Throughout North America, chronic wasting disease (CWD) has emerged as perhaps the greatest threat to wild cervid populations, including white-tailed deer (WTD; Odocoileus virginianus). White-tailed deer are the most sought-after big game species across North America with populations of various subspecies in nearly all Canadian provinces, the contiguous US, and Mexico. Documented CWD cases have dramatically increased across the WTD range since the mid-1990s, including in Minnesota, US. CWD surveillance in free-ranging WTD and other cervid populations mainly depends upon immunodetection methods such as immunohistochemistry and enzyme-linked immunosorbent assay (ELISA) on medial retropharyngeal lymph nodes and obex. More recent technologies centered on prion protein amplification methods of detection have shown promise as more sensitive and rapid CWD diagnostic tools. Here, we used blinded samples to test the efficacy of real-time quaking-induced conversion (RT-QuIC) in comparison to ELISA for screening tissues collected in 2019 from WTD in southeastern Minnesota, where CWD has been routinely detected since 2016. Our results support previous findings that RT-QuIC is a more sensitive tool for CWD detection than current antibody-based methods. Additionally, a CWD testing protocol that includes multiple lymphoid tissues (e.g., medial retropharyngeal lymph node, parotid lymph node, and palatine tonsil) per animal can effectively identify a greater number of CWD detections in a WTD population than a single sample type (e.g., medial retropharyngeal lymph nodes). These results show that the variability of CWD pathogenesis, sampling protocol, and testing platform must be considered for the effective detection and management of CWD throughout North America.

Chronic wasting disease (CWD) is a contagious, 100% fatal neurodegenerative disease affecting cervids. Classified as a transmissible spongiform encephalopathy, CWD is caused by a misfolded prion protein (PrPCWD) which is shed through bodily fluids and can remain infectious in the environment for years (Williams and Young 1980; Prusiner 1982). Originally detected in Colorado mule deer (Odocoileus hemionus) in 1967, CWD has been detected in additional cervid species and expanded in geographic distribution (Williams and Miller 2002). As of June 2021, CWD has been found in at least 26 states in the US and three Canadian provinces. The continued expansion of CWD across North America, and recent detections in Scandinavian countries (Mysterud et al. 2020) is changing how cervids are hunted, managed, and consumed. For these reasons, stakeholders tasked with managing CWD must have access to the best diagnostic tools and relevant protocols.

The Minnesota Department of Natural Resources (MNDNR) has surveyed free-ranging white-tailed deer (Odocoileus virginianus) for CWD since detection in 2002 in farmed elk (Cervus canadensis) and free-ranging white-tailed deer in Minnesota and Wisconsin, respectively. Since then, CWD has been detected in 12 captive cervid facilities and 110 free-ranging deer in Minnesota (>90,000 tested), with the disease potentially established endemically in the southeast region, albeit at a low prevalence (La Sharr et al. 2019). Surveillance efforts primarily utilize samples from hunter-harvested animals along with those collected opportunistically from deer found dead in poor body condition, euthanized deer displaying clinical signs of CWD, deer killed by vehicle collisions, and targeted agency culling.

Diagnostic tests for CWD can be classified into two categories: first-generation antibody-based diagnostics such as immunohistochemistry (IHC) and enzyme-linked immunosorbent assay (ELISA) and second-generation prion protein amplification assays such as real-time quaking-induced conversion (RT-QuIC), and protein misfolding cyclic amplification. Management agencies have employed ELISA screening of medial retropharyngeal lymph nodes with confirmatory IHC on samples considered suspect by ELISA, because these have traditionally been considered the gold standard for CWD (Haley and Richt 2017). Both ELISA and IHC, the currently available validated assays for CWD, must be completed at National Animal Health Laboratory Network laboratories. This diagnostic bottleneck has resulted in laboratories being at or beyond testing capacity (Schuler et al. 2021). However, advances in CWD diagnostics allow the implementation of new diagnostic standards and sampling techniques, a critical development as testing pressures and expectations increase (Haley and Richt 2017; McNulty et al. 2019; Bloodgood et al. 2021; Henderson et al. 2020). Amplification assays have been refined over the past decade and show potential as the next gold standard choice of diagnostic tools with increased sensitivity in a high-throughput platform, as well as usefulness with nontraditional sample types such as blood, saliva, feces, and ear biopsies (Elder et al. 2015; Haley and Richt 2017; Tennant et al. 2020; Ferreira et al. 2021).

Few studies have compared the use of prion amplification assays with current immunodetection assays on free-ranging cervids (Haley et al. 2014). We aimed to compare the CWD detection capabilities of RT-QuIC test to ELISA using a population-level sample set of free-ranging deer from southeastern Minnesota.

Study area and experimental design

The MNDNR contracted with US Department of Agriculture–Wildlife Services to conduct culling from 22 January to 29 March 2019 within areas of known CWD-positive deer detections near Preston and Winona, Minnesota; a karst topography region with mixed upland hardwoods, swamp, and agricultural lands (Fig. 1). Priority areas were designated spatially by Public Land Survey System sections (1 mi2; 2.29 km2) with a high number of total CWD-positive deer, positive female deer (considered to be disease anchors), or areas with high deer densities in close proximity to known positives. Intact carcasses were transported to the Preston Department of Natural Resources Forestry office where MNDNR staff collected tissue samples: medial retropharyngeal lymph nodes (RPLN), submandibular lymph nodes (SLN), parotid lymph nodes (PLN), palatine tonsils (PT), feces, whole blood, and neck muscles (not included in this study). Tissue collection tools (scalpel handles, forceps) were wiped of physical debris and cleaned with a bleach solution between animals. A new scalpel blade was used between animals. Collected tissues were handled with tools only and placed directly into 4-oz (118 mL) Whirl-Pak (Nasco, Fort Atkinson, Wisconsin, USA) bags. Blood was collected in an ethylenediaminetetraacetic acid (EDTA) tube as a free flow at time of decapitation. Feces were collected from the anus by hand with new disposable gloves and placed directly into 4-oz Whirl-Pak bags. In some cases, significant coagulation of blood or lack of feces precluded collection. All samples were preserved at –20 C in the field.

