Ecological Applications, 22(3), 2012, pp. 1050–1059 Ó 2012 by the Ecological Society of America Emerging prion disease drives host selection in a wildlife population STACIE J. ROBINSON, 1,6 MICHAEL D. SAMUEL, 2 CHAD J. JOHNSON, 3 MARIE ADAMS, 4 AND DEBBIE I. MCKENZIE 5 1 Department of Forest and Wildlife Ecology, University of Wisconsin–Madison, Room 208 Russell Labs, 1630 Linden Drive, Madison, Wisconsin 53706 USA 2 U.S. Geological Survey, Wisconsin Cooperative Wildlife Research Unit, Room 224 Russell Labs, 1630 Linden Drive, Madison, Wisconsin 53706 USA 3 Department of Soil Science, University of Wisconsin–Madison, 107 Hiram Smith Annex, 1555 Observatory Drive, Madison, Wisconsin 53706 USA 4 Biotechnology Center, University of Wisconsin–Madison, 425 Henry Mall, Madison, Wisconsin 53706 USA 5 Department of Biological Sciences, University of Alberta, Edmonton, Biological Sciences Building, 11455 Saskatchewan Drive, Edmonton, Alberta T6G 2E9 Canada Abstract. Infectious diseases are increasingly recognized as an important force driving population dynamics, conservation biology, and natural selection in wildlife populations. Infectious agents have been implicated in the decline of small or endangered populations and may act to constrain population size, distribution, growth rates, or migration patterns. Further, diseases may provide selective pressures that shape the genetic diversity of populations or species. Thus, understanding disease dynamics and selective pressures from pathogens is crucial to understanding population processes, managing wildlife diseases, and conserving biological diversity. There is ample evidence that variation in the prion protein gene (PRNP) impacts host susceptibility to prion diseases. Still, little is known about how genetic differences might influence natural selection within wildlife populations. Here we link genetic variation with differential susceptibility of white-tailed deer to chronic wasting disease (CWD), with implications for fitness and disease-driven genetic selection. We developed a single nucleotide polymorphism (SNP) assay to efficiently genotype deer at the locus of interest (in the 96th codon of the PRNP gene). Then, using a Bayesian modeling approach, we found that the more susceptible genotype had over four times greater risk of CWD infection; and, once infected, deer with the resistant genotype survived 49% longer (8.25 more months). We used these epidemiological parameters in a multi-stage population matrix model to evaluate relative fitness based on genotype-specific population growth rates. The differences in disease infection and mortality rates allowed genetically resistant deer to achieve higher population growth and obtain a long-term fitness advantage, which translated into a selection coefficient of over 1% favoring the CWD-resistant genotype. This selective pressure suggests that the resistant allele could become dominant in the population within an evolutionarily short time frame. Our work provides a rare example of a quantifiable disease-driven selection process in a wildlife population, demonstrating the potential for infectious diseases to alter host populations. This will have direct bearing on the epidemiology, dynamics, and future trends in CWD transmission and spread. Understanding genotype-specific epidemiology will improve predictive models and inform management strategies for CWD-affected cervid populations. Key words: Bayesian modeling; chronic wasting disease (CWD); disease-driven selection; epidemiol- ogy; evolution; infection rate; mortality rate; population dynamics; prion disease; single nucleotide polymorphism (SNP); white-tailed deer; wildlife disease. INTRODUCTION Infectious diseases are one of the important forces driving population dynamics and natural selection in wildlife populations. It has long been recognized that diseases can play a role in population declines and adversely impact sensitive species (McCallum and Dobson 1995). Short of decimating populations, infec- tious agents may act to constrain population size, growth rates (Anderson and May 1979), and species distributions (Atkinson and LaPointe 2009), and influ- ence migratory escape from pathogens (Altizer et al. 2011). It is well established in agricultural systems (Goldmann 2008) and human populations (Van Bler- kom 2003) that adaptation to disease can drive host evolution; but empirical evidence of quantifiable natural selection in free-ranging wildlife populations is rare. Although population studies have identified selection in some systems, it is notoriously difficult to identify specific genetic components under selection and quantify disease-related selective forces (Dwyer et al. 1990). For example, high diversity in major histocompatibility Manuscript received 20 May 2011; revised 24 October 2011; accepted 28 October 2011. Corresponding Editor: N. T. Hobbs. 6 E-mail: stacie.j.robinson@gmail.com 1050