Introduction

Irritable bowel syndrome (IBS) is a functional gastrointestinal (GI) disorder characterized by abdominal pain or discomfort associated with abnormal bowel habit [1]. The prevalence of IBS in the industrialized world is approximately 10–15 % in the general population, which makes IBS one of the most common GI disorders [2]. Individuals suffering from IBS have a reduced quality of life and report more co-morbidities than the general population [3, 4]. IBS is commonly subtyped according to predominant bowel habit into diarrhoea-predominant, constipation-predominant, or mixed/alternating phenotypes (IBS-D, IBS-C, or IBS-M) [1, 5].

The pathogenesis and pathophysiology of IBS is incompletely understood, but abnormal GI motility, visceral hypersensitivity, altered brain–gut function, low-grade inflammation, and psychosocial factors are considered to contribute [6]. During recent years, the evidence for involvement of gut microbiota in the pathogenesis and pathophysiology of irritable bowel syndrome (IBS) has increased, and this is the focus of this review, where direct and indirect evidence for the importance of gut microbiota alterations in IBS will be discussed.

Post-infectious IBS

The onset of IBS symptoms after a bout of gastroenteritis comprises one of the strongest indications for the importance of gut microbiota for the induction of IBS symptoms. There have been numerous studies assessing the risk of developing IBS after a bout of gastroenteritis and meta-analyses have established that there is a six- to sevenfold increased risk of developing IBS after an infectious gastroenteritis [7, 8••]. However, GI infections are extremely common around the world and, in the vast majority of affected people, a full recovery rapidly occurs. Therefore, a bout of gastroenteritis is obviously not enough to develop IBS, and several risk factors are associated with development of post-infectious (PI) IBS. Hence, psychological distress at the time of the gastroenteritis increases the risk of developing long-standing GI symptoms, and the same holds true for being of young age and of female gender [9]. Moreover, evidence suggests that factors related to the infectious event or to the inflammatory reaction per se might also be of importance [6, 10].

Immune Activity and Recognition of Luminal Antigens in IBS

The potential role of low-grade inflammation in IBS has been acknowledged during recent years, and a link between altered gut microbiota composition and inflammation in IBS has been suggested [6]. A clinical observation indirectly supporting the potential role of low-grade inflammation in the generation of symptoms compatible with IBS comes from patients with inflammatory bowel disease, i.e. ulcerative colitis and Crohn’s disease. A substantial proportion of these patients have IBS-like symptoms when they are in remission, hence when there are no obvious signs of active inflammation [11]. Moreover, IBD patients without signs of active inflammation but with IBS-like symptoms had more severe psychological symptoms and reduced quality of life, pointing indirectly towards an interaction between peripheral and central factors in the genesis of these symptoms, in line with risk factors identified for development of PI-IBS. Also, mucosal dysbiosis in patients with inflammatory bowel disorders has been reported [12], which suggests that this might also be a factor of relevance for remaining symptoms in patients in remission, but this has to be confirmed in future studies.

Also of great interest, several genes involved in the host response to intraluminal antigens or bacterial invasion, or the immune responses to these antigens, are differently expressed in IBS patients, and especially so in patients with PI-IBS [13]. Recent studies have identified several susceptibility genes for IBS-D and PI-IBS involved in the innate immunity and recognition of bacteria but also maintaining the integrity of the intestinal barrier [14]. Interestingly, TNFSF15 encoding the TLA1 protein is identified in both Crohn´s disease and IBS [15, 16•], underlining the possibility of common pathways in these diseases.

Cells of the immune system have specific receptors, for example toll-like receptors (TLR), which recognize danger signals, such as bacterial and viral components (Fig. 1). TLRs are widely expressed, both on the cell surface and within the cytoplasm, by dendritic cells, macrophages, and epithelial cells among others. In a recently published paper, it was reported that IBS-M patients displayed a significant up-regulation of TLR2 and TLR4 on epithelial cells in the colonic mucosa, which resulted in increased secretion of the pro-inflammatory cytokines IL-8 and IL1-β [17]. Additionally, increased levels of mucosal TLR4 and TLR5 and decreased levels of TLR7 and TLR8 in IBS patients have been reported [18]. IBS patients have also been demonstrated to have increased TLR2 expression on monocytes [19], and TLR-induced cytokine secretion patterns are altered in patients relative to healthy individuals [20]. Also, antibodies specific for flagellin, the primary component of bacterial flagellae, are more frequently found in IBS patients than in the healthy population [21]. Taken together, these data support an engagement between gut microbiota and the mucosal immune system, involving both innate and adaptive immunity, which may generate a low-grade inflammatory response that can be of importance for symptom generation in IBS.

