Chemical profiling and classification of cannabis through electrospray ionization coupled to Fourier transform ion cyclotron resonance mass spectrometry and chemometrics
Article
Borille, BT, Ortiz, RS, Mariotti, KC et al. (2017). Chemical profiling and classification of cannabis through electrospray ionization coupled to Fourier transform ion cyclotron resonance mass spectrometry and chemometrics
. ANALYTICAL METHODS, 9(27), 4070-4081. 10.1039/c7ay01294b
Borille, BT, Ortiz, RS, Mariotti, KC et al. (2017). Chemical profiling and classification of cannabis through electrospray ionization coupled to Fourier transform ion cyclotron resonance mass spectrometry and chemometrics
. ANALYTICAL METHODS, 9(27), 4070-4081. 10.1039/c7ay01294b
Cannabis sativa is chemically characterized as containing terpenophenolic structures, named cannabinoids, which are exclusively found in this plant. In Brazil, the international trafficking of small amounts of cannabis seeds by transport companies has significantly grown in recent years. In this context, combining the chemical profiling data of cannabis with chemometric techniques provides investigative forces with information towards interrupting such illegal activities. In this paper, 68 samples of cannabis seeds from seizures performed by the Brazilian Federal Police were germinated, planted, and cultivated under controlled conditions in a greenhouse, and analyzed by positive and negative electrospray ionization coupled to Fourier transform ion cyclotron resonance mass spectrometry (ESI(±)-FT-ICR MS and ESI(±)MS/MS) techniques considering different growth periods. The chemical profiling using ESI(+)FT-ICR MS enabled the detection of 123 species as cannabinoid compounds or metabolites and 8 non-cannabinoid constituents. The multivariate techniques applied to FT-ICR MS data yielded satisfactory results to predict the plant growth time by means of a combination of the genetic algorithm with partial least squares regression (GA-PLS). The combined information of the positive and negative modes allowed the construction of the PLS regression model with a prediction error of approximately 1 week to determine the growth time of the plants.
