Fecal sample spiking assays
Individual fecal samples were handled separately, using fresh disposable, single use nitrile gloves, and disposable weigh boats to prevent cross-contamination of samples. Using CWD-negative fecal samples as determined by RT-QuIC described below in data analysis, spiking experiments were used to demonstrate recovery of PrPCWD from feces of 9 different scavenger and predator species: gray wolf (hereafter wolf), cougar, coyote, American crow (hereafter crow), American black bear (Ursus americanus, hereafter bear), raccoon (Procyon lotor), common raven (Corvus corax, hereafter raven), red fox (Vulpes vulpes, hereafter fox), and bald eagle (Haliaeetus leucocephalus, hereafter eagle; sources listed above and in Table 1). Spiking assays were carried out by using 50 mg of each fecal sample combined with 50 μL of a tenfold dilution of a 10% (w/v) CWD-positive WTD brain homogenate (BH) from the obex region of tagged WTD #5219 in the late stages of CWD (hereafter referred to as the reference sample), provided by the Wisconsin Department of Natural Resources (WDNR). The reference sample was prepared in 1× phosphate-buffered saline (PBS) to achieve a range of final concentrations of 10–3 to 10–8 mg/mL of CWD-positive brain. CWD-positive and negative BH were determined by RT-QuIC. For the negative control spike, 50 μL of a 10–3 mg/mL dilution of a 10% (w/v) CWD-negative WTD BH was used. Each spike dilution was added to feces and allowed to soak in for 2 h, then placed in a 50 °C incubator without agitation for 16 h, to keep assay temperature conditions consistent with those used for RT-QuIC. Spiked fecal samples were then prepared as described below.
Fecal sample extract preparation for RT-QuIC assay
Fecal samples (Table 1) were stored at − 80 °C prior to preparation for use in the RT-QuIC assay. Individual fecal samples and subsamples were handled separately, using fresh disposable/single use nitrile gloves, and disposable weigh boats to prevent cross-contamination of samples. To prepare field collected fecal samples for detection of PrPCWD, each individual sample had one subsample collected from three different areas/regions consisting of a slice from the middle and both ends of the sample, totaling three biological replicates (50 mg each). A separate scalpel blade was used to cut each section, that included surface material as well as inner material up to a depth of 1 cm to ensure sufficient sampling of surface and inner portions. These 3 subsamples were then extracted individually, for a total of 3 separate extractions, using 1 mL of sterile 1× PBS, then processed in a thermomixer (1400 rpm, 30 min, 25 °C; Eppendorf ThermoMixer F1.5) followed by centrifugation at 16,000×g for 15 min. Supernatants (~ 750 μL) were collected, followed by addition of 750 μL 1× PBS to each feces pellet and remaining buffer. Samples were vortexed, thermomixed (1400 rpm, 30 min, 25 °C), and centrifuged again at 16,000×g for 15 min, after which an additional ~ 750 μL supernatant was removed and added to the first supernatant. Resultant supernatants were then centrifuged again for further clarification at 16,000×g for 15 min, followed by collection of 1 mL of clarified supernatant into a new sterile tube and 500 μL of 23.1 mM sodium phosphotungstate hydrate (Na-PTA; Sigma-Aldrich, Cat. # 496626) was added. Samples were incubated without agitation for 16 h at 4 °C, then centrifuged (4 °C, 30 min, 5000×g). Sample supernatants were carefully removed from each tube and discarded. Pellets were retained and washed with a 1:1 solution of 18 MΩ H2O and 23.1 mM Na-PTA followed by centrifugation (4 °C, 30 min, 5000×g) and aspiration of the wash solution. Pellets were resuspended in 30 µL of RT-QuIC sample buffer (0.1 g/mL sodium dodecyl sulfate in phosphate-buffered saline with N-2 cell culture supplement; ThermoFisher, Waltham, MA) and reconstituted using a Qsonica Q700 cup horn ultrasonicator (Amplitude 36 for 1 min). A volume of 2 µL of each reconstituted sample was used to seed each reaction well for 8 technical replicates.
