Detailed analysis of gut bacteria types and their functions in irritable bowel syndrome

Irritable bowel syndrome (IBS) is a chronic gastrointestinal disorder of gut-brain interaction which affects approximately 4-10% of the population, with higher prevalence in females^1,2. Symptoms include abdominal pain, bloating, and disturbed bowel function in the absence of structural or organic pathology^3–5. In addition to gastrointestinal symptoms, IBS is often accompanied by fatigue⁶, and psychological conditions such as anxiety and depression^7,8. Evidence shows that psychological stressors can affect bowel function^8–12 or conversely, bowel symptoms can in turn, affect behavior and mood¹³. This results in a heterogeneous and complex disease model for which there are still no identified distinct biological mechanisms or markers^14–16. It has been demonstrated that signals in the gut generated by the gut microbiota, alter nerve signals (neurotransmitters such as serotonin, glutamate, acetylcholine, γ-aminobutyric acid (GABA)) and immune and hormonal (endocrine factors and enzymes) responses between the gut and the brain¹⁷. These findings have attracted the attention of researchers towards understanding the role of the microbiota in conditions associated with gut-brain axis (GBA) dysfunction, such as IBS¹⁸.

Recent gut microbiome studies in IBS research have attempted to identify bacterial profiles or “signatures” in the microbiota at the phyla or genera level¹⁹. Still, the identification of specific species that clearly contribute to IBS symptomology have rarely been successful. The inconsistency in sequencing and analysis methods, limited sample size, type of specimen, gut-brain-imaging protocols, lack of details on the use of probiotics and antibiotics and diet habits, and high heterogeneity between healthy controls (HC) have made the identification of IBS-associated bacterial signatures and gut-brain interactions challenging. For example, the PCR-based 16S ribosomal RNA (rRNA) is the most common microbiome assessment technique but it is limited in depth and scope for analysis of fecal microbiota and has the potential to be biased as it requires the design of specific primers to target known phyla and genera and is thereby not optimal for the discovery of previously unknown gut bacteria at the species level^20–22.

In a meta-analysis, Pittayanon et al.²³ systematically assessed studies conducted from 2011 to 2018 and reported IBS-specific subtypes based on intestinal microbiota characterized by an increase in families Enterobacteriaceae, Lactobacillaceae, and Bacteroidales and a decrease in genus Bifidobacterium, Fecalibacterium, and Clostridiales compared to healthy individuals²³. Though most of the studies in this meta-analysis used 16S rRNA analysis, other studies utilized methods such as denaturing gel gradient electrophoresis (DGGE), fluorescent in situ hybridization (FISH), terminal restriction fragment length polymorphism (TRFLP), quantitative polymerase chain reaction (qPCR) and microarray. Shortly following this meta-analysis, a Swedish random population study on the fecal and mucosa-associated microbiome of IBS patients based on 16S rRNA analysis reported no IBS-specific bacterial signatures²⁴. Subsequent metagenome studies have aimed to identify unique bacterial signatures across IBS cohorts based mostly on 16S methods. To date, the overall findings have been inconsistent, with high variability between methods and study subjects’ fecal bacterial composition^25–32.

Other approaches are clearly necessary to detangle conditions related to the GBA, such as a more recent study which jointly evaluated both the fecal microbiome and the metabolome. Jeffery et al. have reported significant differences in the fecal microbiome and metabolomes of individuals with or without IBS.³³ Several other studies have shown that the gut microbiota, rich with Firmicutes and Bacteroidetes phyla, plays a vital role in carbohydrate, protein, and fatty acid metabolism. Furthermore, it has been shown that the impaired production of anti-inflammatory short-chain fatty acids (SCFAs) is associated with IBS^34,35.

During the last decade, research has provided an increasing amount of evidence that supports the theory that the gut microbiome influences not only gastrointestinal functions but behavior and cognition as well.^8,36–44 Most of this evidence comes from studies which demonstrate that the microbiota and its metabolites are associated not only with digestion and gut motility but also with stress, emotional behavior, and pain modulation^45–47. As the role of the gut microbiota in IBS remains inconclusive, further exploratory investigations are required to examine potential relationships between specific bacteria and symptoms in IBS. Recent advances in ultrahigh-resolution metagenomic sequencing enable simultaneous interrogation of microbial taxonomic diversity and functional metabolic capacity, providing opportunities to move beyond single-taxon associations toward systems-level understanding of microbiota–host interactions. Such approaches are particularly relevant for heterogeneous disorders of gut–brain interaction such as IBS, where microbial contributions may arise from distributed community functions rather than discrete microbial signatures.