Figure 1

Locations of the 519 white-tailed deer (Odocoileus virginianus) culled from 22 January to 29 March 2019 by the Minnesota Department of Natural Resources near Preston and Winona, Minnesota, USA. Thirteen deer identified as chronic wasting disease (CWD) enzyme-linked immunosorbent assay, immunohistochemistry, and real-time quaking-induced conversion (RT-QuIC) positive by medial retropharyngeal lymph node samples are indicated by large circles. Four additional deer identified as CWD putative positives by RT-QuIC are indicated by large triangles.

Figure 1

Locations of the 519 white-tailed deer (Odocoileus virginianus) culled from 22 January to 29 March 2019 by the Minnesota Department of Natural Resources near Preston and Winona, Minnesota, USA. Thirteen deer identified as chronic wasting disease (CWD) enzyme-linked immunosorbent assay, immunohistochemistry, and real-time quaking-induced conversion (RT-QuIC) positive by medial retropharyngeal lymph node samples are indicated by large circles. Four additional deer identified as CWD putative positives by RT-QuIC are indicated by large triangles.

Close modal

In accordance with surveillance, RPLN samples from all deer were tested by CWD ELISA at Colorado State University Veterinary Diagnostic Laboratory (CSU VDL) using the Bio-Rad TeSeE Short Assay Protocol (SAP) Combo Kit (BioRad Laboratories Inc., Hercules, California USA). For each animal, a homogenate was produced using three subsamples from each RPLN and tested as a pooled sample of both RPLN subsamples. Any suspect-positive RPLN from ELISA was confirmed through IHC of the prion protein as described (Hoover et al. 2016). The same procedures were used for all ELISA lymphoid tissue testing at CSU VDL.

RT-QuIC analysis

Following transport from the field, bilateral PLN, SLN, and PT, blood, and feces were preserved at –80 C until RT-QuIC analysis. All RT-QuIC research staff were blinded to the RPLN ELISA and IHC results.

Animal and sample identification: RT-QuIC screening began with sampling and testing of one PLN from each of the 519 deer. Samples that exhibited amyloid seeding activity (ASA) were tested a second time to confirm the laboratory procedures and both results were reported to the MNDNR. Due to limited resources, a 60-animal subset of the original 519 animals was created by MNDNR staff for further testing. This subset provided the following: 1) three times as many known negative animals as known positive animals (by ELISA testing of RPLN); 2) 300 individual samples across six sample types; and 3) nearly 1,800 individual RT-QuIC reactions. The subset included the following: 1) all deer that were CWD-positive by ELISA and IHC on RPLN; 2) all deer that exhibited ASA by RT-QuIC on unilateral PLN; and 3) a randomly chosen set of deer that were CWD not detected by ELISA on RPLN to reach 60 animals. In light of the study by Bloodgood et al. (2021) in which unilateral sampling of RPLN was less sensitive than bilateral sampling of RPLN, the 60-sample subset were subjected to RT-QuIC testing following bilateral sampling of PLN, SLN, and PT, as well as available whole blood and feces. Tissue samples were tested as a pool of tissue type per animal (e.g., bilateral PLN subsamples pooled). Within this sample subset, the 13 ELISA and IHC CWD-positive RPLN were provided by CSU VDL and tested by RT-QuIC. After samples were unblinded, bilateral PLN, SLN, and PT subsamples from all deer that exhibited ASA on at least one tissue type by RT-QuIC were provided to CSU VDL for ELISA-only testing. Note that the BioRad ELISA is only validated for retropharyngeal lymph nodes and obex. Staff at CSU VDL were blinded to the RT-QuIC results.

Substrate preparation: Recombinant hamster PrP (HaPrP90-231; provided by National Institutes of Health Rocky Mountain Laboratory) was cloned into the pET41a(+) expression vector and was expressed in Rosetta™ (DE3) Escherichia coli cells (Millipore Sigma, Darmstadt, Germany). Expression and purification of the recombinant substrate was performed following a modified version of the protocol from Orrù et al. (2017). Specifically, protein expression was induced using 0.75 mM isopropylthio-β-galactoside in place of Overnight Express™ autoinduction (Novagen, Darmstadt, Germany), and cells were cracked with two passes at 16,000 psi on a microfluidizer rather than using a homogenizer.

Lymph tissue preparation: Lymph tissue (RPLN, PLN, SLN, and PT) dissections were initiated by a cross-sectional cut with sample collection along the cut face to reduce potential cross-contamination that might have originated during field collection. We used disposable forceps and scalpels and surface decontamination between samples (1:1.5; 5.25% sodium hypochlorite). Samples were dissected on fresh disposable benchtop paper. A 10% (w:v) suspension was made by adding 100 mg of tissue to 900 µL of phosphate buffered saline (PBS). In the case of bilateral samples, 50 mg of each tissue was dissected and added to one tube. Tissue suspensions were homogenized using 1.5-mm-diameter zirconium oxide beads (Millipore Sigma, Burlington, Massachusetts, USA) and a BeadBug homogenizer (Benchmark Scientific, Sayreville, New Jersey, USA), maximum speed for 90 s. Homogenized samples were stored at –80 C. Samples were diluted further to 10–3 in dilution buffer (0.1% sodium dodecyl sulfate, 1× PBS, N-2 Supplement [Life Technologies Corporation, Carlsbad, California, USA]), and 2 µL were added to 98 µL of RT-QuIC master mix. The final tissue dilution factor in the reaction was 1:50,000.

