Go to contents (site navigation)


Realised by ALMS™
developer of the AIDS-HIV Reference project
Abstract No.: MoP-213
Session: Proteomics: Clinical
Presentation date: Mon, Aug 28, 2006
Presentation time: 09:50 – 11:20

Analysis of E.coli Protein Expression in the Mouse Gastrointestinal Tract by 2D-PAGE and nanoLC-ESI-MS/MS

Wolfram Engst1, Carl Alpert1, Jacqueline Scheel1, Michael Blaut1

1 German Institute of Human Nutrition Potsdam-Rehbruecke, Nuthetal, Germany

Correspondence address: Wolfram Engst, German Institute of Human Nutrition Potsdam-Rehbruecke, Mass Spectrometry, Arthur-Scheunert-Allee 155, Nuthetal, D-14558 Germany.

Keywords: Chromatography, Liquid -, Nanoscale Capillary; Mass Spectrometry, QTOF; MS/MS, Liquid Chromatography; Protein Identification.

Novel aspect: Development of a nanoLC-ESI-MS/MS method for characterization of host-bacteria interactions in vivo.


The human digestive tract is a complex ecosystem, where several hundred bacterial species interact with each other, food components and the host. The gut microbiota fulfills a number of functions in host physiology but is also presumed to contribute to the development of diseases, such as inflammatory bowel disease, colon cancer and allergies.
While the influence of the bacteria on the host has been repeatedly demonstrated, the host may also have an influence on the bacteria. At present, knowledge about these interactions is very scarce. Host-bacteria interactions have mostly been investigated with regard to the host response or to activities of pathogenic bacteria. In contrast, we aim to identify reactions of non-pathogenic bacteria that result from their contact with host-cells as they occur in the gastrointestinal tract (GIT) by means of the detection of differentially expressed proteins. In a proteomic approach the response of non-pathogenic human bacteria during transit of the gastrointestinal tract was investigated using a gnotobiotic animal model (ex-germfree mice mono-associated with E. coli) and Nanospray ESI-MS/MS for subsequent identification of the expressed proteins.
Samples from feces and different compartments of the GIT were separated by 2D-PAGE and spots of interest were selected after pattern analysis with the PDQuest software. Proteins were digested with trypsin and the resultant peptide mixtures were analyzed on a Waters capillary LC system (CapLC) coupled to a Waters QTOF-Ultima API mass spectrometer fitted with a Nanolockspray source. The acquisition of MS/MS spectra was performed in an automated data directed switching between the MS and MSMS mode based upon ion intensity and charge state.
Protein identification based on sequence information of the tryptic peptides was achieved using ProteinLynxTM Global Server 2.2 software (Waters corporation) for processing of MS/MS data and subsequent databank searching. Precursor and product ions were automatically lockmass corrected against Glufibrinopeptide B and erythromycin which generally results in a mass accuracy close to 5 ppm. The option of inclusion or exclusion of peptide masses in data acquisition and to merge results by ProteinLynxTM Global Server permitted the identification of additional peptides, thereby increasing the sequence coverage of identified proteins.
The bacterial protein expression varied with the transit along the GIT probably as well as with the nutritional status of the host. However in fecal samples the products of e.g. the genes ptgA, galM, gldA, dhaK, otcC, talB, yjgF, ahpC, rpiA, artI, aspG2, ppiB, ptsN ( = rpoB), dsbA and mao2, several of which are involved in energy metabolism or which have regulatory functions, appear to be constantly expressed.