Capabilities and limitations of using DNA metabarcoding to study plant-pollinator interactions

Area of Science:

• Ecology
• Genomics
• Entomology

Background:

• Pollinator populations are declining globally, necessitating better understanding of plant-pollinator relationships.
• Traditional methods for studying bee foraging behavior have limitations.
• DNA metabarcoding presents a promising alternative for identifying plant species in bee pollen loads.

Purpose of the Study:

• To compare DNA metabarcoding with observational methods for identifying plant species in bee pollen.
• To evaluate diet breadth characterizations derived from both methods.
• To assess the impact of regional versus local reference databases on DNA metabarcoding results.

Main Methods:

• DNA metabarcoding targeting ITS2 and rbcL gene regions.
• Analysis of pollen loads from 404 bees across three eastern Oregon habitats.
• Comparison of DNA metabarcoding data with direct foraging observations.

Main Results:

• 31%-86% consistency between DNA metabarcoding and observational data for plant species identification.
• 8%-50% divergence in diet breadth characterizations between the two methods.
• Regional databases identified non-native plant species in 22%-25% of detections.
• DNA metabarcoding revealed higher sampling completeness and lower specialization in plant-pollinator networks.

Conclusions:

• DNA metabarcoding is a valuable tool for studying plant-pollinator interactions, offering a more comprehensive understanding of bee diets.
• The choice of reference database (regional vs. local) impacts species identification accuracy.
• Further research should consider the strengths and limitations of DNA metabarcoding for ecological inferences.