Quantifying amino acid and amino sugar biomarkers in a single approach to estimate necromass from soil archaea, bacteria, fungi, and plants

Soil organic matter is the largest carbon (C) pool in terrestrial ecosystems, and it is largely composed of microbial necromass. Microbes contribute to the long-term C storage in soils by incorporating C from plants into their biomass and, consequently, microbial necromass becomes stabilized mostly in mineral associated organic matter. So far, most studies have focused on tracing microbial necromass using amino sugar biomarkers, while plant contributions to soil organic matter are predominantly traced using lignin or long-chain alkanes. Glucosamine and muramic acid are amino sugars commonly used as biomarkers of fungal and bacterial necromass, respectively. Amino acids, such as D-enantiomers and non-proteinogenic amino acids have also been used though rarely as microbial and/or plant necromass tracers. For instance, meso(D,L)-diaminopimelic acid can be found in the peptidoglycan peptide chain of gram-negative bacteria, while hydroxyproline is commonly found in glycoproteins of plant cell walls. Currently, only very few studies have measured amino sugars alongside primary and secondary amino acids as biomarkers of plant and microbial necromass. In this study, we propose a new method that allows the simultaneous exploration of microbial and plant residue biomarkers using a single run via 6-aminoquinolyl-N-hydroxysuccinimidyl carbamate (AQC) derivatization, followed by ultra-high performance liquid chromatography (UHPLC) and Orbitrap high resolution mass spectrometry. For this, we analysed 121 species of archaea, bacteria, fungi and plants. We were able to quantify amino acids and amino sugar biomarkers in the biomass of all taxonomic groups, as well as compare how these biomarker contents varied between broad taxonomic groups. We confirmed the biomarker potential of non-proteinogenic amino acids and amino sugars using indicator species analysis as well as supervised multivariate approaches, such as random forest and partial least squares discriminant analysis (PLS-DA). Our results showed that hydroxyproline is a biomarker specific for plants, while L,L-diaminopimelic acid can be used alongside muramic acid as biomarkers specific for bacteria. Talosaminuronic acid represents a biomarker specific for archaea, while glucosamine was a biomarker indicative of archaea, bacteria and fungi, being absent in plants. Our results showcase an unparalleled approach to trace both plant and microbial contributions to soil organic matter which will help improve our understanding of how different organic matter sources contribute to soil carbon formation and stabilization. This approach also allows the quantitation of plant versus microbial contributions to the continuum from litter decomposition to soil organic matter formation though microbial processing, the contribution of plant, fungal and bacterial organic matter to mineral-associated organic matter (MaOM) versus particulate organic matter (POM), and to soil macro- and microaggregate formation.