Use of different RT-QuIC substrates for detecting CWD prions in the brain of Norwegian cervids

Chronic wasting disease (CWD) is a highly contagious prion disease affecting captive and free-ranging cervid populations. CWD has been detected in United States, Canada, South Korea and, most recently, in Europe (Norway, Finland and Sweden). Animals with CWD release infectious prions in the environment through saliva, urine and feces sustaining disease spreading between cervids but also potentially to other non-cervids ruminants (e.g. sheep, goats and cattle). In the light of these considerations and due to CWD unknown zoonotic potential, it is of utmost importance to follow specific surveillance programs useful to minimize disease spreading and transmission. The European community has already in place specific surveillance measures, but the traditional diagnostic tests performed on nervous or lymphoid tissues lack sensitivity. We have optimized a Real-Time Quaking-Induced Conversion (RT-QuIC) assay for detecting CWD prions with high sensitivity and specificity to try to overcome this problem. In this work, we show that bank vole prion protein (PrP) is an excellent substrate for RT-QuIC reactions, enabling the detection of trace-amounts of CWD prions, regardless of prion strain and cervid species. Beside supporting the traditional diagnostic tests, this technology could be exploited for detecting prions in peripheral tissues from live animals, possibly even at preclinical stages of the disease.


Results
teSee tM WB analysis detects prions in brain or lymph nodes of CWD affected animals. The CWD affected animals used in this study were identified through the Norwegian surveillance program for CWD that has been started in 2016. These animals were firstly diagnosed by TeSeE TM ELISA test and then confirmed by WB analysis (diagnostic statuses are summarized in Table 2). To verify the presence of different distribution patterns of CWD prions in both brain and lymph nodes WB analyses were finally performed. Proteinase K resistant PrP (PrP res ) was found in the samples from all the CWD affected moose (Mo1, Mo2 and Mo3) and red deer (Rd1), while PrP res was detected only in 3 out of 7 brain samples of CWD affected reindeer (Fig. 1a). In contrast, PrP res was not found in the lymph nodes of CWD affected moose and red deer but always detected in lymph nodes of the CWD affected reindeer (Re1-Re7) (Fig. 1b). PrP res was not detected in any of the samples (brain or lymph node) collected from healthy animals.
RT-QuIC analysis with bank vole PrP enables efficient prion detection in brain samples from cervids where PrP Sc was biochemically detected. With the aim of evaluating the efficiency of different substrates in detecting CWD prions in different cervid species, RT-QuIC experiments were performed using recombinant PrP proteins with amino acidic sequences belonging to the following animal species: Syrian hamster, bank vole (109 M), deer (96 G), reindeer (176D) and elk (132 M) (Fig. 2). According to the most recent publications (see Table 1), we decided to use truncated proteins since the C-terminal protein domain allows prion detection with high sensitivity and specificity 96,97 . Evaluation of the overall RT-QuIC performance was based on brain samples collected from CWD affected animals where PrP Sc was detected by means of TeSeE TM ELISA and WB. Brain homogenates of Rd1, Mo1, Mo2, Mo3, Re1, Re2 and Re3 were diluted from 10 −5 to 10 −7 and subjected to RT-QuIC analysis. Brain samples of healthy animals (Rd2, Rd3, Mo4, Mo5, Re8 and Re9) were used as controls.
Regardless of the animal species or brain dilution, bank vole PrP enabled PrP Sc detection with a higher sensitivity and specificity compared to the other tested substrates (Fig. 3a). Particularly, all CWD affected animals' samples induced RT-QuIC seeding activity within 10 hours while those of healthy controls did not. The reaction was stopped at 16 hours because at this point negative controls started to induce unspecific seeding activity.
Compared to the bank vole PrP, Syrian hamster PrP substrate did not detect all CWD affected animals' samples and the sensitivity decreased at higher dilutions. In particular, one (Re2), two (Re1 and Re2) or three (Re1, Re2 and Rd1) CWD affected animals' samples were not detected at 10 −5 , 10 −6 or 10 −7 dilutions, respectively (Fig. 3b). Notably, this substrate hardly detected Norwegian reindeer CWD prions compared to the moose and red deer ones. The reaction was stopped at 53 hours and none of the negative controls induced unspecific reaction. We analyzed whether the prion protein sequence homology between CWD prions and substrate could increase the power of discrimination between prion affected and healthy animals. We then analyzed the samples using PrP substrates with amino acid sequences of deer, reindeer and elk. Surprisingly, while deer PrP (Fig. 3c) was still able to detect CWD affected animals' samples with sensitivity and specificity quite comparable to that of Syrian hamster PrP, reindeer (Fig. 3d) and elk PrP (Fig. 3e) were characterized by a very rapid aggregation in both the positive and the negative samples. Even with a time threshold of 3 hours, we could not clearly discriminate between CWD affected animals and healthy controls. Thus, PrP substrates with cervid sequences appear to be less efficient in detecting CWD prions than those with bank vole and Syrian hamster sequences. Two brain samples from CWD affected white-tailed deer (WTd) from North America were included in the RT-QuIC analysis to verify whether the overall performance of the assay could have been influenced by the origin of the CWD prions (Norway vs North America). The two isolates were provided to the Norwegian Veterinary Institute in Oslo in 2006 as part of a ring trial (courtesy of Aru Balachandran, Canadian Food Inspection Agency, Alberta, Canada). In this case, their seeding activities were similar to that of Norwegian CWD affected deer, in terms of lag phase and fluorescence intensity ( Fig. 4 and see Supplementary Fig. S1).

