Metabolic markers for the yield of lipophilic indole alkaloids in dried woad leaves (Isatis tinctoria L.)

Abstract

The pharmacologically active dichloromethane extracts of dried woad leaves (Isatis tinctoria L.), and the methanol extracts of comparable fresh leaves of the same plants, were analyzed by LC-MSⁿ. The fresh leaf metabolite profile revealed a complex pattern of indolic compounds. Besides the known indigo precursors, isatan A, isatan B and indican, seven previously unreported indole derivatives were characterized: acetylindican, malonylindican, two dioxindole glucosides, dioxindole malonylglucoside, 6-hydroxyindole-3-carboxylic acid 6-O-glucoside and 6-hydroxyindole-3-carboxylic acid glucose ester. The integration of 122 compounds in fresh leaves and of five selected compounds (indoxyl, isatin, indigo, indirubin, and tryptanthrin) in dried leaves, formed the input data for a stepwise modelling procedure generating five predictive linear models. The structure of the predictive models and a cross validation provide evidence that the models could predict well or moderately well the accumulation of the selected lipophilic compounds, and were simple enough to be used in a woad cultivation program. PLS regression models relating each of the five selected dry leaf indolics to the fresh leaf metabolome were then fitted in order to deduct potential precursors and mechanisms leading to the formation of these lipophilic indolics in drying woad leaves. The models suggested glucobrassicin, isatan A and isatan B as the main candidate precursors of these compounds, besides a minor contribution of other fresh leaf indolics, including malonylindican, actylindican and dioxindole malonylglucoside. Dioxindole malonylglucoside was identified here as isatan C. The models further suggested that the accumulation of phenylpropanoid antioxidants in woad leaves has a negative impact on the formation of indoxyl, isatin, indigo, indirubin and tryptanthrin.

Keywords: Brassicaceae; Indigo; Isatin; Isatis tinctoria L.; LC-MS; Metabolic markers; PLS regression; Stepwise variable selection; Tryptanthrin.