Using metagenome shotgun sequencing, we comprehensively investigated the gut microbiome compositional alterations in IBS patients (n=63) and healthy controls (HC) (n=34). We hypothesized that the fecal bacterial composition of IBS patients differs from HCs. Therefore, our primary aim was to identify specific bacterial taxa that differed between IBS and HCs that could potentially be associated with microbial functional pathways that contribute to alterations of gut function seen in IBS. To accomplish this, we further investigated the functional features of identified bacterial species and their associated metabolic pathways to predict the role of microbiota in IBS pathogenesis and host metabolism.

This study compared the composition of the gut microbiome of 97 female participants with IBS (n=63) to HCs (n=34). Literature suggests that IBS pathogenesis is linked with either reduced or altered bacterial diversity and richness^33,69,70. In our metagenomic study, although we observed a slight reduction in bacterial richness and diversity values in the IBS group relative to HCs, we were unable to robustly distinguish the two groups to a significant degree. Our diversity measures revealed high intra-group variability, which agrees with many reports in the literature^71,72 and is often the explanation given for a lack of robust differentiation between groups.

The betadispersion analysis indicated a trend toward greater inter-individual variability within the IBS group compared to healthy controls, consistent with the broader spread of IBS samples observed in the NMDS ordination. Although this difference did not reach statistical significance, sPLS-DA likewise failed to demonstrate clear separation between groups, suggesting that between-subject variability exceeds group-level differences in this dataset. In this context, IBS-associated differences are best described as modest shifts in the relative abundance of multiple taxa and metabolic pathways rather than a single dominant microbial signature. While the cross-sectional nature of the data precludes inference about coordinated or dynamic community behavior, the concurrent compositional and functional differences observed across multiple features are consistent with a distributed pattern of microbiome alteration rather than a single defining taxonomic marker.

According to many theories, the gut microbiota plays a key role in IBS, but existing studies have shown conflicting results due to inconsistencies in cohort sizes, intra-group variability, sampling protocols, and methods of analysis^{23,33,73–78}. Although we observed no clear separation between IBS patients and HCs in the measures of gut microbiota richness or diversity indices, by analyzing differential abundances, we identified several bacterial taxa which potentially act as IBS-associated bacterial signatures. Unlike previous 16S studies, our metagenome results did not find a significant phylum-level difference in Bacteroidetes or Firmicutes abundance between the IBS and HCs, but rather a significant difference in the abundance of the phylum Euryarchaeota, which was significantly lower in IBS patients compared to HCs. IBS patients also showed a significantly higher abundance in the genera Streptococcus and lower abundance of Bifidobacterium and Methanobrevibacter when compared to healthy participants^79–81. At the species level we found 68 differentially abundant bacterial species, 17 of which were found to similarly differentiate IBS from HCs using two separate methods of analysis, both sPLS-DA and Deseq2.

Of the bacterial species found to be differentially abundant in IBS, some are commensal and commonly found in the human microbiome such as Lachnoanaerobaculum umeaense⁸², but others have been implicated in overgrowth and disease particularly in immunocompromised individuals such as Acinetobacter haemolyticus and Clostridium perfringens^83,84. Of the group of 24 differentially abundant species in Supplementary Table S2, some have been implicated in oral diseases, including Streptococcus mutans and australis, Rothia dentocariosa, and Actinomyces oris^85–87. Those more abundant in IBS which have been implicated in bowel disease include Desulfovibrio piger, and Streptococcus infantarius and equinus (S. bovis complex)^88–91. ^88,89Notably, some bacterial strains such as Bifidobacterium animalis, Lactobacillus plantarum and Lactobacillus gasseri are thought to play an important role in host metabolism and are widely found in fermented foods and probiotics⁹², and were also relatively high in abundance in IBS patients compared to HCs.

Both sPLS-DA and DESeq2 provide complementary perspectives on microbial community differences. DESeq2 identifies taxa that are individually differentially abundant between groups using a univariate framework, whereas sPLS-DA captures combinations of taxa whose joint variation contributes to group discrimination through multivariate co-abundance patterns. As such, the two approaches are inherently comparable in their aim to detect group-associated features, but differ in the structure of the signals they are designed to identify. The limited overlap between taxa identified by these methods likely reflects these methodological differences: DESeq2 is sensitive to strong, consistent changes in individual taxa, while sPLS-DA can detect weaker but coordinated variation across multiple taxa that may not reach significance in isolation. Additionally, the high inter-individual variability and modest effect sizes observed in this dataset may further reduce concordance between univariate and multivariate approaches. Taxa identified uniquely by sPLS-DA should therefore not be dismissed as false positives, but rather interpreted as candidate features reflecting community-level structure, albeit with caution given the potential for model overfitting and the need for independent validation.