Species
Sample ID Status  Table 2. Demographic information and TeSeE TM WB results of the Norwegian animals included in this study. www.nature.com/scientificreports www.nature.com/scientificreports/ RT-QuIC analysis with bank vole PrP enables efficient prion detection in reindeer brain samples where PrP Sc was not biochemically detected. Finally, we have analyzed samples collected from CWD affected reindeer (Re4, Re5, Re6, Re7) where PrP Sc had only been detected in the lymph nodes and not the brain with TeSeE TM ELISA and WB. We decided not to perform serial dilutions given the lack of WB signal in the brain samples and chose a dilution of 10 −5 . As previously observed, the bank vole PrP was the most efficient at detecting prions in all the CWD affected reindeer (Fig. 5a). The Syrian hamster PrP did not detect 2 out of 4 CWD samples (Re6 and Re7) (Fig. 5b); whilst deer ( Fig. 5c) and reindeer PrP (Fig. 5d) did not detect 3 out of the 4 CWD samples (Re4, Re6, Re7 and Re4, Re5, Re6, respectively). Elk PrP did not detect any of them (Fig. 5e).

Discussion
First discovered in deer in Colorado many decades ago, CWD rapidly spread to many other American states 98 and Canada 10 . In April 2016, the disease was diagnosed in a Norwegian reindeer from the Nordfjella area 3 . This was the first case of CWD in Europe, and the first reindeer reported with naturally occurring CWD. This disease is contagious within cervid populations and can efficiently transmit directly between animals or through the environment. Healthy animals can be infected after close contact with saliva, urine and feces of affected ones (direct horizontal transmission) or after being exposed to environments contaminated with excreta or carcasses of diseased animals (indirect horizontal transmission). Prions persist in the environment for long time and contribute significantly to disease spreading.
In Europe surveillance programs are aimed at detecting the presence of CWD in wild and farmed cervids. The validated and approved diagnostic tests require animals to be sacrificed for sampling the CNS (e.g. brainstem) and/or lymphoid tissue for ELISA, WB and IHC based analyses 61 . Although these tests reach high levels of diagnostic accuracy for CWD, PrP Sc accumulation in brain and lymphoid tissues can be lower than the detection threshold of the tests, especially in the early stages of disease. In addition, different CWD prion strains can affect test performance, as in the case of Nor98/atypical scrapie 99 .
In this work we evaluated the efficiency of the highly sensitive RT-QuIC assay in detecting low amounts of CWD prion in different Norwegian cervid species. Our aim was to set up optimal conditions for PrP Sc detection, regardless of prion strains and animal species. For this reason, brain homogenates of CWD affected moose, red deer and reindeer were serially diluted and subjected to RT-QuIC analysis performed using Syrian hamster, bank vole, deer, reindeer and elk PrP as reaction's substrates.
Our results indicated that the bank vole PrP enabled CWD prion detection in every brain dilution of all cervid species, especially in reindeer where PrP Sc detection was more challenging compared to the other species and we could clearly discriminate CWD affected animals from healthy controls. A slightly less efficient detection of CWD prions was observed using Syrian hamster PrP. By using bank vole PrP we could detect prions in brain samples of reindeer that had tested negative with traditional diagnostic tests. Moreover, we efficiently detected PrP Sc in samples from North American cervids, which have CWD prion strains that might be different from those found in Norway 9 . Thus, the use of bank vole PrP overcomes strain-related effects which are known to influence the efficiency of the RT-QuIC. The capability of bank vole PrP to detect a wide range of prion strains has already been reported 100 and here we demonstrate for the first time that this substrate enables high efficient detection of multiple CWD strains in different Norwegian CWD affected cervids.
