Skip to main content
Browse Subject Areas

Click through the PLOS taxonomy to find articles in your field.

For more information about PLOS Subject Areas, click here.

  • Loading metrics

Prion-Seeding Activity in Cerebrospinal Fluid of Deer with Chronic Wasting Disease

  • Nicholas J. Haley ,

    Affiliation Department of Microbiology, Immunology, and Pathology, College of Veterinary Medicine and Biomedical Sciences, Colorado State University, Fort Collins, Colorado, United States of America

  • Alexandra Van de Motter,

    Affiliation Department of Microbiology, Immunology, and Pathology, College of Veterinary Medicine and Biomedical Sciences, Colorado State University, Fort Collins, Colorado, United States of America

  • Scott Carver,

    Affiliation School of Zoology, University of Tasmania, Hobart, Tasmania, Australia

  • Davin Henderson,

    Affiliation Department of Microbiology, Immunology, and Pathology, College of Veterinary Medicine and Biomedical Sciences, Colorado State University, Fort Collins, Colorado, United States of America

  • Kristen Davenport,

    Affiliation Department of Microbiology, Immunology, and Pathology, College of Veterinary Medicine and Biomedical Sciences, Colorado State University, Fort Collins, Colorado, United States of America

  • Davis M. Seelig,

    Affiliation Department of Veterinary Clinical Sciences, University of Minnesota, St. Paul, Minnesota, United States of America

  • Candace Mathiason,

    Affiliation Department of Microbiology, Immunology, and Pathology, College of Veterinary Medicine and Biomedical Sciences, Colorado State University, Fort Collins, Colorado, United States of America

  • Edward Hoover

    Affiliation Department of Microbiology, Immunology, and Pathology, College of Veterinary Medicine and Biomedical Sciences, Colorado State University, Fort Collins, Colorado, United States of America


Transmissible spongiform encephalopathies (TSEs), or prion diseases, are a uniformly fatal family of neurodegenerative diseases in mammals that includes chronic wasting disease (CWD) of cervids. The early and ante-mortem identification of TSE-infected individuals using conventional western blotting or immunohistochemistry (IHC) has proven difficult, as the levels of infectious prions in readily obtainable samples, including blood and bodily fluids, are typically beyond the limits of detection. The development of amplification-based seeding assays has been instrumental in the detection of low levels of infectious prions in clinical samples. In the present study, we evaluated the cerebrospinal fluid (CSF) of CWD-exposed (n=44) and naïve (n=4) deer (n=48 total) for CWD prions (PrPd) using two amplification assays: serial protein misfolding cyclic amplification with polytetrafluoroethylene beads (sPMCAb) and real-time quaking induced conversion (RT-QuIC) employing a truncated Syrian hamster recombinant protein substrate. Samples were evaluated blindly in parallel with appropriate positive and negative controls. Results from amplification assays were compared to one another and to obex immunohistochemistry, and were correlated to available clinical histories including CWD inoculum source (e.g. saliva, blood), genotype, survival period, and duration of clinical signs. We found that both sPMCAb and RT-QuIC were capable of amplifying CWD prions from cervid CSF, and results correlated well with one another. Prion seeding activity in either assay was observed in approximately 50% of deer with PrPd detected by IHC in the obex region of the brain. Important predictors of amplification included duration of clinical signs and time of first tonsil biopsy positive results, and ultimately the levels of PrPd identified in the obex by IHC. Based on our findings, we expect that both sPMCAb and RT-QuIC may prove to be useful detection assays for the detection of prions in CSF.


Chronic wasting disease (CWD) is an efficiently transmitted prion disease of cervids (e.g. deer, elk, and moose), and the only known prion disease affecting free-ranging, non-domestic animals. While the origins of CWD are uncertain, the disease has been present in wild cervid populations of northern Colorado and southern Wyoming for over 40 years [1,2] and has now been identified in both captive and free-ranging cervids in 22 states, 2 Canadian provinces, and the Republic of Korea [3]. With intensified national surveillance efforts, CWD continues to be identified in areas previously thought to be free of infection, including recent discoveries in Iowa, Texas, and Pennsylvania in 2012 [46]. The prevalence of CWD varies across North America, but can be as high as 30% in some areas of Colorado and up to 100% in captive populations [7,8].

Identification of infected cervids requires a sensitive and specific detection assay. Immunohistochemistry (IHC) of designated target tissues is currently considered the gold standard diagnostic test for CWD and other prion diseases of animals and man. The true sensitivity and specificity of IHC in the detection of infected individuals is unknown, though it has been demonstrated that the assay may underestimate the level of abnormal pathogenic prion protein (PrPd) in a given sample due to the necessity of a proteolytic pre-treatment step to abolish cellular prion protein (PrPC) cross-reactivity [911]. More recently, reports have focused on in vitro PrP conversion assays that involve either amplification of the protease-resistant prion protein (e.g. serial protein misfolding cyclic amplification - sPMCA) [10,12], fluorometric quantitation of recombinant PrP seeding activity as measured by thioflavin T binding (e.g. real time quaking-induced conversion – RT-QuIC) [13], or detection of altered PrP conformation (e.g. the conformation-dependent immunoassay - CDI) [14,15].

Definitive identification of infected individuals also relies on selection of an appropriate diagnostic specimen [1618]. In cervids, the retropharyngeal lymph nodes have been shown to identify infected animals at earlier stages of disease [16,18], though in other species, including cattle and man, the obex (a region of the caudal medulla) is generally considered to be the tissue of choice for the detection of TSE infection [19]. As retropharyngeal lymph node and obex samples are only available postmortem, efforts over the past decade have been made to identify peripherally accessible samples from which an antemortem diagnosis may be made. To this end, several investigators have examined peripheral lymphoid tissues of sheep and deer (e.g. tonsil and recto-anal mucosal associated lymphoid tissue, RAMALT) for antemortem detection of prion infection and found sensitivities (when compared to obex IHC) approaching 100% in both pre-clinical and clinical animals [2025]. More recently, using cerebrospinal fluid (CSF), amplification or fluorometric seeding activity assays have shown promise in distinguishing TSE-infected and uninfected individuals, including CWD-infected elk [26], scrapie-infected hamsters [27], and humans with Creutzfeldt-Jakob disease [28,29].