Blood preparation: We modified a protocol for blood preparation from Elder et al. (2015) and the phosphotungstic acid precipitation was first described by Safar et al. (1998). One mL of EDTA whole blood was placed in a tube with 1.5-mm-diameter zirconium oxide beads and underwent four cycles of flash freeze-thaw consisting of 3 min in dry ice and 3 min at 37 C. It was then homogenized using a BeadBug homogenizer at top speed for two cycles (a total of 180 s). The homogenate was centrifuged at 500 × G for 2 min. We incubated 100 µL of supernatant with 7 µL of 4% (w/v) phosphotungstic acid (Sigma-Aldrich, St. Louis, Missouri, USA) in 0.2 M magnesium chloride. The product was then incubated in a ThermoMixer (Eppendorf, Enfield, Connecticut, USA) at 37 C for 1 h (1,500 rpm) and subsequently centrifuged for 30 min at 21,100 × G. The pellet was resuspended in 20 µL of dilution buffer. We added 2 µl of the 10–2–diluted suspension to the RT-QuIC reaction described next.

Feces preparation: We modified a fecal preparation protocol developed by Tennant et al. (2020). The fecal pellet was manually homogenized into 10% homogenates using 1× PBS. The solution was centrifuged at 1,000 × G for 15 min at 4 C. We centrifuged 500 µL of supernatant at 21,100 × G for 30 min. The pellet was resuspended in 100 µL of 1× PBS and incubated with 7 µL of phosphotungstic acid solution as described earlier, then centrifuged for 30 min at 21,100 × G. The pellet was resuspended in 10 µL 0.1% sodium dodecyl sulfate in PBS. We added 2 µL of this suspension to the RT-QuIC reaction described herein.

Assay parameters: Recombinant hamster PrP substrate was filtered at 3,000 × G through a 100 kDa molecular weight cutoff spin column. A master mix was made to the following concentrations: 1× PBS, 500 µM EDTA, 50 µM Thioflavin T, 300 mM NaCl, and 0.1 mg/mL rPrP. All reagents were filter-sterilized through 0.22-µm polyvinyl difluoride membrane filters. We pipetted 98 µL of the master mix into each well on a black 96-well plate with clear bottoms. Twelve to 16 wells were used for controls: four to six with CWD-negative WTD lymph nodes, four to six with CWD-positive WTD lymph nodes, and four with no sample. Plates were sealed with clear tape then shaken on a BMG FLUOstar® Omega microplate reader (BMG LABTECH Inc., Cary, North Carolina, USA) at 700 rpm, double orbital for 57 s, then rested for 83 s, repeated 21 times, then the fluorescence was recorded. For all tissue types, the temperature was set to 42 C. The whole shake-and-rest then read cycle was repeated 58 times for a total of about 46 h. Readings were recorded with an excitation filter of 450 nm and an emission filter of 480 nm. The gain was set to 1,600. The machine performed 21 flashes per well.

RT-QuIC data analysis: Rate of amyloid formation was determined as the inverse of hours (1/h) for ASA to surpass a threshold of 10 standard deviations above the average baseline readings after 4.5 h for each plate (Hoover et al. 2016). A minimum of four replicates were performed for each sample. Samples were considered an indeterminate (putative positive) result when at least 50% of the replicates gave a fluorescence signal higher than the threshold cut-off value. P values were calculated based on rate of amyloid formation versus negative controls using a Mann-Whitney Utest as described by Tennant et al. (2020) to determine the statistically significant difference in rate of amyloid formation between samples tested and negative controls on the respective plates. Statistical significance was established at 0.05 (α=0.05) and P values below 0.05 were considered statistically different. We additionally examined statistical significance for all RT-QuIC data using a maxpoint ratio analysis (Vendramelli et al. 2018; see Supplementary Methods). All statistical analyses were performed using GraphPad Prism software (version 9, San Diego, California, USA). We report 95% Wilson score confidence limits for proportions, which is appropriate for small sample size and sample proportions close to 0 or 1 (Brown et al. 2001). Confidence limits were estimated using Epitools Epidemiological Calculators (Ausvet, Bruce, Australian Capital Territory, Australia).

First generation surveillance: ELISA

Of the 519 culled deer, RPLN from 13 deer (0.025; 95% Wilson confidence limits [CL], 0.015–0.042) were reported as CWD positive by CSU VDL through ELISA screening and IHC confirmation (Fig. 1 and Table 1).

Table 1

Samples from a 61 white-tailed deer (Odocoileus virginianus) subset of the 519 deer culled from 22 January to 29 March 2019 by the Minnesota Department of Natural Resources (DNR) near Preston and Winona, Minnesota, USA. Samples listed are from deer that indicated chronic wasting disease enzyme-linked immunosorbent assay (ELISA) positivity and/or exhibited amyloid seeding activity (ASA) by real-time quaking-induced conversion (RT-QuIC) on at least one sample type. LN = lymph node; X = statistically significant ASA determined by Mann-Whitney U-test and Dunnet's test (P<0.05; see Supplementary Methods); ND = not detected; O = none or not statistically significant ASA determined by Mann-Whitney U-test and Dunnet's test (P<0.05; see Supplementary Methods); — = data not available because samples were not available.

Samples from a 61 white-tailed deer (Odocoileus virginianus) subset of the 519 deer culled from 22 January to 29 March 2019 by the Minnesota Department of Natural Resources (DNR) near Preston and Winona, Minnesota, USA. Samples listed are from deer that indicated chronic wasting disease enzyme-linked immunosorbent assay (ELISA) positivity and/or exhibited amyloid seeding activity (ASA) by real-time quaking-induced conversion (RT-QuIC) on at least one sample type. LN = lymph node; X = statistically significant ASA determined by Mann-Whitney U-test and Dunnet's test (P<0.05; see Supplementary Methods); ND = not detected; O = none or not statistically significant ASA determined by Mann-Whitney U-test and Dunnet's test (P<0.05; see Supplementary Methods); — = data not available because samples were not available.
Samples from a 61 white-tailed deer (Odocoileus virginianus) subset of the 519 deer culled from 22 January to 29 March 2019 by the Minnesota Department of Natural Resources (DNR) near Preston and Winona, Minnesota, USA. Samples listed are from deer that indicated chronic wasting disease enzyme-linked immunosorbent assay (ELISA) positivity and/or exhibited amyloid seeding activity (ASA) by real-time quaking-induced conversion (RT-QuIC) on at least one sample type. LN = lymph node; X = statistically significant ASA determined by Mann-Whitney U-test and Dunnet's test (P<0.05; see Supplementary Methods); ND = not detected; O = none or not statistically significant ASA determined by Mann-Whitney U-test and Dunnet's test (P<0.05; see Supplementary Methods); — = data not available because samples were not available.

