Fig 1.
Pollen collection methods and timeline.
(A) Photo of a honeybee hive with pollen trap at the entrance. (B) Diagram of the pollen trap. These traps are designed to remove the pollen grain from the bees’ corbicula once they return from a foraging trip by forcing workers to enter the hive through a small grid that dislodges the pollen, which then falls into the collection basket. (C) A timeline of pollen collections during the Formic Acid or control treatment period. The photo depicts the placement of the pads in the hive. Half of the pads were soaked with Formic Acid and the other half were lightly soaked with water as a placebo. Pollen was collected three times: (1) April 26, 2022, before treatment application, (2) April 28, 2022, three days after treatment, and (3) May 5, 2022, seven days later, immediately before treatment removal.
Fig 2.
Relative read abundances of taxonomic composition from individual pollen samples throughout treatment application.
Relative abundance bar chart showing the taxonomic composition of plant genera identified from DNA sequences of pollen samples. The x-axis represents individual Sample IDs of different pollen collections: different locations of the hives on the University of Utah campus are marked with either an ‘H’ or an ‘K’, and the following number denotes the sample number (e.g., Ke1, Ke2, and Ke3 are replicates from the same hive). Each color represents a different plant genera identified in the pollen, and the size of the color for each sample indicates the relative abundance of the plant genus in relation to the entire sample. The chart is organized by different treatment groups, hives that received Formic Acid treatment and hives that received the control (placebo) treatment and is organized in the order the pollen was collected to allow comparison over time.
Fig 3.
Heatmap of plant genera detected from individual pollen samples throughout treatment application.
Heatmap illustrates the change in presence of plant genera per pollen sample across different collection dates. Color gradient changes from blue (for a higher percentage presence of a plant genus in a sample) to grey (a lower percentage presence). The chart is organized by different treatment groups (Formic Acid vs control) over time (‘Pre’, ‘During’, and ‘Post’).
Fig 4.
Non-metric multidimensional scaling (NMDS) ordination plots illustrate differences in foraging choices based on (A) treatment groups and (B) individual hives.
Distances were calculated using the Hellinger dissimilarity index. In these plots, points that are closer together represent more similar pollen compositions. Ellipses enclose a 95% confidence limit and NMDS stress values indicate how well the visual representation preserves the original distance relationships. Stress values below 0.2 indicate a good fit, meaning the NMDS plot reasonably represents the original data structure.
Fig 5.
Bar charts showing the taxonomic presence of plant genera identified from DNA sequences of pollen samples collected from honeybees after foraging.
Each of the plots represent individual honeybee hives before and after Formic Acid or control treatment. Hives (A) ‘Km’ and (B) ‘Hm’ represent the control group, while hives (C) ‘Ke’ and (D) ‘Ks’ receive Formic Acid treatment. For each hive, the left side of the plot represents pollen collected before treatment, and the right side represents pollen collected at the end of the treatment period. Bar length corresponds to the relative read abundance of each plant genus, log10-normalized. The y-axis lists all plant genera identified in the study. Difference between the left and right sides of each plot indicates changes in genera identified from the pollen following treatment.
Fig 6.
Shannon diversity index (SDI) and plant genera richness by treatment or individual hives.
Box plot diversity measurements are calculated using plant genera data identified from DNA sequences of pollen collected from honeybees under different treatments and from individual hives (‘Hive ID’). The dark line at the center indicates median value, the top and bottom of the box indicate 1st and 3rd quartile, respectively. Each dot is the calculated value of an individual pollen collection sample. The top row depicts the SDI (higher the value, lower overall diversity) between plot (A) the two treatment groups, Formic Acid Treatment and control hives before, during and post treatment and plot (B) by Hive ID in which all samples collected from each hive are included. The bottom row represents richness diversity, (the number of plant genera present in the group) of the two treatment groups, plot (C), and richness by Hive ID, plot (D). See S9 Table for comprehensive statistics extracted from boxplots in this figure. See S12 Table for post-hoc analysis results..