7th Annual BMRP Investigator Meeting - Abstract

Horace T. Williams^1,2,b, I. Jane Cox², Sara E. Marshall³, Derek P. Jewell⁴, Subrata Ghosh¹, David Taylor-Robinson^1,2 and Timothy R. Orchard^1,a

 

¹Departments of Gastroenterology & Hepatology and ²Imaging Sciences, Imperial College London (London, England); ³Department of Immunology, University of Dundee (Dundee, Scotland); ⁴Gastroenterology Unit, University of Oxford (Oxford, England)

 

The pathogenesis of the inflammatory bowel diseases (IBD), Crohn’s disease (CD) and ulcerative colitis (UC) is thought to involve a genetically-determined, abnormal host immune response to an environmental stimulus, likely to be bacterial. The intestinal flora differ between CD and UC patients and controls. Metabolic profiles of biofluids such as urine can be obtained using nuclear magnetic resonance (NMR) spectroscopy: these urinary profiles are influenced by the intestinal flora, since gut bacterial metabolism generates specific, quantifiable products. It was hypothesised that such metabolites would differ between CD and UC patients and healthy individuals, and that multivariate pattern-recognition techniques could distinguish the cohorts.

 

Significant differences were found between the groups with respect to metabolites influenced by the gut flora: hippurate, formate and 4-cresol sulphate. Multivariate analysis, using principal components analysis and partial least squares discriminant analysis, could differentiate the cohorts. The results were independent of medication and diet, and were confirmed through longitudinal sampling.

 

Hippurate, a bacterial co-metabolite, differed the most between cohorts (p<0.0001), and exerted considerable leverage in the multivariate models. The difference was most significant in patients with CD. There are two stages in the biosynthesis of this metabolite: i. the metabolism of dietary aromatic compounds to benzoate by the gut microflora, and ii. conjugation of benzoate with glycine to form hippurate, which is excreted in the urine. An experiment was undertaken to investigate whether the differences in hippurate levels were due to differences in the gut flora, rather than abnormalities in host conjugation, through the administration of sodium benzoate to CD patients and healthy controls, with subsequent quantification of hippurate excretion. The conjugating ability of the CD cohort was not impaired, strongly implicating the microflora in the differences found.

 

In conclusion, specific urinary metabolites, influenced by gut microfloral metabolism, differ between CD patients, UC patients and controls; the biosynthesis of hippurate in CD has also been elucidated. The emerging technique of urinary metabolic profiling with multivariate analysis was able to distinguish these cohorts.

 

^aPrincipal Investigator, ^bCo-investigator and Presenter