Evaluation of different protocols for the analysis of lipophilic plant metabolites by gas chromatography–mass spectrometry using potato as a model

Abstract

In the analysis of lipophilic plant metabolites by gas chromatography–mass spectrometry a step is required to release fatty acids and other analytes from complex molecules. Seven alternative methods were compared to the standard method of 1% H2SO4/50°C/16 h using Desiree and Phureja potato tubers as models. With two sodium methoxide alkali-catalysed methods (0.5 M NaOCH3/50°C/1 and 16 h) recoveries of ferulic acids increased, long chain fatty acids and sterols decreased, 2-hydroxy acids were negligible, solanidine was absent and Δ5-avenasterol isomerisation was minimal. Using a harsh alkali hydrolysis (1.0 M KOH/120°C/24 h) followed by a mild methylation (1% H2SO4/50°C/1.5 h), recoveries of polyunsaturated fatty acids were poor, sterols decreased but Δ5-avenasterol isomerisation was minimal. With a mild alkali hydrolysis (0.5 M NaOH/100°C/5 min) followed by methylation with boron trifluoride (14%BF3/100°C/30 min) recoveries of sterols and 2-hydroxy fatty acids were similar to the standard method and Δ5-avenasterol isomerisation was high. Lower ferulic acid recoveries, absence of solanidine and overestimation of fatty alcohols were evident in both methods involving alkali hydrolysis. Three different methods using hydrochloric acid (1.00 M HCl/70°C/5 h, 0.63 M HCl/110°C/2 h and 2.00 M HCl/50°C/24 h) all gave increased recoveries of 2-hydroxy acids, ferulic acids, solanidine and sterols, although Δ5-avenasterol isomerisation increased. Hydrochloric acid methods are recommended for studies requiring quantitative determinations (i.e. concentration of metabolite in sample). Either the hydrochloric acid methods or the standard sulphuric acid method are suggested for determining relative concentrations between samples, although there is a requirement for further studies.