Rapid end-point quantitation of prion seeding activity with sensitivity comparable to bioassays

Abstract

A major problem for the effective diagnosis and management of prion diseases is the lack of rapid high-throughput assays to measure low levels of prions. Such measurements have typically required prolonged bioassays in animals. Highly sensitive, but generally non-quantitative, prion detection methods have been developed based on prions' ability to seed the conversion of normally soluble protease-sensitive forms of prion protein to protease-resistant and/or amyloid fibrillar forms. Here we describe an approach for estimating the relative amount of prions using a new prion seeding assay called real-time quaking induced conversion assay (RT-QuIC). The underlying reaction blends aspects of the previously described quaking-induced conversion (QuIC) and amyloid seeding assay (ASA) methods and involves prion-seeded conversion of the alpha helix-rich form of bacterially expressed recombinant PrP(C) to a beta sheet-rich amyloid fibrillar form. The RT-QuIC is as sensitive as the animal bioassay, but can be accomplished in 2 days or less. Analogous to end-point dilution animal bioassays, this approach involves testing of serial dilutions of samples and statistically estimating the seeding dose (SD) giving positive responses in 50% of replicate reactions (SD(50)). Brain tissue from 263K scrapie-affected hamsters gave SD(50) values of 10(11)-10(12)/g, making the RT-QuIC similar in sensitivity to end-point dilution bioassays. Analysis of bioassay-positive nasal lavages from hamsters affected with transmissible mink encephalopathy gave SD(50) values of 10(3.5)-10(5.7)/ml, showing that nasal cavities release substantial prion infectivity that can be rapidly detected. Cerebral spinal fluid from 263K scrapie-affected hamsters contained prion SD(50) values of 10(2.0)-10(2.9)/ml. RT-QuIC assay also discriminated deer chronic wasting disease and sheep scrapie brain samples from normal control samples. In principle, end-point dilution quantitation can be applied to many types of prion and amyloid seeding assays. End point dilution RT-QuIC provides a sensitive, rapid, quantitative, and high throughput assay of prion seeding activity.

Figure 1. RT-QuIC sensitivity: analysis of dilutions of a scrapie brain homogenate stock.

The designated dilutions of 263K scrapie or normal BH were used to seed RT-QuIC reactions containing recombinant hamster PrP^c 90-231 substrate. Each 0.75 h time point for each dilution is represented as an average ThT fluorescence from 8 replicate wells on a 96-well plate. The 50% seeding dose (SD₅₀) is defined as the amount giving sufficiently enhanced ThT fluorescence in half of the replicate wells. In this case, the approximate SD₅₀ was achieved with a 2 µl aliquot (the seed volume) of a 10⁻⁹ dilution of the scrapie BH stock. The Spearman-Kärber estimate of the SD₅₀/2 µl neat brain tissue is shown, along with that adjusted to SD₅₀/gram brain tissue.

Figure 2. SDS-PAGE, circular dichroism (CD), and Fourier transform infrared spectroscopy analyses of RT-QuIC products.

RT-QuIC reactions were seeded with 5×10⁻⁶ dilution of either NBH or 263K BH (containing 100 fg PrP^Sc). (A) Products were PK digested and analyzed by SDS-PAGE. The gel was stained with a non-specific protein stain (Deep Purple). The circle indicates the ∼18 and 19 kDa bands while the bracket represents the ∼13 and 14 kDa bands in the PK-digested products of the scrapie-seeded reaction. (B) CD spectrum of the initial hamster rPrP^C 90-231 substrate for RT-QuIC reactions. (C) FTIR spectra of hamster rPrP^C 90-231, 263K-seeded rPrP 90-231 RT-QuIC product, and PK-treated brain-derived 263K PrP^Sc.

Figure 3. Combined RT-QuIC end-point dilution analyses from four scrapie BH stocks.

Each 263K stock (Brain #1-#4 corresponding to Panels A-D) was serially diluted and used to seed RT-QuIC reactions. The percentage of replicate wells (n = 4–8) that are ThT-positive according to the criteria in Material and Methods was calculated and the mean percentage ± standard deviation from three separate experiments with each stock is shown. Animal bioassay data is represented as the percentage of replicate animals reaching the clinical stage of disease at each dilution of 263K scrapie BH (4–22 animals per dilution per stock).

Figure 4. RT-QuIC end-point dilution analysis of three 263K-inoculated preclinical 10 dpi hamster BHs.

Hamsters were i.c.-inoculated with 263K scrapie BH. Animals were sacrificed at 10 dpi and brains were analyzed with RT-QuIC dilution analysis. Eight replicate wells were used for each BH dilution. The average Spearman-Kärber estimates of the SD₅₀/2 µl and SD₅₀/g neat brain tissue from three animals are shown.

Figure 5. RT-QuIC detection and end point dilution analysis of nasal lavages from HY-TME-infected hamsters.

At the clinical stage of disease in hamsters i.c.-inoculated with HY-TME, nasal lavage samples (∼1 ml) were collected from individual hamsters, serially diluted and used to seed RT-QuIC reactions. (A) Detection of HY-TME in nasal lavages is shown. As positive and negative controls, 5×10^-6-fold dilutions of hamster scrapie BH (containing ∼100 fg PrP^Sc) and NBH, respectively, were used as seeds. Dilutions of HY-TME brain were also tested for comparison. The data points show the average ThT fluorescence of 8 replicate wells. (B) Three representative HY-TME infected hamster lavages were serially diluted and analyzed by RT-QuIC. Each data point represents the percentage of replicate wells (n = 4) that are ThT-positive according to the criteria in Material and Methods as a function of nasal lavage dilution factor. The tabular data from lavages tested here can be found in Table 2.

Figure 6. RT-QuIC end-point dilution analysis of CSF from scrapie-infected hamsters.

CSF was collected at the clinical stage of disease from two individual hamsters (Panel A and B) i.c.-inoculated with 263K scrapie BH. CSF was serially diluted 5-fold and analyzed by RT-QuIC. Each 0.75 h time point for each dilution is represented as an average ThT fluorescence from 8 replicate wells. The average Spearman-Kärber estimates of the SD₅₀/2 µl (the seed volume) and SD₅₀/ml neat CSF are shown.

Figure 7. Tissue tolerance of RT-QuIC assay.

RT-QuIC reactions with recombinant hamster 90-231 PrP^c substrate were seeded with 2 µL neat or diluted samples of cerebral spinal fluid (A), 10% muscle (B), or 10% spleen (C), each spiked with a 10^−6.3 dilution of normal or 263K scrapie-infected (clinical) hamster BH. Each 0.75 h time point for each dilution is represented as an average ThT fluorescence from 4 replicate wells.

Figure 8. Sodium chloride titration for sheep, deer, and hamster RT-QuIC reactions.

Dilutions of sheep scrapie (10^−4.9; ∼10 fg PrP^Sc) (A), deer CWD (10^−6.3; ∼4 fg PrP^CWD) (B), and hamster scrapie (10^−7.3; ∼10 fg PrP^Sc) (C) BH's and corresponding dilutions of NBH of the same species were used to seed RT-QuIC reactions. The NaCl concentration in the reactions was varied as designated. All reactions utilized the homologous full length rPrP^C substrates. The data points show the average ThT fluorescence of 4–8 replicate wells.