Crossing species barrier by PrPSc replication in vitro generates new infectious prions

doi:10.1016/j.cell.2008.07.030

Cell. Author manuscript; available in PMC 2009 September 9.

Published in final edited form as:

Cell. 2008 September 5; 134(5): 757–768.

doi: 10.1016/j.cell.2008.07.030

PMCID: PMC2740631

NIHMSID: NIHMS70686

Crossing species barrier by PrP^Sc replication in vitro generates new infectious prions

Joaquín Castilla,¹ Dennisse Gonzalez-Romero,¹^# Paula Saá,^1,²^# Rodrigo Morales,^1,³ Jorge De Castro,¹ and Claudio Soto¹^*

¹Dept of Neurology, University of Texas Medical Branch, Galveston, TX 77555

²Centro de Biología Molecular, Universidad Autónoma de Madrid, Madrid, Spain.

³University of Chile, Santiago, Chile.

^*Correspondence to Dr Claudio Soto, Email: clsoto/at/utmb.edu, Tel 409-7470017, Fax 409-7470020

^#These authors contributed equally to this study.

The publisher's final edited version of this article is available at Cell

See other articles in PMC that cite the published article.

Publisher's Disclaimer

Summary

Prions are unconventional infectious agents composed exclusively by the misfolded prion protein (PrP^Sc), which transmits the disease by propagating its abnormal conformation to the cellular prion protein (PrP^C). A key characteristic of prions is their species barrier, by which prions from one species can only infect a limited number of other species. Here we report the generation of novel infectious prions by inter-species transmission of PrP^Sc misfolding in vitro. Hamster PrP^C misfolded by mixing with mouse PrP^Sc generated new prions that were infectious to wild type hamsters. Similarly, new mouse prions were generated by crossing the species barrier in the opposite direction. A detailed characterization of the infectious, biochemical and histological properties of the disease produced indicate that the in vitro generated material across the species barrier correspond to new prion strains. Successive rounds of PMCA amplification result in a progressive adaptation of the in vitro produced prions, in a process reminiscent to the strain stabilization process observed upon serial passage in vivo. Our results indicate that PMCA is a valuable tool to investigate cross-species transmission and suggest that species barrier and strain generation are determined by the propagation of PrP misfolding.

Introduction

Prion diseases also known as Transmissible Spongiform Encephalopathies (TSEs) are infectious neurodegenerative diseases affecting the brain of humans and several species of mammals (Collinge, 2001). Creutzfeldt-Jakob disease (CJD) is the most common TSE in humans, and scrapie in sheep, Bovine spongiform encephalopathy (BSE) in cattle and chronic wasting disease (CWD) in cervids are the most prevalent prion diseases in animals. Unlike conventional infectious micro-organisms, the TSE agent appears to be devoid of genetic material and instead composed exclusively by a misfolded form of the prion protein (PrP^Sc) (Prusiner, 1982; Prusiner, 1998). PrP^Sc has the unprecedented ability to replicate in the body by inducing the misfolding of the cellular form of the prion protein (PrP^C).

One of the characteristics of the agent responsible for prion diseases is its ability to infect some species and not others (Chen and Gambetti, 2002; Hill and Collinge, 2004; Moore et al., 2005). This phenomenon is known as species barrier. Even between close species, the species barrier is manifested as an incomplete attack rate and a prolongation of the time it takes for animals to develop the clinical disease when injected with another species infectious material (Hill and Collinge, 2004). Primary inter-species transmission is usually not very efficient and takes a long time for the prion replication process to reach the point in which full-blown clinical disease appears. After sequential passages, the PrP^Sc in the new host adapt, resulting in a shortage of the incubation period and stabilization of the new strain (Hill and Collinge, 2004).

Compelling evidence indicate that the species barrier is largely controlled by the sequence of PrP (Moore et al., 2005). Unfortunately, we cannot predict the degree of a species barrier simply by comparing the prion proteins from two species. The barrier has to be measured by experimental studies in animals. These studies are long and costly and in the case of the human species barrier, the studies have to be done with experimental models, which validity is not absolutely guarantied. Evaluation of the species barrier is of tremendous medical importance for risk assessment and to implement regulatory measures to avoid spreading of diseases (Moore et al., 2005). At this time, the epidemiological evidence suggest that among animal TSEs only cattle BSE has been transmitted to humans, generating a variant form of CJD (vCJD) (Bruce, 2000; Collinge, 1999; Ironside and Head, 2004). It is unlikely that sheep scrapie is a concern for humans, because the disease has been described for centuries and no increased prevalence of human prion diseases has been found in scrapie endemic areas (Caramelli et al., 2006; Hunter, 1998). However, the appearance on new “atypical” strains of scrapie as well as the known transmission of BSE to sheep has generated new concerns of human infections with sheep-derived material (Buschmann and Groschup, 2005; Hunter, 2003). Similarly, the risk that transmission to humans of some of the newer animal prion diseases, such as CWD cannot be ruled out at the present time (Williams, 2005; Xie et al., 2005).

