2018 ESA Annual Meeting (August 5 -- 10)

OOS 32-9 - Multiscale models for predicting parasite life history

Thursday, August 9, 2018: 4:20 PM
345, New Orleans Ernest N. Morial Convention Center
Megan A. Greischar, Ecology & Evolutionary Biology, University of Toronto, Toronto, ON, Canada, Lindsay M. Beck-Johnson, Biology, Colorado State University, Fort Collins, CO and Nicole Mideo, Ecology and Evolutionary Biology, University of Toronto, Toronto, ON, Canada
Background/Question/Methods

Within a host, malaria parasites tradeoff between proliferation and the production of specialized stages required for onward transmission. All else being equal, parasite strains that invest more into proliferation will exploit resources faster, imposing greater virulence on the host. A multitude of ecological factors within the host lead to selection for proliferation at the expense of transmission investment, including host immunity and within-host competition. In addition, the abundance of mosquito vectors and susceptible hosts can change dramatically over the lifespan of a malaria infection, dynamics that could either intensify or oppose within host selection for greater proliferation. We pair a within host model of malaria infection—allowing host infectiousness to vary with transmission investment and age of infection—with a temperature-forced model of mosquito population dynamics to simulate the spread of a malaria epidemic through a host population.

Results/Conclusions

We identify transmission investment strategies that maximize either (1) transmission from an infected host; (2) the number of vectors that survive to become infectious; or (3) the number of hosts that become infected over the course of a transmission season. We find that explicitly tracking vector population dynamics leads to important ecological feedbacks that influence the evolution of parasite traits that are expressed within a host. Comparing across these biological scales shows how adding multiple layers of ecological realism into evolutionary models alters predictions about optimal traits and, in our specific case, lays the groundwork for understanding how geography influences virulence.