Efficient transmission of TSE infection requires a close similarity between the primary amino acid sequence of the PrP in the donor and in the recipient animal. This allows PrP Sc to interact specifically with and convert the host's PrP C into the disease-associate isoform and could explain why CWD is so easily transmissible between different cervid species 26,101,102 . Our results showed that the efficiency of the RT-QuIC test was reduced by the use of deer PrP, especially in reindeer, and the sensitivity dropped drastically when using reindeer and elk PrP. Nevertheless, similar observations have been made in the field of human prion diseases: the use of human PrP substrate for RT-QuIC analyses results in lower sensitivity and specificity compared to Syrian hamster or bank vole PrP for detecting PrP Sc in peripheral tissues (e.g. cerebrospinal fluid and olfactory mucosa) of prion diseased patients [103][104][105][106][107] .  www.nature.com/scientificreports www.nature.com/scientificreports/ environment. It is estimated that more than 60% of Americans have eaten deer or elk meat or their derived products 108 while a large number of cattle, sheep and goats have grazed in CWD contaminated environments. Thankfully, controlled natural exposure studies and targeted surveillance programs currently indicate no cases of natural interspecies transmission of CWD 10,109-113 .
Experimentally, CWD can efficiently be transmitted by intracerebral inoculation in mice, mink, squirrel monkeys, ferrets, sheep and some cattle 10,109,[114][115][116][117][118] . But attempts to transmit to transgenic mice overexpressing human prion protein 119 or to Cynomologus macaques 120 , which are evolutionarily closer to humans, were unsuccessful. This suggests little or no zoonotic potential. The efficiency of CWD transmission to humans has been also evaluated in vitro with highly sensitive PMCA and RT-QuIC techniques. Some studies showed that the human PrP can be converted to the pathologic form by CWD prions. It was suggested that the efficiency of this conversion is highly influenced by (i) human PrP polymorphism (recipient), (ii) cervid PrP polymorphism (donor) and (iii) isolates origin (strain) [121][122][123][124][125] . Overall, experiments performed using in vitro amplification techniques suggest that the species barrier between cervids and human is not absolute. However, although these techniques mimic in vitro the process of prion conversion, they lack many of the biological interactions occurring in vivo and the results, regarding the study of the complex phenomenon of the species barrier, should be carefully interpreted. In addition, the species barrier does not only depend on the PrP sequence homology between host and recipient, but also the prion strain. It is therefore conceivable that different CWD strains may have different abilities at crossing the species barrier. Many other factors may play a pivotal role in driving this phenomenon. CWD prions can undergo to processes of selection and adaptation once the interspecies transmission has occurred, with the generation of new prion conformers more prone to propagate in the new host and likely easier to transmit within the species 126,127 . For instance, prions from cattle affected by bovine spongiform encephalopathy (BSE) crossed the species barrier (although with low efficiency) and infected humans, generating a new disease named variant Creutzfeldt-Jakob disease (vCJD) 128 . There is therefore considerable concern that CWD prions could cross the species barrier, adapt to humans and result in new forms of prion disease. The ongoing surveillance has not reported any documented cases of CWD transmission to humans at present. However, the lack of interspecies transmission cannot definitively be ruled out 112,129 .
Our optimized RT-QuIC performed with bank vole PrP could be used as first step screening assay followed by traditional confirmatory TeSeE ELISA, WB or IHC assays to increase the accuracy of CWD detection in affected animals. After a process of validation where many more samples of CWD affected animals and negative controls will be analyzed with this technique, it could be employed as new tool for the diagnosis of CWD either at clinical or preclinical stage of the disease. In addition, considering its elevated analytical sensitivity and rapidity, RT-QuIC might also be exploited for a quick and efficient PrP Sc detection in tissues and biological fluids, such as urine, saliva or feces. These samples are easier to collect than CNS and lymphoid tissues and do not require immobilization or euthanasia of animals. This test could also be used to confirm the absence of infection in animals prior to restocking. Finally, other than monitoring the spreading of CWD prions between cervid species, RT-QuIC with bank vole PrP can be further extended to evaluate the presence of prions in tissues collected from other animals (e.g. sheep, goats, cattle) eventually exposed to contaminated environment.
In conclusion, we provide evidence that RT-QuIC performed with bank vole PrP as reaction substrate is capable of detecting CWD prions, regardless of the cervid species, strains and geographical origin, with good analytical sensitivity and specificity. This rapid and useful technique is, in combination with traditional diagnostic tests, ideal for screening samples containing low concentrations of CWD prions.

