial metabolite production of anti-inflammatory compounds, such as the short chain fatty acid (SCFA) butyrate (Pozuelo et al. 2015; Borycka-Kiciak et al. 2017).

There is also contradictory evidence for differences in the composition of the gut microbiota between different IBS subtypes. Some studies have associated IBS-C, not IBS-D, with an increased rate of breath methane production (Ghoshal et al. 2016). Breath methane has been suggested as a potential diagnostic test for IBS-C (Hwang et al. 2010), however methane is also detectable in healthy people and is not produced by all IBS-C patients. Methane has been shown to delay intestinal transit, and while a causative role has not yet been established it is believed that methane may act as a neuromuscular transmitter (Triantafyllou et al. 2014). There has been no consensus on a microbial signature for IBS-D, although several studies have observed decreases in the relative abundance of Lactobacillus species and Faecalibacterium prausnitzii (both generally considered ‘good’ bacteria) and an increase in the Enterobacteriaceae family (generally considered pathogenic) (Carroll et al. 2011, 2012; Liu et al. 2017). This has led to speculation that a disruption in the ratio of ‘good’ and ‘bad’ bacteria in IBS-D could contribute to the pathophysiology of IBS. However there is currently no mechanistic evidence for this, nor is this disruption considered sufficient in and of itself to explain the symptoms and altered physiology of IBS-D (Maharshak et al. 2018).

The ability of commensal bacteria to alter the expression of genes involved in the construction of the gut epithelial barrier is another potential mechanism of IBS pathophysiology (Hooper et al. 2001; Ma et al. 2004; Jacob et al. 2005). IBS patients have significantly lower mRNA expression of tight junction proteins involved in the construction of the gut epithelial barrier when compared to healthy controls (Piche et al. 2009). The resulting increased exposure of neuromuscular junctions to bacterial antigens leads to induction of an immune response (Collins 2002) which produces low grade inflammation and CNS mediated visceral hypersensitivity (O’Mahony et al. 2005; Zhou et al. 2009).

Gut microbiota and the BGA

Evidence for microbial involvement in the BGA dates back to the 1970s when anatomical and immune system abnormalities of the GIT were noted in germ-free mouse models (Gustafsson and Maunsbach 1971; Wannemuehler et al. 1982), suggesting that the gut microbiota was directly affecting the tissues it came in close contact with. Altered GI motility was also noted in germ-free animal models which was ‘normalised’ by the introduction of conventional gut microbiota (Caenepeel et al. 1989; Husebye et al. 2001). Subsequently bacterial components, such as Flagellin, have been shown to induce an immune system reaction as well as ANS mediated altered motility (Vrees et al. 2002; Rhee et al. 2004). Another notable anatomical difference found in germ-free animal models in is the significantly increased EC density (Uribe et al. 1994), which could potentially explain the differences in GI motility observed in these models.

The far-reaching effects of the gut microbiota have been demonstrated on the CNS and on the pathogenesis of depression and anxiety. Morphological differences in regions of the brain that control fear, anxiety and stress have been demonstrated in germ-free animal models, as well as reduced anxiety-like behaviour highlighting the vital role that the microbiota play in the development of these structures(Heijtz et al. 2011; Luczynski et al. 2016). Furthermore, GI infections have been demonstrated to induce anxiety-like behaviour in mouse models; activation of brain regions involved with mood, fear, and anxiety have been noted in these animals (Lyte et al. 2006; Goehler et al. 2007). Modulation of the gut microbiota by probiotics (Bravo et al. 2011), diet (Ledochowski et al. 2000) and antibiotics (Bercik et al. 2011) have all been shown to modulate anxiety- and depressive-like behaviours in mouse models. In human trials, participants reported reduced depression symptoms when taking probiotics which correlated decreased amygdala and limbic system activation on fMRI. This suggests that a probiotic organism could modulate the CNS and the HPA axis in humans (Pinto-Sanchez et al. 2017). Furthermore some components of the inflammasome complex, which as mentioned previously, mediates immune responses to the presence of pathogens as well as being a key mediator of physical and psychological stress, have recently been shown to be altered in patients with depression. When caspase-1, a mediator of the inflammasome, was pharmacologically inhibited in mice, decreased exhibition of depressive behaviour was observed. Increases in gut bacteria known to ameliorate inflammation were also noted in these animals (Wong et al. 2016).

