Gut microbial taxonomy and metabolism in paediatric Crohn's disease during exclusive and maintenance enteral nutrition using OMICS technologies

Clark, Clare Martha (2020) Gut microbial taxonomy and metabolism in paediatric Crohn's disease during exclusive and maintenance enteral nutrition using OMICS technologies. PhD thesis, University of Glasgow.

Full text available as:
[img] PDF
Download (88MB)


Objectives: Treatment with exclusive enteral nutrition (EEN) is an effective therapy, successfully helping children with Crohn's disease achieve remission without the need for corticosteroids. The use of maintenance enteral nutrition (MEN) following induction of remission using EEN, is thought to help maintain clinical remission. Although the efficacy of EEN is well established, the mechanism remains unknown. There is now strong evidence to support an aetiopathogenesis in Crohn's disease which implicates an interaction between environmental factors and the indigenous gut microbiota in genetically susceptible individuals.

The aim of this prospective observational study was to test the hypothesis that clinical response to EEN and reduction of colonic inflammatory markers are associated with a characteristic bacterial taxonomy (composition) and short/medium chain fatty acids (C2-C8) (functionality); and that maintenance of these proles with MEN, while returning to habitual diet, reduces the risk of relapse. The idea that gut bacteria can be altered using diet, is an important area of research in paediatric Crohn's disease, hence this study also aimed to link the gut microbiota with dietary intake.

Methods: Thirty-four children with Crohn's disease; 10 with ulcerative colitis (UC); 11 patients without inflammatory bowel disease (non-IBD); and 25 healthy controls were asked to provide faecal samples. Children with Crohn's disease also provided samples at 4-weeks, and 8-weeks of treatment with EEN; and then at 2-weeks and 8-weeks of MEN. Dietary intake was estimated using a food frequency questionnaire (FFQ), at week-0 and at 2-weeks and 8-weeks post-EEN. Post-EEN, children were treated with either 20% MEN, an immunosuppressant, or both. Bacterial DNA was extracted using the chaotropic method followed by amplification of the 16s rRNA gene (V4) for Illumina MiSeq sequencing. SCFA extraction was carried out with diethyl ether followed by gas chromatography.

Results: Baseline. Children with Crohn's disease had retarded growth and weight gain compared with healthy children (p<0.01). BMI z-scores correlated with inflammatory markers: albumin (rho=0.611; p<0.001), CRP (rho=-0.536; p<0.001) and ESR (rho=-0.407; p<0.03), showing disease severity led to an increased risk of poor growth outcomes. Faecal calprotectin was higher in children with Crohn's disease and UC than non-IBD patients and healthy children (p<0.001). Multi-dimensional scaling using euclidean distance of white blood cell counts; ALT/AST; CRP; ESR and albumin, shows children with Crohn's disease have a blood marker prole which has 20% variance from children with UC and non-IBD conditions (R2=0.197; p<0.001). Patients with Crohn's disease and UC had reduced microbiota diversity compared with healthy children (p<0.001). Many gram-positive commensal bacteria, including butyrate producing species of Firmicutes (particularly Clostridiales) were decreased; while gram-negative potential pathobionts including Gamma-proteobacteria; Fusobacterium and Veillonella increased in children with Crohn's disease compared with healthy children. Akkermansia, a genus associated with healthy gut mucosa, was also reduced in Crohn's disease patients (mean log2 difference p<0.05).

The faecal short chain fatty acid valerate (p=0.02) and medium chain fatty acids, hexanoate and octanoate (p<0.001), were reduced in children with Crohn's disease compared with healthy children. The profile of short/medium chain fatty acids in children with Crohn's disease differed from healthy children (p=0.01). Reduced hexanoate was associated with reductions in a number of Firmicutes in children with Crohn's disease (p<0.05). Estimated dietary intake suggested children with Crohn's disease also had reduced fibre intake, particularly fruit, along with reduced intake of vitamins (A, E, B7 and C) (p<0.05).

EEN. Of 32 children who took EEN, 23 (72%) went into remission (wPCDAI <12.5). Nine (28%) failed to respond and went on to corticosteroids. No differences were seen for baseline faecal calprotectin or blood markers, between children who responded to EEN and those who failed EEN. In children who responded to 8-weeks of EEN calprotectin, ESR and CRP were reduced (p<0.001, p=0.002 and p=0.02 respectively); mirrored by an increase in albumin (p<0.001). Blood inflammatory marker profiles after treatment with EEN were similar to non-IBD controls (R2=0.225; p<0.001).

