Behavioral fever in the house fly (Musca domestica L.): Effects on host fitness and the efficacy of fungal biopesticides

The progression of disease in ectotherms is strongly influenced by environmental temperature. To this end, many ectotherms, including insects, have evolved the ability to invoke a behavioral fever; upon sensing an infection, the host will alter its thermoregulatory behavior to maintain a body temperature above its normal thermal optimum. This behavior imposes a sub-optimal thermal environment on the pathogen, and may enhance the efficiency of the host immune system, thereby reducing the virulence of the infection. However, the actions required to reach and maintain a higher body temperature can simultaneously impose significant fitness costs on the host. In the context of biological control, these dynamics can have major influences on the efficacy of a biopesticide. Using the house fly (Musca domestica L.), a major economic pest of intensive animal units, we explore the nature and consequences of behavioral fever to flies infected with entomopathogenic fungi. We began by investigating how dose and growth kinetics of different isolates of entomopathogenic fungi impacted virulence to house flies. We found that virulence was positively correlated with dose, and that growth kinetics were qualitatively similar between isolates. The most virulent of these isolates, Beauveria bassiana (Bals.) Vuillemin, was then used to examine the effects of fungal infection on the thermoregulatory behavior of house flies. Using thermal gradient assays, flies infected with B. bassiana were found to invoke a behavioral fever by remaining at higher temperatures than uninfected flies. We found that fevering benefitted infected flies by extending their survival and allowing them to lay more eggs, while simultaneously imposing costs in terms of reduced egg viability. To explore how house flies might limit the costs associated with fevering, we characterized the thermoregulatory behavioral fever of flies that were exposed to different doses of B. bassiana spores on free-choice thermal gradients. Infected flies were found to invoke a behavioral fever, the degree of which was positively correlated with dose. Moreover, both uninfected flies and cultures of B. bassiana were shown to suffer fitness costs when subjected to simulated fly fever temperatures. Lastly, we investigated the role of fungal infection and the resultant behavioral fever on the vector competence of house flies by measuring the frequency and bacterial loads in excreta collected from infected flies under both forced-fever and constant temperature conditions. Fungal infection was found to significantly decrease the frequency of fly excretion in the time before death, although fever delayed this reduction by one day. Fevered flies also showed a general increase in the levels of bacteria in their excreta over time, indicating that fever may serve to increase fly vector competence. These results add insight into the adaptive value of behavioral fever in insects and highlight how the efficacy of biological control agents can be mediated by environmental temperature and insect behavior.