Plant–animal interactions in the era of environmental DNA (eDNA)—A review

Plant–animal interactions (PAI) represent major channels of energy transfer through ecosystems, where both positive and antagonistic interactions simultaneously contribute to ecosystem functioning. Monitoring PAI therefore increases the understanding of environmental health, integrity, and functioning, and studying complex interactions through accurate, cost-effective sampling can aid in the management of detrimental anthropogenic impacts. Environmental DNA (eDNA)-based monitoring represents an increasingly common, nondestructive approach for biodiversity monitoring, which could help to elucidate PAI. Here, we aim to provide an overall discussion on the potential of using eDNA to study PAI. We assessed the existing literature on this subject from 2009 to 2021 using a freely accessible web search tool. The search was conducted by using keywords involving eDNA and PAI, including both species-specific and metabarcoding approaches, recovering 43 studies. We summarized the advantages and current limitations of such approaches, and we outline research priorities to improve future eDNA-based methods for PAI analysis. Among the 43 studies identified using eDNA to measure PAI such as pollination, herbivory, mutualistic, and parasitic relationships, they have often identified higher taxonomic diversity in several direct comparisons with DNA-based gut/bulk sampling and conventional survey methods. Research needs include the following: better understanding of the influencing factors of eDNA detection involved in PAI (e.g., eDNA degradation, origin, and types), methodological standardization (sampling methods and primer development), and more inclusive sequence reference databases. If these research priorities are addressed, it will have a significant impact to enable PAI biodiversity monitoring with eDNA. In the future, the implementation of eDNA methods to study PAI can particularly benefit the scalability of environmental biomonitoring surveys that are imperative for ecosystem health assessments.