Interval multivariate curve resolution in the dereplication of HPLC-DAD data from Jatropha gossypifolia.

INTRODUCTION Jatropha gossypifolia has been used quite extensively by traditional medicine for the treatment of several diseases in South America and Africa. This medicinal plant has therapeutic potential as a phytomedicine and therefore the establishment of innovative analytical methods to characterise their active components is crucial to the future development of a quality product. OBJECTIVE To enhance the chromatographic resolution of HPLC-UV-diode-array detector (DAD) experiments applying chemometric tools. METHODS Crude leave extracts from J. gossypifolia were analysed by HPLC-DAD. A chromatographic band deconvolution method was designed and applied using interval multivariate curve resolution by alternating least squares (MCR-ALS). RESULTS The MCR-ALS method allowed the deconvolution from up to 117% more bands, compared with the original HPLC-DAD experiments, even in regions where the UV spectra showed high similarity. The method assisted in the dereplication of three C-glycosylflavones isomers: vitexin/isovitexin, orientin/homorientin and schaftoside/isoschaftoside. CONCLUSION The MCR-ALS method is shown to be a powerful tool to solve problems of chromatographic band overlapping from complex mixtures such as natural crude samples.