Plant-pollinator interactions over time: Pollen metabarcoding from bees in a historic collection

Area of Science:

• Ecology
• Evolutionary Biology
• Genomics

Background:

• Pollination is crucial for agriculture and ecosystems, making plant-pollinator interactions a key research area.
• Natural history collections offer valuable historical samples for ecological and evolutionary studies.
• Identifying pollen on pollinators reveals past plant communities and pollinator feeding preferences.

Purpose of the Study:

• To assess the feasibility of using DNA metabarcoding to identify pollen from historic bee specimens.
• To evaluate the effectiveness of different DNA barcode regions (ITS1, ITS2, rbcL) for pollen identification.
• To explore the potential of historic collections for studying long-term ecological changes.

Main Methods:

• Pollen was sampled from 93-year-old bee specimens (Megachile venusta) from South Africa.
• DNA metabarcoding was performed on mixed pollen samples using next-generation sequencing (Illumina MiSeq).
• Sequenced reads were compared against reference databases (GenBank) for plant identification.

Main Results:

• Internal Transcribed Spacer (ITS) regions (ITS1 and ITS2) were successfully amplified, while rbcL was inconsistent.
• ITS2 provided more informative plant classification than ITS1.
• Age of the specimen did not affect sequencing success.
• Limited local plant sequence data in reference databases impacted classification confidence.

Conclusions:

• Pollen metabarcoding from historic bee specimens is a viable method for obtaining pollen DNA barcodes.
• DNA metabarcoding offers a rapid and accurate alternative to traditional palynology for determining pollen provenance.
• Historic collections can be leveraged for large-scale spatial and temporal ecological investigations, including climate change impacts.