In the present study, we have applied a standardized sPMCA incorporating polytetrafluoroethylene (PTFE, Teflon®) beads (i.e. sPMCAb) [30] and RT-QuIC methodologies to examine the CSF of 48 CWD-exposed and naïve white-tailed deer (Odocoileus virginianus) with known inoculum source (e.g. blood, saliva, brain, etc.), genotype, survival periods, and serial tonsil biopsy IHC results. We hypothesized that sPMCAb and RT-QuIC results would correlate positively with IHC results, as well as to one another, suggesting CSF amplification may be a reliable means of antemortem detection of prion infection. We also sought to identify which specific historical correlates (e.g. incubation periods, etc.) were the strongest predictors of assay results derived from sPMCAb and RT-QuIC. We hypothesized that deer showing protracted signs of CWD, based on longitudinal clinical and lymphoid biopsy history, were more likely to show positive CSF amplification in either assay.

Materials and Methods

Ethics Statement

All animals were inoculated and maintained in dedicated, restricted access, indoor CWD research facilities in accord with Colorado State University Institutional Animal Care and Use Committee guidelines.  The animal care and use committee has approved this study under ACUC approval number 08-175A-01.

Infected cervids

In previous studies, forty eight white-tailed deer (Odocoileus virginianus) were exposed to CWD from positive or negative sources in various forms (e.g. urine and feces, saliva, environmental fomites, blood or blood components, or brain tissue) and by various routes (e.g. orally, intravenously, intracranially, or through environmental exposure) [3133]. The sources of inoculum included terminally-ill white-tailed or mule deer (Odocoileus hemionus) of unknown PrP genotype (courtesy of Michael Miller, Colorado Division of Parks and Wildlife; Terry Spraker, Colorado State University; the National Park Service; and the Wisconsin Department of Natural Resources), and sub-passage studies in white-tailed deer of either of two genotypes: homozygous for glycine (i.e. G/G) at cervid Prnp position 96, or heterozygous at that position, with alleles for glycine and serine (G/S). Study deer (summarized in Table 1 and Table S1) were selected on the basis of availability of CSF for analysis. Cerebrospinal fluid samples were collected cleanly early in the postmortem examination through aspiration from the ventral subarachnoid space at the atlanto-occipital site and frozen at -80°C for a range of 2-5 years prior to analysis.

Animal IDInoculumInitial tonsil + date (months)Duration of Clinical Signs (weeks)Mean IHC ScoreMean PMCA ScoreMean RT-QuIC Score
106Positive brain123110.52
108Whole blood121600.50
110Whole blood1228110.67
114Whole blood190000.2
123Negative brainNA000.170
133Whole blood1214110.38
136Positive brain1252110.38
137Whole blood1900.1666710
143Positive brain1240.3333300
302Whole blood1290.8333300
303B cells1200.6666700
322B cells1900.8333300
346B cells120100
347Whole blood690.3333300
371B cells190000
372Whole blood128110.22
374Whole blood68100
4116Whole blood120110.52
4119Whole blood12100.510.3
4128Blood cells120110.24
4488Negative brainNA0000
4502Whole blood61210.50.57
4506Blood cells190100.38
4509Negative brainNA0000
4513Whole blood62111
4516Negative brainNA0000
4521Blood cells1900.1666700

Table 1. Summary of background information and sPMCAb, RT-QuIC, and IHC results from individual deer.

Data on the inoculum, earliest tonsil biopsy positive result (in months), and duration of clinical signs (in weeks) prior to necropsy were collected on experimentally inoculated deer. Scores from each experimental assay were normalized and correlated to clinical histories.
Download CSV

Preparation of normal brain homogenate for sPMCA

To reduce the risk of contamination, normal brain homogenate (NBH), the substrate for prion conversion in vitro by sPMCA, was prepared from Tg(CerPrP)5037 mice in a room that had not previously been used for prion research [10,3436]. Following euthanasia and perfusion with 5mM EDTA in phosphate-buffered saline (PBS), whole brain was collected from naïve Tg(CerPrP)5037 mice and placed on ice. Brain homogenates were prepared as a 10% (w/v) solution in PMCA buffer (1% triton-X 100 [v/v], 5mM EDTA, and 150mM NaCl in PBS adjusted to a pH of 7.2) with the addition of Complete Protease Inhibitors (Roche Pharmaceuticals, Indianapolis, IN) using a dounce homogenizer. Homogenates were then centrifuged for 1 minute at 2000rpm and the supernatant frozen in single-experiment aliquots at –80°C in a room in which neither prion containing or potentially exposed materials or equipment were ever present until use in sPMCA. Each preparation was composed of brain from 4-6 mice to minimize the potential influence of expression variation in CerPrP or other co-factors [37,38]; multiple preparations of NBH were utilized over the course of the experiments.

Serial PMCA of cerebrospinal fluid

Cerebrospinal fluid was initially coded to allow for blinded evaluation by sPMCA, then thawed briefly with 10μl of test or control CSF added to 50μl of NBH with the addition of three 1/16” polytetrafluoroethylene (Teflon®) beads (i.e. sPMCAb, McMaster-Carr). Each test sample was evaluated in duplicate and in parallel with positive and negative controls, in adjacent wells of a 96-well plate (USA Scientific, Ocala, FL); along with NBH prepared from unexposed Tg(CerPrP)5037 mice as additional, unseeded negative controls. In each experimental run, between 25-35% of samples evaluated were from sham-inoculated deer (n=56/171 total replicates), and an additional 10-20% (n=25/171 total replicates) of the samples were unspiked, NBH negative controls. After sample loading, plates were sonicated using an ultrasonic processor (Qsonica, Newtown, CT) and incubated at 37°C. Sonication parameters entailed 40 second bursts at power level 7.0, followed by 30 minutes of incubation. Each duplicate sample was processed through 92 cycles of sonication performed over 48 hours, followed by a transfer of a 10µl aliquot of sonicated material into 50μl of fresh NBH with PTFE beads for serial amplification over a total of three rounds. Following each round of amplification, samples were evaluated by western blotting, as described below, for the presence of PrPd. For each sample, the number of positive rounds in each duplicate were tallied to allow for relative quantification of PrPd; thus, each sample could have a maximum score of 6 (e.g. three positive rounds in duplicate experiments), and a minimum score of zero. Scores were then normalized by dividing by the total number of rounds run (i.e. six for each sample) to arrive at a final score between 0-1.

