PMCA-replicated PrPD in urine of vCJD patients maintains infectivity and strain characteristics of brain PrPD: Transmission study

The presence of abnormal, disease-related prion protein (PrPD) has recently been demonstrated by protein misfolding cyclic amplification (PMCA) in urine of patients affected with variant Creutzfeldt-Jakob disease (vCJD), a prion disease typically acquired from consumption of prion contaminated bovine meat. The complexity and multistage process of urine excretion along with the obligatory use of PMCA raise the issue of whether strain characteristics of the PrPD present in vCJD brains, such as infectivity and phenotype determination, are maintained in urine excreted PrPD and following amplification by PMCA. We inoculated transgenic mice expressing normal human PrP with amplified urine and brain homogenate achieving the same 100% attack rate, similar incubation periods (in both cases extremely long) and histopathological features as for type and severity of the lesions. Furthermore, PrPD characteristics analyzed by immunoblot and conformational stability immunoassay were indistinguishable. Inoculation of raw vCJD urine caused no disease, confirming the extremely low concentration of PrPD in vCJD urine. These findings show that strain characteristics of vCJD brain PrPD, including infectivity, are preserved in PrPD present in urine and are faithfully amplified by means of PMCA; moreover, they suggest that the PrPD urine test might allow for the diagnosis and identification of disease subtype also in sporadic CJD.

The mechanisms leading to the presence of PrP D in urine of vCJD patients remain to be determined 11,12 . The complex, multistage process of urine formation, raises the possibility that, while spreading to urine, PrP D undergo subtle conformational changes altering strain characteristics. Moreover, this issue is further compounded by the extreme under-representation of PrP D in urine that requires extensive amplification or enrichment procedures 11,12 . Transmission to appropriate hosts has been shown to be a suitable approach to define and compare prion strain properties 13 . To this aim, we inoculated transgenic (Tg) mice expressing human PrP 129 M (Tg40) with PrP D obtained from vCJD urine following PMCA and untreated vCJD brain homogenate (BH). Raw vCJD urine was injected as control. Mice injected with PMCA-treated products and those injected with untreated vCJD BH developed prion diseases that were phenotypically indistinguishable and were characterized by similar attack rates and extended incubation periods. Moreover, resPrP D displayed similar conformational features. By contrast, none of the animals inoculated with untreated vCJD urine developed a prion disease.

Results transmission characteristics.
Inoculation of Tg40 mice with PMCA-treated vCJD urine resulted in prion disease with an attack rate of 100% and incubation periods, as measured by days post inoculation (dpi), of 661 ± 45 and 713 ± 36, depending on the final dilution (1x or 1:10, respectively) of the inoculated PMCA product (Table 1). These results were comparable to those obtained with the transmission of vCJD BH where attack rate again was always 100% and incubation periods varied between 574 ± 104 and 648 ± 39 dpi, according to the 10% or 1% BH concentration of the inoculum ( Table 1). The only statistically significant difference was observed when comparing dpi following inoculation of the more diluted PMCA-treated urine (1:10) to the dpi of the more concentrated vCJD BH (10%) inoculum ( Table 1). Inoculation of PMCA-treated unseeded-substrate, as negative control, does not have any effect on mice life span. None of the 96 Tg40 mice inoculated with raw vCJD urine tested positive for prion disease up to 707 dpi.
Histopathology and immunohistochemistry. The histopathological features of the Tg40 mice challenged with PMCA-treated urine essentially reproduced type and distribution of the lesions associated with the inoculation of vCJD BH. In both conditions, spongiform degeneration (SD) consisted mostly of large vacuoles that predominantly affected subcortical formations including thalamus, hypothalamus, basal ganglia and brain stem (Figs 1i, iii, row l, and 2A). By contrast, florid plaques (FP) populated mostly the cerebral cortex and, to a lesser extent, subcortical regions and cerebellum (Fig. 1i, iii, rows 2 and 4). Non-FP plaques of various sizes were also seen often lined up along the remnants of the lateral ventricles above the hippocampus (Fig. 1i, iii, row 3). Diffuse PrP immunostaining codistributed with the SD, with intensities that appeared to match the SD severity (Fig. 1ii, iv, row 1). FP and non-FP were strongly immunostained while stellate apparently pericellular PrP deposits were occasionally seen in the cerebral cortex (Fig. 1ii, iv, rows 2 and 3). PrP immunostaining affected also the cerebellum, including the molecular layer and, occasionally, the granular layer and the deep white matter (Fig. 1ii, iv, row 4).
