Chemical Profile and Antibacterial Activity in Hydrodistilled Oil from Aerial Parts of Prangos ferulacea (L.) Lindl. and Prediction of Gas Chromatographic Retention Indices by Using Genetic Algorithm Multiple Linear Regressions

Abstract

The essential oil obtained by hydrodistillation from aerial parts of Prangos ferulacea growing wild plant in the marginal brackish regions of Semnan province from Iran was analyzed by means of GC and GC/MS instrumentations. The main components were b-phellandrene (20.39 %), a-terpinolene (15.26 %), a-pinene (11.59 %), d-3-carene (11.06 %), a-phellandrene (9.09 %), trans-b-ocimene (9.67 %). The oil shows antibacterial activity against Escherichia coli and Staphylococcus saprophytacus with minimum inhibitory concentration values of 3.27 and 8.19 μg/mL, respectively. Furthermore, quantitative structure-property relationships were developed using a genetic algorithm variable-selection approach. The correlation between the experimental and calculated Kovats indices (R²_train = 0.95, REP=1.5; R²_test = 0.9952, REP=0.9) as well as reliable cross validation results (Q²_LOO = 0.98, Q²_LGO = 0.96) demonstrate the high applicability of the proposed model to predict RIs of a diverse set of natural compounds.