Recently, we reported the generation of infectious prions in vitro by amplification of PrP^Sc misfolding in the test tube (Castilla et al., 2005). For these experiments we used a technology termed PMCA (protein misfolding cyclic amplification) that mimic in vitro some of the fundamental steps involved in PrP^Sc replication in vivo at an accelerated kinetic (Saborio et al., 2001; Soto et al., 2002). During PMCA, small quantities of PrP^Sc are mixed with excess of PrP^C and through a cyclical process involving incubation and sonication, prion propagation occurs in an autocatalytic way. Using this procedure, prions can be maintained replicating indefinitively in the test tube and after successive rounds of dilutions followed by PMCA amplification, PrP^Sc used to begin the reaction can be eliminated and only in vitro generated misfolded protein remains in the sample (Castilla et al., 2005). Inoculation of PMCA-generated prions into wild type animals resulted in a disease with the same clinical, neuropathological and biochemical features as the disease produced by brain-derived infectious material (Castilla et al., 2005). The conclusion of these findings that all the information required to propagate the infectious properties is enciphered in the structure of PrP^Sc is further supported by recent studies from Supattapone and coworkers in which infectious prions were generated in vitro by PMCA using purified PrP^C and PrP^Sc with the sole addition of synthetic polyanions (Deleault et al., 2007).

The goal of this study was to attempt crossing the species barriers in vitro to generate novel infectious prions in a cell-free system. For these studies we used mice and hamsters, two experimental rodent systems widely employed in TSEs studies and for which several prion strains are available (Bruce, 2003; Kimberlin and Walker, 1988; Morales et al., 2007). The PrP sequence shows 9 differences between these two animal species (Fig. 1A). Infectivity studies have shown that there is large barrier for prion transmission between these species (Kimberlin et al., 1989; Kimberlin and Walker, 1988; Race et al., 2002). Our findings show that incubation of PrP^C from one of the species with PrP^Sc from the other resulted in new PrP^Sc that was infectious to wild type animals. Interestingly a detailed examination of the infectious, neuropathological and biochemical features of the disease produced revealed characteristics that were different from other known prion strains. These results indicate that the novel prions generated in vitro by crossing the mice-hamsters barrier represent a new strain. Strikingly, studies of the infectious characteristics of these newly generated prions after different rounds of PMCA showed that the procedure not only enable crossing the species barrier, but also resulted in stabilization of the new strain in vitro by successive rounds of amplification. Our findings represent the first time in which prions have been propagated in vitro across the species barrier leading to the generation and adaptation of novel prion strains.

Figure 1

In vitro conversion of hamster PrP^C induced by mouse RML PrP^Sc

Results

Crossing the Mice-Hamsters species barrier to generate new hamster’s prions

To assess whether prions can be generated in vitro across the species barrier, we used hamsters and mice, two widely studied rodent experimental models of TSEs (Bruce, 2003; Kimberlin and Walker, 1988; Morales et al., 2007). A PMCA experiment done using our standard conditions for amplification of mouse RML prions showed no detectable formation of PrP^Sc when hamster PrP^C was used as a substrate (Fig 1A). Conversely, a robust PrP^Sc generation was observed using mouse PrP^C substrate. For this experiment we mixed a 1000-fold dilution of RML PrP^Sc into 10% brain homogenates of healthy hamsters and mice, respectively. We reasoned that if in vivo it takes longer for prions to replicate across species barriers, in PMCA we should also encounter more difficulties to convert PrP^C using PrP^Sc from a different species. To attempt forcing the in vitro conversion, we added a higher proportion of PrP^Sc-containing mice brain homogenate into the hamster substrate. A range of dilutions from 50- to 800-fold were tested, but the problem with these experiments is that the large concentration of RML PrP^Sc used as inoculum, makes it difficult to estimate convincingly if new PrP^Sc generation was obtained (Fig 1B). Fortunately, the 3F4 monoclonal antibody can recognize hamster, but not mouse PrP (Lund et al., 2007). Using this antibody for western blot, we could clearly observe that protease-resistant hamster PrP^Sc was being produced when the reaction was done with low dilutions (from 1:50 to 1:200) of mouse RML PrP^Sc (Fig. 1C). When the amplification was attempted with 800-fold diluted PrP^Sc-containing mouse brain homogenate, only a very faint signal was observed confirming the results obtained in figure 1A and the idea that the combination of PrP^C and PrP^Sc from different species impairs PMCA efficiency.

