The use of chemometrics to study multifunctional indole alkaloids from Psychotria nemorosa (Palicourea comb. nov.). Part I: Extraction and fractionation optimization based on metabolic profiling.

Extraction methods evaluation to access plants metabolome is usually performed visually, lacking a truthful method of data handling. In the present study the major aim was developing reliable time- and solvent-saving extraction and fractionation methods to access alkaloid profiling of Psychotria nemorosa leaves. Ultrasound assisted extraction was selected as extraction method. Determined from a Fractional Factorial Design (FFD) approach, yield, sum of peak areas, and peak numbers were rather meaningless responses. However, Euclidean distance calculations between the UPLC-DAD metabolic profiles and the blank injection evidenced the extracts are highly diverse. Coupled with the calculation and plotting of effects per time point, it was possible to indicate thermolabile peaks. After screening, time and temperature were selected for optimization, while plant:solvent ratio was set at 1:50 (m/v), number of extractions at one and particle size at ≤180μm. From Central Composite Design (CCD) results modeling heights of important peaks, previously indicated by the FFD metabolic profile analysis, time was set at 65min and temperature at 45°C, thus avoiding degradation. For the fractionation step, a solid phase extraction method was optimized by a Box-Behnken Design (BBD) approach using the sum of peak areas as response. Sample concentration was consequently set at 150mg/mL, % acetonitrile in dichloromethane at 40% as eluting solvent, and eluting volume at 30mL. Summarized, the Euclidean distance and the metabolite profiles provided significant responses for accessing P. nemorosa alkaloids, allowing developing reliable extraction and fractionation methods, avoiding degradation and decreasing the required time and solvent volume.