RESUMO
A headspace solid phase microextraction (HS-SPME) method combined with gas chromatography-mass spectrometry (GC/MS) was developed and optimized for extraction and analysis of volatile organic compounds (VOC) of leaves and galls of Myrcia splendens. Through a process of optimization of main factors affecting HS-SPME efficiency, the coating divivnilbenzene-carboxen-polydimethylsiloxane (DVB/Car/PDMS) was chosen as the optimum extraction phase, not only in terms of extraction efficiency, but also for its broader analyte coverage. Optimum extraction temperature was 30°C, while an extraction time of 15min provided the best compromise between extraction efficiencies of lower and higher molecular weight compounds. The optimized protocol was demonstrated to be capable of sampling plant material with high reproducibility, considering that most classes of analytes met the 20% RSD FDA criterion. The optimized method was employed for the analysis of three classes of M. splendens samples, generating a final list of 65 tentatively identified VOC, including alcohols, aldehydes, esters, ketones, phenol derivatives, as well as mono and sesquiterpenes. Significant differences were evident amongst the volatile profiles obtained from non-galled leaves (NGL) and leaf-folding galls (LFG) of M. splendens. Several differences pertaining to amounts of alcohols and aldehydes were detected between samples, particularly regarding quantities of green leaf volatiles (GLV). Alcohols represented about 14% of compounds detected in gall samples, whereas in non-galled samples, alcohol content was below 5%. Phenolic derived compounds were virtually absent in reference samples, while in non-galled leaves and galls their content ranged around 0.2% and 0.4%, respectively. Likewise, methyl salicylate, a well-known signal of plant distress, amounted for 1.2% of the sample content of galls, whereas it was only present in trace levels in reference samples. Chemometric analysis based on Heatmap associated with Hierarchical Cluster Analysis (HCA) and Principal Component Analysis (PCA) provided a suitable tool to differentiate VOC profiles in vegetal material, and could open new perspectives and opportunities in agricultural and ecological studies for the detection and identification of herbivore-induced plant VOC emissions.
