Spatial and evolutionary predictability of phytochemical diversity

To cope with environmental challenges, plants produce a wide diversity of phytochemicals, which are also the source of numerous medicines. Despite decades of research in chemical ecology, we still lack an understanding of the organization of plant chemical diversity across species and ecosystems. To address this challenge, we hypothesized that molecular diversity is not only related to species diversity, but also constrained by trophic, climatic, and topographical factors. We screened the metabolome of 416 vascular plant species encompassing the entire alpine elevation range and four alpine bioclimatic regions in order to characterize their phytochemical diversity. We show that by coupling phylogenetic information, topographic, edaphic, and climatic variables, we predict phytochemical diversity, and its inherent composition, of plant communities throughout landscape. Spatial mapping of phytochemical diversity further revealed that plant assemblages found in low to midelevation habitats, with more alkaline soils, possessed greater phytochemical diversity, whereas alpine habitats possessed higher phytochemical endemism. Altogether, we present a general tool that can be used for predicting hotspots of phytochemical diversity in the landscape, independently of plant species taxonomic identity. Such an approach offers promising perspectives in both drug discovery programs and conservation efforts worldwide.

Phytochemical diversity describes the richness and abundance of the specialized metabolites produced by vegetation. It is a key aspect of plant functional diversity and, thus, affects plant fitness (1), ecosystem functioning (2), and services to humankind (3). Despite its relevance, chemical ecologists still struggle to understand both the evolutionary origin of phytochemical diversity and its variation across ecosystems (4). Only a small fraction of the >300,000 currently described phytochemicals (5) has been ascribed to a known ecosystem function or process (6). This is because most identification work has been undertaken on model organisms, such as crop plants (7), and because drug discovery programs have so far been based on prior ethno-medicinal knowledge or random sampling, rather than systematic sampling from the tree of life (8) or guided by ecologically relevant information (9). The ability to better predict the presence and diversity of phytochemicals of interest from phylogenetic information, or from specific environments or habitat types, could uncover the full spectrum and function of phytochemicals in the landscape while also orienting drug discovery research (10). Moreover, documenting landscape variability in phytochemical diversity is particularly important in the context of land use change, which is causing losses of plants that possess a yet-unknown value to medicine and science (11).The plant metabolome includes both primary functions, expected to be conserved across species, and specialized functions, associated to specific lineages or environments (1). Thus, phytochemical variation in the landscape is expected to arise from a combination of evolutionary (12, 13) and ecological (14, 15) constraints. From a macroevolutionary standpoint, some classes of phytochemical compounds are specific to plant clades (e.g., glucosinolates in Brassicaceae, or tropane alkaloids in Solanales; ref. 16). Such lineage-dependent variation is thought to be driven by chemical defense innovations followed by coevolutionary dynamics with herbivores (17, 18). In particular, the escape-and-radiate model (13) predicts that plant lineages diversify by creating novel, more potent, or complex chemical mixtures in response to biotic pressure (19). Therefore, plant lineages that have experienced more evolutionary split events are predicted to have evolved higher levels of phytochemical diversity (13). From an ecological perspective, phytochemical diversity is expected to be the result of plant adaptation to abiotic and biotic conditions, both of which vary along ecological gradients in landscapes (2, 20). For example, habitats that impose constraints on plant growth, such as cold and resource-poor environments, may be expected to drive selection toward potent chemical defense mechanisms that reduce tissue loss (21). At the same time, it is well established that herbivores and pathogens can promote divergent selection between plant congeners, leading to increasing chemical dissimilarity (22). As such, species relatedness alone is a poor predictor of site-level phytochemical diversity.Here, we questioned whether phytochemical diversity can be predicted from the phylogenetic and ecological heterogeneity observed in the landscape. We hypothesized that phytochemical diversity is not only related to local plant species diversity but is also constrained by other ecological factors, especially trophic, climatic, edaphic, and topographic variation. We developed a methodological framework involving: 1) comprehensive sampling of plant species along ecological transects that cover the entire range of regional vegetation ecological boundaries; 2) assessing species-level phytochemical composition and combining it with species distribution models (SDMs) for extrapolating phytochemical diversity across the landscape based on species occurrences; 3) extracting climatic and topographical variables associated with each unique molecule observed across all species to build molecular distribution models (MDMs); and 4) projecting phytochemical diversity and composition across the landscape based on these MDMs (SI Appendix, Fig. S1). Here, we consider phytochemical diversity both as the richness of clustered metabolic features and the presence/absence of the families of compounds they represent. We expected that phytochemical diversity values compiled from the projected MDMs would better explain plot-level phytochemical diversity and composition than phytochemical diversity calculated from plant species composition alone.