In-field pollen consumption by orchard predatory mites detected by metabarcoded molecular gut content analysis

DNA metabarcoding has become a relatively common tool for studying consumption of plants by arthropod herbivores and identifying pollen collected by pollinating insects. However, studies using molecular gut content analysis to detect pollen-feeding in generalist predators are exceedingly rare. Pollen-feeding is known to be important to the nutritional ecology of many generalist predators and is particularly well studied in phytoseiid mites. In apples, generalist phytoseiids such as Typhlodromus caudiglans Schuster (Mesostigmata: Phytoseiidae) and the stigmaeid predator, Zetzellia mali (Ewing) (Trombidiformes: Stigmaeidae), feed on both pest mites and pollen trapped on foliage. It is unknown whether these mites commonly pollen-feed or if they exhibit in-field preferences for pollen type. To detect likely pollen-feeding in these predatory mites, we sampled tree foliage and the ground cover (vacuum sampling) from an apple orchard for 9 weeks in 2019 and 2020, and plant DNA was detected through metabarcoded gut content analysis using the trnF and internal transcribed spacer 2 (ITS-2) regions. We detected DNA from 31 plant genera in our samples. Malus was the most common taxon, followed by Pinus, Brassica, Festuca, and Lactuca. Forty-six percent of phytoseiid and 34% of Z. mali samples were positive for non-Malus DNA. Changes in detection of several taxa (Pinus, Festuca, Hyoscyamus, Citrullus, Poa, and Solanum) throughout the monitoring period likely reflected temporal variation in pollen availability. Pollen availability may play a role in determining the composition of predatory mite communities. Metabarcoding to detect possible pollen-feeding has the potential to improve our understanding of the role of floral resources in conservation biological control.