Characterizing the demographic history and prion protein variation to infer susceptibility to chronic wasting disease in a naïve population of white-tailed deer (Odocoileus virginianus)

Abstract

Assessments of the adaptive potential in natural populations are essential for understanding and predicting responses to environmental stressors like climate change and infectious disease. Species face a range of stressors in human-dominated landscapes, often with contrasting effects. White-tailed deer (Odocoileus virginianus; deer) are expanding in the northern part of their range following decreasing winter severity and increasing forage availability. Chronic wasting disease (CWD), a prion disease affecting deer, is likewise expanding and represents a major threat to deer and other cervids. We obtained tissue samples from free-ranging deer across their native range in Ontario, Canada, which has yet to detect CWD in wild populations. We used high-throughput sequencing to assess neutral genomic variation and variation in the prion protein gene (PRNP) that is partly responsible for the protein misfolding when deer contract CWD. Neutral variation revealed a high number of rare alleles and no population structure, and demographic models suggested a rapid historical population expansion. Allele frequencies of PRNP variants associated with CWD susceptibility and disease progression were evenly distributed across the landscape and consistent with deer populations not infected with CWD. We estimated the selection coefficient of CWD, with simulations showing an observable and rapid shift in PRNP allele frequencies that coincides with the start of a novel CWD outbreak. Sustained surveillance of genomic and PRNP variation can be a useful tool for guiding management practices, which is especially important for CWD-free regions where deer are managed for ecological and economic benefits.

Keywords: Canadian wildlife; PRNP; RADseq; population genetics; prion; ungulate.

FIGURE 1

Distribution of the subsample of free‐ranging white‐tailed deer samples obtained between 2003 and 2018 by the Ontario Ministry of Natural Resources and Forestry (OMNRF) that were used for the reduced representation genome analysis (n = 190; cream) and the prion protein genetic analysis (n = 631; orange and cream). The natural distribution of free‐ranging white‐tailed deer is shown for Ontario with a darker shade of gray

FIGURE 2

A 771 bp region of the white‐tailed deer prion protein gene was analyzed from free‐ranging white‐tailed deer in Ontario, Canada (n = 631). The stacked polymorphisms across 19 variable loci were organized by broad management in Ontario and are shown. Circles indicate loci previously described as variable in the white‐tailed deer prion protein gene. Triangles indicate novel variable loci in the white‐tailed deer prion protein gene

FIGURE 3

A plot of PC1 and PC2 from the principal component analysis (PCA) on the reduced representation white‐tailed deer genome. PC1 and PC2 were able to explain a total of 4.3% of the genomic variation observed. Gray circles represent scores from samples in northern Ontario. Orange circles represent scores from samples in southern Ontario

FIGURE 4

Demographic parameter estimates from δaδi for the optimal 1D model for white‐tailed deer in Ontario. Parameter estimates are the ancient population size (Na), the ratio of population size after instantaneous change to ancient population size (nuB); the ratio of contemporary to ancient population size (nuF); and the time in the past at which instantaneous change happened and growth began (TC; in units of 2*Na generations). Included are the optimized log‐likelihood (LL) and bootstrap uncertainties (BU)

FIGURE 5

Simulated allele frequency projections for nucleotide positions (a) 285 A/C and (b) 743286 G/A in the PRNP gene under selection. Green and red lines are the selection coefficients provided by Robinson et al. (2012), while the black dots are those from our simulations, with blue line representing the mean selection coefficient