Wildlife Diseases I

Contributed Oral

Vaccine Baiting Strategy Influences Rabies Seroprevalence Rates in Urban Raccoon Populations
Emily Beasley, Kathleen Nelson, Richard Chipman, Amy Gilbert, Amy Davis

The raccoon variant of the rabies virus (RABV) is intensely managed in the eastern United States and Canada with the goal of eventual elimination. RABV spread is contained via distribution of oral rabies vaccine (ORV) in specific management zones. ORV bait distribution in urban areas is primarily conducted through hand baiting from vehicles, which limits coverage patterns to roads. In Burlington, Vermont, previous estimates of seroprevalence rates of rabies virus neutralizing antibodies (an index of target population immunity) in raccoons (Procyon lotor) are below thresholds necessary for RABV elimination. However, cases of raccoon rabies are infrequent, suggesting that actual seroprevalence rates may be higher. Understanding how ORV management impacts seroprevalence rates and case reduction in urban areas is critical for long-term RABV elimination. Using raccoon capture data from twelve sites located in the greater Burlington area in 2016–2017, we used a Bayesian multinomial N-mixture model to 1) jointly estimate site-level seroprevalence, abundance, and capture rate, and 2) evaluate the effects of trapping methods, baiting strategies, and environmental covariates on seroprevalence, abundance, and capture rates. Observed seroprevalence rates and abundance in Burlington were likely unbiased, as the probability of capturing an individual raccoon at least once during the 10-day trapping periods was high (mean 0.88, 95% credible interval 0.57–1.00). Seroprevalence tended to increase in areas with greater bait density, but seroprevalence was highest in sites classified as highly developed urban habitat. This was likely caused by higher spatial coverage of ORV baits in these areas due to a greater density of roads compared to areas with less developed habitats. These results suggest that ORV bait density and bait coverage are both important.  We recommend baiting strategies that promote a more comprehensive spatial distribution of ORV baits to refine management of RABV in urban areas.

Effect of Urban Habitat Use on Parasitism in Mammals: A Meta-Analysis
Courtney S. Werner, Charles L. Nunn

Rates of urbanization are increasing globally, with consequences for the dynamics of parasites and their wildlife hosts. A small subset of mammal species have the dietary and behavioural flexibility to survive in urban settings. The changes that characterize urban ecology—including landscape transformation, modified diets and shifts in community composition—can either increase or decrease susceptibility and exposure to parasites. We used a meta-analytic approach to systematically assess differences in endoparasitism between mammals in urban and non-urban habitats. Parasite prevalence estimates in matched urban and non-urban mammal populations from 33 species were compiled from 46 published studies, and an overall effect of urban habitation on parasitism was derived after controlling for study and parasite genus. Parasite life cycle type and host order were investigated as moderators of the effect sizes. We found that parasites with complex life cycles were less prevalent in urban carnivore and primate populations than in non-urban populations. However, we found no difference in urban and non-urban prevalence for parasites in rodent and marsupial hosts, or differences in prevalence for parasites with simple life cycles in any host taxa. Our findings therefore suggest the disruption of some parasite transmission cycles in the urban ecological community

Risky Business: Relating Risk of Direct Contact with Disease Risk
Maria Dobbin, Evelyn Merrill, Peter Smolko

Chronic wasting disease (CWD) is a fatal, prion disease of cervids that was first detected in Alberta in 2005. Transmission of CWD occurs by direct contact with infected individuals and via contaminated environments. We investigate the seasonal effects of grouping patterns and landscape heterogeneity on direct, pair-wise contacts within and between sex-specific (same or mixed sex) groups of mule deer (Odocoileus hemionus) in eastern Alberta. First, we establish criteria based on spatial-temporal movements of collared deer to define sex-specific group membership. Second, we classify whether sex-specific dyads of collared deer are in different groups and model the relative risk of a sex-specific contact occurring in a locale based on landscape characteristics. Third, we relate seasonal predictions of the spatial relative risk of contacts to the risk of deer being CWD-infected in an area based on hunter-harvest, CWD surveillance data. We predict that a harvested male deer being CWD+ is more strongly related to risk of contact with other between-group males than females, whereas the risk of a harvested female deer is most related to within-group contacts between females. Further, we predict that winter contact risk is more related to the risk of deer being CWD+ than summer contact risk across all group and pair types.

