Metabolic Assessment of Gradual Development of Moderate Experimental Colitis in IL-10 Deficient Mice

Abstract

Evidence has linked genetic predisposition and environmental exposures to the worldwide pandemic of inflammatory bowel diseases (IBD), but underlying biochemical events remain largely undefined. Here, we studied the gradual development of colitis in Interleukin 10 deficient mice using a combination of (i) histopathological analysis of intestinal sections, (ii) metabolic profiling of blood plasma, and (iii) measurement of plasma inflammatory biomarkers. Data integration using chemometric tools, including Independent Component Analysis, provided a new strategy for measuring and mapping the metabolic effects associated with the development of intestinal inflammation at the age of 1, 8, 16, and 24 weeks. Chronic inflammation appeared at 8 weeks and onward, and was associated with altered cecum and colon morphologies and increased inflammatory cell infiltration into the mucosa and the submucosa. Blood plasma profiles provided additional evidence of loss of energy homeostasis, impaired metabolism of lipoproteins and glycosylated proteins. In particular, IL-10-/-mice were characterized by decreased levels of VLDL and increased concentrations of LDL and polyunsaturated fatty acids, which are related to the etiology of IBD. Moreover, higher levels of lactate, pyruvate, citrate and lowered glucose suggested increased fatty acid oxidation and glycolysis, while higher levels of free amino acids reflected muscle atrophy, breakdown of proteins and interconversions of amino acids to produce energy. These integrated system investigations demonstrate the potential of metabonomics for investigating the mechanistic basis of IBD, and it will provide novel avenues for management of IBD.

1H NMR-based metabolomic study of metabolic profiling for systemic lupus erythematosus

“…18,23 The alteration in serum metabolites might reflect an increased energy demand relative to the decreased energy availability under inflammatory conditions. 30 Formate is a potential alternative single-carbon source for the production of N5,N10-methylene THF required for biosynthesis of nucleic acid. The serum level of formate was markedly decreased in the SLE patients, which might result from high consumption of formate due to a significant increase in the biosynthesis of purine nucleotides in an inflammatory condition, such as the septic state.…”

GCMS-based metabolomic study in mice with colitis induced by dextran sulfate sodium

“…These results may have been caused by the incorporation of glutamine into mucosal cells during the resolution phase. 24 In this study PLS-DA scores plots based on 92 metabolites detected in the colon tissue exhibited differences among the acute phase of inflammation (day 7), the resolution phase of inflammation (day 10), and the normal state, and the corresponding PLS-DA loadings plots revealed that several metabolites strongly influenced the separation of each group, i.e., succinic acid, indole-3-acetic acid, glutamic acid, and glutamine (Fig. 2).…”

Independent component analysis in non-hypothesis driven metabolomics: Improvement of pattern discovery and simplification of biological data interpretation demonstrated with plasma samples of exercising humans

“…In previous investigations [19,21,30] the data 'x' were modelled to 'A × s' by two different approaches: if x was variables × observations, then the independent component 'x' represents the weight matrix of the metabolites, and the mixing matrix 'A' represents the impact factors in the metabolic process [21]; if x was observations × variables, then the mixing matrix 'A' represents the weight matrix of the metabolites, and the independent component 'x' represents the impact factors in the metabolic process [19,30]. These two different input approaches may both be applied in metabolomics in theory, however we applied an alternative strategy.…”

“…ICA has been widely and successfully utilized in signal processing [6,7], image feature extraction [8,9], medical imaging [10,11], genomics and protein profiling [12,13], process monitoring [14,15]. ICA can separate the source signals with non-Gaussianity and statistically independent data, thus it has also successfully been applied in the areas of proteomics [13,16,17], transcriptome [18] and metabolomics [19][20][21][22][23][24]. Although the numbers of principal components after whitening and the independent components are important for the explanations of the data and the biological process [18], few papers were focused on the optimization of ICA [18].…”