Second generation surveillance: RT-QuIC

The first objective for RT-QuIC analysis was a blinded screening of the 519 deer using unilateral PLN samples. The first analysis identified 11 (0.021; 95% CL, 0.012–0.038) samples exhibiting significant (P<0.05) ASA with repeated results upon retesting (Table 1). Of the approximately 4,150 individual RT-QuIC reactions underlying these 519 PLN samples, we observed significant ASA within the PLN samples of the 11 deer noted earlier and in a single well (nonsignificant) in only 11 other PLN samples. We next examined the diagnostic agreement between these 11 animals with significant ASA, animals previously classified as CWD-positive by ELISA and IHC (n=13), and a series of ELISA-negative controls (total n=60). One animal that was positive by ELISA and IHC (MN100528) was mistakenly not included in the 60-deer sample set, but had the same tissue set screened subsequently (following unblinding) bringing the sample set to 61 deer.

The RT-QuIC results indicated 16 (26%) of 61 deer exhibited significant (P<0.05) ASA in at least one sample type. We observed ASA in the following: 12/61 (0.20; 95% CL, 0.12–0.31) bilateral PLN, 12/61 (0.20; 95% CL, 0.12–0.31) bilateral SLN, 13/61 (0.21; 95% CL, 0.13–0.33) bilateral PT, 13/13 (1.0; 95% CL, 0.77–1.0) bilateral RPLN, 7/51 (0.14; 95% CL, 0.07–0.26) whole blood, and 5/47 (0.11; 95% CL, 0.05–0.23) feces (Table 1 and Fig. 2). Sample prevalences estimated from this subset and the original 519 are biased toward CWD-positive detections (originating from deer social groups in areas of documented CWD cases) and do not reflect the true population prevalence.

Figure 2

Relative rates of amyloid formation (1/h) demonstrated through real-time quaking-induced conversion of lymphoid tissue samples collected from a 61 white-tailed deer (Odocoileus virginianus) subset of the 519 deer culled from 22 January to 29 March 2019 by the Minnesota Department of Natural Resources (MN DNR) near Preston and Winona, Minnesota, USA. Samples exhibiting amyloid seeding activity were deemed positive by Mann-Whitney U-test and Dunnet's test (***, P<0.001; **, P<0.01; *, P<0.05; see Supplementary Methods). Parotid=parotid lymph node; submandibular=submandibular lymph node; retropharyngeal=medial retropharyngeal lymph node.

Figure 2

Relative rates of amyloid formation (1/h) demonstrated through real-time quaking-induced conversion of lymphoid tissue samples collected from a 61 white-tailed deer (Odocoileus virginianus) subset of the 519 deer culled from 22 January to 29 March 2019 by the Minnesota Department of Natural Resources (MN DNR) near Preston and Winona, Minnesota, USA. Samples exhibiting amyloid seeding activity were deemed positive by Mann-Whitney U-test and Dunnet's test (***, P<0.001; **, P<0.01; *, P<0.05; see Supplementary Methods). Parotid=parotid lymph node; submandibular=submandibular lymph node; retropharyngeal=medial retropharyngeal lymph node.

Close modal

We observed intraindividual variability in the types of tissues that exhibited ASA for each of the 16 animals (Figs. 2, 3). We detected ASA beneath levels of significance (P<0.05) in lymphoid and blood samples (n=6) from several of these 16 animals and considered them as indeterminate results (Table 1 and Fig. 2). Additionally, amyloid seeding was detected in 2/4 replicates (indeterminate result; not significant) within a palatine tonsil sample from a single animal (MN145287; adult, female) that had no indication of ASA in any other sample type and was ELISA negative across all sample types (Table 1 and Fig. 2). No other samples from the 61 deer demonstrated ASA.

Figure 3

Relative rates of amyloid formation (1/h) demonstrated through real-time quaking-induced conversion (RT-QuIC) of all sample types from six of the 519 white-tailed deer (Odocoileus virginianus) culled from 22 January to 29 March 2019 by the Minnesota Department of Natural Resources near Preston and Winona, Minnesota, USA. Note the variability in amyloid seeding activity (ASA) across animals. (A) demonstrates a highly CWD-positive animal across all sample types. (B) is probably an animal in earlier disease progression when compared to (A). (C–F) are probably animals with very early infection in that only one sample type is statistically significant. Note that the deer shown in (D–F) did not have medial retropharyngeal lymph nodes (RPLN) available to test because they were not detected as CWD-positive by enzyme-linked immunosorbent assay and thus had been discarded at Colorado State University Veterinary Diagnostic Laboratory. Samples exhibiting ASA were deemed positive by Mann-Whitney U-test and Dunnet's test (***, P<0.001; **, P<0.01; *, P<0.05; see Supplementary Methods). nt=not tested because those samples were not available for RT-QuIC testing; parotid=parotid lymph node; RPLN=medial retropharyngeal lymph node; subm.=submandibular lymph node; tonsil=palatine tonsil.