Sociodemographic and Regional Determinants of Rabies Submission Bias in North Carolina
Rachael Urbanek, Carson Hicks, Colleen Olfenbuttel, CWB, OiTF Organizer

Between 2008-2018, North Carolina animal control offices submitted 300-1,000 wild terrestrial animals for rabies testing each year, however, only 30-46% of total submissions tested positive annually. Given the low percentage of positive tests and high number of submissions in some counties, we used one-way ANOVAs to determine if a submission bias existed across the state for wild species. Additionally, using multiple regression analyses and data from all 100 counties, we regressed the number of total submissions and the percent positive rabies cases on sociodemographic and regional variables to determine what factors influence total submissions and percent positive cases. Counties across the state differed in the average annual total submissions and average annual percent positive cases, indicating bias in both measures. Counties that are in the Piedmont region, have higher percentages of White residents, denser human populations and housing density, or more tourism income had the largest number of total submissions for rabies testing. Alternatively, counties that had higher percentages of residents whom had attended some college submitted fewer animals for testing. Counties that had higher percentages of positive rabies cases included those within the Piedmont and Coastal Plain regions and counties with higher percentages of Black and Hispanic residents. Alternatively, Mountain region counties and counties with higher percentages of White residents had lower percentages of positive rabies cases. Median age also had a negative effect on the percent of positive rabies cases.  Determining which factors may influence submissions will help both wildlife and public health professionals identify where targeted educational rabies and wildlife programs are needed. We suggest rabies education should be focused towards the Mountain region, White residents, and older individuals in North Carolina since these groups of residents were more likely to request rabies testing on non-rabid wildlife.

Disease Prevalence in Striped Skunks Across an Urban-Rural Gradient
Katelyn Amspacher, Agustin Jimenez, Clayton Nielsen

Striped skunks (Mephitis mephitis) are generalist mesopredators that act as hosts to many common diseases and parasites that also infect other wildlife, domestic animals, and humans. Typically, these infections are diagnosed through tests that detect exposure. To concentrate on active infections, we collected tissue samples during necropsies of 98 striped skunks from southern Illinois during April 2018-March 2021. We tested tissues for canine distemper virus (CDV), canine parvovirus (CPV), and leptospirosis via rtPCR in 84, 66, and 84 individuals, respectively. We tested 51 individuals for toxoplasmosis via fluorescent antibody tests. We established a 1-km buffer around each collection location, calculated the mean human modification score (scale 0-1) from a raster layer, and tested whether anthropogenic development affected striped skunk infection. No striped skunks tested positive for CPV or toxoplasmosis (1.3-82% CPV and 0-71.4% toxoplasmosis in previous studies). Twenty-four striped skunks (28.6%) tested positive for leptospirosis (13-62% in previous studies), and 5 striped skunks (6%) tested positive for CDV (8-80% in previous studies). Leptospirosis was detected throughout the sampling period, but CDV was exclusively detected in February-April of 2019 and 2020. Human modifications scores surrounding collection sites ranged from 0.2-0.8, but no correlation with infection was detected. Leptospirosis was the most prevalent pathogen in this study, likely due to the persistence of the bacteria in water and soil and their wide diversity of hosts. Prevalence of CDV was influenced by season and year, likely due to its direct transmission and lethality of the disease in striped skunks. Our study illustrates the prevalence of pathogens in striped skunks across a human-modified landscape.

