Investigating untargeted metabolomics for its use in integrative taxonomy – Linking metabolomics, DNA marker-based se-quencing and bioimaging of phenotypes

Integrative taxonomy is a fundamental part of biodiversity and combines traditional morphology-based taxonomy with additional methods like DNA sequencing or biochemistry. Here, we aim to establish untargeted metabolomics methods for use in chemotaxonomy. We use three thallose liverwort species Riccia glauca, R. sorocarpa and R. warnstorfii (order Marchantiales, Ricciaceae) and Lunularia cruciata (order Marchantiales, Lunulariacea) as outgroup. Liquid chromatography high-resolution mass-spectrometry (UPLC/ESI-QTOF-MS) with data-dependent acquisition (DDA-MS) are integrated with DNA-marker based sequencing of the trnL-trnF region and high-resolution bioimaging. Our untargeted chemotaxonomy methodology allows to distinguish taxa based on chemophenetic markers at different levels of complexity: (1) molecules, (2) compound classes, (3) compound superclasses, and (4) molecular descriptors. We are aligning the results of the clade Riccia to the outgroup species Lunularia cruciata and are comparing the results at different levels with the information obtained from DNA sequencing. For the investigated Riccia species, we identified 71 chemophenetic markers at the molecular level, a characteristic composition in 21 compound classes, and 21 molecular descriptors largely indicating electron state, presence of chemical motifs and hydrogen bonds. Our untargeted approach revealed many chemophenetic markers that can provide more mechanistic insight into phylogenetic delimitation of species within a clade than genetic-based methods coupled with traditional morphology-based information.