PNNL

Abstract

While genomic approaches have been applied for the detection and identification of individual bacteria within microbial communities, analogous proteomics approaches have been effectively precluded due to their inherent complexity. An in silico assessment of peptides that could potentially be present in the proteomes of artificial simple and complex communities was performed to evaluate the effect of proteome complexity on species detection. A mass spectrometry-based proteomics approach was employed to experimentally detect and validate the predicted tryptic peptides initially identified as distinctive within the simple community. An assessment of peptide distinctiveness and the potential for mapping to a particular bacterium within a community was made throughout each step of the study. A second in silico assessment of peptide distinctiveness for a complex community of 25 microorganisms was conducted to investigate the levels of instrumental performance that would be required to experimentally detect these peptides, as well as how performance varied with complexity (e.g., the number of different microorganisms). The experimental data for a simple community showed that it is feasible to predict, observe, and to quantify distinctive peptides from one organism in the presence of at least a 100-fold greater abundance of another, thus yielding putative markers for identifying a bacterium of interest. This work represents a first step towards quantitative proteomic characterization of complex microbial communities and the possible development of community wide markers of perturbations to such communities.