Honey bee propolis is a complex, resinous mixture created by bees using plant sources such as leaves, flowers, and bud exudates. This study characterized how cropland surrounding apiaries affects the chemical composition and antimicrobial effects of propolis. The chemical composition and compound abundance of the propolis samples were analyzed using Gas Chromatography-Mass Spectrometry (GC-MS) and the antimicrobial effects were analyzed using the 50% minimum inhibitory concentration (MIC50) assay against four relevant bee pathogens, Serratia marcescens, Paenibacillus larvae, Lysinibacillus sphaericus, and Klebsiella pneumoniae. Propolis composition varied significantly with apiary, and cropland coverage predicted mean sum abundance of compounds. The apiary with the highest cropland coverage exhibited significantly higher MIC50 values for S. marcescens and K. pneumoniae compared to other apiaries. These results demonstrate that agricultural land use surrounding honey bee apiaries decreases the chemical quality and antimicrobial effects of propolis, which may have implications for the impacts of land use on hive immunity to potential pathogens.
Propolis Collection and Extraction
Propolis was collected from five different apiaries of a commercial beekeeping operation in Fayette and Clayton counties in Iowa. Samples were collected from 5–10 colonies at each location, yielding a total of 45 samples. Propolis was collected from the front rim of the top lid of each hive. This location on the hive was chosen because it is the area of most active recent propolis placement by honey bees, due to frequent beekeeper disturbance. Each propolis sample was collected into sterile plastic tubes and stored at −80 °C until further use. Propolis samples were then pulverized with a mortar and pestle, and diluted in 100% ethanol (0.2 g propolis in 1.8 g ethanol). Samples were then mixed at 500 rpm at 30 °C for 48 h in the dark. Samples were filtered using a 0.22 μm syringe filter (Fisher Scientific, Hampton, NH, USA). Filtered samples were then diluted according to specific needs for chemical or antimicrobial analysis.
The antimicrobial effects of the propolis samples were measured by performing a MIC50
assay, according to Wiegand et al. [34
] with modifications. Briefly, propolis samples were serially diluted with phosphate buffered saline (PBS) to yield a diluted range of 0.1 mg propolis/mL to 17 mg propolis/mL, a range previously observed to be inhibitory to bacterial growth [7
]. Ethanol serially diluted in PBS was used as the negative control. Fifty μL of each propolis sample or negative control dilution was then plated on a 96 well plate in triplicates.
Paenibacillus larvae (ATCC 9545) and Lysinibacillus sphaericus (ATCC 4525) were obtained from American Type Culture Collection (ATCC, Manassas, VA, USA), and Serratia marcescens and Klebsiella pneumoniae were obtained from Carolina Biological Supply Company (Burlington, NC, USA). Bacterial suspensions were prepared by transferring two colonies of bacteria from agar media into tryptic soy broth (S. marcescens and K. pneumoniae) or brain heart infusion broth (P. larvae and L. sphaericus). Viability of bacteria was determined on a hemocytometer using erythrosine B. A 1 × 105 cells/ml bacterial solution was then prepared using the appropriate liquid broth medium. Next, 100 μL of each bacteria suspension was added per well in a 96 well plate. The plates were then incubated at 30 °C (S. marcescens and L. sphaericus) and 37 °C (K. pneumoniae and P. larvae) for 24 h. Following incubation, 0.75 μL from each well was diluted in 50 μL of PBS and transferred to a well in a 12 well plate with the appropriate agar media for each bacterium. These plates were then incubated at 30 °C (S. marcescens and L. sphaericus) and 37 °C (K. pneumoniae and P. larvae) for 24 h. Colonies of bacteria were then counted and the MIC50 of each sample for each bacterium was determined as the lowest propolis concentration to inhibit 50% of the bacterial growth.
Effectiveness of Propolis against Pathogens
In this analysis, we constructed a series of linear models between apiary and MIC50 scores, using apiary as the explanatory variable and the MIC50 score for each hive’s propolis against each pathogen as the response variable (resulting in one linear model for each pathogen). In the event of significant models, we used a Tukey HSD test to identify which apiaries were significantly different from one another.