Quantitative measures of myo-IP6 in soil using solution 31P NMR spectroscopy and spectral deconvolution fitting including a broad signal

Abstract

Inositol phosphates, particularly myo-inositol hexakisphosphate (myo-IP₆), are an important pool of soil organic phosphorus (P) in terrestrial ecosystems. To measure concentrations of myo-IP₆ in alkaline soil extracts, solution ³¹P nuclear magnetic resonance (NMR) spectroscopy is commonly used. However, overlap of the NMR peaks of myo-IP₆ with several other peaks in the phosphomonoester region requires spectral deconvolution fitting (SDF) to partition the signals and quantify myo-IP₆. At present, two main SDF approaches are in use; the first fits a Lorentzian/Gaussian lineshape to the myo-IP₆ peaks directly to the baseline without an underlying broad signal, and the second fits a Lorentzian/Gaussian lineshape to the myo-IP₆ peaks simultaneously with an underlying broad peak. The aim of this study was to compare the recovery of added myo-IP₆ to soil extracts using both SDF procedures for six soil samples of diverse origin and differing concentrations of organic P (112 to 1505 mg P per kg_soil). The average recovery of total added myo-IP₆ was 95% (SD 5) and 122% (SD 32) using SDF with and without an underlying broad signal, respectively. The recovery of individual peaks of myo-IP₆ differed, most notably, the C5 phosphate peak of myo-IP₆ was overestimated by up to 213% when a broad peak was not included in SDF. Based on the SDF procedure that includes a broad peak, concentrations of myo-IP₆ ranged from 0.6 to 90.4 mg P per kg_soil, which comprised 1–23% of total phosphomonoesters. Our results demonstrate that the SDF procedure with an underlying broad signal is essential for the accurate quantification of myo-IP₆ in soil extracts.