Proteomics screening for honey bee health


Moravcova R, Moon K-M, Stacey GR, Rogalski JC, Yuan X, Pernal SF, Guarna MM, Hoover S, Conflitti IM, Zayed A, Currie R, Giovenazzo P, Pepinelli M, Foster LJ (2023) Proteomics screening for honey bee health. 48th International Apimondia Congress, p. 51, 4-8 Sep 2024, Santiago, Chile.


Although honey bees are crucial to world agriculture, their health has declined over recent decades due to different stressors. Unfortunately, identifying the stressors impacting bee health without post-mortem analysis within a colony is difficult. By measuring changes in the transcriptome, proteome, and microbiome induced by known stressors, we aim to identify diagnostic markers to help develop a health assessment tool powered by stressor-specific biomarkers (BeeCSI). This study focuses on the proteomics contingent of a project for Canadian honey bee health monitoring and management. Laboratory and field studies were carried out in Canada across five provinces. Honey bees were naturally or experimentally exposed to different stressors, including parasites, pathogens, agrochemicals, nutrition restriction, and the eleven most common two-way combinations of stressors cross-compared for effects. Five biological replicates (colonies) per condition and eight crop systems (four of which were studied longitudinally over two years) were studied, totaling >2500 samples. Both control and affected bees were sampled with collected metadata. To obtain quantitative data, dissected, digested bee tissues (head, abdomen, gut) were analyzed by state-of-the-art mass spectrometry.
Data collected for the first subset (1377 of 2577 samples) includes treatments with five (of six) parasites and pathogens, five agrochemicals (themselves and within mixtures), two nutrition restriction stressors, and all eight crop systems. The data identified ~5800 protein groups with an average of ~4700 protein quantifications per sample, leading to the richest proteomics data set on honey bees to date. The first analyses show no significant effects on protein abundance due to provinces of origin. However, including cage/colony and tissue specificity highlighted differences between tested relationships (protein amount~dose+colony+tissue). When tested within specific tissues, we observed ~2300 (q<0.05) differentially expressed proteins across all experiments, where the maximum number of significant proteins within an experiment was 600. However, when testing without tissue specificity, only ~24 significant proteins were detected across all experiments, emphasizing the importance of tissue-specific analysis. Upon completion, we will compare the effects of stressors based on collected metadata (e.g., province origin, type of experiment – colony/cage, treatment dosage, disease severity), identifying significant stressor-specific markers and conditions to develop a new bee health assessment tool.

Date de publication