Fig. 1
figure 1

Mucosal dendritic cells (DC) recognize microbial danger signals through pathogen recognition receptors, for example TLRs, migrate to mesenteric lymph nodes (MLN), and present danger signals, antigens, to T cells. The T cells leave the blood stream and migrate to the intestinal mucosa to fight the danger. Circulating monocytes (Mo) maturating into macrophages (Ma), which fagocytose bacteria and are important for clearing infectious agents, are also recruited to the intestinal mucosa in response to pro-inflammatory cytokines produced by DC on site

Full size image

Effects of Antibiotics and Probiotics in IBS

The previous understanding has been that antibiotic treatment may induce or worsen IBS symptoms. This view was supported by two observational studies, showing that systemic antibiotic treatment could in fact increase functional GI symptoms [22, 23]. In contrast, recent randomized controlled trials demonstrate improved symptoms in IBS patients after treatment with a non-absorbable antibiotic, Rifaximin [24•, 25]. The therapy seems to be effective, especially for flatulence and bloating, and in the most recent trial a better effect than with placebo was observed during the 10 weeks follow-up period, but the therapeutic advantage over placebo was merely 10 % [24•]. Further studies are needed to evaluate whether Rifaximin should be used for IBS, but the results indicate a potential role for gut microbiota alterations at least in a subset of IBS patients. Another line of indirect evidence for a role of gut microbiota alterations in IBS comes from studies on the effect of treatment with probiotics for IBS patients. A systematic review, comprising 19 randomized controlled trials including more than 1,600 patients with IBS, concluded that probiotics appear to be efficacious in IBS, although the magnitude of benefit and the most effective species and strain are still uncertain [26•]. However, as one plausible effect of probiotics is to alter the gut microbiota composition [27–29], the positive effect of probiotics in IBS, although admittedly modest, supports the involvement of gut microbiota in the genesis of symptoms in IBS.

Gut Microbiota Composition of IBS Patients

In recent years, many research groups have focused on identifying the gut microbiota composition of the large intestine of IBS patients, using modern culture-independent techniques [30]. Several of the available studies demonstrate that IBS patients, or subgroups thereof, have an altered microbiota composition relative to healthy individuals [31, 32•, 33, 34, 35•]. A finding that has been seen in several studies is a gut microbial composition enriched with Firmicutes together with a reduced abundance of Bacteroidetes in the IBS subjects relative to healthy individuals [32•, 35•, 36].

Most studies have been performed in adult patient cohorts, but there are also studies reporting an altered gut microbiota in pediatric IBS patients [37•, 38]. In a study of children with IBS by Saulnier et al., it was reported that the fecal microbiome composition differed between IBS-C and IBS-U patients [37•]. There were not a sufficient number of IBS-D patients in the study to include these in the analysis. Likewise, it has been reported that abundance of bifidobacteria is lower in adult IBS-D patients than in IBS-C patients [39]. In contrast, our group demonstrated no difference of fecal microbiota composition according to IBS subtypes based on Rome criteria [1, 5]. Instead, we could clearly identify two separate clusters of IBS patients, not associated with IBS subtypes, whose microbiota composition was different from healthy individuals. Moreover, a third cluster of IBS patients with normal-like microbiota composition was identified [35•]. In general, the results of current studies on large intestinal microbiota in IBS patients lack consensus. It seems likely that IBS patients, or subgroups of patients, may have an altered microbiota, but it is far from clear whether the microbiota changes are related to IBS subtypes according to the dominant symptoms.

Diversity of Fecal and Mucosal Microbiota

The majority of the hitherto published studies on gut microbiota of IBS patients are based on analysis of fecal microbiota, as a proxy for the intestinal microbial communities. Several studies have demonstrated lower fecal microbial diversity [31, 33] of IBS patients as compared to healthy individuals. Additionally, our group demonstrated that, among the IBS patients with an altered microbiota composition, one cluster of IBS patients had increased phylogenetic diversity whereas another cluster had lower phylogenetic diversity relative to healthy individuals [35•].