Figure 3

Relative rates of amyloid formation (1/h) demonstrated through real-time quaking-induced conversion (RT-QuIC) of all sample types from six of the 519 white-tailed deer (Odocoileus virginianus) culled from 22 January to 29 March 2019 by the Minnesota Department of Natural Resources near Preston and Winona, Minnesota, USA. Note the variability in amyloid seeding activity (ASA) across animals. (A) demonstrates a highly CWD-positive animal across all sample types. (B) is probably an animal in earlier disease progression when compared to (A). (C–F) are probably animals with very early infection in that only one sample type is statistically significant. Note that the deer shown in (D–F) did not have medial retropharyngeal lymph nodes (RPLN) available to test because they were not detected as CWD-positive by enzyme-linked immunosorbent assay and thus had been discarded at Colorado State University Veterinary Diagnostic Laboratory. Samples exhibiting ASA were deemed positive by Mann-Whitney U-test and Dunnet's test (***, P<0.001; **, P<0.01; *, P<0.05; see Supplementary Methods). nt=not tested because those samples were not available for RT-QuIC testing; parotid=parotid lymph node; RPLN=medial retropharyngeal lymph node; subm.=submandibular lymph node; tonsil=palatine tonsil.

Close modal

ELISA analysis of animals exhibiting ASA by RT-QuIC

To provide tissue-specific comparison between ELISA and RT-QuIC, PLN, SLN, and PT samples (n=51) from the 17 RT-QuIC positive or indeterminate animals were blind-tested by ELISA at CSU VDL. Fifty of the 51 samples demonstrated RT-QuIC and ELISA results agreement (Table 1). Deer MN145273 PT exhibited significant (P<0.05) ASA by RT-QuIC and was not detected by ELISA. Four samples (MN100528 SLN, MN137219 PT, MN137346 PLN, MN145287 PT) demonstrated indeterminate ASA (beneath levels of significance) and were not detected by ELISA. Samples positive on both testing platforms showed a moderate positive correlation of the semiquantitative measures of prion content (R2=0.581; Fig. 4). Samples with an ELISA optical density (OD) value of 3.999 did not bias the correlation, because a similar R2 was derived from the sample set with those samples removed (data not shown).

Figure 4

Relationship between the relative rate of amyloid formation (1/h) demonstrated through real-time quaking-induced conversion (RT-QuIC) and enzyme-linked immunosorbent assay (ELISA) optical density (OD) for lymphoid tissues examined from a 61 white-tailed deer (Odocoileus virginianus) subset of the 519 deer culled from 22 January to 29 March 2019 by the Minnesota Department of Natural Resources near Preston and Winona, Minnesota, USA. Depicted samples revealed both amyloid seeding activity by RT-QuIC and had OD values between 0.100 and 3.999. A moderate positive correlation is appreciated across all lymphoid tissue samples (dotted line; Pearson's correlation coefficient, R2=0.581). The relationship between medial retropharyngeal lymph node (RPLN) and palatine tonsil (PT) samples is appreciated in that RPLN samples had a generally lower rate of amyloid formation than PT samples. LN=lymph node.

Figure 4

Relationship between the relative rate of amyloid formation (1/h) demonstrated through real-time quaking-induced conversion (RT-QuIC) and enzyme-linked immunosorbent assay (ELISA) optical density (OD) for lymphoid tissues examined from a 61 white-tailed deer (Odocoileus virginianus) subset of the 519 deer culled from 22 January to 29 March 2019 by the Minnesota Department of Natural Resources near Preston and Winona, Minnesota, USA. Depicted samples revealed both amyloid seeding activity by RT-QuIC and had OD values between 0.100 and 3.999. A moderate positive correlation is appreciated across all lymphoid tissue samples (dotted line; Pearson's correlation coefficient, R2=0.581). The relationship between medial retropharyngeal lymph node (RPLN) and palatine tonsil (PT) samples is appreciated in that RPLN samples had a generally lower rate of amyloid formation than PT samples. LN=lymph node.

Close modal

The goal for diagnostic assays is to optimize target detection so that animals that are diseased or infected are identified as true positives (sensitivity) and animals that are not diseased are identified as true negatives (specificity). An array of factors affects this optimization including, but not limited to, sample type and integrity, sampling method, assay elements, and individual animal (e.g., physiologic or immune state, genetics) and population characteristics (e.g., pathogen prevalence). For CWD, disease detection in wild populations has primarily been based on ELISA and IHC and the use of postmortem RPLN and/or obex samples. This approach might be limited with respect to the detection of low levels of prion burden as seen in early stages of disease (Hoover et al. 2016; Haley and Richt 2017; Hoover et al. 2017; McNulty et al. 2019). Our results support RT-QuIC's potential to detect low levels of CWD prion by using a diverse sample set (multiple lymphoid tissues, blood, and feces) that might have value for earlier detection of the disease.

A growing body of research collectively provides evidence that second-generation amplification-based assays have greater sensitivity in the detection of misfolded PrP than first-generation tests (Hoover et al. 2016; Haley and Richt 2017; Soto and Pritzkow 2018; McNulty et al. 2019). Our results support this conclusion as varied tissue types exhibited ASA with RT-QuIC, where, in some cases, ELISA testing resulted in no detection of PrPCWD in the specific tissue type or within the animal. This led to the discovery of four putative CWD-positive animals by RT-QuIC out of a pool of 48 deer with CWD not detected by ELISA on RPLN, as well as additional putative positive sample types from two deer that were CWD positive by ELISA on RPLN. Collectively, these six animals demonstrated intra-animal variability in CWD-positive tissues (ELISA and/or RT-QuIC). Based on relatively low ELISA OD values of the positive tissues in these six animals, and given that RT-QuIC identified tissues that exhibited ASA that were not detected by ELISA, we posit that these six animals were either in the early stages of CWD infection or represent RT-QuIC false-positives. We believe that the former is the case because zero ASA was observed in the 44 “negative” animals (i.e., zero of four replicates) and there was no indication of inter-animal or intra-animal cross-contamination from the sample collection, preparation, or RT-QuIC processes based on the spatial and chronological distribution of CWD detections and the reliability of plate-level controls. It is also possible that these six animals exhibited natural CWD prion strain variability to the degree that effective antibody-based detection via ELISA was hampered (discussed soon).