Distribution and Risk Factors of Lymphoproliferative Disease Virus (LPDV) in Iowa Wild Turkeys
Kelsey Smith, Julie Blanchong

First identified in the United States in 2009, Lymphoproliferative Disease Virus (LPDV) is an avian retrovirus found in Wild Turkeys (Meleagris gallopavo) that has since been documented across the eastern and central U.S. and multiple Canadian provinces. Examination of hunter-harvested turkeys suggests that most birds are asymptomatic, but little is known about population-level effects or impacts on poults. Our goal was to investigate the occurrence of LPDV in Iowa, identifying potential risk factors and spatial patterns of LPDV distribution. In collaboration with the Iowa Department of Natural Resources, tarsi from hunter-harvested turkeys were collected in 2019 and 2020 across Iowa. DNA was extracted from bone marrow and tested for LPDV using the gag gene and results were visualized using agarose gel electrophoresis with confirmation of a subset of positive samples by sequencing. LPDV infected turkeys were identified in 68 of 99 counties.  Logistic regression was used to examine the relationship between the risk of LPDV infection and the ratio of agriculture to forest, total forest edge, forested area, presence of water bodies, historic translocation data, harvest trends, and brood survey data. Cluster analysis identified areas of elevated prevalence of LPDV in Iowa. While more research is needed to fully understand the effects of LPDV on Wild Turkey populations, these results contribute to ongoing surveillance and monitoring of LPDV in North America and may help managers identify localized regions for additional surveillance and management.

How Far Do They Really Go? the Long and Tortuous Path of Dispersing White-Tailed Deer in Michigan
Jonathan Trudeau, David Williams, Sonja Christensen, Dwayne Etter

The dispersal behavior of white-tailed deer (WTD) is typically reported as one metric, the Euclidean distance between a natal and post-natal range. However, dispersal distance is a single metric describing behavior that is widely accredited for the natural spread of diseases, such as chronic wasting disease (CWD), and likely underrepresents the complexity and extent of the behavior. The objective of this study was to determine other dispersal characteristics that improve our ability to capture the risk WTD dispersal poses to disease spread and how dispersal behavior varies across landscape types. Between January 2018 and April 2021, we equipped 153 WTD across a human-development gradient with GPS radio-tracking collars and identified 26 dispersers in south-central Michigan. Males within the first 1.5 years of life were the primary dispersers (n=20) from both urban and rural environments. We found a similar dispersal distance to previous studies with a mean of 12.88 km (± 2.47) and a maximum of 54.40 km. However, unlike previous studies that were restricted by temporally coarse location data, our high-resolution GPS location data allowed us to measure the dispersal path more accurately and determine the tortuosity of each event. Though many dispersal events were relatively straight, we determined that some paths were upwards of 128 km and 14x longer than the Euclidean distance reported for the same event. We found correlative trends between circuity and development type, suggesting human development plays a crucial role in the dispersal of WTD. Knowing the tortuosity of dispersal events better describes the area a deer can cover and the potential risk they pose to facilitating pathogen spread. Further, our study informs what segment of the deer population is more likely to spread a pathogen at a landscape-scale, and where management efforts may be most effective in reducing CWD spread.

Evaluation of Oral Iophenoxic Acid Biomarkers in Raccoons to Measure Bait Uptake
Shylo Johnson, Amy Davis, Molly Selleck, Alison Barbee, Steven Volker, David Goldade, Amy Gilbert