Diet

Diet plays an important role in the generation of symptoms in IBS patients, around 70% of IBS patients report that their symptoms are triggered by food (Rej et al. 2018). Abnormal small intestinal motility after food ingestion has been observed in a subset of IBS patients however the mechanism of this is unknown (Manousos et al. 1967; Deiteren et al. 2010). Early studies implicated reductions in fibre intake to IBS symptoms (Thompson 1974). This observation was supported by studies that reported the beneficial effects of wheat bran expediting GI transit (Kirwan and Smith 1977). However, IBS patients often report an intolerance to fibre containing foods (Floch 2005; Ford et al. 2008; Heizer et al. 2009) and insoluble fibre may worsen IBS symptoms in some patients (Bijkerk et al. 2004). Levels of butyrate (an end product of fibre fermentation in the gut) are reportedly decreased in IBS patients (Farup et al. 2016), which could reflect low fibre intake and/or fewer butyrate-producing bacteria in IBS patients (Duan et al. 2019). Reductions in butyrate production are thought to effect GI motility and sensation as well as increase gut barrier permeability (Pryde et al. 2002; Soret et al. 2010; Wang et al. 2012). Non-coeliac gluten sensitivity has been suggested as a driver of symptoms in some patients (Barmeyer et al. 2017). However, recent work has elegantly demonstrated that many IBS patients are intolerant to fructans, a member of a group of carbohydrates called FODMAPs (Fermentable Oligosaccharide, Disaccharide, Monosaccharide, and Polyols) (Skodje et al. 2018). Significantly, fructans are often present in gluten containing food.

Food intolerances have been studied intensely in terms of IBS, which are generally produced by Immunoglobulin E (IgE) mediated immune reactions, but studies of IgE in IBS patients are inconclusive (Lessof et al. 1980; Jun et al. 2006; Choung and Murray 2019). Instead, what has been shown is that the direct application of food antigens to the gut mucosa results in increased intra-epithelial lymphocyte (IEL) recruitment and loss of epithelium integrity (Fritscher-Ravens et al. 2014) that enables luminal bacterial antigens to enter the systemic circulation and induce a low grade inflammatory response (through MC and IEL recruitment and activation), impacting visceral hypersensitivity and GI motility.

The idea that food intolerances are an important trigger in the pathogenesis of IBS has stemmed from the efficacy of elimination diets in controlling IBS symptoms (Ong et al. 2010). However, despite elimination diet studies consistently demonstrating a reduction in GI symptoms, food intolerance as a trigger for IBS is still controversial in part because of small study numbers, not taking into account IBS subtype and/or inappropriate controls. Moreover, when participants are blinded and challenged with their ‘trigger’ foods, a significant placebo effect is demonstrated (Farah et al. 1985). This may, in part, reflect the inter-individual differences in gut microbiota that, in turn, reflect how people might respond differently to potential dietary triggers. IBS patients have consistently demonstrated significantly higher breath hydrogen and methane concentrations, a proxy of gut microbiota fermentation, which is reduced after elimination diets (King et al. 1998). It was hypothesised that changes in gas production rates were caused by altered bacterial fermentation in the gut but that other factors associated with bacterial fermentation, for example local effects of SCFAs or amines may drive GI symptoms such as altered motility (King et al. 1998).

FODMAPs

Over the last 15 years there has been increasing evidence concerning the efficacy of elimination diets for the treatment of IBS symptoms. In particular the elimination of FODMAPs has shown consistent alleviation of IBS symptoms (Nanayakkara et al. 2016). FODMAPs are thought to induce GI symptoms through two central mechanisms. Firstly, through osmotic actions in the small intestine (Madsen et al. 2006; Barrett et al. 2010) and secondly through rapid fermentation by the gut microbiota in the proximal colon (Major et al. 2017). This causes increased water content and gas production and, therefore, changes in colonic transit and pain in patients with visceral hypersensitivity. A high FODMAP diet has been shown to increase gut epithelial barrier permeability in animal models, potentially through increased abundance of inflammation causing bacterial antigens. Decreases in the abundance of these bacteria have been observed in humans undertaking low FODMAP diets (Zhou et al. 2018).

However the restriction of FODMAPs in the diet has ramifications on the gut microbiota. Breath hydrogen excretion has been demonstrated to be significantly reduced after undertaking the low FODMAP diet indicating a shift in gut microbiota fermentation patterns (Staudacher et al. 2012). One study has shown that a 50% reduction in FODMAP ingestion led to a 6 fold reduction in the relative abundance of Bifidobacteria compared to controls (Staudacher et al. 2012) and other studies showed a reduction in Bifidobacteria, Faecalibacterium prausnitzii and Clostridium cluster IV during the low FODMAP diet (Staudacher et al. 2012, 2016). Most low FODMAP diet studies are for less than six months duration and the long term consequences of this disruption to the composition and potentially the metabolism to gut microbiota are unknown (Staudacher and Whelan 2017).

