P147 Metabolomics discriminate children with Crohn's disease from ulcerative colitis and from healthy controls – preliminary study
Daniluk U.1, Ciborowski M.2, Daniluk J.*3, Pietrowska K.2, Kretowski A.2, Lebensztejn D.1
1Medical University of Bialystok, Department of Pediatrics, Gastroenterology and Allergology, Bialystok, Poland 2Medical University of Bialystok, Clinical Research Centre, Bialystok, Poland 3Medical University of Bialystok, Department of Gastroenterology and Internal Medicine, Bialystok, Poland
Ulcerative colitis (UC) and Crohn's disease represent inflammatory bowel disease (IBD) with multifactorial pathogenesis. Metabolic profiling might be used to understand interactions between nutrients, the intestinal metabolism and the microbiota composition in these diseases. The aim of our study was to determine the usefulness of untargeted metabolomics in detection of metabolic differences and similarities between children with Crohn's disease (CD) or ulcerative colitis (UC) in comparison to healthy individuals.
Metabolic fingerprinting of serum samples was estimated with liquid chromatography coupled to mass spectrometry (LC-QTOF-MS) in newly diagnosed children with CD (n=9, median age 14 years), ulcerative colitis (n=9, median age 13.5 years) and controls (n=10, median age 12.5 years). Statistical analysis was used to find metabolic differences between Controls and IBD patients as well as between UC and CD groups. Depending on data distribution t-test or U-test were used to select significant metabolites. Multivariate statistical analysis was used for samples classification.
Detected metabolites were used to build good quality partial least square discriminant analysis model (R2=0.99, Q2=0.76) which showed clear separation between studied groups. 46 metabolites were significantly discriminating between IBD patients and Controls. Several lysophospholipids – lysophosphatidylcholine (LPC), lysophosphatidylethanolamine (LPE) were found decreased (from 30 to 60%) in IBD patients (p-value 0.005–0.03). Phosphatidylcholine (PC) 42:6 was 46% increased (p=0.02) in IBD group. 44 metabolites were significantly discriminating CD from UC patients. Among them phospholipids (Phosphatidylethanolamine, PE 38:5 +271%, p=0.003 and PC 35:5 +46%, p=0.03) and anandamide (+48%, p=0.02) were increased in UC group. While PC 42:6 (+40%, p=0.00006) and LPC 15:0 (+79%, p=0.0006) were increased in CD group.
Metabolic fingerprinting allows for discrimination between Controls and two studied inflammatory bowel diseases. Obtained results indicate for disruption mainly of lysophospholipids metabolism in IBD patients. On the other hand phospholipids and anandamide were mainly responsible for discrimination between CD and UC groups. Although promising, obtained results are preliminary and require validation on larger group of patients and with the use of targeted analysis of significant metabolites.