Biomarker metaproteomics for relative taxa abundances across soil organisms

Soil organisms are often classified using methods targeting individual groups of taxa (e.g., bacteria, fungi and invertebrates), which hampers our ability to directly compare the relative abundance of different groups across environmental gradients. We posit that the use of protein biomarkers could help to provide a more real representation of the cross-kingdom soil microbial populations. Here, we tested if the abundant proteins ATP synthase F(0) complex (ATPS), elongation factors (EF), glyceraldehyde-3-phosphate dehydrogenase (GAPDH), GroEL, pyruvate dehydrogenase (PyrDH), RNA polymerase beta chain (RNAP), and translation initiation factor 2 (TIF) could be used to describe the taxonomic composition of microbial communities. As positive control, we used a mock community with different relative abundances of algae, archaea, bacteria, and viruses. We tested this approach on a previously published soil metaproteomes from which we randomly selected samples from forests, grasslands, and shrublands (each n = 10). Unfortunately, the biomarker approach is not feasible for viruses as these organisms do not share single genes. All biomarkers showed decent accuracy to determine the relative abundances of archaea, bacteria, and eukaryota in the mock community. However, false positive hits dominated on phylum level probably due to sequence homology. Archaeal proteins were only detected in the soil samples when EF was used as biomarker at an abundance of 0.7%. Bacteria dominated the EF-metaproteome and were most abundant in shrublands (64.4%) while eukaryotes were more abundant in forests (25.6%). In compliance with previously published results, the correlation analysis revealed the impact of mean annual temperature and pH on both bacteria and eukaryota. Our approach not only shows the potential to use biomarker metaproteomics to unveil the relative taxa abundances across soil organisms but also the need to create mock communities comprising members of all soil taxa. • Screening of seven biomarkers for equal cross-kingdom abundances in soil metaproteomics. • Biomarker metaproteomics is able to identify domain-level abundances of soil taxa. • Elongation factors showed highest accuracy and identified archaeal proteins. • Bacteria are positively correlated to pH and negatively to mean annual temperature. • Eukaryotes are negatively correlated to pH and positively to mean annual temperature.