Western blotting (WB) of sPMCAb amplified cerebrospinal fluid

Following each round of amplification, an aliquot of each sonicated sample was subjected to western blotting for evaluation of PrPd signal. Fifteen µl of sample homogenate were mixed with 7µl of sample buffer (0.1% [v/v] triton-X 100 and 4% [w/v] SDS in PBS), digested with 3µl proteinase-K at 500µg/ml (final concentration: 60µg/ml) for 20’ at 37°C followed by 10’ at 45°C. Twenty µl of this preparation were run on a pre-cast 12% bis-tris SDS-PAGE gel (Invitrogen) in a Bio-Rad electrophoresis apparatus for 1 hour at 160mV. Samples were then transferred to a PVDF membrane using a Transblot Turbo transfer system (Bio-Rad). PVDF membranes were subsequently probed with a PrP-specific monoclonal antibody (BAR224 conjugated to horse radish peroxidase, HRP) diluted 1:20,000 in 5% (w/v) powdered milk in 0.2% Tween-20 in tris-buffered saline (TBST) for 1 hour. Following washing with 0.5% Tween 20 in tris-buffered saline, immunoreactivity was detected using an enhanced chemiluminescent detection system (ECL-plus, Amersham Biosciences) in an LAS 3000 imaging system. (Fuji Photo Film, Fuji Inc, Valhalla, NY)

Preparation of recombinant PrP for RT-QuIC

Real-time quaking-induced conversion assays were performed with recombinant Syrian hamster PrP (SHrPrP) encoding residues 90-231 in pET41b and expressed and purified as previously described [39]. In brief, 1 liter cultures of lysogeny broth (LB) containing Auto Induction™ supplements (EMD Biosciences) were inoculated with SHrPrP-expressing Rosetta strain E. coli, grown overnight, and harvested when optical density (OD, 600 nm) of ~3 was reached. Cells were lysed with Bug Buster™ reagent with supplemented Lysonase™ (EMD Biosciences) and inclusion bodies (IB) were harvested by centrifugation of the lysate at 15,000 x g. IB pellets were washed twice and stored at -80 °C until purification (typically 24 hours or less). IB pellets were solubilized in 8 M guanidine hydrochloride (GuHCl) in 100 mM NaPO4 and 10 mM Tris pH 8.0, clarified by centrifugation at 15,000 g for 15’ and added to Super Flow nickel-nitrilotriacetic acid (Ni-NTA) resin (Qiagen) pre-equilibrated with denature buffer (6.0 M GuHCl, 100 mM NaPO4, 10 mM Tris pH 8.0). Denatured SHrPrP and Ni-NTA resin was incubated by rotating at room temperature for 45 min and then added to an XK fast protein liquid chromatography column (GE Healthcare). Refolding was achieved on column using a linear refolding gradient of denature buffer to refold buffer (100 mM NaPO4, 10 mM Tris pH 8.0) over 340 ml at 0.75 ml/min. SHrPrP was eluted with a linear gradient of refold buffer to elution buffer (100 mM NaPO4, 10 mM Tris pH 8.0 500 mM imidazole pH 5.5) over 100 ml at 2.0 ml/min. Fractions were pooled and dialyzed against two changes of 4 liters of dialysis buffer [20 mM NaPO4 pH 5.5). Recovered SHrPrP was adjusted to a final concentration of ~0.5 mg/ml.

RT-QuIC analysis of cerebrospinal fluid

Blinded cervid CSF samples were diluted 1:10 in RT-QuIC dilution buffer (PBS with 0.05% sodium dodecyl sulfate, SDS). Five µl of this 10-1 dilution were added to 95µl of RT-QuIC reaction buffer, consisting of 50mM NaPO4, 250mM NaCl, 1.0mM ethylenediaminetetraacetic acid tetrasodium salt (EDTA), 10µM Thioflavin T (ThT), and 0.1mg/ml Syrian hamster rPrPC, to yield a final volume of 100µl. Each test sample was repeated in triplicate in three separate experiments. Positive controls, consisting of 5µl of a 10-3 dilution of pooled CWD-positive brain from six experimentally infected white-tailed deer (cervid brain pool 6, CBP6) spiked into 95µl of RT-QuIC reaction buffer, were included in triplicate in each experiment. Negative controls, also repeated in triplicate, consisted of CSF samples from sham-inoculated white-tail deer prepared in a manner consistent with study deer, as well as untreated RT-QuIC reaction buffer spiked with 5µl of RT-QuIC dilution buffer. Reactions were prepared in a black 96-well, optical-bottom plate, which were then sealed and incubated in a BMG Labtech PolarstarTM fluorimeter at 42°C for 48 hours (192 fifteen minute cycles) with intermittent shaking cycles; specifically 1 minute shakes (700rpm, double orbital pattern) interrupted by 1 minute rest periods. Thioflavin T fluorescence measurements (450nm excitation and 480nm emission) were taken every 15 minutes with the gain set at 1200, with the relative fluorescence units (RFU) for each triplicate sample progressively monitored against time with orbital averaging and 20 flashes/well at the 4mm setting.

Replicates were considered positive when they crossed a pre-defined positive threshold, calculated as five standard deviations above the mean fluorescence of all negative controls (CSF from sham-inoculated deer and untreated replicates spiked with RT-QuIC dilution buffer) in the assay. Times at which a sample crossed the positive threshold (Ct) were recorded; samples remaining below the threshold were considered negative. Ct scores were derived by dividing the average Ct time of positive controls in an experiment by the average Ct times of individual triplicates from unknown CSF samples, resulting in RT-QuIC Ct scores typically between 0-1. Each sample was repeated in triplicate in three separate experiments to estimate the reproducibility of RT-QuIC scores. The fluorometric curves generated from sample amplification were analyzed using GraphPad Prism 5.0, with the real-time data from each well fit with a Boltzmann sigmoidal curve. From these curves, the slope and final (maximal) plateau fluorescence were determined. Parameters for each triplicate sample were averaged across the 3 experiments.

