Use of ion mobility mass spectrometry to enhance cumulative analytical specificity and separation to profile 6-C/8-C-glycosylflavone critical isomer pairs and known-unknowns in medicinal plants.

INTRODUCTION: Plant medicine/herbal extracts are typically complex, encompassing a wide range of flavonoid diversity and biological benefits. Combined with a lack of standards; species authentication profiling is a challenge. A non-targeted screening strategy using two-dimensional (2D) separation and specificity of ultra-high-performance liquid chromatography ion mobility collision-induced dissociation mass spectrometry (UHPLC-IM-CID-MS) has been investigated, to identify the 6-C and 8-C-glycosylflavone isomer orientin/isoorientin and vitexin/isovitexin pairs in Passiflora species. Utilising available standards and "known-unknowns" a reference CCS (collision cross-section) speciation finger print for Passiflora extracts could be generated to illustrate species profiling.

MATERIAL AND METHODS: SPE was performed to extract flavonoids of interest from powdered and ground Passiflora leaf. Chromatographic separation was achieved via UHPLC and analysis performed using positive/negative ion electrospray coupled with linear T-wave IM-MS (calibrated to perform accurate mass and CCS measurements).

RESULTS: Comparative phytochemical screening of Passiflora alata, P. edulis, P. incarnata and P. caerulea leaf extracts has generated CCS, CID IM product ion spectra, 2D separation with UHPLC-IM-MS, enabling the unequivocal identification of flavone C-glycosides in complex extracts. A phytochemical reference CCS library was generated comprised of "knowns" and "known-unknowns". Isomers have been differentiated using a CCS metric enabling novel CCS specific isomeric quantitation of co-eluting isomers.

CONCLUSIONS: The screening approach illustrated has the potential to play an important role in the profiling of medicinal plants to determine phytochemical make-up and improve consumer safety through generation of highly specific speciation profiles.