The RT-QuIC in vitro prion amplification assay was performed as described by Orru et al.39, with sodium iodide as described by Metrick et al.40 with minor modifications. Briefly, 2 µL of sample extracts were added to a given well of a 96-well format optical-bottom black microplate (Thermo Scientific, Fair Lawn, NJ, USA), each already containing 98 µL of RT-QuIC reaction mixture (0.1 mg∙mL−1 90–231 recombinant hamster prion protein, produced as previously described39, 300 mM sodium iodide, 20 mM sodium phosphate, 1.0 mM ethylenediaminetetraacetic acid, and 10 µM thioflavin T). Microplate-compatible spectrophotometers capable of heating, shaking, and fluorescence monitoring (BMG FLUOstar, Cary, North Carolina) were used with the following instrument settings: 50 °C for all samples double orbital pattern shaking at 700 rpm with 60-s shake/60-s rest cycles, fluorescent scans (λexcitation = 448 nm, λemission = 482 nm) every 15 min, at a gain of 1600, and a total run time of 48 h.
Collection of fecal samples
Fecal samples included in this study were obtained from either the field or from wildlife rehabilitation centers (Table 1). Individual fecal samples were collected using sterile Whirl–Pak bags and sterile, single-use nitrile gloves (changed between sample collects), then stored at − 80 °C prior to analysis. Below we describe location information and methods used to locate and collect fecal samples.
Northern Yellowstone ecosystem
Wolf, cougar, fox, coyote, and raven fecal samples included in this study from the Northern Yellowstone ecosystem (NYE) were provided by the Yellowstone Center for Resources, Yellowstone National Park, Wyoming, U.S.A. The NYE is a mountainous region (ranging approximately 1500–2400 m) located in northwestern Wyoming, USA, and south-central Montana, USA and is characterized primarily by lower elevation steppe and grassland, and higher elevation coniferous forests, with relatively few wetland areas41. The NYE is inhabited by several cervid prey species on which wolves and other predators prey42. CWD had not been detected in the study area during sampling periods, nor had CWD been detected in cervids within 30 km of the study area. Fecal collection methods for wolf43, followed by fox, coyote, cougar, and raven from the NYE are described briefly. Feces from GPS radio collared wolves (IACUC IMR YELL Smith wolves 2012) were collected in 2020 during early winter (30 days: November 15–December 15) approximately 17 days following deposition. Cougar, coyote, and fox fecal samples were located and collected from cougar GPS clusters near cougar kill sites or by tracking fox and coyotes opportunistically during cougar cluster investigations. Raven fecal samples were collected during capture, tagging, and handling of ravens for a monitoring study following permits and animal handling protocols permitted by state (Montana Scientific Collector’s Permit 2022-020-W), federal (Master Banding Permit 22489), NPS (Yell-2022-SCI-8072), and University of Washington animal care and use committee (Protocol 3077-01). All methods for feces collection were performed in accordance with the relevant guidelines and regulations.
As part of a multiyear CWD study centered in southwest Wisconsin, USA the Wisconsin Department of Natural Resources (WDNR) captured and fit 763 deer with GPS collars from 2017 to 2020 in Iowa, Grant, and Dane counties in southwestern Wisconsin, following protocols approved by WDNR’s Animal Care and Use Committee (Protocol: 16-Storm-01). This region is characterized by high CWD prevalence44—an area where CWD was first established in WI, has been estimated to have been present in the environment for over 20 years45, and studied extensively in prior research46,47,48. Habitat in this area is characterized by steep hills, forested ridges, deep river valleys, karst geology, and cold-water trout streams. Elevations range from 184 to 524 m. Coyote are reported to leave numerous fecal deposits within 80 m of carrion they are consuming49, therefore fecal samples were collected by the field crew opportunistically within 0–80 m of collared WTD mortality sites from 2021 to 2022. Deer GPS collars notified the field crew when the collar had been motionless for 4 or more hours. Deer mortality sites were then identified by GPS points and/or VHF telemetry. Morality sites were investigated to determine cause of death and those that had signs of predator activity were searched for feces.
The cougar fecal samples from the Pine Ridge and Niobrara Valley regions in northern Nebraska, U.S.A. used in this study were provided by Nebraska Game and Parks Commission (NGPC). Habitat in the Rine Ridge area is characterized by meadows, pine and deciduous forests, steep buttes, small streams, and minor peaks (ranging approximately 900–1600 m). Habitat in the Niobrara Valley is characterized by steep hills, bluffs, pine forests and canyons, boreal forests, grasslands, and the Niobrara River. As part of ongoing carnivore studies in Cherry, Dawes, and Sheridan counties in Nebraska, U.S.A., cougar fecal samples were collected by NGPC wildlife staff with the help of a trained detection dog and handler. While not part of ongoing CWD surveillance efforts in Northwest Nebraska, cougar fecal samples included in this study were collected within areas that are also designated big game research deer management units (DMUs) where CWD has been detected.