Lesion profiles showed an overall similar distribution of lesions in Tg mice inoculated with PMCA-treated urine or vCJD BH. However, lesions were slightly more severe in mice challenged with 1x PMCA-treated urine than in those receiving vCJD BH (1% or 10%) or 1:10 diluted PMCA-treated urine ( Fig. 2A). Neither prion-related lesions nor PrP D deposits were observed in Tg40 mice inoculated with PMCA-unseeded substrate and raw vCJD urine up to 673 dpi and 707 dpi, respectively ( Fig. 2A and data not shown). Morphometric analysis of the density and size of FP in the cerebral cortex revealed no significant differences between the two groups of inoculated mice (Fig. 2B,C). prp D characterization. Regardless of the inoculum, Tg40 animals injected with PMCA-amplified vCJD urine or with untreated vCJD brain homogenate showed the presence of a PK-resistant PrP D (resPrP D ) displaying the same electrophoretic profile, which was characterized by the prevalence of the diglycosylated band and the mobility to 19 kDa of the unglycosylated isoform (type 2B) (Fig. 3). resPrP D representation in undiluted  Table 1. Transmission to Tg40 mice of PMCA-treated vCJD urine, PMCA negative control, untreated vCJD urine and vCJD brain homogenate (BH). a Six rounds of PMCA have been performed on the resuspended pellet derived from 1 ml urine. b Inoculum 1x and 1:10 correspond to undiluted and to diluted 1:10, respectively. c dpi values are express as mean ± standard deviation. d T2B means type 2B and identifies the typical electrophoretic profile of vCJD resPrP D . *Statistically significant difference (p < 0.002) between adjacent T2B values.
The stability of resPrP D species extracted from Tg40 mice inoculated with either PMCA-treated vCJD urine or vCJD BH was tested with the conformational stability immunoassay (CSI). No significant difference was detected between the two resPrP D species in their rate of denaturation following treatment with increasing amounts of the denaturant guanidine hydrochloride (GdnHCl). When the GdnHCl treatment was expressed as GdnHCl molar quantity reducing the amount of resPrP D to half ([GdnHCl] 1/2 ), the values obtained were 1.31 ± 0.1 M and 1.4 ± 0.1 M (Fig. 4).

Discussion
We have recently shown that urine of patients affected by vCJD harbors minute amounts of PrP D estimated at 1 × 10 −16 g/mL, which is approximately 12 orders of magnitude smaller than the PrP D amount in the vCJD brain tissue (1 × 10 −4 g/g) 11 . Despite the small amount, urine PrP D can be used to diagnose the presence of prion disease in vCJD with 93% sensitivity and 100% specificity 11 . This low PrP D content is consistent with our finding that none of the 96 Tg40 mice inoculated with raw urine formed detectable resPrP D especially considering that at least in sCJD the Tg40 limit detectability of brain PrP D is approximately 1 × 10 −5 equivalent concentration 14 . www.nature.com/scientificreports www.nature.com/scientificreports/ The complex physiology of urine formation raises the question of how and at which stage PrP D spreads to urine. Although the precise mechanism remains unidentified, three propagation processes have recently been considered 12 . Given its significant presence in vCJD blood, PrP D may invade urine during the stage of glomerular filtration, tubular secretion or both. Indeed, increase PrP D representation in urine has been observed experimentally in the presence of kidney inflammatory processes associated with increase of urine protein excretion 15 . Urine contains relatively high amounts of PrP that is truncated at the N-terminus and harbors an incomplete glycosylphosphatidylinositol anchor suggesting that it is shed from cell surfaces 16 . However, this truncated PrP has been identified as C-terminal fragment C1 16 , a product of normal PrP C metabolism that is an unlikely substrate for conversion to PrP D since the N-terminal cleavage breaks the 100-130 residue domain considered important  www.nature.com/scientificreports www.nature.com/scientificreports/ for the conversion of PrP C to PrP D17-21 . It is also noteworthy in this context that contrary to many visceral organs, no PrP D has been demonstrated in the bladder 9 . Finally, urine PrP D might result from traditional contiguous PrP D propagation to urine from kidney, blood or both especially at the level of the nephrons 9,10,12 .