Optimization of sPMCAb and RT-QuIC for use with cervid CSF

To optimize each of the seeded amplification assays, we selected CSF from six study deer, inoculated with a variety of substrates through various routes, which were found to have large accumulations of PrPd in their obex (i.e. immunohistochemical scores of 3/3 – deer 106, 110, 133, 144, 346, and 4116), and a sham inoculated deer (deer #4488) for dilutional analysis. For sPMCAb, we selected a range of spike volumes including 20μl, 10μl, and 1μl into 50μl of NBH. Samples were amplified as described above for sPMCAb, over a total of three rounds and evaluated by western blotting. For RT-QuIC, we selected a range of dilutions of CSF into RT-QuIC dilution buffer: 10° (i.e. no dilution), 10-1, and 10-2; 5μl of each dilution were spiked into 95μl of RT-QuIC amplification buffer and subjected to amplification over a 24hr period as described above.

Immunohistochemistry of cervid obex

Immunohistochemistry (IHC) was performed using the protocol described by Spraker et al [40]. Briefly, 3–5 mm sections of formalin fixed formic acid treated tissues were deparaffinized at 60–70°C for 1 hour, rehydrated via a series of xylene/ethanol baths, and treated in formic acid a second time (5 minutes) prior to a 20 minute antigen retrieval (Dako Target Retrieval Solution 10×) cycle in a 2100 Retriever™ (PickCell Laboratories). Slides were further processed with the aid of a Ventana Discovery™ autostainer utilizing the Ventana Red Map™ stain kit, the PrPd specific primary antibody BAR224, and a biotinylated secondary goat anti mouse antibody (Ventana). After autostaining, the slides were quickly rinsed in a warm water detergent solution, passed through a series of graded alcohols and cover-slipped. Stained slides were blindly evaluated by two independent evaluators and assigned scores between 0-3, with zero being a complete absence of staining and three indicating intense PrPd-specific staining throughout the dorsal motor nucleus of the vagus and remainder of the obex. Scores were then averaged and divided by the maximum score (i.e. 3), to arrive at normalized IHC scores between 0-1.

Statistical analyses of experimental results

To firstly evaluate if there was a relationship between sPMCAb, RT-QuIC, and IHC scores, we correlated normalized scores against one another using Spearman rank correlation. We then evaluated sPMCAb and RT-QuIC relative to IHC for sensitivity, specificity, positive and negative predictive values.

We next evaluated and contrasted how well sPMCAb and RT-QuIC test scores were predicted by inoculum, genotype (GG/GS), month at which the individual was tonsil biopsy positive (by IHC), survival period (months), duration of clinical signs (weeks), and IHC obex scores. To undertake this we employed an Information Theoretic approach [41], whereby all single and pair-wise combinations of predictor variables were evaluated, using logistic regression, and ranked based on Akaike Information Criterion corrected for small sample size (AICc) [41]. This approach is of value because it enables determination of the most parsimonious model or set of models to explain sPMCAb and RT-QuIC scores, and also calculation of Variable Importance weights to determine the relative importance of one predictor variable over others [41]. Because Information Theory departs from frequentist based statistical approaches, which are dependent on P-values, we also calculated McFadden r2 values (appropriate for calculating fit of logistic regression models) so that the relative fit of models to the data could be assessed.

The above analyses were also extended to evaluate how well predictor variables performed at predicting Boltzmann slope and maximum fluorescence for deer with positive RT-QuIC diagnoses. These values could only be assessed for deer returning a positive RT-QuIC result, causing the dataset to be reduced (n=14) and, accordingly, analyses were restricted to all pair-wise combinations of predictor variables so that the number of models did not greatly exceed the amount of data [41]. All analyses were conducted using the program R ( Models were conducted based on a binomial data distribution, with the exception of models predicting maximum fluorescence which was based on a Gaussian distribution. Because models of maximum fluorescence were based on a Gaussian distribution, a regular coefficient of variation was computed, rather than McFadden r2.


Immunohistochemical detection of PrPd in cervid obex

Using previously reported methods for IHC detection of PrPd in cervid tissues, we analyzed obex collected at necropsy from each individual deer. Independent scores (0-3) from two separate evaluators were averaged and divided by the maximum score (i.e. 3), to arrive at a normalized score between 0-1. IHC demonstrated PrPd in 26/48 individuals (Table 1), with normalized scores from positive animals ranging from 0.17-1 (mean: 0.77). (Figure 1)

Figure 1. Immunohistochemistry of white-tailed deer obex.

Immunohistochemistry results from deer #4502 (A), 4129 (B), 144 (C) and 4488 (D). Normalized scores from IHC+ animals ranged from 0.17 (e.g. #4129) to 1.0 (e.g. #4502 and 144). Sham-inoculated deer (e.g. #4488, D) were consistently negative by IHC.

Optimal spike volumes vary for sPMCAb and RT-QuIC

To evaluate the optimal volumes for seeded amplification in both sPMCAb and RT-QuIC, we evaluated a range of starting volumes and dilutions of CSF from six CWD-positive deer in each assay: 20μl, 10μl and 1μl of CSF into 50μl of NBH with sPMCAb, and 10°, 10-1, and 10-2 dilutions of CSF in RT-QuIC. For sPMCAb, we observed the most efficient seeded amplification with 10μl of spike volume (5/6 showing positive amplification). Less efficient amplification was observed with 20μl (2/6 positive) and 1μl (1/6 positive) spike volumes. For RT-QuIC, seeded amplification had the highest precision when using 10-1 dilutions of CSF, showing seeded amplification in 3/3 replicates in 5/6 CSF samples; higher and lower dilutions of CSF showed less repeatable amplification. (Figure 2)

Figure 2. Optimization of sPMCAb and RT-QuIC.

Various spike volumes and dilutions were chosen for optimization of each assay for use with cervid CSF. Cerebrospinal fluid from six CWD-positive deer (#106, 110, 133, 144, 346, and 4116) and one CWD-negative deer (#4488) were chosen for optimization. With sPMCAb (A.), spike volumes including 20μl (Lanes 2, 5, 8, 11, 14, 17, and 20), 10μl (Lanes 3, 6, 9, 12, 15 18, and 21), and 1μl (Lanes 4, 7, 10, 13, 16, 19, and 22) were evaluated using three rounds of sPMCAb. For RT-QuIC (B.), dilutions of CSF including 10°, 10-1, and 10-2 were evaluated in 48hr experiments. The threshold for positive amplification (Ct) is represented by the horizontal hashed red line. Data encompassing results from IHC, sPMCAb, and RT-QuIC from CSF samples from seven deer (C.) were evaluated to estimate the optimum spike volumes and dilutions of CSF, and revealed that 10μl of whole CSF appeared to be optimal for sPMCAb, while a 10-1 dilution of CSF had the greatest precision in RT-QuIC analysis.