It is important to highlight that it is not well known to what extent fecal microbiota reflects the mucosa-adherent microbiota community. Some reports have addressed both fecal and mucosal microbiota composition of IBS patients [31, 40–42]. Differences in both fecal and mucosa-associated gut microbiota between patients and healthy individuals seem to exist, although variations between these two ecological niches are present [31]. Also, lower bifidobacteria counts [42] and higher prevalence of Pseudomonas aeruginosa [41] have been described in both fecal and mucosal samples of IBS patients.

Gut Microbiota Associated with IBS Symptoms

There is little direct proof that gut microbiota or their metabolites may control intestinal function in humans, even though several animal studies provide evidence supporting this [43–45]. Hence, there are relatively few reports of association between microbiota abundance and IBS symptoms. However, our group recently reported several clear associations between clinical characteristics and different taxa [35•]. Moreover, in the same study, we demonstrated that the prevalence of depression was increased in IBS patients with normal-like fecal microbiota, whereas the group of IBS patients with an altered fecal microbiota (high Firmicutes:Bacteroidetes ratio) displayed normal prevalence of depression, implicating that symptom generation may be different in these two groups of IBS patients [46]. However, this hypothesis needs to be confirmed in future studies. Another group has described that stool frequency was negatively correlated with number of mucosa associated bifidobacteria and lactobacilli [39]. The abundance of gammaproteobacteria has also been positively correlated with IBS symptoms [32•]. Additionally, a Ruminococcus torques-like phylotype has also been reported to positively correlate with IBS symptom scores, including emotional and social function, and systemic and bowel symptoms [47]. Furthermore, in a recent, relatively large study of IBS patients, correlation analyses of microbial groups and IBS symptoms indicated the involvement of several groups of Firmicutes and proteobacteria in the genesis of IBS symptoms [32•]. In one of the few studies of pediatric IBS patients, an association between high/moderate pain phenotype and taxa belonging to the Alistipes genera was identified. Also, taxa of the genera Akkermansia, Parabacteriodes, and Ruminococcaceae were found in children reporting frequent pain episodes [37•]. Interestingly, in a study of healthy Finnish adults, an association between low bifidobacteria counts and pain was described, when the gut microbiota composition and health perception and the occurrence of intestinal symptoms was followed during a 7-week study period [48]. However, even though available studies suggest an association between microbes and symptoms in IBS, the relative importance of different taxa for IBS symptoms is not consistent between existing studies. Further studies in large, phenotypically well-characterized cohorts of IBS patients are needed to clarify this association.

Effects of Bacterial Metabolites on Intestinal Function

Metabolites produced by the gut microbiota community may have effects on intestinal function. For example, short chain fatty acids (SCFA), such as butyrate and acetate, produced by endogenous bacteria, have been demonstrated to activate the ileal motor pattern [49], through specific G protein receptor signaling [50]. It has also been shown that Lactobacillus paracasei metabolites modulate contractility of intestinal smooth cells [51], and E. coli Nissle secretions modulate contractility of human muscle strips [52]. Moreover, Lactobacillus acidophilus and L. paracasei have been reported to modulate pain and visceral hypersensitivity perception, respectively [53, 54]. Interestingly, an increased sulfate-reducing microbiota population in the gut of C-IBS patients has been reported, which could lead to enhancement in toxic sulfide production, which in turn could influence gut physiology and contribute to IBS pathogenesis [55]. Also, an altered gut microbiota community producing less SCFA, in an in vitro fermentation system in response to incubations with various carbohydrates and fibers, has been described in patients with D-IBS [56]. Thus, an altered production of metabolites of the colonic bacterial microbiota may be related to the development of gastrointestinal symptoms in patients with IBS.

Small Intestinal Microbiota in IBS

The question whether IBS patients have small intestinal bacterial overgrowth (SIBO) has been extensively debated during the last decade [57••, 58, 59]. Initial reports suggesting an important role for SIBO in IBS [60, 61], have been hard to replicate by other groups [62, 63]. One of the problems with the SIBO hypothesis has been that it is mainly based on demonstrating SIBO with the lactulose hydrogen breath test, which has poor sensitivity and specificity [58]. Moreover, in a recent study using combined oro-caecal scintigraphy and hydrogen breath test, it was demonstrated that the lactulose hydrogen breath test merely reflects rapid oro-caecal transit and not SIBO in IBS patients [64]. Also, potential confounding factors, such as proton pump inhibitors (PPI) [65, 66], and altered motility may explain discrepancies between studies. Further, in a study from our group where we used the gold standard for diagnosing SIBO, i.e. culture of jejunal aspirate, we were unable to find an increased prevalence of SIBO in IBS, even though a mild increase of small intestinal bacteria was more commonly found in patients with IBS than in healthy controls [62]. So, to summarize, the relevance of small intestinal microbiota alterations in IBS remains to be proven, and further studies addressing this controversy, using valid techniques, are much awaited.

