BEEDET - Identification of wild bees and honey bees using non-lethal molecular methods

The decline of pollinating insects is now a proven fact reported in media and scientific journals. To halt this rapid and massive disappearance of pollinating insects, various actions have been taken in France, such as the Pollinators Plan (plan pollinisateurs) (2021-2026), and in Europe with the drafting of Article 8 of the European Commission (June 2022), which provides for “the obligation to reverse the decline of pollinators and achieve an upward trend in pollinator populations until satisfactory levels are reached. This obligation will be based on the implementation of a method for monitoring pollinators”.

However, at present, inventories and monitoring of pollinating insects face significant methodological barriers due to a lack of specialised entomologists and reliable, standardised methods for identifying species. In addition to the institutional obligations, there is also a strong societal demand for monitoring pollinator populations in different environmental contexts, using methods that preserve living specimens.

In this context, we have proposed developing a non-lethal protocol based on eDNA amplification and sequencing, that can be easily replicated by institutes or conservation associations to identify wild bees from the traces such as hairs, faeces or saliva, they leave on flowers when foraging.

Approaches

The experimental setup consisted in four strawberry plants, commonly used as pollinometers, placed in insect-proof enclosures or exposed to the open air for three days.

• In condition A, the strawberry plants enclosed with four bees that had been caught in a net, photographed and identified; These were from species Antophora plumipes, Andrena sp. and Lasioglossum sp.
• In condition B, the plants were exposed to the open ai
• In condition C, the strawberry plants were not in contact with pollinating insects, being protected by nets.

After three days, the flowers were harvested individually, then frozen and stored at -20°C before DNA extraction from traces.

Results

Optimisation of DNA extraction and amplification conditions

• Development of a protocol for extracting DNA from traces of insects deposited on flower
• Definition of 16S minibarcode amplification conditions suitable for several genera, starting from very small amounts of potentially degraded DNA.
• Optimisation of MiSeq sequencing conditions (read depth, cost).

Bioinformatic data processing

Prior to metabarcoding identification from our partially degraded DNA from environmental samples (eDNA), a reference barcode database for wild bees in the Occitanie region. Minibarcodes (250 bps insect 16S) had been built.

Two of the species introduced manually in condition A were detected: Antophora and Andrena (A). However, several species of wild bees belonging to different genera, as well as sequences from other pollinating insects such as hoverflies (Philantus), were also detected.

Sequences were detected under conditions C, in which the plants were protected from insects from the environment. This could be due to the fact that the strawberry plants used in this experiment came from external nurseries, where insects were present.

We also found a lot of contamination from Homo sapiens and unsurprisingly sequences corresponding to 16S DNA from strawberry plants. For further work, the effect of these contaminations will need to be limited, for example by using blocking oligonucleotides to reduce the amplification of contaminations.

The proof of concept is conclusive: under controlled conditions, it is possible to detect the passage of insects from foraged flowers.

INRAE units involved

• UMR DYNAFOR - Dynamiques et écologie des paysages agriforestiers
• UMR GenPhySE - Génétique, Physiologie et Systèmes d'Elevage