Identification of Potential Host Plants of Sap-Sucking Insects (Hemiptera: Cicadellidae) Using Anchored Hybrid By-Catch Data

SIMPLE SUMMARY: Plant–insect interactions are a significant driver of terrestrial biodiversity. Evolutionary studies of such interactions have been hindered by the lack of reliable host plant records, which have primarily been obtained through field observations. More recently, traditional or next-generation sequencing methods have been used successfully to detect host plant DNA in DNA extracted from plant-feeding insects, but most such studies have focused on chewing insects that ingest plant tissues with large quantities of DNA. In this study, next-generation sequencing data were used to determine the feasibility of detecting plant genes in sap-sucking insects, which may ingest very little plant DNA. Although no plant-specific probes were used to generate the sequence data, multiple plant genes were retrieved in the by-catch data. Our results suggest that next-generation sequencing may present a powerful tool for detecting and characterizing potential host plants using DNA extracted from phytophagous insects, including sap feeders. ABSTRACT: Reliable host plant records are available for only a small fraction of herbivorous insect species, despite their potential agricultural importance. Most available data on insect–plant associations have been obtained through field observations of occurrences of insects on particular plants. Molecular methods have more recently been used to identify potential host plants using DNA extracted from insects, but most prior studies using these methods have focused on chewing insects that ingest tissues expected to contain large quantities of plant DNA. Screening of Illumina data obtained from sap feeders of the hemipteran family Cicadellidae (leafhoppers) using anchored hybrid enrichment indicates that, despite feeding on plant fluids, these insects often contain detectable quantities of plant DNA. Although inclusion of probes for bacterial 16S in the original anchored hybrid probe kit yielded relatively high detection rates for chloroplast 16S, the Illumina short reads also, in some cases, included DNA for various plant barcode genes as “by-catch”. Detection rates were generally only slightly higher for Typhlocybinae, which feed preferentially on parenchyma cell contents, compared to other groups of leafhoppers that feed preferentially on phloem or xylem. These results indicate that next-generation sequencing provides a powerful tool to investigate the specific association between individual insect and plant species.