Evolutionary metabolomics of specialized metabolism diversification in the genusNicotianahighlights allopolyploidy-mediated innovations inN-acylnornicotine metabolism

Abstract

Specialized metabolite (SM) diversification is a core process to plants adaptation to diverse ecological niches. Here we implemented a computational mass spectrometry (MS)-based metabolomics approach to explore SM diversification in tissues of 20 species covering Nicotiana phylogenetics sections. To drastically increase metabolite annotation, we created a large in silico fragmentation database, comprising more than 1 million structures, and scripts for connecting class prediction to consensus substructures. Altogether, the approach provides an unprecedented cartography of SM diversity and section-specific innovations in this genus. As a case-study, and in combination with NMR and MS imaging, we explored the distribution of N-acyl nornicotines, alkaloids predicted to be specific to Repandae allopolyploids, and revealed their prevalence in the genus, albeit at much lower magnitude, as well as a greater structural diversity than previously thought. Altogether, the novel data integration approaches provided here should act as a resource for future research in plant SM evolution.

Understanding the phytochemical diversity of plants: Quantification, variation and ecological function

“…Studies quantifying chemodiversity using a diversity index most often calculate diversity as a function of richness and evenness, by calculating the Shannon's or Simpson diversity. Similar to richness, variation in diversity has been documented at different levels of biological organization, including for different plant tissues (Eilers 2021, Elser et al 2022), populations (Bravo-Monzón et al 2014 and species (Ortiz et al 2019, Peguero et al 2021. There is also variation along altitudinal gradients (Glassmire et al 2016, Volf et al 2020, Philbin et al 2021) and for different levels/types of herbivory (Li et al 2020, Philbin et al 2022.…”

“…Phytochemical richness is a frequently reported measure. Variation in it has been documented on all levels of biological organization, including between tissues (Whitehead et al 2013, Elser et al 2022), individuals (Ziaja and Müller 2022), populations (Zeng et al 2022, Eisen et al 2022), species (Macel et al 2014, Züst et al 2020), orders (Courtois et al 2009), and communities (Peguero et al 2021), as well as across herbivory treatments (Agrawal 2000), phylogenies (Becerra et al 2009, Cacho et al 2015) and landscapes (Defossez et al 2021). Here, richness varies from a handful of compounds of a specific biosynthetic class, to several thousands metabolic features that are assumed to represent individual unidentified compounds.…”

Mass2SMILES: deep learning based fast prediction of structures and functional groups directly from high-resolution MS/MS spectra

“…We then also examined Mass2SMILES on a metabolomics dataset acquired for 20 Nicotiana species (Elser et al, 2022) to better judge of its performance on a real use case study. For this, we first predicted all the structures across the whole dataset.…”

“…Table 6. Selected predictions obtained for the metabolomics dataset on Nicotiana species (Elser et al, 2022). First, predicted SMILES were converted into molecular formulas and then compared with the predictions from SIRIUS, if consistent, SMILES were further converted with ClassyFire into chemical classes.…”

“…Some examples of class predictions that did not match with CANOPUS ones are additionally depicted. If blank, no annotation was generated in the study from Elser et al (2022). Numbers indicate the number of predicted aromatic hydroxyl groups, this number is not in line with the structure predicted by MSNovelist (bottom left).…”