Discrimination of Eugenia uniflora L. biotypes based on volatile compounds in leaves using HS-SPME/GC–MS and chemometric analysis

• It is a simple and efficient method for discrimination plant biotypes based on volatile compounds in leaves using HS-SPME/GC-MS and chemometric analysis
• The method was successfully applied to real samples of Eugenia uniflora L. biotypes and under the optimized extraction conditions 33 VOCs were identified.
• Principal component analysis (PCA) and hierarchical clustering analysis (HCA) provided a suitable tool to differentiate the volatile profiles in plants leaves with different kinds of fruit color biotypes.

The volatile profile of Eugenia uniflora leaves from different fruit colors biotypes was established using headspace solid-phase microextraction combined with gas chromatography–quadrupole mass spectrometry (HS-SPME/GC–MS). The optimization of the extraction conditions was carried out using multivariate strategies, which were factorial design and response surface methodology. The highest extraction efficiency was achieved with 0.1 g of leaf sample, using a Carboxen/Polydimethylsiloxane (CAR/PDMS, 75 μm) fiber for 60 min at 54 °C. After optimization, all samples were analyzed with the optimized extraction conditions that allowed for identify 33 VOCs, belonging to different chemical groups namely hydrocarbon monoterpenes, hydrocarbon sesquiterpenes, oxygenated monoterpenes, oxygenated sesquiterpenes and other compounds. The compounds found in higher relative abundance in leaves samples with biotypes orange, red and purple were germacrene B (9.55, 9.04 e 9.60%), γ-elemene (8.55, 8.14 e 7.97%), β-elemene (7.37, 4.99 e 9.26%), germacrene D (5.48, 6.08 e 6.46%), γ-muurolene (5.28, 7.67 e 5.02%) e β-caryophillene (7.09, 3.44 e 5.17%), respectively. Heatmap showed that the orange group was clearly distinct from the red and purple groups, with higher amounts of all hydrocarbon monoterpenes. While samples of red and purple biotypes presented great similarity in VOCs profile. Principal component analysis (PCA) and hierarchical clustering analysis (HCA) provided a suitable tool to differentiate the volatile profiles in plants leaves with different kinds of fruit color biotypes, pointing to the possibility of different varieties for this kind of plant.