We additionally examined microbiome functional metabolic pathways and found 39 differentially abundant pathways between the two groups (Supplementary Table S4). IBS microbiota functional synthesis pathways were found to be abundant in probiotic strains and lactic-acid producers such as Lactobacillus plantarum, Streptococcus mutans, Streptococcus infantarius, Streptococcus equinus, Streptococcus parasanguinis, Actinobacterial species Rothia dentocariosa and Gordonibacter urolithinfaciens and Proteobacteria species Desulfovibrio piger. Many of these species have been found to be abundant and commensal in the healthy human gut microbiome, as they are functionally linked with amino acid synthesis, carbohydrate degradation, and SCFAs such as acetate, butyrate, and propionate production that provide energy to intestinal epithelial cells and enhance the gut barrier function, yet some commensal microbes can still act as opportunistic pathogens in dysbiosis. Interestingly, a significantly high abundance of a major sulfate-reducing bacteria Desulfovibrio piger indicates possible changes in metabolic output in IBS patients⁹⁰. This further suggests that excessive sulfide production could influence the intestinal structure, regularity, and digestion and contribute to IBS pathogenesis⁹¹.

The lactic-acid-producing bacteria Streptococcus thermophilus and potentially Streptococcus salivarius have been reported to produce GABA via glutamate decarboxylase in a strain-dependent manner. While microbially derived GABA has been proposed to contribute to gut–brain signaling, its physiological relevance in humans remains unclear. Although increased abundance of Streptococcus has been observed in IBS cohorts, the functional implications for enteric nervous system (ENS) activity are not yet established⁹³. In the present study, the higher relative abundance of Streptococcus in IBS samples may be consistent with altered microbial metabolic potential but does not provide direct evidence of ENS modulation.

The species Collinsella aerofaciens is involved in carbohydrate metabolism, including lactose utilization and fermentation, and has been linked to bile acid metabolism and host–microbe interactions⁹⁴. Although Collinsella abundance varied across samples and was higher in IBS patients in this cohort, the factors underlying this difference remain unclear. While these taxa may contribute to the functional pathways identified, such as amino acid biosynthesis, our results do not support attribution of these pathways to individual species, but rather suggest that they arise from the combined activity of multiple members of the microbial community. Although we identified 39 metabolic pathways that differed between IBS and HCs, their heterogeneous and low-abundance profiles limited our ability to pinpoint specific bacterial drivers of IBS pathogenesis and symptoms. Understanding such distributed functional alterations may inform future microbiota-targeted strategies, including dietary modulation or probiotic interventions, even when taxonomic biomarkers alone show limited discriminatory power.

As metagenomic analysis moves toward clinical application, the field still faces challenges in resolving microbes at fine taxonomic scales. The gut microbiota is an exceptionally diverse and metabolically active ecosystem, and advances in high-throughput sequencing are only beginning to reveal its intricate relationship with human health. By combining complementary analytical approaches with metagenomic profiling, as demonstrated in this study, we can begin to identify key metabolic features of the microbiome that distinguish health from disease. Looking ahead, the integration of multi-omics data linking microbiome, metabolome, and patient symptomology and physiology holds great promise for unraveling the microbial and metabolic mechanisms underlying complex disorders such as IBS.

Although we observed some significant differences in gut microbiota abundance and metabolic pathway abundances between IBS patients and HCs, we were only able to classify 38% of the whole-genome readouts and 62% of reads remained “unclassified”. The classified proportion is relatively low, and our statistics indicate the partiality of reference databases on previously sequenced genomes and the lack of sensitivity in classification algorithms for detecting novel microbes. A major limitation of this study relates to the metagenomic taxonomic classification pipeline used at the time of analysis (2019). The analyses were performed using Kraken2 (v2.0.8 beta) in combination with the MiniKraken database, reflecting the state of available tools at that time. While the core classification algorithm of Kraken2 has remained stable, both software versions and reference databases have since undergone substantial development. In addition, related tools used for taxonomic and functional profiling, such as HUMAnN and MetaPhlAn and their associated reference databases, have also been significantly updated, and these advances are not reflected in the current results. Limited current knowledge pertaining to the functional features of IBS-related bacteria also restricts our understanding of their metabolic role in IBS progression and overall gut health. In addition, greater sample size and documentation of concurrent use of probiotics or diet would have been essential in understanding between-group differences between IBS patients and healthy participants. Lastly, although the focus on high-severity females with IBS reduces group and sex-related confounds, it limits generalizability to the IBS in low-severity and male subjects. Our study highlights the need for future studies that focus more on the clinical investigation of gut microbiota and metabolites to understand the potential routes of IBS pathophysiology.

Supplementary Information.