The Interplay between Mucosal Microbiota Composition and Host Gene-Expression is Linked with Infliximab Response in Inflammatory Bowel Diseases

Abstract

Even though anti-TNF therapy significantly improves the rates of remission in inflammatory bowel disease (IBD) patients, there is a noticeable subgroup of patients who do not respond to treatment. Dysbiosis emerges as a key factor in IBD pathogenesis. The aim of the present study is to profile changes in the gut microbiome and transcriptome before and after administration of the anti-TNF agent Infliximab (IFX) and investigate their potential to predict patient response to IFX at baseline. Mucosal biopsy samples from 20 IBD patients and nine healthy controls (HC) were examined for differences in microbiota composition (16S rRNA gene sequencing) and mucosal gene expression (RT-qPCR) at baseline and upon completion of IFX treatment, accordingly, via an in silico pipeline. Significant differences in microbiota composition were found between the IBD and HC groups. Several bacterial genera, which were found only in IBD patients and not HC, had their populations dramatically reduced after anti-TNF treatment regardless of response. Alpha and beta diversity metrics showed significant differences between our study groups. Correlation analysis revealed six microbial genera associated with differential expression of inflammation-associated genes in IFX treatment responders at baseline. This study shows that IFX treatment has a notable impact on both the gut microbial composition and the inflamed tissue transcriptome in IBD patients. Importantly, our results identify enterotypes that correlate with transcriptome changes and help differentiate IFX responders versus non-responders at baseline, suggesting that, in combination, these signatures can be an effective tool to predict anti-TNF response.

Keywords: anti-TNF; biomarkers; host transcriptome; inflammatory bowel disease; infliximab; microbiome; microbiota; response to therapy.

Figure 1

Overall design of the study. This protocol allowed us to showcase IBD vs healthy controls dysbiosis, to highlight the influence of Infliximab on microbiota composition and identify response-related microbial and transcriptional biomarkers.

Figure 2

(A) Microbiota composition changes at phylum level among healthy controls, Crohn’s disease and ulcerative colitis patients. (B) α-diversity (quantification of biodiversity) differences in the 3 groups. (C) β-diversity (qualitative enterotype differences) of the 3 groups.

Figure 3

(A) Relative abundance changes of the microbial genera among healthy controls (control), Crohn’s disease (CD) and ulcerative colitis (UC) patients. (B) LEfSe analysis showing microbial genera associated with the 3 groups. (C) Venn diagram depicting the microbial genera constantly present (core microbiome) in the samples of the 3 groups versus those found exclusively in HC, CD and UC.

Figure 4

(A) Microbiota composition at phylum level among Crohn’s disease patients before treatment (CD_PRE) and after treatment in non-responders (CD_NONR) and responders (CD_R). (B) α-diversity (quantification of biodiversity) differences of the 3 groups. (C) β-diversity (qualitative enterotype differences) of the 3 groups.

Figure 5

(A) Relative abundance of microbial genera among Crohn’s disease patients before treatment (CD_PRE) and after treatment in non-responders (CD_NONR) and responders (CD_R). (B) LEfSe analysis showcasing microbial genera associated with the sample groups. (C) Venn Diagram depicting the microbial genera constantly present (core microbiome) in the samples of the 3 groups versus those found only in responders, non-responders and at baseline (before initiation of treatment).

Figure 6

Microbiome analysis of Crohn’s disease samples before treatment of which we know the response outcome: response (CD_PRE_R) and non-response (CD_PRE_NONR). (A) β-diversity (qualitive enterotype differences). (B) Relative abundance changes of the microbial genera (C) LEfSe analysis showcasing microbial genera associated with the sample groups. (D) Venn diagram depicting the microbial genera constantly present (core microbiome) in the samples of the 2 groups.

Figure 7

Overview of the genera with differential abundance between responders and non-responders and how these correlate with specific differentially expressed genes. IL18, CCR3, CXCL8, TLR6, TLR9 and TNFSF14 are upregulated and CCR4 is downregulated in both groups (2/2 genera abundant in responders and 3/4 abundant in non-responders) and appear to meet the criteria to be characterised as biomarkers for prediction of response to IFX therapy.