Materials and Methods
Compliance with Ethical Standards. All animal samples included in this study were provided by the Norwegian surveillance program for CWD in compliance with ethical standards.
Animals. The following animals from Norway were included in the study: (i) 5 Moose (Alces alces) (3 affected by CWD and 2 healthy animals), (ii) 3 red deer (Cervus elaphus) (1 affected by CWD and 2 healthy animals) and (iii) 9 reindeer (Rangifer tarandus tarandus) (7 affected by CWD and 2 healthy animals). The information concerning the animal's geographical origin, sex, age and diagnostic status is summarized in Table 2. teSee tM ELISA and WB tests for CWD diagnosis. All the CWD affected animals were first detected by the Norwegian surveillance program for CWD, using commercially available tests for the detection of PrP Sc . Brain tissues and a piece of lymph node were homogenized at 20% (weight/volume) in individual grinding tubes. Rapid test TeSeE TM SAP ELISA (Bio-Rad) was carried out according to the manufacturer's instruction. Positive ELISA samples were then analyzed with TeSeE TM Western blot (Bio-Rad) for confirmation. teSee tM Western blot analysis of brain and lymph nodes samples. The homogenates submitted to Western blot were collected from the grinding tubes primarily analyzed by rapid test TeSeE ™ ELISA. The WB test was performed, with slight modifications, according to the manufacturer's instructions. Briefly, PrP C was www.nature.com/scientificreports www.nature.com/scientificreports/ digested by incubating the homogenates with Proteinase K (20 µl per ml) for 10 min at 37 °C. Electrophoresis was performed using mini-PROTEAN ® TGX ™ Precast Gels (Bio-Rad) and Power Pac Universal (first 10 min at 60 V followed by approximately 35 min at 120 V). Gels were then electroblotted using semi-dry transfer apparatus (Trans-Blot ® Turbo ™ Transfer System, Bio-Rad) onto polyvinylidene fluoride (PVDF) membrane (Bio-Rad). The immunoblotting process began by blocking the membrane, to prevent unspecific bindings, with the kit's block solution for 30 min, then a second 30 min incubation was carried out using monoclonal antibodies SHa31 (AbI from the kit) and an additional monoclonal antibody (P4) at a dilution of 1:1000. Lastly, a 20 min incubation with goat anti-mouse immunoglobulin G (IgG) antibody conjugated with horseradish peroxidase (AbII from the kit) was carried out. The test's chemiluminescent substrate ECL (Western blotting detecting reagents, Amersham ECL TM ) was then added and the chemiluminescent signals were visualized using ChemiDoc System (Bio-Rad). The samples were declared positive if characteristic banding patterns of PK-resistant core of PrP Sc were present. Preparation of the samples for RT-QuIC analyses. Brain tissues were homogenized at 10% (weight/ volume) in Bio Rad buffer (from TeSeE TM grinding tubes), serially diluted (from 10 −5 to 10 −7 ) and subjected to RT-QuIC analysis. Two brain tissues of CWD affected white tailed deer (WTd) collected from North America (used for a ring trial for CWD diagnosis in 2006) were homogenized, diluted at 10 −5 and 10 −6 and included in the analysis.

RT-QuIC experimental procedures.
Protein substrate solutions were allowed to thaw at room temperature and filtered through a 100 kDa Nanosep centrifugal device (Pall Corporation). Samples were analyzed in triplicate in a black 96-well optical flat bottom plate (ThermoScientific). The final reaction volume was 100 µL and the reagents (Sigma) were concentrated as follow: 150 mM NaCl, 0.002% SDS, 10 mM PBS, 1 mM EDTA, 10 µM ThT and 0.13 mg/ml of recPrP. To avoid contamination, reaction mixes were prepared and loaded (98 µL) onto the microplate in a prion-free laboratory. After the addition of 2 µL of diluted brain homogenates (from 10 −5 to 10 −7 ), the plate was sealed with a sealing film (ThermoScientific) and inserted into a FLUOstar OPTIMA microplate reader (BMG Labtech). The plate was incubated at 55 °C with cycles of 1 min shaking (at 600 rpm, double orbital) and 1 min incubation. Fluorescence readings (480 nm) were taken every 15 min (30 flashes per well at 450 nm). A sample was considered positive if the two highest fluorescence values (AU) of the replicates were greater than 10.000 AU and at least two, out of three replicates, crossed the time threshold that was set for each recombinant substrate. We set the following time thresholds for each PrP: (i) Syrian hamster 48 hours, (ii) bank vole 10 hours, (iii) deer 3.5 hours, (iv) reindeer 3 hours and (v) elk 3 hours. In particular, we have evaluated the time at which the unspecific aggregation of each PrP template occurred in the presence of negative samples (analyzed at least three different times). We have then set this value as time-threshold. Therefore, all samples able to promote PrP aggregation before this time-threshold were considered able to exert a seeding activity while the others were considered unable to promote a seeding activity for each PrP substrate. Data are plotted in graphs showing the time taken for each replicate (black dots) to reach the fluorescence threshold (lag phase).
Statistical analyses and graphic representation. Statistical analysis (mean and standard error of the mean (S.E.M.)) and graphic representations of our data were performed using the Prism software (v5.0 GraphPad).

Data availability
All data generated or analyzed during this study are included in this published article.