Although 25 bacterial species discriminated responders from non-responders at baseline, there was no pattern of taxonomic relatedness between these. Successful treatment with EEN changed the microbiota community structure further from that of healthy children (R2=0.070; p<0.001). During EEN gram-negative bacteria including Pasteurellaceae, Bacteroidales S24-7, Fusobacteriaceae and Veillonellaceae (Negativicutes) were reduced, while increases mostly came from gram-positive Clostridiales (p<0.05). During EEN short/medium chain fatty acid proles moved towards that of healthy children (R2=108; p=0.002), while the concentration of short/medium chain fatty acids decreased, particularly butyrate (p=0.001). Neither energy intake nor composition of diet at baseline predicted response to EEN.

MEN. Of the 23 children who responded to EEN, 17 (74%) were treated with MEN post-EEN. Although children responding to treatment had age appropriate growth during EEN (p=0.056), growth velocity was not maintained once children went back onto normal diet (p=0.96). No differences in growth were seen between types of maintenance therapy. The use of MEN was not associated with an increase in length of remission, however the number of children in the non-MEN group was too small (n=6) to reject the null hypothesis.

Faecal calprotectin returned to near pre-treatment levels, having a significant increase after only 8-weeks post-EEN (p=0.004), showing MEN failed to maintain the reduction of inflammation achieved during EEN. Bacterial changes during EEN, were also not maintained post-EEN. Patients using MEN, had increases in gram-negative bacteria (Veillonellaceae, Enterobacteriaceae, Bacteroidaceae, Prevotellaceae and Verrucomicrobiaceae). Within 2-weeks post-EEN, short chain fatty acids including butyrate returned towards pre-treatment levels (p<0.001). The use of MEN did not appear to prevent the ratio of short/medium chain fatty acids returning to pre-treatment levels. One year after the start of treatment with EEN, only 8/23 (35%) original respondents had maintained remission.

Conclusions: Reduced fibre intake in children with Crohn's disease appears to lead to reduced diversity of Firmicutes, particularly Clostridiales. This reduction in commensal bacteria opens a niche for gram-negative bacteria like the Enterobacteriaceae. These changes could lead to decreased short/medium chain fatty acids, valerate, hexanoate and octanoate, which provide antimicrobial protection against gram-negative bacteria; again opening a niche for Enterobacteriaceae and other gram-negative pathobionts. Decreased intake of fibre may trigger some bacteria to switch from fermenting fibre to feeding on mucin glycans, consequently damaging the protective layer of the gut mucosa. It is thus possible that low fibre intake, under certain conditions, drives increased gut inflammation. Malabsorption, of vitamins A, C, D and E, which have immunoregulatory roles, could further exacerbate inflammation. As faecal calprotectin levels rise this limits bacterial access to zinc, manganese and iron, thus further depleting Clostridiales and increasing dysbiosis. Therefore a chronic escalation of calprotectin, could contribute in part to increased inflammation.

This study does not support the hypothesis that 20% MEN can extend remission times in children with Crohn's disease. The fact that 74% of children went onto MEN post-EEN, but only 35% remained in disease remission after 1-year, suggests that either MEN is dose dependent or that exclusion of the normal diet leads to reduced inflammation. Altering normal diet in order to induce and maintain remission in children with Crohn's disease is an attractive option, for both patients and treatment centres. Further studies are needed to explore whether increased fibre or vitamin intake can protect against inflammation and relapse in Crohn's disease. Studies also need to examine the role calprotectin plays in modifying the gut microbiota.

Item Type: Thesis (PhD)
Qualification Level: Doctoral
Keywords: Crohn's Disease, Inflammatory Bowel Disease, IBD, microbiota, microbiome, OMICS, 16S rRNA, Short Chain Fatty Acids, SCFA, diet, Exclusive Enteral Nutrition, Maintenance Enteral Nutrition.
Subjects: Q Science > QR Microbiology
R Medicine > R Medicine (General)
R Medicine > RJ Pediatrics
Colleges/Schools: College of Medical Veterinary and Life Sciences > School of Cardiovascular & Metabolic Health
Supervisor's Name: Ijaz, Dr. Umer Z.
Date of Award: 2020
Depositing User: Dr Clare M Clark
Unique ID: glathesis:2020-81261
Copyright: Copyright of this thesis is held by the author.
Date Deposited: 30 Apr 2020 19:09
Last Modified: 02 Sep 2022 11:10
Thesis DOI: 10.5525/gla.thesis.81261
Related URLs:

Actions (login required)

View Item View Item


Downloads per month over past year