Dereplication of Flavonoid Glycoconjugates from Adenocalymma imperatoris-maximilianii by Untargeted Tandem Mass Spectrometry-Based Molecular Networking

The interpretation of large datasets acquired using high performance liquid chromatography coupled with tandem mass spectrometry represents one of the major challenges in natural products research. Here we propose the use of molecular networking to rapid identify the known secondary metabolites from untargeted MS/MS analysis of Adenocalymma imperatoris-maximilianii plant extracts. The leaves, stems and roots of A. imperatoris-maximilianii were extracted using different solvents according to Snyder selectivity triangle. The samples were analyzed by HPLC coupled with ion trap mass spectrometer in a collision-induced dissociation MS/MS configuration in both positive and negative electrospray ionization modes. Molecular networking simultaneously organized the spectra by cosine similarity. The chemical identification was performed based on the systematic study of the main fragmentation pathways observed for the resulting network. The untargeted tandem mass spectrometry-based molecular networking allowed for the identification of 63 metabolites, mainly mono-, di-and tri-, C-and/or O-glycosyl flavones. Molecular networking was capable not only to dereplicate known flavonoids, but also to point out related prenyl derivatives, described for the first time in Adenocalymma species. The gasphase reaction route to form the characteristic [M-H2O-(30/60/90)](+) fragments in C-glycosyl flavones was suggested as sequential sugar ring opening followed by retro-aldol elimination involving aldose-ketose isomerization. The use of molecular networking with LC-CID-MS/MS assisted the identification of various isomeric and isobaric flavonoid glycoconjugates by establishing clusters according to the fragmentation similarities. Additionally, the proposed cross-ring sugar cleavages can contribute to the identification of C-glycosides by MS/MS analysis.