Botrytis cinerea: Coconut or Catastrophe? Quantification of γ-Nonalactone in Botrytised and Non-Botrytised New Zealand Wines

Abstract

g-Nonalactone has been identified as a significant contributor to the aroma profile of a range of wines and is associated with stonefruit and coconut descriptors. The exact route of g-nonalactone biosynthesis in wine has not been fully elucidated; however, precursors including 4-oxononanoic acid, linoleic acid, 13-hydroxyoctadeca-9,11-dienoic acid (13-HODE) and 9-hydroxyoctadeca-10,12-dienoic acid (9-HODE) have been identified in incubation experiments. Wines produced from grapes infected with “noble rot” caused by Botrytis cinerea fungus generally show higher concentrations of g-nonalactone compared to non-botrytised white wines, but the relative contribution of potential formation pathways has not been elucidated. To assess the effect of linoleic acid on the production of g-nonalactone in wine, fermentations with and without added linoleic acid were carried out in synthetic grape must (SGM) at 28 °C using commercial Saccharomyces cerevisiae EC1118. Prior to g-nonalactone quantitation in the finished wines and in a subset of six Australian and New Zealand commercial wines, several routes for the synthesis of a deuterated analogue of g-nonalactone were attempted, before the deuterated d6-analogue of g-nonalactone from its non-deuterated analogue was produced successfully. Subsequently, attempts were made to utilise the d6-analogue as an internal standard for the measurement of g-nonalactone using gas chromatography-mass spectrometry. However, the synthetic deuterated g-nonalactone analogue proved to be an inappropriate internal standard for this purpose, due to incomplete incorporation of deuterium atoms. 2-Octanol was instead used as a surrogate internal standard. g-Nonalactone was successfully identified (above the limit of detection, 4.12g/L) in two commercial New Zealand botrytised wine samples, and one fermentation sample to which linoleic acid (132 mg/L) had been added. This suggests a possible link between the effect of Botrytis cinerea and/or linoleic acid, and increased levels of g-nonalactone in wine. The promising results in these preliminary experiments have led to an improved internal standard being sought out for the quantification of g-nonalactone in wine, and further investigation of its biosynthesis. A 13C4-labelled g-nonalactone analogue was successfully synthesized based on a previous method, with four 13C atoms being introduced into the molecule via two Wittig olefination steps. This standard will be used for a much larger survey of approximately 40 botrytised and non-botrytised New Zealand wines, in addition to further fermentation experiments assessing the effects of the addition of a wider range of putative precursors.