Viruses are ubiquitous within all habitats that support cellular life and represent the most important emerging infectious diseases of plants. Despite this, it is only recently that we have begun to describe the ecological roles of plant viruses in unmanaged systems and the influence of ecosystem properties on virus evolution. We now know that wild plants frequently harbor infections by diverse virus species, but much remains to be learned about how viruses influence host traits and how hosts influence virus evolution and vector interactions. To identify knowledge gaps and suggest avenues for alleviating research deficits, we performed a quantitative synthesis of a representative sample of virus ecology literature, developed criteria for expanding the suite of pathosystems serving as models, and applied these criteria through a case study. We found significant gaps in the types of ecological systems studied, which merit more attention. In particular, there is a strong need for a greater diversity of logistically tractable, wild dicot perennial study systems suitable for experimental manipulations of infection status. Based on criteria developed from our quantitative synthesis, we evaluated three California native dicot perennials typically found in Mediterranean-climate plant communities as candidate models: Cucurbita foetidissima (buffalo gourd), Cucurbita palmata (coyote gourd), and Datura wrightii (sacred thorn-apple). We used Illumina sequencing and network analyses to characterize viromes and viral links among species, using samples taken from multiple individuals at two different reserves. We also compared our Illumina workflow with targeted RT-PCR detection assays of varying costs. To make this process accessible to ecologists looking to incorporate virology into existing studies, we describe our approach in detail and discuss advantages and challenges of different protocols. We also provide a bioinformatics workflow based on open-access tools with graphical user interfaces. Our study provides evidence that dicot perennials in xeric habitats support multiple, asymptomatic infections by viruses known to be pathogenic in related crop hosts. Quantifying the impacts of these interactions on plant performance and virus epidemiology in our logistically tractable host systems will provide fundamental information about plant virus ecology outside of crop environments.

Plant viruses are ubiquitous throughout plant communities, but research on viral impacts largely focuses on crops. Little is known about how viruses influence wild plants in their native habitats. To address this gap, we examined virus interactions with wild drought-tolerant perennials in California desert natural areas encroached upon by agriculture. We used metagenomics, targeted diagnostics, and phylogenetics to assess virus diversity and clade relationships, and experiments to investigate viral influence on hosts. We focused on three herbaceous perennials ( Cucurbita foetidissima, C. palmata, and Datura wrightii) and tested the hypothesis that these wild species accumulate virus infections typically found in crops and transmitted by polyphagous insects. We predicted that such infections might be retained across seasons and potentially impair plant performance. Virome profiling revealed a rich community of previously characterized virus species (12 total), with virus community structure varying by site and host species. The dominant viruses in the wild hosts were non-native crop pathogens, including cucurbit aphid-borne yellows virus (CABYV) and cucurbit yellow stunting disorder virus (CYSDV). Targeted testing revealed that CABYV infected as many as 88% of sampled wild Cucurbita individuals, with dual CABYV–CYSDV infections common in natural areas adjacent to desert agriculture. CABYV infections reduced shoot and root production in greenhouse experiments with the two wild Cucurbita species. Phylogenetic analyses suggest that CABYV was introduced to California multiple times from other continents. Our findings provide concerning evidence of ways in which human activities can alter virus pressure on wild plants and potentially contribute to plant decline.