In the present study, we show that PMCA-treated urine PrP D is competent to transmit to Tg mice expressing human PrP a prion disease that, as to the histopathological phenotype and resPrP D properties, is indistinguishable from the disease transmitted by inoculation with vCJD BH. These findings suggest that urine resPrP D preserves the strain characteristics of the brain resPrP D . Therefore, the mechanism of urine invasion by PrP, whatever that might be, as well as the PMCA procedure, must preserve the conformational characteristics of brain-generated resPrP D in vCJD. The present study also confirms the competence of PMCA to accurately replicate PrP D strain features enciphered in the original seed. This PMCA property, originally reported for the 263K hamster scrapie agent [22][23][24] , is here demonstrated for the first time in a human prion strain. An additional corollary to the present finding is that the non-invasive PMCA-based urine diagnostic test may be used reliably to determine not only the presence of prion disease but also to establish the diagnosis of vCJD.
Presence of PrP D in urine has been reported in 40% of patients affected with various subtypes of sCJD, by far the most prevalent type of human prion diseases 1 , using a solid-state binding matrix to capture and concentrate urine PrP D12 . By contrast, no PrP D signal was detected in sCJD urine following the same experimental conditions as those used for PMCA analysis of vCJD urine samples 11 . It has been reported that PrP D associated with sCJDMM1, the most common subtype of sCJD, is a very inefficient seed for PMCA when using the same conditions described in this article (Camacho et al., Prion 2017; May 23-26, 2017; Edinburgh, UK). Finding a solution to this technical hurdle may afford a urine-based diagnostic test that, along with codon 129 PrP genotype determination, may provide the diagnosis of sCJD subtype.

Materials and Methods
Biological safety. Prion material was handled in a biosafety level 2 (BSL-2) laboratory and all mice were housed in a BSL-2 room of Case Western Reserve University animal facility.
Urine samples. Urine samples were obtained from the National Prion Disease Pathology Surveillance Center (NPDPSC). Samples were collected, during the symptomatic period, from a vCJD patient, who died at 25 years of age, 32 months after clinical onset 9,25 . Urine samples were stored at −80 °C until use.
Protein misfolding cyclic amplification (PMCA). PMCA of vCJD urine and unseeded substrate was carried out as previously reported 11 . Briefly, 1 ml of urine was centrifuged at low speed (5000 × g for 20 minutes at 4 °C) to separate debris. Two fractions, supernatant (S1) and pellet (P1), were obtained. S1 was then centrifuged at high speed (100,000 × g for 1 hour at 4 °C) and the pellet (P2a) directly resuspended in 10% homogenate (BH) prepared from brains of transgenic (Tg) mice expressing human PrP-129M (Tg line with PMCA efficiency higher than that of Tg40), which also served as substrate for PMCA. P1, containing cellular debris, was resuspended in 1 ml of water and subjected to high-speed centrifugation (100,000 × g for 1 hour at 4 °C). The pellet (P2b) was resuspended in Tg mice-BH, and identified as P2a. The two resuspended pellets (P2a and P2b) were pooled and processed through PMCA cycles. After 96 cycles, the Tg mouse PrP C substrate was refreshed, and sample subjected to a new round of PMCA cycles. The final sample was obtained after 6 rounds of PMCA. The same Tg mouse BH used as a substrate, left unseeded and similarly treated with 6 rounds of PMCA served as negative control.