Serial PMCA prion seeding activity in cerebrospinal fluid of white-tailed deer

Using a modified sPMCA technique incorporating PTFE beads (sPMCAb) [30], amplification was observed in 16/48 cerebrospinal fluid samples (Table 1). The mean number of rounds positive in amplified samples was 5.1 out of 6, ranging from 0-3 rounds for each duplicate and yielding sPMCAb scores from 0.17-1.0 (mean: 0.85). A single aliquot (i.e. 1/56 replicates from CWD-negative control samples) of CSF from a sham-inoculated deer (deer 123) yielded a false positive result in a single replicate in sPMCAb, though failed to demonstrate amplification in RT-QuIC (see below). The remaining controls (CSF from additional negative control deer, additional aliquots of CSF from deer 123, and unspiked negative controls) failed to amplify PrPd during the course of sPMCAb analysis. (Figure 3)

Figure 3. Serial PMCAb detection of prion seeding activity in CSF from CWD-positive deer.

Cerebrospinal fluid samples, along with positive and negative controls, were assayed in duplicate for three rounds of sPMCAb, with the total number of positive rounds tallied for each sample. Positive control samples were positive in each duplicate through all rounds (Lanes 2 and 3, for a total of 6/6 positive rounds and a normalized score of 1), while CSF samples varied in their total number of positive rounds. CSF from deer #4502 (Lanes 4 and 5) and #4129 (Lanes 8 and 9) demonstrated amplification in 3/6 rounds of sPMCAb, for a normalized score of 0.5, while CSF from deer 144 was positive in 6/6 rounds (Lanes 6 and 7). Negative controls, including deer #4488 (Lanes 11 and 12) and unspiked brain homogenate (NBH – lanes 12 and 13) remained negative throughout all rounds.

RT-QuIC detects prion seeding activity in cerebrospinal fluid of white-tailed deer

An RT-QuIC assay employing truncated Syrian hamster recombinant prion protein [13] as a substrate successfully amplified CWD prion seeding activity in 14/48 CSF samples (Table 1). By comparing threshold times to those of a 10-3 dilution of CBP6 positive control, resultant RT-QuIC scores from positive animals ranged from 0.2-1.0, with a mean score of 0.47, inferring that the level of prion-seeding activity was multiple orders of magnitude lower than that found in the CNS. All samples were repeated in three separate experiments, and negative controls (CSF from sham-inoculated deer and untreated controls spiked with RT-QuIC dilution buffer) remained negative throughout the course of analysis. Over the course of multiple experiments, we found that RT-QuIC scores were fairly reproducible, with standard error of Ct scores between separate experiments ranging from 0.01-0.07. (Figure 4)

Figure 4. RT-QuIC seeded amplification detected by ThT fluorescence in CSF of deer.

A 10-3 dilution of cervid brain pool (red), as well as 10-1 dilutions of CSF from deer #4502 (blue) and 144 (orange) showed evidence of prion seeding activity (ThT binding) in 48 hour RT-QuIC experiments. The time to reaching threshold fluorescence (Ct – red horizontal dashed line) varied amongst RT-QuIC positive animals, and correlated with IHC scores. CSF from some CWD-positive deer (e.g. deer #4129), all sham-inoculated deer (e.g. deer #4488), and all untreated controls failed demonstrate seeding activity.

Inter-assay correlation and relationship to available historical data

All three assays (IHC, sPMCAb, and RT-QuIC) were positively correlated (Table 2). The strength of agreement was greatest between sPMCAb and RT-QuIC and less strong vs. IHC. Prion seeding activity in the CSF was detected by sPMCAb and RT-QuIC in ~50% of obex IHC-positive deer (Table 3). Both amplification assays were more likely to align with IHC-negative results (specificity), and deer with a positive sPMCAb or RT-QuIC score had a high probability of detection by IHC (positive predictive value). However, absence of amplification by sPMCAb or RT-QuIC was only moderately likely to align with absence of detection by IHC (negative predictive value). Overall, sPMCAb and RT-QuIC were broadly comparable.


Table 2. Spearman correlation of diagnostic methods.

Upper diagonal represents correlation coefficients (ρ) and lower diagonal corresponding P-values.
Download CSV
Positive predictive value87.592.9
Negative predictive value62.561.8

Table 3. Evaluation of how PMCA and QuIC compare to IHC in identification of CWD-positive deer.

Download CSV

Using regressions and Information Theory, we found that sPMCAb and RT-QuIC diagnostic scores were predicted by single and pair-wise combinations of inoculum, genotype, month tonsil biopsy positive, incubation period, duration of clinical signs and IHC diagnosis (Table S2). IHC-positivity was the best (most parsimonious) predictor of sPMCAb-positivity (Figure 4a, slope coefficient ± SE 2.371 ± 1.108). Next in variable importance weight was duration of clinical signs (slope coefficient ± SE 0.112 ± 0.065), while the remaining predictors were less important. IHC score was also the best predictor of RT-QuIC-positivity (Figure 4, slope coefficient ± SE 4.190 ± 2.576). The month at which an animal was tonsil biopsy positive was the second best predictor (slope coefficient ± SE -0.212 ± 0.117), inferring that a longer period for which prions might accumulate in the CNS favored prion entry into the CSF. For both sPMCAb and RT-QuIC, McFadden r2 values were generally related to variable importance weights (Figure 4a), except for inoculum. However, inoculum source and type was a relatively unimportant predictor variable, owing to the number of categories within this predictor causing low model parsimony (Figure 5a, Table S2). Boltzmann slope, a modeled, best fit slope based on available data, for RT-QuIC yielded low McFadden r2 values (Figure 5b, Tables S1 and S3). A similar pattern was observed for predictors of Boltzmann maximum fluorescence for RT-QuIC, although inoculum was distinctly less important than the other variables (Figure 5b, Tables S1 and S3).

Figure 5. The relative importance of predictor variables for (a) sPMCAb (blue circles) and RT-QuIC (green circles) scores, and (b) Boltzmann slope (blue circles) and maximum fluorescence (green circles).