It is problematic to confirm the status of RT-QuIC-positive samples that are negative by first-generation methods because amplification assays approach attogram levels of misfolded prion protein detection, multifold lower than the limits of immunodetection methods (Haley et al. 2018b). However, similar studies focused on longitudinal analyses of CWD infection support claims that amplification-based assays can effectively detect CWD infections prior to both ELISA and IHC (Haley et al. 2018a, 2020; Denkers et al. 2020; Henderson et al. 2020). Previous studies also indicate that the unequal tissue distribution of CWD prion protein in early stages of the disease, coupled with sectioning or sampling technique variability (concerns with detection assays in general), contributes to the apparent reduced sensitivity of ELISA and IHC (Hoover et al. 2017; Bloodgood et al. 2021). Another complicating factor when comparing first- and second-generation CWD assays is the potential for variable strains of PrPCWD to be detected by RT-QuIC or protein misfolding cyclic amplification but missed by immunodetection methods due to strain sensitivity to enzymatic digestion. The ELISA (BioRad TeSeE specifically), IHC, and western-blotting methods use antibodies that do not distinguish between PrP and PrPCWD, necessitating Proteinase-K digestion to identify PrP strains resistant to degradation. Recent data indicate that a variety of PrPCWD strains are circulating in cervids, raising the possibility that diagnostic methods not using enzymatic digestion have greater potential for identifying a broader family of PrPCWD (Duque Velásquez et al. 2015; Osterholm et al. 2019). Simultaneously, the potential for naturally occurring, noninfectious conformations of healthy PrP to cause false-positives in amplification-based analyses must be considered.

Sample type also influences CWD test sensitivity, particularly early in infection. Amyloid seeding activity has been documented by RT-QuIC in feces and blood of experimentally infected deer in the preclinical stages and shortly after inoculation, respectively (Elder et al. 2015; Tennant et al. 2020). In both studies, optimal sample conditions—limiting freeze-thaw of feces and heparin preservation of blood—presented the most consistent results. We did not have these ideal conditions, which might have contributed to the variable RT-QuIC results from feces and blood. The pathogenesis of CWD provides valuable insight into the prion seeding activity we observed in PT. Tonsillar tissue is one of the first tissues to demonstrate prion immunodeposition, which fits with observed prion traffic through the lymphatic system and is supported in this study (Haley and Richt 2017; Hoover et al. 2017; Henderson et al. 2020). Palatine tonsil tissue from multiple animals, possibly in early stages of the disease, exhibited ASA by RT-QuIC and had either low OD values or were not detected by ELISA. This might be a limitation of the application of ELISA to this particular tissue type; the BioRad ELISA is only validated for retropharyngeal lymph nodes and obex. Either way, our results support the value of PT as a tissue of choice for CWD surveillance, particularly when using RT-QuIC. Based on our results, PT demonstrated higher levels of ASA (i.e., “hotter” samples) than did RPLN, yet a majority of both sample types were at the upper limits of detection by ELISA. This leads to the possibility that although RPLN can be good for ELISA-based surveillance, PT might be the ideal tissue type for RT-QuIC-based surveillance. Although the feasibility of identifying PT versus RPLN in field extraction should be considered, further investigation is warranted. We also document the potential for RT-QuIC screening of PLN to identify early stages of CWD infection, and our results support the continued importance of RPLN for CWD surveillance. Potential limitations to these results lie in the sampling scheme of this study. Although RT-QuIC and ELISA were performed on the same tissue types across and within animals, different subsamples of each tissue were used for each assay. Regardless, there was a high-level of correspondence between the testing methods. Collectively, our RT-QuIC data provide evidence that multitissue sampling would enhance detection and removal of CWD-positive animals through postmortem CWD surveillance, although fiscal limitations must be considered when considering sample type(s) and testing protocols beyond the current standards.

Early outbreak detection through CWD surveillance is imperative. This is particularly true for geographic locations where CWD has not been detected previously, because earlier disease discovery on the landscape leads to earlier implementation of management and more effective control (Miller and Fisher 2016). The benefit of enhanced test sensitivity is a reduction in the sample number needed for disease detection and monitoring, particularly when population prevalence is low. We posit that the growing number of studies documenting the utility of RT-QuIC for the surveillance of a variety of protein-misfolding and prion diseases, including CWD, collectively demonstrate that the method is robust and will greatly aid our understanding and control of CWD (Wilham et al. 2010; Orrú et al. 2015; Caughey et al. 2017; Franceschini et al. 2017; Cooper et al. 2019; Saijo et al. 2019; Henderson et al. 2020; Rossi et al. 2020). Our study provides additional data on RT-QuIC as the assay is further developed and validated for future CWD surveillance, management, and regulatory initiatives.

Future investigation, building on these discoveries, will include optimization of a sampling protocol for CWD amplification assays to increase detection sensitivity, elucidating potential impacts of prion protein allele variation on testing for particular CWD strains, and continued analyses of multitissue samples secured from CWD-positive regions. These efforts, combined with robust epidemiological validation to characterize diagnostic test performance (sensitivity and specificity) of new CWD diagnostic tests will help to establish a more complete depiction of the CWD landscape. Our study, using a blinded sample set of free-ranging WTD from documented CWD hotspots, and multiple sample types, indicate that RT-QuIC might be a more powerful tool than current methods in identifying CWD-positive animals in coordination with a multitissue sample collection protocol, particularly for early infections and in previously CWD-free locations, whether free-ranging or farmed. These findings have direct implications for the effective surveillance and management of CWD across all cervids.