Histopathology, immunohistochemistry, lesion profiles and morphometric analysis. Mouse
brains were harvested and bisected along the inter-hemispheric fissure. One half was stored at −80 °C until needed for biochemical studies; the other was fixed in formalin and used for histopathological and immunohistochemical examinations. Four coronal slices, obtained from the formalin-fixed half brain, at approximately bregma 0.5 mm, −1.7 mm, −3.8 mm and −6.0 mm, were processed for microscopic examination following staining with hematoxylin and eosin (H.E.) or immunostaining with monoclonal antibody (mAb) 3F4 as previously described 27,28 .
Lesion profiles were generated by semi-quantitative evaluation of the severity of spongiform degeneration (SD) on a 0-4 scale, in eight brain regions examined on H&E stained sections (also see Fig. 1). Morphometric analysis to assess density of florid plaques was performed as previously described 29 . Briefly, morphometric analysis was carried out on dorsal, intermediate and ventral cerebral cortices at the three anterior bregma levels. Density was expressed as percentage of the cortical area occupied by florid plaques, whereas size was expressed in µm 2 , and calculated by dividing the area occupied by the florid plaques for their number. To obtain a comparable sample size, the random function (Excel software) was used to randomly select areas of cerebral cortex.
Conformational stability immunoassay (CsI). CSI was performed as previously described [30][31][32] with minor modifications. Aliquots of 10% BH in LB100 pH 8.0 (100 mM Tris HCl pH 8.0, 100 mM NaCl, 0.5% Nonidet P-40, 0.5% sodium deoxycholate, 10 mM EDTA) were centrifuged for 10 min at 1,000 × g at 4 °C and pellets discarded. 100 µl of the supernatants were diluted with an equal volume of GdnHCl to obtain a final concentration ranging from 0 to 4 M and incubated for 1.5 h at room temperature. GdnHCl was subsequently removed by precipitation with 5 volumes of methanol with overnight incubation at −20 °C and centrifuged for 30 min at 18,000 × g. Pellets were resuspended in 100 µl LB 100 pH 8.0 by sonication. Each aliquot was digested with 5U/ ml PK for 1 h at 37 °C. The reaction was stopped by addition of 2 mM PMSF. After denaturation, samples were analyzed by WB (Ab: 3F4) and developed by Odyssey Classic infrared imaging system (LI-COR Biosciences). PrP densitometry was performed by Odyssey V3.0 software (LI-COR Biosciences). After normalization, the data were plotted and expressed as mean ± standard deviation. Conformational stability, evaluated as a function of resPrP D conversion to PK-sensitive PrP following exposure to increasing concentrations of GdnHCl, was best fitted to a sigmoidal dose-response equation by GraphPad Prism 7 (GraphPad Software Inc.). The formula [GdnHCl] 1/2 ± SD was calculated for the two groups and compared by statistical analysis. statistical analysis. Statistical analyses were performed with (i) Student's T-test, for lesion profile, florid plaque density-size analyses, and conformational stability immunoassay, and (ii) analysis of variance (One-Way ANOVA) followed by Tukey's multiple comparison test, for incubation periods (days post inoculation, dpi). ethics statement. The human samples used in the current study were provided by NPDPSC. Written informed consent for research was obtained from the patient according to the Declaration of Helsinki. No minor participants were included in the study. All procedures were performed under protocols approved by the Institutional Review Board for Human Investigation of University Hospitals, Case Medical Center of Cleveland, OH, and as per regulations of the Declaration of Helsinki. All animal studies were carried out in accordance with the recommendations in the Guide for the Care and Use of Laboratory Animals of the National Institutes of Health. The protocols were approved by the Institutional Animal Care and Use Committee of the Case Western Reserve University (Protocol number 2012-0199).