Nature of soil organic phosphorus: an assessment of peak assignments in the diester region of 31P NMR spectra

This study involved a critical assessment of peak assignments in 31P NMR spectra of alkaline soil extracts and included 31P NMR spectroscopy of (1) 500 mM NaOH solutions of RNA, DNA, and lecithin; (2) 50 mM H2SO4, 500 mM NaHCO3 (pH 8.5), and 100 mM NaOH extracts from Pseudomonas putida (Gram-negative bacterium), Bacillus subtilis (Gram-positive bacterium), Penicillium citrinum (a fungus), Aspergillus niger (a fungus), and leaves of Betula pubescens, Picea abies, and Pinus cembra; (3) 100 mM NaOH solutions of evaporated methanol–chloroform extracts from B. subtilis, P. putida, B. pubescens leaves, and from organic and mineral soil horizons; (4) 100 mM NaOH and 50 mM H2SO4 extracts from delipidized bacterial cells, and 100 mM NaOH extracts from delipidized birch leaves and soil samples. Results showed that the resonance at 0 ppm, previously assigned to phospholipids and nucleic acids, was caused by DNA-P. Resonances of phospholipids of plant and microbial origin were observed at about 1.5–1.7 and 0.6–0.7 ppm, in regions previously assigned to teichoic acid-P by various authors. Non-lipidic compounds extracted from B. subtilis and resonating at 1.9 ppm probably were represented by teichoic acids. 31P NMR spectroscopy cannot differentiate signals derived from P of phospholipids and non-lipidic compounds in the low field of the diester region of spectra of alkaline extracts from soils, while quantitative differentiation of DNA-P and the other diesters seems quite a simple task. An unknown resonance at −1.5 ppm remained unidentified. The real concentrations of diester-P in soils can be considerably underestimated and those of monoester-P overestimated, when analyzed in alkaline extracts, because of almost complete hydrolysis of RNA and partial hydrolysis of phospholipids to monoesters.