We thank National Institutes of Health Rocky Mountain Labs, especially Byron Caughey, Andrew Hughson, and Christina Orru for training and assistance with the implementation of RT-QuIC. Fred Schendel, Tom Douville, and staff of the University of Minnesota Biotechnology Resource Center provided critical support with respect to large-scale production of recombinant proteins. Kathi Wilson of the Colorado State University Veterinary Diagnostic Laboratory kindly provided assistance with ELISA and IHC testing of samples reported herein. We thank the Minnesota Supercomputing Institute for secure data storage of computational products stemming from our work. Kristen Davenport helped guide our statistical analysis. Lon Hebl graciously provided access to animals housed at the Oxbow Park and Zollman Zoo. This project would not have been possible were it not for the collection of biological samples, and we are grateful to the following persons for their assistance in the field: MNDNR Wildlife staff, Roxanne J. Larsen, Negin Goodarzi, Devender Kumar, Jeremy Schefers, as well as USDA APHIS Wildlife Services staff. Funding for research performed herein was provided by the Minnesota State Legislature through the Minnesota Legislative-Citizen Commission on Minnesota Resources (LCCMR), Minnesota Agricultural Experiment Station Rapid Agricultural Response Fund, University of Minnesota Office of Vice President for Research, Minnesota Department of Natural Resources, and start-up funds awarded to P.A.L. through the Minnesota Agricultural, Research, Education, Extension and Technology Transfer (AGREETT) program.

Supplementary material for this article is online at http://dx.doi.org/10.7589/JWD-D-21-00033.