Size of circles represents McFadden r2 values (range <0.001 – 0.292). Where r2 values were <0.03 a blue or green arrow indicates their position for clarity. Although inoculum r2 values were often relatively high (larger circles) compared to other predictors, their importance was low indicating that the high r2 values for this variable was a statistical artifact of the high number of inoculum categories.


As the only known prion disease affecting free-ranging animals, the early and antemortem detection of chronic wasting disease infection in cervids is of paramount importance in disease management. The detection of CWD and other prions in bodily fluids, including saliva and CSF, has been the subject of recent research [26,42]. In the present study, we compared sPMCAb and RT-QuIC analysis of CSF from CWD-exposed cervids in relation to IHC detection of PrPd in the obex region of the brain and other available historical information (e.g. inoculum source, incubation period, genotype, etc.).

We found that results from both in vitro detection assays (sPMCAb and RT-QuIC) were highly correlated, with ~50% of deer with IHC-positive PrPd in the obex showing demonstrable prion seeding activity in the CSF. Assay agreement was fairly high between sPMCAb and RT-QuIC, and while sPMCAb identified slightly more IHC-positive animals than RT-QuIC (16/26 vs. 14/26 for RT-QuIC), assay scores seemed to correlate slightly better between RT-QuIC and IHC using Spearman correlation (0.690 vs. 0.593 for sPMCAb). These results are consistent with a previous study evaluating the CSF of CWD-infected Rocky Mountain elk (Cervus elaphus nelsoni), in which elk with higher levels of PrPd demonstrated by IHC of the obex were more likely to be sPMCA positive [26]. In addition to our findings in samples from CWD-positive animals, sPMCAb produced a false positive result from a single aliquot of CSF from a negative control deer (1/56 total replicates), implying slight deficits in specificity for this assay. Previous experiments in our lab and elsewhere have found specificities of <100% for three-round sPMCA [36], so these results perhaps are not surprising. As amplification was only observed in a single aliquot of CSF from one of the negative control deer, while additional aliquots from the same deer remained negative, sample contamination seems probable.

A number of previous studies employing amplification assays for the detection of PrPd in CSF samples have been reported, predominantly in IHC-confirmed cases of CJD in humans and scrapie of hamsters [27,28]. In these reports, seeded amplification was observed in 75-80% of CJD or scrapie-positive individuals. Another study evaluating the CSF of Rocky Mountain elk by sPMCA was only able amplify protease-resistant prion from 3/15 (20%) samples from CWD-positive animals, somewhat lower than the percentage of prion-seeding activity seen in the CSF of white-tailed deer in the present study. Taken collectively, this seems to suggest that prionorrhachia (shedding of prions into the CSF) is either intermittent or inconsistent in CWD-infected cervids, or levels of prion seeding activity are very low, compared to scrapie-infected hamsters or CJD-infected humans. A systematic evaluation of TSE-infected CNS tissues of deer and other species may help pinpoint the probable source of CSF shedding (e.g. ependymal cells [43]) via correlating histopathological or PrPd-deposition analysis to amplification results.

In addition to comparing inter-assay results, we sought correlations between amplification results and other available historical data from experimental animals, a study component our laboratory is uniquely suited to undertake. We found that two aspects of the clinical history – duration of clinical signs and earliest time point of tonsil biopsy positivity – were positively correlated to sPMCAb and RT-QuIC results, respectively. This finding seems to imply that the longitudinal accumulation of CWD prions in the central nervous system of deer, either following extended incubation periods or early access to the lymphoreticular system – the predominant amplification site in affected deer – has an important impact on the accumulation of detectable prions in the cerebrospinal fluid. In previous studies, we have found a correlation between genotype and successful amplification of CWD prions in lymphoreticular tissues [11]; this finding does not appear to translate to seeded amplification in the CSF of deer as there was little correlation between genotype and likelihood of detection via either sPMCAb or RT-QuIC.

Apart from simply investigating the presence or absence of seeding activity, we also examined relationships between available historical data and advanced analysis of RT-QuIC amplification curves, to determine whether additional insights could be gained into the relationships of seeding activity curves and sample source. We focused on amplification amplitude and slope, and attempted to correlate these to each of the a priori variables. In this examination, we found no substantive relationship of any of the variables to either mean slope or peak fluorescence. These results may imply that these metrics of RT-QuIC analysis do not arise from inherent properties of the amplifiable prions present in the CSF, or may be due to the small sample size available for the statistical analyses. We are presently analyzing additional samples from multiple cervid sources to further examine whether advanced amplification metrics may correlate to certain aspects of an individual’s clinical history.

In summary, the present study evaluated two amplification assays – sPMCAb and RT-QuIC – for their ability to amplify PrPd in the CSF of CWD-exposed and naïve white-tailed deer. Results between the two assays correlated well with each other and to IHC results from obex collected at necropsy, albeit with reduced sensitivity. A priori variables, notably date of tonsil biopsy positivity and duration of clinical signs, influenced the likelihood of a sample being positive by either assay, while our post-amplification analyses (slope of curve or peak fluorescence) did not correlate with clinical histories. Based on our findings, we believe that amplification assays hold continued promise in the detection of prion-infection using post- or ante-mortem samples and our future work will continue to evaluate the utility of these assays in detecting seeding activity in these tissues and biological fluids.

Supporting Information

Table S1.

Summary of genotype, incubation period and advanced RT-QuIC metrics from study deer. Neither genotype nor incubation period were significantly correlated to sPMCA or RT-QuIC results; while advanced metrics derived from RT-QuIC experiments had no direct correlation with any of the historical variables from CWD-exposed or sham inoculated deer.


Table S2.

Comparison of predictor models to estimate sPMCAb and RT-QuIC scores. Models represent logistic regressions of all individual and pair-wise combinations of predictors, and are ranked from most parsimonious to least parsimonious by ∆AICc. 1=Inoculum (blood and blood products, environment, urine/feces, negative, brain, saliva), 2=IHC, 3=Genotype (GG/GS), 4=Month sample was tonsil biopsy positive (by IHC), 5=Incubation period (months), and 6=Duration of clinical signs (weeks). K=number of model parameters, w=model weight, r2=McFadden’s r2.


Table S3.

