Given the complexity of the plant metabolome exhibiting a wide spectrum of physico-chemical properties, finding the best compromise for GC–MS analysis is a challenging exercise. In this study, the GC–MS protocol for extracting and analyzing polar metabolites from apple tissue is optimized.We found pure methanol extraction to be slightly better as compared to the two phase methanol/chloroform/water extraction in terms of introducing less degradation of the extracted metabolites while being comparable
in extraction efficiency and repeatability. The methanol extraction was superior to the single phase methanol/chloroform/water extraction in all performance measures. A multi-response optimization based on a desirability function was applied to optimize the derivatization. This procedure allows searching for optimal parameters while simultaneously considering overall detection enhancement of metabolites from various metabolic classes. A short oximation at a high temperature in combination with a low silylation temperature gave results similar to a longer oximation at low temperature in
combination with a high silylation temperature. Increasing silylation time from 0.5 h to 2 h resulted in an
improvement of the silylation reaction.