Colonic metaproteomic signatures of active bacteria and the host in obesity

Obesity is associated with the intestinal microbiota in man but the underlying mechanisms are yet to be fully understood. Our previous phylogenetic study showed that the faecal microbiota profiles of non-obese versus obese and morbidly obese individuals differed. Here, we have extended this analysis with a characterisation of the faecal metaproteome, in order to detect differences at a functional level. Proteins were extracted from crude faecal samples of 29 subjects, separated by 1D gel-electrophoresis and characterized using reversed phase liquid chromatography-tandem mass spectrometry. The peptide data were analysed in database searches with two complementary algorithms, OMSSA and X!Tandem, to increase the number of identifications. EggNOG database searches resulted in the functional annotation of over 90% of the identified microbial and human proteins. Based on both bacterial and human proteins, a clear clustering of obese and non-obese samples was obtained that exceeded the phylogenetic separation in dimension. Moreover, integration of the metaproteomic and phylogenetic data sets revealed notably that Bacteroidetes were metabolically more active in the obese than non-obese subjects. Finally, significant correlations between clinical measurements and bacterial gene functions were identified. This study emphasizes the importance of integrating data of the host and microbiota to understand their interactions. This article is protected by copyright. All rights reserved.