Sex-specific differences in pathogen susceptibility in honey bees (Apis mellifera)

Abstract

Sex-related differences in susceptibility to pathogens are a common phenomenon in animals. In the eusocial Hymenoptera the two female castes, workers and queens, are diploid and males are haploid. The haploid susceptibility hypothesis predicts that haploid males are more susceptible to pathogen infections compared to females. Here we test this hypothesis using adult male (drone) and female (worker) honey bees (Apis mellifera), inoculated with the gut endoparasite Nosema ceranae and/or black queen cell virus (BQCV). These pathogens were chosen due to previously reported synergistic interactions between Nosema apis and BQCV. Our data do not support synergistic interactions between N. ceranae and BQCV and also suggest that BQCV has limited effect on both drone and worker health, regardless of the infection level. However, the data clearly show that, despite lower levels of N. ceranae spores in drones than in workers, Nosema-infected drones had both a higher mortality and a lower body mass than non-infected drones, across all treatment groups, while the mortality and body mass of worker bees were largely unaffected by N. ceranae infection, suggesting that drones are more susceptible to this pathogen than workers. In conclusion, the data reveal considerable sex-specific differences in pathogen susceptibility in honey bees and highlight the importance of ultimate measures for determining susceptibility, such as mortality and body quality, rather than mere infection levels.

Figure 1. Cumulative survival of drones and workers during the 14 days of the cage trial.

Bees that were terminated on day 14 were treated as censured in the analyses. Drones showed significant higher mortality in the treatment groups that were challenged with N. ceranae (Ps = 0.037), indicated by different letters (a, b) in the figure. Workers of the different treatment groups showed no differences in mortality. The groups with BQCV-inoculation differ from the other groups in terms of treatment from day 7 onwards.

Figure 2. Body mass of drones and workers of the four treatment groups on the last day of the cage trial after 14 days.

The boxplots show interquartile range (box), median (black line within the interquartile range), data range (dashed vertical lines) and outliers (open dots). Significant differences (all Ps<0.05) were detected between the N. ceranae and control (*  = P<0.05) as well as N. ceranae and BQCV (**  = P<0.01) group in the drones as indicated by black horizontal bars between the respective treatment groups under the boxplots. No differences were detected between the treatment groups of the workers.

Figure 3. N. ceranae spores per bee in drones and workers of the four treatment groups after termination of the bees on day 14.

The boxplots show interquartile range (box), median (black line within the interquartile range), data range (dashed vertical lines) and outliers (open dots). Significant differences within the bee type (drones and workers) are marked through different letters. Significant differences between the treatment groups of workers and drones are indicated by a black horizontal bar (***  = P<0.001).

Figure 4. BQCV loads expressed as viral copies per bee (log10-transformed) of drones and workers of the four treatment groups after termination of the bees on day 14.

The boxplots show interquartile range (box), median (black line within the interquartile range), data range (dashed vertical lines) and outliers (open dots). Significant differences within the bee type (drones and workers) are marked through different letters. Significant differences between the treatment groups of workers and drones are indicated by black horizontal bars (***Ps<0.001).