A novel Ruminococcus gnavus clade enriched in inflammatory bowel disease patients

Abstract

Background: Inflammatory bowel disease (IBD) is characterized by chronic inflammation of the gastrointestinal tract that is associated with changes in the gut microbiome. Here, we sought to identify strain-specific functional correlates with IBD outcomes.

Methods: We performed metagenomic sequencing of monthly stool samples from 20 IBD patients and 12 controls (266 total samples). These were taxonomically profiled with MetaPhlAn2 and functionally profiled using HUMAnN2. Differentially abundant species were identified using MaAsLin and strain-specific pangenome haplotypes were analyzed using PanPhlAn.

Results: We found a significantly higher abundance in patients of facultative anaerobes that can tolerate the increased oxidative stress of the IBD gut. We also detected dramatic, yet transient, blooms of Ruminococcus gnavus in IBD patients, often co-occurring with increased disease activity. We identified two distinct clades of R. gnavus strains, one of which is enriched in IBD patients. To study functional differences between these two clades, we augmented the R. gnavus pangenome by sequencing nine isolates from IBD patients. We identified 199 IBD-specific, strain-specific genes involved in oxidative stress responses, adhesion, iron-acquisition, and mucus utilization, potentially conferring an adaptive advantage for this R. gnavus clade in the IBD gut.

Conclusions: This study adds further evidence to the hypothesis that increased oxidative stress may be a major factor shaping the dysbiosis of the microbiome observed in IBD and suggests that R. gnavus may be an important member of the altered gut community in IBD.

Fig. 1

Dynamics of longitudinal microbial composition and facultative anaerobic microbial profiles in inflammatory bowel disease. a Individual microbial trajectories of 20 inflammatory bowel disease (IBD) patients and 11 control individuals with sufficient longitudinal data over time, from the current study (Massachusetts General Hospital, p-identifiers) and (Emory University, S-identifiers). Each subject exhibits an individualized microbiome signature. Each phylum has an overall color while genera are represented as different shades of the overall color. b Maximal relative abundances of facultative anaerobes across all subjects in the LSS (n = 266), Lewis et al. (n = 368) [13], and HMP (n = 80) [15] cohorts. Overall, facultative anaerobes are significantly higher in IBD patients compared to controls (nested ANOVA LSS p = 0.0478, Lewis et al. p = 0.005)

Fig. 2

R. gnavus transiently dominates the gut microbiome in IBD. a The maximum relative abundance of R. gnavus across samples (time courses) is shown for all subjects in the LSS (n = 266), Lewis et al. (n = 368), and HMP (n = 80) cohorts. While the abundance of most anaerobes are lower in IBD patients, the abundance of R. gnavus is significantly higher in IBD patients compared to controls. b Relative abundance of R. gnavus over time for IBD patients in the LSS cohort. The abundance of R. gnavus is not constantly high, but rather has transient increases in the IBD gut. c A principle coordinate analysis (PCoA) of the Bray-Curtis distance of species-level microbial communities of LSS IBD patient p8808 over 9 months. The dominant R. gnavus strain in months 4–10 and month 12 is R. gnavus clade 1, while in month 11 the dominant strain of R. gnavus is R. gnavus clade 2 (Fig. 3). Inset shows the Harvey-Bradshaw Index (HBI) score, a clinical indication of active disease and inflammation, for this patient over time. The dramatic, transient increased abundances of R. gnavus in month 11 corresponds to an increase in HBI values (i.e., disease activity). d Colony forming units of R. gnavus, Eubacterium elegans, and E. coli at 0, 1, and 3 h post-transfer to atmospheric oxygen conditions (see “Methods”). Error bars represent standard deviation. Dotted line signifies limit of detection. No colonies were detected for the obligate anaerobe E. elegans at the 1- and 3-h time points. As expected, E. coli showed growth during oxygen exposure, and interestingly, despite being classified as an obligate anaerobe, R. gnavus was able to tolerate atmospheric oxygen for several hours, which may partially explain its increased abundance in the increased oxidative stress of the IBD gut

Fig. 3

R. gnavus metagenomic strain phylogeny. A phylogenetic tree of R. gnavus strains, calculated from SNP profiles of R. gnavus marker genes (see “Methods”), where each tree leaf is a sample from the LSS or Lewis et al. cohorts (subject plus time point). Subject SKST012 is the only included control individual, as no other control metagenomes contained enough R. gnavus reads for detection and strain assignment by StrainPhlAn. Bootstrap values are indicated on branches and reveal two distinct clades of R. gnavus strains

Fig. 4

Functional profiles of IBD-related R. gnavus strains; 199 IBD-specific genes within R. gnavus (rows) and their depth of coverage across metagenomic samples (columns). All samples with at least 1× coverage of the R. gnavus pangenome are shown. Cluster of genes indicate that different strain groups (clades; Fig. 3) have different subsets of IBD-specific genes. Several gene families of interest are highlighted (see text), and the full R. gnavus pangenome can be found in Additional file 5: Figure S1