Fecal short-chain fatty acids in non-constipated irritable bowel syndrome: a potential clinically relevant stratification factor based on catabotyping analysis

Abstract

The gut microbiota is believed to be a critical factor in the pathogenesis of IBS, and its metabolic byproducts, such as short-chain fatty acids (SCFAs), are known to influence gut function and host health. Despite this, the precise role of SCFAs in IBS remains a topic of debate. In this study, we examined the bacterial community structure by 16S rRNA gene profiling and SCFA levels by UPLC-MS/MS in fecal samples from healthy controls (HC; n = 100) and non-constipated patients (IBS-D and IBS-M; NC-IBS; n = 240) enrolled in 19 hospitals in Italy. Our findings suggest a significant difference between the fecal microbiomes of NC-IBS patients and HC subjects, with HC exhibiting higher intra-sample biodiversity. Furthermore, we were able to classify non-constipated patients into two distinct subgroups based on their fecal SCFA levels (fecal catabotype "high" and "low"), each characterized by unique taxonomic bacterial signatures. Our results suggest that the fecal catabotype with higher SCFA levels may represent a distinct clinical phenotype of IBS that could have implications for its diagnosis and treatment. This study provides a new perspective on the intricate relationship between the gut microbiome and bowel symptoms in IBS, underscoring the importance of personalized strategies for its management.

Keywords: Gastrointestinal disorder; abdominal pain; butyrate; catabotyping; diarrhea; fecal microbiome; fecal type; metataxonomics; succinate; valerate.

Figure 1.

Diagram reporting the number of patients for which specific indicated data are available. (a), quantification of organic acids in feces; (b), metataxonomics of fecal samples through 16S rRNA gene profiling; (c), abdominal pain and bowel habits data.

Figure 2.

Comparison of the intra-sample biodiversity of the fecal bacterial communities between non-constipated patients (NC-IBS; n = 235) and healthy control subjects (HC; n = 100) according to four different α-diversity indexes. Statistics is according to the Mann-Whitney test. **, P < .01; ****, P < .0001).

Figure 3.

PLS discriminant analysis (PLSDA) with prediction background for non-constipated IBS patients (NC-IBS) and healthy control subjects (HC). The Receiver Operating Characteristic (ROC) curve of the PLSDA model is shown in supplementary Figure S1.

Figure 4.

Bacterial taxa in fecal samples exhibiting a significantly different abundance between individuals with non-constipated irritable bowel syndrome (NC-IBS) and healthy subjects (HC). (a) cladogram based on LEfSe analysis (LDA scores for the single taxa are reported in Supplementary Figure S3). (b) significantly different taxa determined through the Mann-Whitney test carried out with CLR-transformed bacterial abundances. Higher and lower abundances for each taxon are reported with a red and cyan background, respectively. The black and yellow heatmap represents the mean CLR-transformed abundances of the reported taxonomic units. The taxonomic lineage of each taxon is shown: p, phylum; c, class; o, order; f, family; g, genus; s, species.

Figure 5.

Organic acids in fecal samples exhibiting a significantly different abundance between individuals with non-constipated irritable bowel syndrome (NC-IBS) and healthy subjects (HC). The statistical analysis was performed using the Mann-Whitney test and indicated a statistically significant difference compared to HC samples. **, P < .01; ***, P < .001; ****, P < .0001. Red numbers refer to the σ variance of the corresponding data.

Figure 6.

Stratification analysis of non-constipated IBS patients (NC-IBS) based on the fecal levels of the organic acids. (a) principal-coordinate analysis (PCoA) generated through catabotyping based on the fecal concentrations of acetate, butyrate, propionate, lactate, succinate, and valerate. (b) spider chart based on the abundances of bacterial catabolites expressed as relative abundance.

Figure 7.

Bacterial taxa in fecal samples of NC-IBS patients exhibiting a significantly different abundance between catabotypes FC-H and FC-L. Significantly different taxa have been determined through the Mann-Whitney test carried out with CLR-transformed bacterial abundances. The black-yellow heatmap represents the mean CLR-transformed abundances of the reported taxonomic units. Higher and lower abundances for each taxon are reported with a red and cyan background, respectively. The taxonomic lineage of each taxon is shown: p, phylum; c, class; o, order; f, family; g, genus; s, species. Taxonomic names written in blue were determined through a manual BLASTN search in GenBank using the sequence of the corresponding reads.

Figure 8.

Correlation analysis between fecal organic acids and IBS symptoms in catabotypes FC-L (n = 105) and FC-H (n = 125). The heatmap represents the τ coefficient of Kendall rank correlation. Asterisks indicate significance in the Kendall rank correlation: *, P < .05; **, P < .01. NRS, numeric rating scale.

Figure 9.

Progression of abdominal pain over 16 weeks in the PROBE-IBS/2 cohort of non-constipated IBS patients stratified in the fecal catabotype FC-L and FC-H. Each point refers to the median value of the numeric rating scale (NRS) for pain assessment. Statistics is according to the Mann-Whitney test between FC-L and FC-H at each specific time point. The numbers in gray close to each point refer to the number of patients. **, P < .01; *, P < .05; n.s., not significant.