Comparison of predictor models to estimate Boltzmann slope and maximum fluorescence of samples diagnosed as CDW positive by QuIC. Models represent logistic regressions for slope and linear regressions for fluorescence of all pair-wise combinations of predictors, and are ranked from most parsimonious to least parsimonious by ∆AICc. 1=Inoculum (blood and blood products, environment, urine/feces, negative, brain, saliva), 2=IHC, 3=Genotype (GG/GS), 4=Month sample was tonsil biopsy positive (by IHC), 5=Incubation period (months), and 6=Duration of clinical signs (weeks). K=number of model parameters, w=model weight, r2=McFadden’s r2 for slope and r2 for fluorescence.



We sincerely thank all of those who have made important contributions to this manuscript, including David Osborn and Sally Dahmes, without whom the primary experiments in cervids could not have been performed, and Drs. Michael Miller, Terry Spraker, and Margaret Wild, who provided initial samples of CWD+ brain and bodily fluids. For long term care and sample collection from source deer and transgenic mice, we sincerely thank Sheila Hays, Kelly Anderson, and Jeanette Hayes-Klug. Without each of their contributions and unique skills, this work could not have been completed.

Author Contributions

Conceived and designed the experiments: NJH DMS AV. Performed the experiments: NJH AV KD. Analyzed the data: NJH AV DH KD SC EH. Contributed reagents/materials/analysis tools: NJH DMH SC KD CM EH. Wrote the manuscript: NJH AV KD SC DMS CKM EH DH.


