Quantifying functional microbiomes: an integrated, quantitative metaproteomics approach
Year: 2018
Type: other
Abstract: Microbiome research is offering promising insights into microbial contributions to human health and environmental systems. Microbiome responses can be studied by metaproteomics, which identifies proteins actively expressed by the microbial community providing insights into its functional state. The information obtained using mass spectral counts alone, however, allow only for semi-quantitative analysis. As a result, subtle, but important, dynamics of microbiome composition and function remain undetected. In this study, we demonstrate the use of precursor ion intensity-based quantification (MS1) and how, coupled to analyses within open-source platform - Galaxy, MS1 intensity-based quantification reveal connections between taxa, biochemical function and protein abundances. For this investigation, two datasets were used: (1) a clinical dataset with microcosms from plaque samples taken from subjects with a high risk for dental caries were grown in biofilm reactors in the presence and absence of sucrose ( doi:10.1186/s40168-015-0136-z ) and (2) an environmental microbial dataset from sediments of an anoxic lake incubated in bioreactors. The datasets were subjected to high-pH fractionation and Orbitrap acquisition (clinical dataset) or gas-phase fractionation and Q-Exactive acquisition (environmental dataset) to generate mass spectrometry data. The mass spectra were assigned to peptides by searching against appropriate metaproteomics databases. Subsequently, both datasets were subjected to MS1 intensity-based quantitation (moFF), taxonomic assignments (Unipept) and functional analysis (EggNOG mapper). From these outputs, a peptide tabular file was generated with corresponding taxa, functional group and quantitative MS1 intensity values from replicates. The MS1 values were normalized and assigned to a) taxa based on method described by Cheng et al. ( doi:10.1186/s40168-017-0375-2 ) and b) functional groups based on methods described by Riffle et al ( doi:10.3390/proteomes6010002 ). Normalized values were further processed, revealing taxonomic distribution of functional expression. Protein-level expression analysis was also performed to identify differentially expressed orthologous groups at a 5% false discovery rate threshold. All analyses were done within Galaxy-P platform, and the results were compared with an alternative non-Galaxy workflow to benchmark the methods used. The alternative workflow used MaxQuant for quantitation and MEGAN6 for taxonomy and functional analysis. Our initial application of this quantitative metaproteomics workflow reveals connections between specific taxa and their contributions to the functional profile of the system, measuring even subtle changes in abundance of peptides derived from specific classes of microbes. Second, it quantifies a significant number of peptide sequences that cannot be assigned confidently to taxa/functional groups. Characterization based on quantification and potential validation of ‘unassigned’ peptides present an important opportunity for future research. By using a clinical and an environmental dataset, we demonstrate how our analysis workflow provides deeper and more accurate biological insight in microbiome studies. We also anticipate that this will lay the groundwork for an advanced multi-omics integration that will enable a multidimensional and higher resolution to the nuances of the expressed functional microbiome.
Institution: University of Minnesota
Language: English
Subfield: Spectroscopy
Field: Chemistry
Domain: Physical Sciences
Open Access status: gold
OpenAlex ID: W2896548817