Microbial imbalance in inflammatory bowel disease patients at different taxonomic levels

Abstract

Background: Inflammatory bowel disease (IBD), is a debilitating group of chronic diseases including Crohn's Disease (CD) and ulcerative colitis (UC), which causes inflammation of the gut and affects millions of people worldwide. At different taxonomic levels, the structure of the gut microbiota is significantly altered in IBD patients compared to that of healthy individuals. However, it is unclear how these IBD-affected bacterial groups are related to other common bacteria in the gut, and how they are connected across different disease conditions at the global scale.

Results: In this study, using faecal samples from patients with IBD, we show through diversity analysis of the microbial community structure based on the 16S rRNA gene that the gut microbiome of IBD patients is less diverse compared to healthy individuals. Furthermore, we have identified which bacterial groups change in abundance in both CD and UC compared to healthy controls. A substantial imbalance was observed across four major bacterial phyla including Firmicutes, Bacteroidetes, Proteobacteria and Actinobacteria, which together constitute > 98% of the gut microbiota. Next, we reconstructed a bacterial family co-abundance network based on the correlation of abundance profiles obtained from the public gut microbiome data of > 22,000 samples of faecal and gut biopsies taken from both diseased and healthy individuals. The data was compiled using the EBI metagenomics database (Mitchell et al. in Nucleic Acids Res 46:D726-D735, 2018). By mapping IBD-altered bacterial families to the network, we show that the bacterial families which exhibit an increased abundance in IBD conditions are not well connected to other groups, implying that these families generally do not coexist together with common gut organisms. Whereas, the bacterial families whose abundance is reduced or did not change in IBD conditions compared to healthy conditions are very well connected to other bacterial groups, suggesting they are highly important groups of bacteria in the gut that can coexist with other bacteria across a range of conditions.

Conclusions: IBD patients exhibited a less diverse gut microbiome compared to healthy individuals. Bacterial groups which changed in IBD patients were found to be groups which do not co-exist well with common commensal gut bacteria, whereas bacterial groups which did not change in patients with IBD were found to commonly co-exist with commensal gut microbiota. This gives a potential insight into the dynamics of the gut microbiota in patients with IBD.

Keywords: Crohn’s disease; Gut microbiota; Inflammatory bowel disease; Microbial imbalance; Ulcerative colitis.

Fig. 1

Microbial diversity and richness. a Species richness is substantially less in CD patient samples compared to the healthy control and UC patient samples. Shannon alpha diversity plot demonstrate that CD patients samples are less diverse compared to the healthy control and UC patient samples (inset figure). b Phylum level richness in the gut microbiota from our samples (left side Y-axis shown in green) compared to the global gut microbial species richness obtained across more than 20 K samples from a variety of conditions (right side Y-axis shown in black)

Fig. 2

The gut microbial abundance. a Phylum, b Class and c Family level abundance in different conditions. Classes and families belonging to the four most abundant phylum across conditions are grouped according to phylum

Fig. 3

The human gut bacteria family co-abundance. a, i The network represents the global human gut bacterial family coexistence considering the four most highly abundant gut bacterial phyla. The network connections are based on correlation test (P-value < 1e−10 and Pearson’s correlation coefficient > 0.3). Edge connection between families from the same phylum is shown in black lines whereas the family connection between different phyla is shown in grey lines. Family nodes are coloured according to phylum. a, ii This graph demonstrates the number of families (i.e. nodes) belonging to a different phylum in the global bacterial family coexistence. a, iii The percentage connection between families from the same phylum is substantially higher in the global bacterial family coexistence compared to a random network of the same size. b A subnetwork of the global human gut bacterial family coexistence network where the abundance level of at least one family node in a connection is > 1.5-fold higher in either CD or healthy condition against each other. c Similarly, a subnetwork of the global human gut bacterial family coexistence network where the abundance level of at least one family node in a connection is > 1.5 fold higher in either UC or healthy condition against each other. Edge connection between families from the same phylum is shown in black lines whereas the family connection between different phylum is shown in grey lines. Family belonging to different phyla are shown in different shapes. The node colour shows the increased abundance level in a disease (CD or UC) or healthy condition compared to each other. The size of the node represents the abundance level in a healthy condition