  1. 1. Williams ES, Young S (1980) Chronic wasting disease of captive mule deer: a spongiform encephalopathy. J Wildl Dis 16: 89-98. doi: PubMed: 7373730.
  2. 2. Williams ES, Young S (1982) Spongiform encephalopathy of Rocky Mountain elk. J Wildl Dis 18: 465-471. doi: PubMed: 7154220.
  3. 3. Sigurdson CJ (2008) A prion disease of cervids: chronic wasting disease. Vet Res 39: 41. doi: PubMed: 18381058.
  4. 4. (2012) Chronic wasting disease, cervid; ——USA (Iowa). Archive number 20120721.1210369.
  5. 5. (2012) Chronic wasting disease, cervid; ——USA (Texas). Archive number 20120711.1197183.
  6. 6. (2012) Chronic wasting disease, cervid; ——USA (Pennsylvania). Archive number 20121014.1341794.
  7. 7. Keane DP, Barr DJ, Bochsler PN, Hall SM, Gidlewski T et al. (2008) Chronic wasting disease in a Wisconsin white-tailed deer farm. J Vet Diagn Invest 20: 698-703. doi: PubMed: 18776116.
  8. 8. Miller MW, Conner MM (2005) Epidemiology of chronic wasting disease in free-ranging mule deer: spatial, temporal, and demographic influences on observed prevalence patterns. J Wildl Dis 41: 275-290. doi: PubMed: 16107661.
  9. 9. Safar JG, Geschwind MD, Deering C, Didorenko S, Sattavat M et al. (2005) Diagnosis of human prion disease. Proc Natl Acad Sci U S A 102: 3501-3506. doi: PubMed: 15741275.
  10. 10. Haley NJ, Mathiason CK, Zabel MD, Telling GC, Hoover EA (2009) Detection of sub-clinical CWD infection in conventional test-negative deer long after oral exposure to urine and feces from CWD+ deer. PLOS ONE 4: e7990. doi: PubMed: 19956732.
  11. 11. Haley NJ, Mathiason C, Carver S, Telling GC, Zabel MC et al. (2012) Sensitivity of protein misfolding cyclic amplification vs. immunohistochemistry in antemortem detection of CWD infection. J Gen Virol.
  12. 12. Saborio GP, Permanne B, Soto C (2001) Sensitive detection of pathological prion protein by cyclic amplification of protein misfolding. Nature 411: 810-813. doi: PubMed: 11459061.
  13. 13. Wilham JM, Orrú CD, Bessen RA, Atarashi R, Sano K et al. (2010) Rapid end-point quantitation of prion seeding activity with sensitivity comparable to bioassays. PLoS Pathog 6: e1001217. PubMed: 21152012.
  14. 14. Thackray AM, Hopkins L, Bujdoso R (2007) Proteinase K-sensitive disease-associated ovine prion protein revealed by conformation-dependent immunoassay. Biochem J 401: 475-483. doi: PubMed: 17018021.
  15. 15. Safar J, Wille H, Itri V, Groth D, Serban H et al. (1998) Eight prion strains have PrP(Sc) molecules with different conformations. Nat Med 4: 1157-1165. doi: PubMed: 9771749.
  16. 16. Spraker TR, Balachandran A, Zhuang D, O'Rourke KI (2004) Variable patterns of distribution of PrP(CWD) in the obex and cranial lymphoid tissues of Rocky Mountain elk (Cervus elaphus nelsoni) with subclinical chronic wasting disease. Vet Rec 155: 295-302. doi: PubMed: 15478500.
  17. 17. Keane DP, Barr DJ, Keller JE, Hall SM, Langenberg JA et al. (2008) Comparison of retropharyngeal lymph node and obex region of the brainstem in detection of chronic wasting disease in white-tailed deer (Odocoileus virginianus). J Vet Diagn Invest 20: 58-60. doi: PubMed: 18182509.
  18. 18. Monello RJ, Powers JG, Hobbs NT, Spraker TR, O'Rourke KI et al. (2013) Efficacy of antemortem rectal biopsies to diagnose and estimate prevalence of chronic wasting disease in free-ranging cow elk (Cervus elaphus nelsoni). J Wildl Dis 49: 270-278. doi: PubMed: 23568902.
  19. 19. Wells GA, Hancock RD, Cooley WA, Richards MS, Higgins RJ et al. (1989) Bovine spongiform encephalopathy: diagnostic significance of vacuolar changes in selected nuclei of the medulla oblongata. Vet Rec 125: 521-524. doi: PubMed: 2512718.
  20. 20. O'Rourke KI, Baszler TV, Besser TE, Miller JM, Cutlip RC et al. (2000) Preclinical diagnosis of scrapie by immunohistochemistry of third eyelid lymphoid tissue. J Clin Microbiol 38: 3254-3259. PubMed: 10970367.
  21. 21. van Keulen LJ, Schreuder BE, Meloen RH, Mooij-Harkes G, Vromans ME et al. (1996) Immunohistochemical detection of prion protein in lymphoid tissues of sheep with natural scrapie. J Clin Microbiol 34: 1228-1231. PubMed: 8727908.
  22. 22. Hill AF, Zeidler M, Ironside J, Collinge J (1997) Diagnosis of new variant Creutzfeldt-Jakob disease by tonsil biopsy. Lancet 349: 99-100. doi: PubMed: 8996424.
  23. 23. Sigurdson CJ, Williams ES, Miller MW, Spraker TR, O'Rourke KI et al. (1999) Oral transmission and early lymphoid tropism of chronic wasting disease PrPres in mule deer fawns (Odocoileus hemionus). J Gen Virol 80 ( 10): 2757-2764. PubMed: 10573172.
  24. 24. González L, Dagleish MP, Bellworthy SJ, Sisó S, Stack MJ et al. (2006) Postmortem diagnosis of preclinical and clinical scrapie in sheep by the detection of disease-associated PrP in their rectal mucosa. Vet Rec 158: 325-331. doi: PubMed: 16531580.
  25. 25. Spraker TR, Gidlewski TL, Balachandran A, VerCauteren KC, Creekmore L et al. (2006) Detection of PrP(CWD) in postmortem rectal lymphoid tissues in Rocky Mountain elk (Cervus elaphus nelsoni) infected with chronic wasting disease. J Vet Diagn Invest 18: 553-557. doi: PubMed: 17121082.
  26. 26. Nichols TA, Spraker TR, Gidlewski T, Powers JG, Telling GC et al. (2012) Detection of prion protein in the cerebrospinal fluid of elk (Cervus canadensis nelsoni) with chronic wasting disease using protein misfolding cyclic amplification. J Vet Diagn Invest 24: 746-749. doi: PubMed: 22621952.
  27. 27. Orrú CD, Wilham JM, Hughson AG, Raymond LD, McNally KL et al. (2009) Human variant Creutzfeldt-Jakob disease and sheep scrapie PrP(res) detection using seeded conversion of recombinant prion protein. Protein Eng Des Sel 22: 515-521. doi: PubMed: 19570812.
  28. 28. McGuire LI, Peden AH, Orrú CD, Wilham JM, Appleford NE et al. (2012) Real time quaking-induced conversion analysis of cerebrospinal fluid in sporadic Creutzfeldt-Jakob disease. Ann Neurol 72: 278-285. doi: PubMed: 22926858.
  29. 29. Atarashi R, Satoh K, Sano K, Fuse T, Yamaguchi N et al. (2011) Ultrasensitive human prion detection in cerebrospinal fluid by real-time quaking-induced conversion. Nat Med 17: 175-178. doi: PubMed: 21278748.
  30. 30. Gonzalez-Montalban N, Makarava N, Ostapchenko VG, Savtchenk R, Alexeeva I et al. (2011) Highly efficient protein misfolding cyclic amplification. PLoS Pathog 7: e1001277. PubMed: 21347353.
  31. 31. Mathiason CK, Hayes-Klug J, Hays SA, Powers J, Osborn DA et al. (2010) B cells and platelets harbor prion infectivity in the blood of deer infected with chronic wasting disease. J Virol 84: 5097-5107. doi: PubMed: 20219916.
  32. 32. Mathiason CK, Hays SA, Powers J, Hayes-Klug J, Langenberg J et al. (2009) Infectious Prions in Pre-Clinical Deer and Transmission of Chronic Wasting Disease Solely by Environmental Exposure. PLOS ONE 4: e5916. doi: PubMed: 19529769.
  33. 33. Mathiason CK, Powers JG, Dahmes SJ, Osborn DA, Miller KV et al. (2006) Infectious prions in the saliva and blood of deer with chronic wasting disease. Science 314: 133-136. doi: PubMed: 17023660.
  34. 34. Haley NJ, Seelig DM, Zabel MD, Telling GC, Hoover EA (2009) Detection of CWD prions in urine and saliva of deer by transgenic mouse bioassay. PLOS ONE 4: e4848. doi: PubMed: 19293928.
  35. 35. Kurt TD, Telling GC, Zabel MD, Hoover EA (2009) Trans-species amplification of PrP(CWD) and correlation with rigid loop 170N. Virology 387: 235-243. doi: PubMed: 19269662.
  36. 36. Haley NJ, Mathiason CK, Carver S, Zabel M, Telling GC et al. (2011) Detection of CWD prions in salivary, urinary, and intestinal tissues of deer: potential mechanisms of prion shedding and transmission. J Virol.
  37. 37. Browning SR, Mason GL, Seward T, Green M, Eliason GA et al. (2004) Transmission of prions from mule deer and elk with chronic wasting disease to transgenic mice expressing cervid PrP. J Virol 78: 13345-13350. doi: PubMed: 15542685.
  38. 38. Angers RC, Seward TS, Napier D, Green M, Hoover E et al. (2009) Chronic wasting disease prions in elk antler velvet. Emerg Infect Dis 15: 696-703. doi: PubMed: 19402954.
  39. 39. Wilham JM, Orrú CD, Bessen RA, Atarashi R, Sano K et al. (2010) Rapid end-point quantitation of prion seeding activity with sensitivity comparable to bioassays. PLoS Pathog 6: e1001217. PubMed: 21152012.
  40. 40. Spraker TR, Miller MW, Williams ES, Getzy DM, Adrian WJ et al. (1997) Spongiform encephalopathy in free-ranging mule deer (Odocoileus hemionus), white-tailed deer (Odocoileus virginianus) and Rocky Mountain elk (Cervus elaphus nelsoni) in northcentral Colorado. J Wildl Dis 33: 1-6. doi: PubMed: 9027685.
  41. 41. Burnham KP, Anderson DR (2002) Model Selection and Inference: A Practical Information-Theoretic Approach. New York: Spring-Verlag.
  42. 42. Henderson DM, Manca M, Haley NJ, Denkers ND, Nalls AV et al. Rapid Antemortem Detection of CWD Prions in Deer Saliva. PLOS ONE 8: e74377. doi: PubMed: 24040235.
  43. 43. Fournier JG, Adjou K, Grigoriev V, Deslys JP (2008) Ultrastructural evidence that ependymal cells are infected in experimental scrapie. Acta Neuropathol 115: 643-650. doi: PubMed: 18369649.