Effects of Diet on Gut Microbiota

Dietary factors have effects on the gut microbiota composition, and this might be of considerable importance for symptoms of IBS patients. Unabsorbed dietary components will serve as nutrients, driving the growth of the microbial community. The effect of dietary habits has recently been elegantly demonstrated: the polysaccharide-rich diet of African children from rural areas resulted in enrichment of Bacteroidetes and depletion of Firmicutes, as compared to Italian urban children [67]. Moreover, in another study, using modern culture independent techniques to demonstrate gut microbiota composition and careful determination of food intake, an association between dietary factors and gut microbiota composition was clearly demonstrated [68]. Current data suggest that gut microbiota composition may be a result of food intake, but the genetic background of the host also seems to be an important factor for shaping the gut microbiota community [69].

Knowledge of the impact of food intake on gut microbiota composition in humans may be achieved by investigating effects of short-term or long-lasting changes in dietary lifestyles. Large intake of fibers and food containing fructans and fructose, and other poorly fermentable dietary carbohydrates (FODMAPs), may give rise to bloating and diarrhea. Therefore, reduction of these food items may have beneficial effects on IBS symptoms [70, 71]. Compared to wheat bran, whole grain cereals increase the numbers of fecal Bifidobacterium spp. and Lactobacillus spp. in healthy subjects [72]. Low calorie diets containing carbohydrates, decreased Roseburia spp. and fecal butyrate concentration in obese subjects [73]. Also, Lactulose acidifies the proximal colon, which has been demonstrated to increase Bifidobacterium spp. counts in healthy humans [74] and patients with chronic idiopathic constipation [75]. A fermentable carbohydrate restriction 4-week diet reduced the abundance of bifidobacteria, and also had positive effects on overall symptoms and bloating of IBS patients [76]. However, whether this diet also had major implications on the gut microbiota composition remains to be determined. Moreover, a diet low in FODMAPs was recently shown to change microbiota composition in parallel with improvement in GI symptoms in a group of IBS patients, further highlighting the role of diet for gut microbiota composition [76].

Sample Collection, Preparation and Analyzing Techniques

The majority of the hitherto published studies on the gut microbiota composition of IBS patients comprise relatively low numbers of patients, especially when considering the heterogeneity of this patient group. The inconsistent data regarding the microbiota composition of IBS patients are probably partly due to the lack of uniformity of the patient group. Yet, the diverging results may also be explained by lack of insight into how sample collection and preparation may affect the results. In a recently published report, it was demonstrated that fecal samples exposed to room or deep freezing temperatures for up to 24 h and 6 months, respectively, exhibit a microbial composition and diversity that shares more identity with its host of origin than any other sample [77]. However, the data from that study implied that the microbiota from IBS patients was less stable at room temperature than that of healthy controls. Moreover, the use of different techniques is also a likely reason for the deviating results. The approach to identify gut microbiota has changed dramatically during the last decade. The gold standard for detection and classification has been based on culture methods, although it has been clear that many of the gut-resident microbes are not possible to culture with traditional methods. Development of new culture-independent techniques based on the 16s RNA characterization of microbial communities has rapidly increased our knowledge of the gut microbiota. Classical 16s RNA-based techniques, such as fingerprinting (DGGE and T-RFLP) and PCR amplification of genes, are now complemented by high throughput 16s RNA techniques, allowing more and better powered observations. Thus, thousands of 16s RNA or their respective genes may be identified and quantified in a single experiment [57••].

Conclusion

The knowledge and understanding of the importance of gut microbiota for symptom generation in IBS is rapidly increasing. Today, alterations in gut microbiota composition are considered to be a plausible factor of relevance for symptoms in at least subsets of these patients. However, due to the fact that the heterogeneity of the patient group is large and that many external factors may contribute to the individual microbiota variation, careful studies of large and phenotypically well-characterized groups of patients are needed in order to more clearly demonstrate the contribution of gut microbiota alterations to the heterogeneous symptom pattern in patients with IBS.