The primary strategy to control wildlife rabies in the US is oral rabies vaccination (ORV), which relies on populations of target species locating and consuming baits across rural and urban landscapes. Biomarkers can be used to assess bait uptake by target and non-target species. We compared marking ability and decay of four different analogues (ethyl, butyl, methyl, and pentyl) of the iophenoxic acid (IPA) biomarker at three different dosages (2 mg, 5 mg, and 10 mg) delivered orally to raccoons (Procyon lotor), an ORV target species.  We also tested marking ability of two IPA analogues (ethyl and propyl) incorporated into the bait matrix of an ORV bait (i.e., placebo Ontario Rabies Vaccine Baits – ONRAB) at a concentration of 1 mg biomarker/bait along with a separate evaluation of taste aversion using a concentration of 10 mg ethyl-IPA/bait. In our initial evaluation, all raccoons readily consumed encapsulated powder IPAs surround by marshmallow. On day 1 post-consumption, 100% of the raccoons were marked at every dosage and every analogue tested. IPAs were still detected at day 572 post-consumption, except for pentyl-IPA in 29% (6/21) of the raccoons. Higher doses of IPA in the bait resulted in a higher concentration of IPA measured by liquid chromatography with tandem mass spectrometry. In our second evaluation, 90% (9/10) of the raccoons consumed ≥ 50% of the coating for each IPA and concentration offered. Both ethyl and propyl 1 mg IPA baits marked the raccoons. The same individuals consumed the coating from both the 1 mg and 10 mg IPA baits. Our analytical method distinguished whether an individual raccoon consumed one or multiple IPA analogues. These IPA analogues have utility in comparing ORV baiting strategies on the landscape enabling more efficient trapping and sampling to refine baiting strategies for wildlife.

Individual-Based Modelling of Chronic Wasting Disease
Kelsey Gritter, Evelyn Merrill, Mark Lewis, Maria Dobbin

Chronic wasting disease (CWD) is an emerging prion disease in Canada that infects mule deer, white-tail deer, elk, and moose by direct and environmental transmission and is invariably fatal.  CWD spread can be promoted at “hotspots” that attract deer, such as those are created intentionally via supplemental feeding or accidentally via grain spills.  Hotspots may increase contacts between- and within-groups depending on how many animals use the site and how long individuals spend at the site.  An individual-based model was created to investigate the effects of different densities and arrangements of hotspots on contact rates between- and within-groups.  The model tracks contacts (when two individuals come within three meters of one another), which are defined as between- or within-group depending on the group membership of the two individuals.  Simulations are run in Netlogo on a heterogeneous landscape and include behaviours such as grouping and home ranges.  Grouping behaviours and home range are simulated via a biased turning angle distribution.  Deer are moved across the landscape model at a two-hour time step based on step-selection movement rules relative to resources and group behaviours.  The step-selection function utilizes GIS layers for environmental weights and GPS-collar movement data for calculating step-selection coefficients, as well as turning angle and step length distributions.  We present preliminary results of how hotspot density and configuration influence contact rates and the potential for disease transmission.

Modeling Host-pathogen Transmission Dynamics To Support Scientific Decision Making
Robin Russell, Toni Rocke, Daniel Walsh, John Grider
Diseases of wildlife are difficult to manage due to limited options, and numerous uncertainties, including uncertainty in wildlife responses to management actions, uncertainty in the identification of host species, and uncertainty in host species demographics and distributions prior to disease emergence. Host-pathogen models can provide guidance for decision makers particularly in the early stages of disease emergence and/or when new tools for management are developed. The development of host-pathogen models requires some knowledge of host distributions and demographics, pathogen transmission routes, and effects of pathogens on host populations. We have developed spatially-explicit, agent-based, host-pathogen models for plague and prairie dogs, rabbit hemmorhagic disease and rabbits, and white-nose syndrome and bats. These models can assist managers with assessing the potential impacts of disease mitigation tactics such as vaccination, the effort /cost needed to achieve desirable outcomes, and the influence of parameters on model results. Additionally, these models can assist with the identification of key uncertainties that require further research as well as mitigation strategies that are optimal across a range of conditions and therefore unaffected by model uncertainty. We discuss the importance of spatial structure (i.e. host distributions) on model outcomes, the commonalities among the host-pathogen systems, and the development of a generalizable framework to address emerging wildlife disease issues.

Contributed Oral
Location: Virtual Date: November 2, 2021 Time: 4:00 pm - 5:00 pm