Development of a bioassay-coupled HPLC-SPE-ttNMR platform for identification of α-glucosidase inhibitors in apple peel (Malus ×domestica Borkh.).

This work describes an analytical platform based on a high-resolution α-glucosidase inhibition assay in combination with hyphenation of high-performance liquid chromatography, solid-phase extraction, and tube-transfer nuclear magnetic resonance spectroscopy, i.e., HPLC-SPE-ttNMR/high-resolution α-glucosidase assay. The platform enables fast screening for individual α-glucosidase inhibitory analytes in complex matrices, followed by structural identification targeted these α-glucosidase inhibitors, as demonstrated by a proof-of-concept study with extract of 'Pink Lady' apple peel. A scout-separation produced a high-resolution biochromatogram and a HPLC chromatogram, which were used for pinpointing HPLC peaks displaying α-glucosidase inhibition. Active analytes were cumulatively trapped on SPE cartridges and the structures identified by (1)H NMR experiments obtained in the HPLC-SPE-ttNMR mode. (-)-Epicatechin (1), reynoutrin (3) and avicularin (4) were identified as active compounds. IC(50) of the active compounds were determined along with six structurally related compounds. Quercetin was the most potent inhibitor with an IC(50) of 8.1±0.4µM. The platform proved to be an efficient method for the identification of α-glucosidase inhibitors.

1H NMR spectroscopy-based interventional metabolic phenotyping: a cohort study of rheumatoid arthritis patients.

1H NMR spectroscopy-based metabolic phenotyping was used to identify biomarkers in the plasma of patients with rheumatoid arthritis (RA). Forty-seven patients with RA (23 with active disease at baseline and 24 in remission) and 51 healthy subjects were evaluated during a one-year follow-up with assessments of disease activity (DAS-28) and 1H NMR spectroscopy of plasma samples. Discriminant analysis provided evidence that the metabolic profiles predicted disease severity. Cholesterol, lactate, acetylated glycoprotein, and lipid signatures were found to be candidate biomarkers for disease severity. The results also supported the link between RA and coronary artery disease. Repeated assessment using mixed linear models showed that the predictors obtained from metabolic profiles of plasma at baseline from patients with active RA were significantly different from those of patients in remission (P=0.0007). However, after 31 days of optimized therapy, the two patient groups were not significantly different (P=0.91). The metabolic profiles of both groups of RA patients were different from the healthy subjects. 1H NMR-based metabolic phenotyping of plasma samples in patients with RA is well suited for discovery of biomarkers and may be a potential approach for disease monitoring and personalized medication for RA therapy.