Co-exposure to a honeybee pathogen and an insecticide: Synergistic effects in a new solitary bee host but not in Apis mellifera

Pesticides and pathogens are major drivers of bee declines. However, their potential interactions are poorly understood, especially for non-Apis bees. This study assessed the combined effects of infestation by the honeybee pathogen Vairimorpha ceranae and chronic exposure to the insecticide flupyradifurone on Osmia bicornis and Apis mellifera. We investigated whether V. ceranae could reproduce in a new solitary bee host (O. bicornis) and assessed sublethal and lethal effects of the pathogen and the pesticide, alone and in combination. We also analyzed the interactive effects of the combined exposure on V. ceranae proliferation and bee survival in the two bee species. Newly-emerged bees were orally infected with 100,000 spores of V. ceranae and then exposed ad libitum to flupyradifurone at field-realistic concentrations. We showed, for the first time, that V. ceranae can replicate in the midgut of O. bicornis, causing histological damage, impaired phototactic response, reduced food consumption, and decreased longevity. The pathogen-pesticide combination caused a synergistic effect in O. bicornis, leading to an abrupt survival decline. In A. mellifera, V. ceranae and flupyradifurone showed antagonistic survival effects, but the pesticide promoted pathogen proliferation. Our results warn against the potential effects of pathogen spillover and multiple stressor exposure on non-Apis bees.

Methods
This dataset was collected during the study investigating the effects of co-exposure to a honeybee pathogen and an insecticide on solitary bees and Apis mellifera. Data were obtained through controlled laboratory experiments, where bees were exposed to different treatment conditions including oral exposure to the insecticide flupyradifurone and oral inoculation with spores of the pathogen Vairimorpha ceranae. Behavioral responses, survival rates, pathogen replication, and cellular alterations were recorded at predefined time points.

Following data collection, raw measurements were processed using R statistical software. Statistical analyses were conducted using generalized linear models (GLMs), survival analysis, and additional exploratory analyses were performed to assess potential interactions between factors.

All raw data, R code, and a README file detailing variable descriptions and data structure are included in the dataset repository.