Plants produce a diverse array of compounds through an extensive, evolutionarily malleable network of metabolic pathways. These metabolites are typically classified into two types-primary and specialized-with most of the diversity occurring among specialized metabolites (Fig. 1A). Metabolite presence-absence and types vary dynamically in an individual plant (within/between tissues/organs, developmental stages, across the circadian cycle), between populations, and also between species. Conservatively, it was estimated that each species harbors ~4.7 unique metabolites, and over a million metabolites were predicted to exist across the plant kingdom (Afendi et al., 2012).The scale of metabolite diversity raises multiple challenges for the study of plant specialized metabolism. First, it creates the challenge of identifying these compounds. While genomic and transcriptomic technologies have advanced significantly in the last decade, the complexity of metabolite structures and their chemical properties hinders adoption of standardized protocols for their detection. Second, the lineage-specific nature of specialized metabolism (Fig. 1A) creates the challenge of identifying biosynthetic enzymes and studying their evolution in non-reference species. Finally, there is still much to be understood about the ecological importance of chemical structural diversity in a single plant and across populations. In this article, we discuss these challenges and potential opportunities in more detail.
CHARACTERIZING THE VAST DIVERSITY OF PLANT METABOLITESPlant metabolites are typically detected using gas or liquid chromatography, coupled with mass spectrometry (GC-MS or LC-MS). GC-MS is appropriate for compounds that can be readily converted to gas phase, such as some terpenoids, sterols, fatty acids, and aromatic volatiles. It is typically used for low molecular weight compounds that are stable at high temperatures, either in their native form or after some chemical derivatization. LC-MS, on the other hand, is applicable for a much broader array of compounds and generally does not require any derivatization.Despite significant advances in MS technologies, several challenges remain for characterizing plant metabolite diversity. First, because of the large diversity of chemical structures, a variety of different protocols must be used to characterize the true "metabolome", i.e., the entire metabolite repertoire, of a plant, unlike the relative uniformity of experimental designs in transcriptomics or genomics. Various factors at the sample preparation, liquid chromatography, mass spectrometry (MS), and data analysis steps can influence which and how well metabolite peaks are detected. The second challenge involves identifying the thousands of metabolites observed in MS data from highly sensitive MS instruments. A recent study found evidence for >300 compounds of a single metabolite class (acylsugars) in a single leaf surface extract (Moghe et al., 2017).