Immune-cognitive system connectivity reduces bumblebee foraging success in complex multisensory floral environments

Abstract

Bumblebees are declining at alarming rate worldwide, posing a significant threat to the function and diversity of temperate ecosystems. These declines have been attributed, in part, to the direct effect of specific pathogens on bumblebee survival. However, pathogens may also have a negative impact on host populations indirectly through immune-induced cognitive deficits in infected individuals. To gain greater insight into mechanisms and potential conservation implications of such 'immune-brain crosstalk' in bumblebees, we non-pathogenetically activated humoral and cellular immune pathways in individuals and then tested for long-term reductions in cognitive performance and foraging proficiency. We show that chronic activation of humoral, but not a cellular, immune pathways and effectors in foragers significantly reduces their ability to flexibly and efficiently harvest resources in multi-sensory floral environments for at least 7 days post-treatment. Humoral defense responses thus have the potential to confer significant foraging costs to bumblebee foragers over timeframes that would negatively impact colony growth and reproductive output under natural conditions. Our findings indicate that fitness effects of immune-brain crosstalk should be considered before attributing wild bumblebee decline to a particular pathogen species.

Figure 1

LPS challenge impairs performance of bumblebee foragers on simple stimulus discrimination learning tasks. (A) The proportion of rewarding flower types selected by treated bees per block of 20 consecutive flower visits to the single task array. Filled circles = mean for Ringer bees; Open circles = mean for LPS bees; Open squares = mean for Elastomer bees. Dotted line represents task performance threshold of 90% visits to the rewarding flower type. (B) Mean number of flower visits required for treated bees to reach 90% task performance threshold over three consecutive blocks of 20 flower visits. Means are shown +/−SE. **P < 0.01.

Figure 2

LPS-challenged bumblebee foragers have reduced cognitive flexibility and reward intake rates. Individual task preferences (A), mean proportion of task switches (B), and mean foraging rates (C) of Ringer, LPS, and Elastomer bees on a mixed array containing color and odor discrimination tasks. Task preference values range from 0 (no preference) to 1 (absolute preference). Foraging rates represent the number of rewarding flower types visited per minute of foraging time. Means are shown +/−SE. *P < 0.05; ***P < 0.001.

Figure 3

Relative expression of apidaecin, abaecin, defensin and hymenoptaecin genes as a function of days post treatment (d.p.i) for LPS-treated bees. Values represent fold increases in gene expression levels for bees in LPS treatment group relative to bees in the Ringer treatment group (denoted by the baseline value of ‘0’). Means are shown +/−SE. Fold changes were calculated using the comparative ∆∆C_T method. **P < 0.01; ***P < 0.001.