Minnesota Wildlife Rehabilitation Center
Feces from adult bear, raccoon, and eagle were collected during the month of September 2022 at the Minnesota Wild and Free Wildlife Rehabilitation Center (MWF) in Garrison, Minnesota, USA. Animals that fecal samples were collected from were originally found at different geographic locations prior to being brought to WMF, varied in time post admittance to MWF, and had differing diets as appropriate for each species. An admitted bear was found in Polk County, Minnesota, USA and the admitted raccoon and eagle were found in Cass County, Minnesota, USA. Individuals whose feces were included in this study were admitted between the months of May–July and were transported to MWF by The Minnesota Department of Natural Resources (MNDNR). The bear diet consisted of dry dog food supplemented with produce (i.e., apples, melon, corn, and assorted berries). The eagle diet typically included fish and rodents (i.e., chipmunks, mice, or gophers). The raccoon diet consisted of dry and/or wet dog food mixed with assorted produce (i.e., apples, melon, corn, and assorted berries).
Dane County Humane Society Wildlife Center
Feces from a juvenile crow was collected during the month of May 2022 at the Dane County Humane Society Wildlife Center in Madison, Wisconsin, USA. The crow was admitted in early May and was fed a diet that consisted of eggs, chicken, meal worms, seeds, nuts, and produce (i.e., apples, melon, corn, and assorted berries).
Amyloid formation rate (AFR) data generated from the RT-QuIC assays were analyzed and visualized using Jmp Pro 15 (SAS Institute, Cary, NC) and Prism 8 (GraphPad, San Diego, CA). To apply a rigorous standard for distinguishing true positive samples from true negatives, the AFR threshold times (i.e., the time at which amplification is determined to have occurred in the RT-QuIC assay) were calculated by adding twenty times the standard deviation of the relative fluorescence unit (RFU) values from cycles 3–14 to the mean of RFU values from cycles 3–14. We previously applied this method to account for baseline fluorescence variation amongst samples in determining if the sample was PrPCWD positive50. Due to false seeding observed in crow, bear, eagle, and raccoon negative control fecal samples, additional analysis was required to distinguish true seeding from false seeding events for these species. To accomplish this, empirical distributions of threshold times in hours of 24 replicates of unspiked fecal samples were used to determine a threshold time that would yield a specificity ≥ 95% for fecal samples from each species. These data were then used to determine a cycle end-time that excluded ≥ 95% of the false seeding that occurred for each species. Cycle end-times for these four feces were determined to be 20 h for crow (specificity of 95%), 24 h for bear (specificity 100%), 25 h for eagle (specificity 96%), and 17 h for raccoon (specificity 100%) (Supplementary Fig. S1). All fecal samples analyzed by RT-QuIC were considered positive if a sample had at least 3 out of 8 technical replicates (or half or more of 24 replicates) with seeding activity and had AFR values that were significantly different from control AFR values based on statistical analysis. Analyses for spiked samples were assessed with Dunnett’s multiple comparison tests, while for the surveillance samples, Kruskal–Wallis tests was used to distinguish which samples had AFRs significantly different from the northern Yellowstone ecosystem (NYE) negative control fecal sample.
Using AFR as a relative measure for prion concentration in a sample, we first evaluated our ability to detect PrPCWD from spiked fecal samples compared to the reference sample, and if the AFR differed among carnivore species. We used a two-way (factorial) ANOVA to compare AFR values between sample types (CWD-positive brain or spiked fecal samples from 9 different species). We included an interaction between species and spike dilution; to assess if detection/recovery in the different feces was sensitive to the concentration of the spike across the tenfold dilution series. Significant interactions were determined using the Tukey HSD for multiple comparisons.
Amyloid formation rates for field collected coyote and cougar fecal samples were found to be non-normally distributed (Supplementary Fig. S2). As such, the nonparametric Kruskal–Wallis test was applied to compare individual fecal samples. If statistical differences were observed across each individual fecal sample, then the non-parametric Steel-dewass post hoc test was used to determine which individual samples differed from the negative control fecal sample.