Ecological release from natural enemies is thought to facilitate biological invasions. For invasive plants, ecological release is hypothesized to occur because introduced plants contain novel secondary chemicals that are repellent to herbivores in the introduced range. We take advantage of novel methods for acquiring and assembling mass spectra (MS) into molecular networks in which similarities in MS fragmentation patterns, and hence chemical structure, are indicated by proximity of compounds within the network. Molecular networks permit the quantification of chemical similarities between samples for which few compounds have been unambiguously identified.
We use metabolomics based on MS molecular networks to assess the similarity of foliar secondary chemistry among 47 native and 18 introduced, invasive plant species that occur at the Smithsonian Environmental Research Center (SERC) in Maryland. These 65 species also widely co-occur in eastern North America. In addition, we estimated the invasiveness of 18 introduced species by recording their occurrence frequency in plots censused by the USDA Forest Inventory and Analysis (FIA) survey. We asked whether invasive plant species were chemically distinct from the native flora, and whether chemical distinctiveness was associated with variation in invasiveness as reflected by occurrence frequency over a 22-state region in eastern North America. Finally, by drawing on published data concerning feeding preference of a generalist caterpillar and the effect of whitetail deer exclosures on plant abundance, we asked whether chemically novel plants are avoided by generalist insect and mammalian herbivores.
Results/Conclusions
Invasive species were less chemically similar than native species to the native plant community at SERC, Maryland, and exhibited greater foliar N, P, and water content and lower leaf toughness as compared to native species. However, we found no difference in caterpillar feeding preference between native and invasive species, nor did we find an association between chemical novelty and herbivore feeding preference. We observed an association between occurrence frequency and chemical novelty only when excluding the extreme outlier Rosa multiflora.
Our results suggest that chemical novelty in the communities to which they are introduced facilitates invasion by some introduced species. However, our results do not support the prediction that avoidance of generalist herbivores is the primary mechanism by which chemical distinctiveness facilitates invasion. This may be due to the presence of multiple, alternative mechanisms that facilitate invasion, as suggested by the runaway success of R. multiflora, despite its chemical similarity to native plants and susceptibility to native generalist herbivores.