Antibiotics

Antibiotic use has long been associated with short-term GI symptoms that may become chronic, despite the cessation of treatment (Goldin and Gorbach 1984). Clostridium difficile infection is most commonly associated with antibiotic use and has become a major cause for morbidity and mortality in hospitalised patients (Ambrose et al. 1985). These observations support to the idea that antibiotic-mediated disruption of the normal gut microbiota has the potential to facilitate the growth of opportunistic pathogenic bacteria in at least a subset of patients with IBS. This is illustrated by the increased incidence of IBS in women given (Balsari et al. 1982) peri-operative metronidazole before undergoing hysterectomy (Alun Jones et al. 1984), and in Italians (Mendall and Kumar 1998) who received antibiotics for an outbreak of Salmonella enteritis (Barbara et al. 2000). A recent study confirmed the association between antibiotic use and IBS incidence in a Danish population (Krogsgaard et al. 2018). Mechanistic information is still lacking but animal studies suggest transient alterations in the gut microbiota triggered by the antibiotics are associated with altered GI motility (Corinaldesi et al. 1999; Barbara et al. 2005), increased inflammatory activity, and increased release of Substance P, a neurotransmitter associated with increased visceral hypersensitivity to colonic distension (Verdu et al. 2006).

Intriguingly, there have been consistent observations that antibiotic usage can lead to clinical improvement in some IBS patients, suggesting antibiotics have potential as a therapeutic as opposed to a trigger (Afdhal et al. 1986). This may reflect an effect of antibiotics on small intestine bacterial overgrowth (SIBO) and associated excessive colonic carbohydrate fermentation (Pimentel et al. 2000). Rifaximin, a nonabsorbed oral antibiotic was found to be associated with a significant reduction in IBS symptoms compared to placebo without the systemic effects produced by other antibiotics (Pimentel et al. 2006). Whilst SIBO is not considered a primary pathophysiological cause of IBS, the consistent reduction of symptoms associated with antibiotic treatment supports a role for the gut microbiota in IBS symptom generation in at least a subset of patients.

Probiotics

Probiotics also offer the ability to modulate IBS symptoms through their capacity to alter the gut microbiota. Defined as ‘live microorganisms which when administered in adequate amounts confer a health benefit on the host’ (Hotel and Cordoba 2001), probiotics have been used to modulate GI health for at least 100 years, predating knowledge that gut bacteria are influenced by diet. In 1907 Elie Metchnikoff advocated the health benefits of adding specific bacterial cultures to the diet (Metchnikoff 1907). He hypothesised that fermented food decreased the amount of ‘putrefactive’ proteolytic bacteria in the gut, and increased ‘good’ saccharolytic bacteria. Metchnikoff identified a bacterium (now known as Lactobacillus acidophilus) in his early experiments that reduced colonic pH as well as the number of bacteria excreted in the stool, both considered at the time to be beneficial for the health. Rettger and Cheplin went on to show that it was possible to change the faecal microbiota by ingesting L. acidophilus cultures, although their study did not record GI symptom improvement (Rettger and Cheplin 1921). The first use of probiotics for the treatment of IBS was reported in the early 1960s (Beck and Necheles 1961), although it has taken the advent of sequencing technologies to more fully appreciate how oral administration of probiotic strains exert their positive effect in the gut. There is evidence to suggest that IBS patients (sorted by constipation or diarrhoea predominant subtype) are characterised by a relative reduction in Lactobacillus and Bifidobacterium species compared to healthy controls, both of which are common probiotic species (Malinen et al. 2005; Kerckhoffs et al. 2009). Early trials of probiotics in IBS found that abdominal pain could be modulated by oral administration of Lactobacillus species (Niedzielin et al. 2001; Sinn et al. 2008). The mechanism of pain reduction was postulated to be due to Lactobacillus induced expression of endogenous opioid and cannabinoid receptors in the GIT (Rousseaux et al. 2007). Further trials of probiotics containing Bifidobacterium infantis led to global improvements in IBS symptom scores (O’Mahony et al. 2005), with increased and reduced plasma concentrations of anti-inflammatory (IL-10) and pro-inflammatory (IL-12) cytokines, respectively, suggesting the immunomodulatory effect of some probiotics. However, other studies found no differences in cytokine levels between IBS patients and healthy controls after probiotic ingestion (O’Sullivan and O’Morain 2000; Niv et al. 2005). Systematic reviews at the time confirmed significant heterogeneity between studies and evidence of publication bias; but the overall consensus was that probiotics are likely beneficial in the treatment of IBS symptoms (Saggioro 2004). A finding reinforced by more recent evidence that certain probiotic organisms, singly or in combination, are associated with reduction of global symptom severity in specific subsets of IBS patients (Distrutti et al. 2016; Ford et al. 2018; Principi et al. 2018). It is not yet possible to conclusively show that probiotics are efficacious in the treatment of IBS (Ford et al. 2018), or whether different probiotics would be more beneficial in different IBS subtypes (Principi et al. 2018; Shapiro et al. 2019). However, there is evidence that profiling volatile organic compounds in the stool of IBS patients shows high predictive value for probiotic efficacy which may be an initial step in resolving these questions (Rossi et al. 2018).

