European Journal of Soil Science, 2010 doi: 10.1111/j.1365-2389.2010.01322.x Extraction of microbial proteome from soil: potential and limitations assessed through a model study L. Giagnoni a , F. Magherini b , L. Landi a , S. Taghavi c , A. Modesti b , L. Bini d , P. Nannipieri a , D. Van der lelie c & G. Renella a a Department of Plant, Soil and Environmental Sciences, University of Florence, Florence, Italy, b Department of Biochemistry Science, University of Florence, Florence, Italy, c Biology Department, Brookhaven National Laboratory, Upton, NY 11973, USA, and d Department of Molecular Biology, University of Siena, Siena, Italy Summary Proteomics is the study of functions and regulation of biological systems based on the analysis of the protein expression profile, and there is a general agreement that soil proteomics may be a tool for better soil management. Because of the ability of soils to stabilize extracellular proteins by various mechanisms, development of soil proteomics needs an assessment of the efficiency of protein extraction from various soil types. We evaluated the possibility of extraction of soil microbial proteome by inoculating Cupriavidus metallidurans CH34, which has a known proteome, into sterile sand, kaolinite, montmorillonite and a mixture of sand, kaolinite, montmorillonite, goethite and humic acids. One hour after inoculation, the viability of C. metallidurans was determined by the colony-forming units method (CFU), the amount of extracted proteins was determined by the Bradford method and the bacterial proteome was analysed by the two-dimensional gel electrophoresis technique (2D-GE). The bacterial number was 2.5 × 10 6 CFU g −1 of soil in all microcosms, whereas the total extracted protein content varied from 98.1 to 1268 μgg −1 in the various microcosms, but was undetectable in the inoculated montmorillonite. The number of protein spots from the bacterial culture and the inoculated microcosms varied between 317 and 591, with 54 variable spots among the pure culture and the microcosms. No protein spots were detected in the 2D-GE from the montmorillonite microcosm. The 2D-GE of artificial soil microcosms showed a protein pattern that was different from those of pure culture and sand and kaolinite microcosms. The results confirm the importance of clay-specific surface area and CEC in protein adsorption as montmorillonite alone had the largest sorptive capacity, and show that the artificial soil used also had a large sorptive capacity for microbial proteins. Globally, the results indicate that the extraction of proteins from soils is strongly influenced by the clay type and organic matter content, and that poor protein extraction efficiency may reduce the potential of soil proteomics. Introduction Proteomics is an analytical approach providing an understanding of the structure, functions and regulation of living organisms by the study of the protein expression profile rather than individual proteins, and is complementary to genomics, transcriptomics and metabolomics (Tyers & Mann, 2003). Progress in protein anal- ysis technologies, methodologies and supporting bioinformatics provides new tools to overcome major problems related to the proteomic analysis of environmental matrices. There is a general consensus that for a better understanding of soil functionality it is necessary to go beyond DNA analysis. Soil DNA characterization provides information about microbial Correspondence: G. Renella. E-mail: giancarlo.renella@unifi.it diversity (Torsvik et al., 2002), whereas soil mRNA isolation can provide information about gene transcription. However, pro- teomics can account better for gene expression in soil and there- fore soil functionality (Nannipieri, 2006). Among the potentials, soil proteomics may lead to the discovery of sublethal microbial biomarkers responding to soil management and pollution, and may improve our understanding of degradation of organic pollutants and organic matter, nutrient cycles, and molecular communica- tion between plants and between plant and microbial communities (Nannipieri, 2006). Moreover, soil may be a reservoir of proteins of biotechnological interest (Nannipieri, 2006). Proteins in soil originate from plants, animals and microor- ganisms inhabiting the soil either through active excretion or passive release. From 96 to 99% of soil total nitrogen (N) is organic and according to the classical acid hydrolysis procedure,  2010 The Authors Journal compilation  2010 British Society of Soil Science 1