Parasitoid-induced behavioral alterations of Aedes aegypti mosquito larvae infected with mermithid nematodes (Nematoda: Mermithidae)

Abstract

Parasite-induced behaviors occur in many insect species and are caused by a wide range of parasites. The majority of studies addressing parasite-induced behavioral alterations have focused on parasites with complex lifecycles and the adaptive nature of such changes. Behavioral changes caused by parasitoids, multi-cellular single-host parasites that kill their host upon emergence, are far less studied and the adaptive nature of these changes is likely to be different. In this dissertation, I investigated behavioral alterations in Aedes aegypti mosquito larvae infected with parasitoid nematodes (family Mermithidae). I conducted several sets of experiments. The first set of experiments allowed me to test the following hypotheses: 1) Mermithid nematodes induce behavioral changes in mosquito larvae and the degree to which they do so depends upon intensity of infection. 2) Different species of mermithid nematodes induce similar changes in mosquito larvae behavior. 3) Behavioral alterations vary with mermithid developmental stage. 4) Mosquito larvae infected with mermithid nematodes behave similarly to uninfected food-deprived mosquito larvae. I found that Ae. aegypti infected with Romanomermis culicivorax or Strelkovimermis spiculatus exhibited resting behaviors significantly more than uninfected controls and that intensity of infection did not affect activity levels. The difference in behavior between infected and uninfected Ae. aegypti was more pronounced in S. spiculatus infections. Early in development, infected mosquito larvae were significantly more active than uninfected controls in some behaviors associated with feeding. Finally, there was no significant difference between infected and uninfected food-deprived mosquito behaviors. Both infected and uninfected food-deprived larvae differed significantly from well-fed uninfected larvae, however. The increase in feeding activity early in development as well as the lack of difference between food-deprived and infected Ae. aegypti behavior may indicate that this increase is a parasitoid adaptation that increases host feeding behaviors in order to acquire sufficient nutrients for successful parasitoid development. The activity of infected Ae. aegypti at specific stages in parasitoid development indicate that these behavioral alterations may be parasitoid adaptations that reduce host predation risk and thus increase pre-emergence host survival. However, without definitive evidence of increased fitness, it is difficult to distinguish between an adaptive explanation for these changes and one that indicates these behaviors may be a consequence of impaired nutrition due to infection. In the second set of experiments I addressed the adaptive nature of mermithid-induced behavioral alterations. I focused specifically on the hypothesis that parasitoids should alter the behaviors of their hosts in ways that decrease the risk of predation, thus increasing the survival of the parasitoid to its free-living stage because my previous results indicated that infection reduced host activity levels. I compared the behaviors of infected Ae. aegypti to those of uninfected larvae in the presence and absence of the predatory mosquito larva, Toxorhynchites rutilus. Studies have shown that differences in activity levels between species of mosquito larvae can lead to differential predation rates by Tx. rutilus and it is reasonable to expect parallel effects of a single species infected with mermithid nematodes. The experiments I conducted allowed me to answer the following questions: 1) Does the recovery time after an artificial alarm stimulus differ between infected and uninfected Ae. aegypti larvae? 2) Do infected and uninfected Ae. aegypti larvae facultatively switch to low-risk behaviors in the presence of Tx. rutilus and do they do so to a similar degree? 3) Does infection result in differential predation on Ae. aegypti by Tx. rutilus! I found that infected Ae. aegypti did not facultatively switch their behaviors in the presence of Tx. rutilus whereas uninfected controls did. This may be because infected larval behaviors in the absence of Tx. rutilus is already biased towards those that are low-risk When the behaviors of the infected and uninfected larvae were compared in the presence of Tx. rutilus, their behaviors did not differ, indicating that both groups maintain low levels of activity in the presence of a predator. These results were consistent with the results of the experiment on predation rate; there was no difference in predation rate on infected or uninfected Ae. aegypti by Tx. rutilus. Based on these results, I found no evidence that mermithid-induced behavioral alterations are adaptive in ways that benefit the parasitoid by decreasing the risk of predation. Eventually, the study of parasite-induced changes of mosquito larvae behavior and the impact these changes could have on various trophic interactions could lead to determining possible mechanisms for how parasite-induced behavioral alterations shape community structure. Because this aspect host-parasite relationship was beyond the investigative scope of my dissertation, I reviewed literature that addresses the ecology and parasites of container dwelling mosquitoes and the potential effect that these parasites may have on the structure of natural container communities.