Faecal microbial transfer

Faecal microbial transfer (FMT) is emerging as a promising means of modulating the gut microbiota in IBS patients, where faecal microbial content of a ‘healthy’, non-symptomatic individual is given to a diseased individual either orally, via colonoscopy or enema. The exact mechanism of action of FMT is not known; putative mechanisms include direct effects such as nutrient competition with the recipients’ gut microbiota and/or the production of bacteriocins and other anti-microbials, as well as indirect effects that include SCFA production, bile acid metabolism and/or maintenance of the GI epithelial barrier (De Palma et al. 2017). However, while clinical trials of FMT in IBS patients have been performed since the 1980s (Borody et al. 1989), inter-study heterogeneity with regards to methodology, screening of faecal microbial donors, and the subtype of IBS patients as well as a lack of long term follow-up have led to inconsistent findings (Aguilar et al. 2015; Mazzawi et al. 2016; Mizuno et al. 2017). Consequently, there is currently no clear understanding of the long-term efficacy of FMT in the treatment of IBS.

A well-designed multicentre, double blind, randomised controlled trial who had carefully selected IBS-D patients reported (Holvoet et al. 2018) lower symptom severity after 24 weeks in patients given a FMT compared to those treated with placebo. Interestingly, IBS patients who responded well to FMT were found to have a higher Bacteroidetes: Firmicutes ratio and larger Prevotella fractions in their faeces than non-responders (Holvoet et al. 2018). A recent study explored the effect that repeated FMT or increasing the dose of FMT would have on IBS patients that had not responded to FMT in a previous trial using the same donor. While this trial only involved 10 participants who had not previously responded to FMT, 70% of these participants responded to the repeated transplant. Improvements to GI and non-GI symptoms were also observed (El-Salhy et al. 2019). This suggests that a large proportion of IBS patients may respond positively to modulation of the gut microbiota via FMT. However, more large-scale studies with long-term follow up are required. It is important to note that not all IBS patients respond positively to FMT (Holvoet et al. 2018), and other factors within the recipient and host, such as genetics, need to be considered (Abadi 2017; Holvoet et al. 2018).

Conclusions

The aetiology of IBS is multifactorial. Convincing evidence for the dysfunction of the immune system, the CNS, the ENS, and the gut microbiota has been found in different subsets of IBS patients. Whether any one pathophysiological mechanism predates the others is unknown, however evidence of a temporal relationship between enteric infection and Pi-IBS development provides an interesting insight into the importance of the disruption of the gut microbiota in the development of IBS symptoms (Collins 2014). IBS patients exhibit a compositionally and metabolically different gut microbiota compared to healthy controls and the functional ramifications of this are unknown, as is the cause. The differences in the composition of gut microflora seen between IBS patients and controls is not consistent across studies and marked heterogeneity remains a feature. Additionally, the drivers of dysbiosis in IBS are not clear but may include host genetics, changes in colonic transit time, or changes in gut hormone secretion. However it seems that a reliable diagnostic gut microbiota ‘fingerprint’ for IBS, or even between IBS subtypes is currently not feasible.

Regardless of pathophysiology, the modulation of the gut microbiota presents exciting possibilities for the treatment of this disorder. Probiotics and antibiotics have been used in IBS treatments for many years, with varying levels of success. The low FODMAP diet has a high level of success in controlling IBS symptoms; however the ramifications for the gut microbiota of adhering to this diet long term are unknown. Increasingly FMT has been trialled for the treatment of IBS symptoms; however long-term follow-up studies are required to assess the efficacy of FMT over long periods of time. Validation tools that could predict the responsiveness of an individual to treatments involving the modulation of the gut microbiota would provide a powerful clinical tool for the treatment of IBS. Initial studies examining faecal biomarkers that have shown initial success in predicting IBS patients responses to probiotic treatments (Rossi et al. 2018) are a promising beginning.

It is likely that a systems biology approach that measures all of the variables discussed in this review may offer the best hope of understanding the aetiology of FGIDs including IBS. However, such an approach will require carefully phenotyped cohorts and significant computational power to understand this complex system.

Acknowledgement

The authors would like to acknowledge Dr Catherine Wall for her input into the final draft of this review as well as funding from High Value Nutrition, National Science Program, Ministry of Business Innovation and Employment, New Zealand.

Disclosure statement

No potential conflict of interest was reported by the authors.

ORCID

Phoebe E. Heenan http://orcid.org/0000-0003-2350-6306

Simone Bayer http://orcid.org/0000-0002-5880-2095

Richard B. Gearry http://orcid.org/0000-0002-2298-5141

Additional information

Funding

This work was supported by the High Value Nutrition, National Science Program, Ministry of Business Innovation and Employment, New Zealand.