PLoS ONE (Jan 2017)
Whitefly attraction to rosemary (Rosmarinus officinialis L.) is associated with volatile composition and quantity.
Abstract
Whitefly (Bemisia tabaci) is an important insect pest, causing severe damage to agricultural crops. The pest was recorded in a commercial rosemary (Rosmarinus officinalis, Lamiaceae) field, colonizing rosemary variety (var.) '2', but not '11'. A series of field and controlled laboratory choice bioassays confirmed the observed phenomenon. Mature potted plants of the two varieties were randomly organized in a lemon verbena (Lippia citrodora) and lemon grass (Cymbopogon spp.) fields. Seven days later var. '2' was significantly more colonized by whiteflies than var. '11'. Under lab conditions, whiteflies were significantly more attracted to var. '2' plantlets than to var. '11' following choice bioassays. Furthermore, cotton plants dipped in an essential oil emulsion of var. '2' had significantly greater colonization than cotton plants dipped in the essential oil emulsion of var. '11'. Similar results were obtained in 'plant-plant', 'plant-no plant' as well as, 'essential oil-essential oil' choice bioassay designs. Analyses of the essential oils of the two varieties identified a set of common and unique volatiles in each variety. Among these volatiles were β-caryophyllene and limonene, two compounds known to be associated with plant-insect interactions. The attraction of B. tabaci to pure (>95%) β-caryophyllene and limonene using a range of concentrations was examined in vitro by choice bioassays. The compounds were attractive to the insect at moderate concentration, but not at the lowest or highest concentrations used, where the insect was not attracted or repelled, respectively. Limonene attracted the insects at rates that were 10-fold lower than β-caryophyllene. The results emphasized the role of host plant volatiles in shaping the structure of B. tabaci populations in nature and in agricultural systems, and provided insights into the factors that contribute to the development of insect populations with unique characteristics. The results could also serve for future development of bio-pesticides and in breeding programs.