Introduction: Two species of oak are dominant in French forests: pedunculate oak (Quercus robur L.) and sessile oak (Quercus petraea Liebl.). Differentiating oak species is difficult, since features such as morphological characters, geographical origin and grain are not always relevant. Even if the former is generally richer in tannin compounds while the latter is often richer in aromatic compounds, the intra-species variability is high. The characterisation of the oak species remains a suitable indicator of the molecular composition and quality of the wood.
Objectives: The aim of this study was to determine differentiating molecules allowing oak species identification in order to assist in a suitable wood selection for a better oak tree valorisation since the selection of the oak wood to be used in the production of barrels plays an essential role in wine ageing.
Materials and methods: Oak wood samples were collected both in forests and in cooperage timber yards. An untargeted metabolomic approach using ultra-high-pressure liquid chromatography qualitative time-of-flight high-resolution mass spectrometry (UHPLC-Q-TOF-HRMS) associated to multivariate statistical analyses (hierarchical ascendant clustering and partial least squares discriminant analysis) was implemented to determine molecular markers of oak species.
Results: Heartwood was identified as the suitable wood part to distinguish oak species. Discriminating molecules did not depend on the sample set. The pedunculate species showed overexpression of bartogenic derivatives while sessile oak presented a higher content in oak lactone precursors and in quercotriterpenosids.
Conclusion: The developed method allowed the identification of relevant compounds for oak species identification to a better wood valorisation and selection.
Keywords: mass spectrometry; multivariate statistical analysis; pedunculate oak; sessile oak; untargeted metabolomics.
© 2020 John Wiley & Sons, Ltd.