RESUMEN
An efficient HPTLC method was developed, which required minimal sample preparation for quantitation of the main anthocyanes in pomace, animal feed as well as various foods. The best separation of 11 anthocyanes was achieved on HPTLC plates silica gel 60 F254 with a mixture of ethyl acetate-2-butanone-formic acid-water for anthocyanins and ethyl acetate-toluene-formic acid-water for anthocyanidins. Due to the high flexibility of the HPTLC method, both anthocyane groups could be developed in a combined 2-step method. The second development was only necessary if anthocyanidins were detected in the samples. This normal phase separation was found superior to the best separation achieved on RP-18 phases with a mixture of water-n-propanol-formic acid. Absorbance measurement was performed using the multi-wavelength scan at 505 (or 510), 520, 530 and 555 nm. The correlation coefficients of the calibrations ranged between 0.9993 and 0.9999 for the 11 anthocyanes. LOQs were all ≤90 ng/zone, most even ≤30 ng/zone and for pn-3-glc and pg-3-glc even ≤7 ng/zone. With regard to the analysis of mv-3-glc in grape seed/marc meal and supplemented animal feed samples, the mean repeatabilities were 1.4% (laboratory 1) and 1.8% (laboratory 2). The intermediate precisions within a laboratory over several months were ≤6.7%. The ruggedness of the method was ≤5.5%. The method was transferred to other sample types. Juice and wine samples, which were from the same plant source, showed a comparable anthocyanin pattern, whereas the pattern was characteristically different between plant sources. Unknown anthocyanin sample components were analyzed via HPTLC-ESI-MS by eluting the zones of interest with the TLC-MS Interface, which was helpful for further characterization of unknowns. An interesting tool was demonstrated by effect-directed analysis with regard to radical scavenging properties and general bioactivity based on detection with Vibrio fischeri bacteria.