A multivariate analysis approach to the study of chemical and functional properties of chemodiverse plant derivatives: lavender essential oils

Six lavender essential oils, L. angustifolia and five Lavandula x hybrida cultivars (Super Z, Abrialis, R.C., Alardii and Ordinario), were evaluated from a phytochemical and biological standpoint, and the results were computed by using multivariate data analysis. The essential oils were analysed by gas chromatography, gas chromatography–mass spectrometry and headspace gas-chromatography. Multivariate analyses (principal component analysis) identified three main phytochemical clusters among lavender essential oils, represented by 1,8-cineole, linalyl acetate and linalool. Functional properties of the essential oils were checked by (1) estimating cytotoxicity and genotoxicity using the Saccharomyces cerevisiae D7 strain; (2) determining antifungal activity against three common phytopathogens (Pythium ultimum, Magnaporthe grisea and Botrytis cinerea), by performing an agar vapour bioassay; and (3) calculating the antioxidant capacity by using the 1, 1-diphenyl-2-picrylhydrazyl (DPPH) and b-carotene bleaching tests. No mutagenic effects were detected, but multivariate analyses (PLS, partial least squares regression) showed that the essential oils belonging to the linalool cluster were the most cytotoxic. Antifungal activity against phytopathogens confirmed the predictive results of PLS. The differences among lavender essential oils regarding weak antioxidant capacity showed a positive relationship between the high polarity compounds and the DPPH method, as determined by PLS. The opposite effect was shown for the same type of compound and b-carotene. Ketones and esters did not exert any significant antioxidant activity. In conclusion, taking lavender essential oils as a model and computing multivariate data of a reduced number of parameters, the proposed approach assured the description of the relationship between a phytocomplex, its constituents and bioactivities, and allowed a comprehensive, predictive approach to be defined, for which the chemical profile provides a possible synergic overall effort in terms of applicative perspectives