Comparison of Protein Extraction Methods and Data Analysis Strategies for Complete Metaproteomic Soil Analysis

Considerable microbial diversity has been discovered in soil through genomic sequencing. Despite its role in biogeochemical cycling, relatively little is known about the proteomic diversity of the soil microbiome as most commercially available soil kits focus on DNA/RNA extractions. Consequently, a plethora of protein extraction techniques have been developed for soil but have yet to be integrated into simplified, modern sample preparation techniques such as the S-Trap. Furthermore, classical data analysis strategies for soil metaproteomics rely on genomically-informed databases for peptide/protein identification. This assumes that DNA/RNA extracts adequately represent the soil proteome. Within this study, we systematically assess several extraction techniques, developing a data processing pipeline which is driven by both proteomics and genomics to fully characterize the soil microbiome. Both pipelines reveal remarkably complementary data, with [~]60% of the protein identifications coming from Proteomically-derived databases. Sodium dodecyl sulfate-based extractions proved to provide the most unique protein identifications ([~]3000 proteins), and by combining both proteomic and genomic-based results, the total protein identifications increased approximately 2-fold for each extraction. Combining these complementary data pipelines with improved extraction techniques can allow for drastically improved proteomic results (12,307 unique protein identifications), even from minute (50 mg) sample volumes. These enhancements to previous workflows can better describe the microbial diversity within soil and provide a deeper functional understanding of the soil microbiome.