Bloodgood
J,
Kiupel
M,
Melotti
J,
Straka
K.
2021
.
Chronic wasting disease diagnostic discrepancies: The importance of testing both medial retropharyngeal lymph nodes.
J Wildl Dis
57
:
194
198
.
Brown
LD,
Cai
TT,
DasGupta
A.
2001
.
Interval estimation for a binomial proportion.
Stat Sci
16
:
101
117
.
Caughey
B,
Orru,
CD,
Groveman
BR,
Hughson
AG,
Manca
M,
Raymond
LD,
Raymond
GJ,
Race
B,
Saijo
E,
Kraus
A.
2017
.
Amplified detection of prions and other amyloids by RT-QuIC in diagnostics and the evaluation of therapeutics and disinfectants.
Prog Mol Biol Transl Sci
150
:
375
388
.
Cooper
SK,
Hoover
CE,
Henderson
DM,
Haley
NJ,
Mathiason
CK,
Hoover
EA.
2019
.
Detection of CWD in cervids by RT-QuIC assay of third eyelids.
PLoS One
14
:
e0221654
.
Denkers
ND,
Hoover
CE,
Davenport
KA,
Henderson
DM,
McNulty
EE,
Nalls
AV,
Mathiason
CK,
Hoover
EA.
2020
.
Very low oral exposure to prions of brain or saliva origin can transmit chronic wasting disease.
PLoS One
15
:
e0237410
.
Duque Velásquez
C,
Kim
C,
Herbst
A,
Daude
N,
Garza
MC,
Wille
H,
Aiken
J,
McKenzie
D.
2015
.
Deer prion proteins modulate the emergence and adaptation of chronic wasting disease strains.
J Virol
89
:
12362
12373
.
Elder
AM,
Henderson
DM,
Nalls
AV,
Hoover
EA,
Kincaid
AE,
Bartz
JC,
Mathiason
CK.
2015
.
Immediate and ongoing detection of prions in the blood of hamsters and deer following oral, nasal, or blood inoculations.
J Virol
89
:
7421
7424
.
Ferreira
NC,
Charco
JM,
Plagenz
J,
Orru
CD,
Denkers
ND,
Metrick
MA
II,
Hughson
AG,
Griffin
KA,
Race
B,
Hoover
EA,
et al.
2021
.
Detection of chronic wasting disease in mule and white-tailed deer by RT-QuIC analysis of outer ear.
Sci Rep
11
:
7702
.
Franceschini
A,
Baiardi
S,
Hughson
AG,
McKenzie
N,
Moda
F,
Rossi
M,
Capellari
S,
Green
A,
Giaccone
G,
Caughey
B,
et al.
2017
.
High diagnostic value of second-generation CSF RT-QuIC across the wide spectrum of CJD prions.
Sci Rep
7
:
10655
.
Haley
NJ,
Carver
S,
Hoon-Hanks
LL,
Henderson
DM,
Davenport
KA,
Bunting
E,
Gray
S,
Trindle
B,
Galeota
J,
LeVan
I,
et al.
2014
.
Detection of chronic wasting disease in the lymph nodes of free-ranging cervids by real-time quaking-induced conversion.
J Clin Microbiol
52
:
3237
3243
.
Haley
NJ,
Donner
R,
Henderson
DM,
Tennant
J,
Hoover
EA,
Manca
M,
Caughey
B,
Kondru
N,
Manne
S,
Kanthasamay
A,
et al.
2020
.
Cross-validation of the RT-QuIC assay for the antemortem detection of chronic wasting disease in elk.
Prion
14
:
47
55
.
Haley
NJ,
Henderson
DM,
Wycoff
S,
Tennant
J,
Hoover
EA,
Love
D,
Kline
E,
Lehmkuhl
A,
Thomsen
B.
2018a
.
Chronic wasting disease management in ranched elk using rectal biopsy testing.
Prion
12
:
93
108
.
Haley
NJ,
Richt
JA.
2017
.
Evolution of diagnostic tests for chronic wasting disease, a naturally occurring prion disease of cervids.
Pathogens
6
:
35
.
Haley
NJ,
Richt
JA,
Davenport
KA,
Henderson
DM,
Hoover
EA,
Manca
M,
Caughey
B,
Marthaler
D,
Bartz
J,
Gilch
S.
2018b
.
Design, implementation, and interpretation of amplification studies for prion detection.
Prion
12
:
73
82
.
Henderson
DM,
Denkers
ND,
Hoover
CE,
McNulty
EE,
Cooper
SK,
Bracchi
LA,
Mathiason
CK,
Hoover
EA.
2020
.
Progression of chronic wasting disease in white-tailed deer analyzed by serial biopsy RT-QuIC and immunohistochemistry.
PLoS One
15
:
e0228327
.
Hoover
CE,
Davenport
KA,
Henderson
DM,
Denkers
ND,
Mathiason
CK,
Soto
C,
Zabel
MD,
Hoover
EA.
2017
.
Pathways of prion spread during early chronic wasting disease in deer.
J Virol
91
:
e00077
17
.
Hoover
CE,
Davenport
KA,
Henderson
DM,
Pulscher
LA,
Mathiason
CK,
Zabel
MD,
Hoover
EA.
2016
.
Detection and quantification of CWD prions in fixed paraffin embedded tissues by real-time quaking-induced conversion.
Sci Rep
6
:
25098
.
La Sharr
K,
Hildebrand
E,
Carstensen
M,
Dexter
M,
Hagen
P,
Jennelle
C,
Cornicelli
L.
2019
.
Surveillance and management of chronic wasting disease in Minnesota.
Minnesota Department of Natural Resources
.
McNulty
E,
Nalls
AV,
Mellentine
S,
Hughes
E,
Pulscher
L,
Hoover
EA,
Mathiason
CK.
2019
.
Comparison of conventional, amplification and bio-assay detection methods for a chronic wasting disease inoculum pool.
PLoS One
14
:
e0216621
.
Miller
MW,
Fisher
JR.
2016
.
The first five (or more) decades of chronic wasting disease: Lessons for the five decades to come.
In:
Transactions of the 81st North American Wildlife and Natural Resources conference, Pittsburgh, Pennsylvania, 13–18 March
, pp.
110
120
.
Mysterud
A,
Benestad
SL,
Rolandsen
CM,
Våge
J.
2020
.
Policy implications of an expanded chronic wasting disease universe.
J Appl Ecol
58
:
281
285
.
Orrù
CD,
Groveman
BR,
Hughson
AG,
Manca
M,
Raymond
LD,
Raymond
GJ,
Campbell
KJ,
Anson
KJ,
Kraus
A,
Caughey
B.
2017
.
RT-QuIC assays for prion disease detection and diagnostics.
Methods Mol Biol
1658
:
185
203
.
Orrú
CD,
Groveman
BR,
Hughson
AG,
Zanusso
G,
Coulthart
MB,
Caughey
B.
2015
.
Rapid and sensitive RT-QuIC detection of human Creutzfeldt-Jakob disease using cerebrospinal fluid.
mBio
6
:
e02451
14
.
Osterholm
MT,
Anderson
CJ,
Zabel
MD,
Scheftel
JM,
Moore
KA,
Appleby
BS.
2019
.
Chronic wasting disease in cervids: Implications for prion transmission to humans and other animal species.
mBio
10
:
e01091
19
.
Prusiner
SB.
1982
.
Novel proteinaceous infectious particles cause scrapie.
Science
216
:
136
144
.
Rossi
M,
Candelise
N,
Baiardi
S,
Capellari
S,
Giannini
G,
Orrù
CD,
Antelmi
E,
Mammana
A,
Hughson
AG,
Calandra-Buonaura
G,
et al.
2020
.
Ultrasensitive RT-QuIC assay with high sensitivity and specificity for Lewy body-associated synucleinopathies.
Acta Neuropathol
140
:
49
62
.
Safar
J,
Wille
H,
Itri
V,
Groth
D,
Serban
H,
Torchia
M,
Cohen
FE,
Prusiner
SB.
1998
.
Eight prion strains have PrPSc molecules with different conformations.
Nat Med
4
:
1157
1165
.
Saijo
E,
Groveman
BR,
Kraus
A,
Metrick
M,
Orrù
CD,
Hughson
AG,
Caughey
B.
2019
.
Ultrasensitive RT-QuIC seed amplification assays for disease-associated tau, α-synuclein, and prion aggregates.
Methods Mol Biol
1873
:
19
37
.
Schuler
K,
Abbott
R,
Mawdsley
J,
McGarvey
K.
2021
.
Getting results in the fight against CWD.
In:
The wildlife professional
,
The Wildlife Society
,
Nashville, Tennessee
,
15
:
36
39
.
Soto
C,
Pritzkow
S.
2018
.
Protein misfolding, aggregation, and conformational strains in neurodegenerative diseases.
Nat Neurosci
21
:
1332
1340
.
Tennant
JM,
Li
M,
Henderson
DM,
Tyer
ML,
Denkers
ND,
Haley
NJ,
Mathiason
CK,
Hoover
EA.
2020
.
Shedding and stability of CWD prion seeding activity in cervid feces.
PLoS One
15
:
e0227094
.
Vendramelli
R,
Sloan
A,
Simon
SLR,
Godal
D,
Cheng
K.
2018
.
ThermoMixer-aided endpoint quaking-induced conversion (EP-QuIC) permits faster sporadic Creutzfeldt-Jakob Disease (sCJD) identification than real-time quaking-induced conversion (RT-QuIC).
J Clin Microbiol
56
:
e00423
18
.
Wilham
JM,
Orrú
CD,
Bessen
RA,
Atarashi
R,
Sano
K,
Race
B,
Meade-White
KD,
Taubner
LM,
Timmes
A,
Caughey
B.
2010
.
Rapid end-point quantitation of prion seeding activity with sensitivity comparable to bioassays.
PLoS Pathog
6
:
e1001217
.
Williams
ES,
Miller
MW.
2002
.
Chronic wasting disease in deer and elk in North America.
Rev Sci Tech
21
:
305
316
.
Williams
ES,
Young
S.
1980
.
Chronic wasting disease of captive mule deer: A spongiform encephalopathy.
J Wildl Dis
16
:
89
98
.

Supplementary data