Using essential oil composition to discriminate between myrtle rust phenotypes in Eucalyptus globulus and Eucalyptus obliqua

Abstract Eucalyptus foliar terpenes have been reported widely to function as chemical defence agents against mammalian herbivores and defoliating insects both via direct toxicity and/or indirect priming of systemic defence mechanisms. Less frequently they have been reported as disease resistance biomarkers. The present study compares composition of foliar essential oils in Eucalyptus globulus Labill. and Eucalyptus obliqua L’Her. between distinct myrtle rust (causal pathogen Austropuccinia psidii Beenken) response phenotypes i.e. completely resistant, hypersensitive and highly susceptible, using canonical discriminant analysis. Stepwise forward variable selection identified six terpene compounds (bicyclogermacrene, globulol, geraniol, β-pinene, cis-p-menth-2-en-1-ol and δ-terpinene) in E. globulus and four terpenes (δ-cadinene, caryophyllene oxide, longifolenaldehyde and α-humulene) in E. obliqua, which in combination, significantly discriminated between rust response phenotypes. The discriminating compounds varied between E. globulus and E. obliqua and were different from those previously identified as rust resistance biomarkers in Eucalyptus grandis W.Hill ex Maiden × Eucalyptus urophylla S.T.Blake hybrids (1,8-cineole and α-terpinyl acetate). While the discriminant functions were statistically significant within each species, cross validation of the marker sets between Eucalyptus species was not significant. These findings indicate that either (a) terpene composition or amount is not a driver of resistance to myrtle rust or (b) that terpene contribution to resistance is through combined synergistic effects of variable terpenes, likely in combination with other plant metabolites or resistance mechanisms.