Newly generated hamster PrP^Sc starting from RML prions was propagated many times in vitro by serial PMCA in order to remove by dilution the initial amount of mouse scrapie brain material added to begin prion replication (Fig. 1D). As described before, using this procedure we can completely remove all molecules of brain-derived PrP^Sc from the sample (Castilla et al., 2005). Hamster PrP^Sc of the RML origin efficiently propagates in vitro at expenses of hamster PrP^C. Interestingly, in the first PMCA round the glycoform distribution pattern of the in vitro generated hamster PrP^Sc was comparable to the RML profile showing the 3 glycoform bands (Fig. 1D). After further PMCA rounds this pattern changed to become undistinguishable from PrP^Sc associated to the typical hamster strains, such as 263K (Fig. 1D) or Hyper, in which the di-glycosylated band is highly predominant. This result suggest that the characteristics of the newly generated PrP^Sc are being adapted to the new specie during successive PMCA cycling, reminiscent of the adaptation process occurring in vivo upon serial passages of the infectious material. After 20 serial rounds of PMCA, representing a dilution equivalent to 10⁻²² with respect to the brain, (since the first round contains a 100-fold dilution of the material), our estimation is that no molecules of mouse brain PrP^Sc should be present in the sample. This in vitro generated material was termed RML-Ha PrP^Sc to emphasis the RML origin of this new hamster misfolded prion protein. To make sure that newly formed PrP^Sc was indeed coming from conversion of hamster PrP^C induced by mouse PrP^Sc, and not just spontaneous “de novo” formation of PrP^Sc in hamsters (Deleault et al., 2007), we did a large experiment to analyze in detail the possibility for spontaneous generation of PrP^Sc and infectivity under our experimental conditions. Samples of healthy brain homogenate from 10 different hamsters were subjected to serial rounds of PMCA amplification in the absence of PrP^Sc seed. The result showed that up to 20 serial rounds of PMCA we did not observed de novo formation of PrP^Sc in any of the samples (Fig. 1E).

Inoculation of wild type hamsters with RML-Ha PrP^Sc (produced after a 10⁻²² dilution of RML scrapie brain homogenate) produced disease in 100% of the animals by both intra-cerebral (i.c.) and intra-peritoneal (i.p.) routes (Fig. 2). The disease exhibits the clinical characteristics typical of hamster scrapie, including hyperactivity, motor impairment, head wobbling, muscle weakness and weight loss. The incubation time in the first passage was 165 ± 6 days by i.c. inoculation (Fig. 2A and C). This in longer than the incubation time obtained with hamster scrapie strains, such as 263K and HY in which a similar quantity of PrP^Sc produces disease at around 100 days by this route (Fig. 2A and C). However, in agreement with our previously reported data (Castilla et al., 2005), when hamster 263K prions were replicated in vitro by PMCA the newly generated PrP^Sc produces disease with a delay similar to that observed with the RML-Ha material (Fig. 2A and C). The delay in our previous study was eliminated upon a second passage in vivo, in which the new infectious material was stabilized to acquire properties undistinguishable from in vivo derived 263K (Fig. 2B and C). Interestingly, in the HY hamster prion strain, PMCA generated material did not show any statistically significant difference compared to in vivo produced prions (Fig. 2A and C). These results suggest that in vitro replication of prions by PMCA maintains the strain characteristics, at least in what respect to the incubation periods. To assess the stability of RML-Ha and estimate the stabilized incubation period, we performed a second passage. As shown in figure 2B, the incubation time of RML-Ha prions was decreased to around 90 days, which is very similar to that obtained with 263K and HY, but different from the DY strain. These results suggest that RML-Ha prions behave similarly to the 263K strain; both in vitro generated prions show a delay in the first passage that gets corrected upon a second in vivo passage. This feature is not displayed by other hamster prions strains, such as HY, or other species of prions (see below the results in mouse), where PMCA generated prions exhibited the same incubation period in the first passage as in vivo produced infectious material. As expected hamsters inoculated with RML prions did not develop disease during the time of the experiment (>400 days). Animals inoculated with hamster brain homogenate subjected to 20 rounds of PMCA in the absence of PrP^Sc (control for the de novo generation of PrP^Sc) did not develop disease >400 days after inoculation (Fig. 2A and C). Intraperitoneal inoculations of the infectious material showed a clear difference between the three prion strains used as reference, with 263K being the fastest and DY not producing disease by this route (Fig. 2D). The incubation period produced by i.p. inoculation of RML-Ha prions was longer than 263K and HY strains, with an average of 254 days in the first passage. This is also longer than 263K prions amplified in vitro by PMCA, which produced disease after 199 days post-inoculation in the first passage (Fig. 2D and F). A second in vivo passage again stabilized PMCA-generated 263K prions to produce disease at a time indistinguishable from brain-derived 263K infectious material. The second passage of RML-Ha prions showed that the stabilized incubation period for the i.p. route was in average around 140 days, which is significantly higher than 263K or 263K-PMCA material, but shorter than HY prions (Fig. 2E and F). The differences remained stable in a third passage (data not shown). These results indicate that in some aspects RML-Ha prions are similar to the agent in the 263K strain, but in others features is intermediate between 263K and HY prions, providing a first indication that the material obtained by crossing the mouse-hamster species barrier represents a new hamster prion strain.

Figure 2

Infectivity of newly generated RML-Ha PrP^Sc after crossing the species barrier

To further assess the characteristics of the disease produced by in vitro generated RML-Ha prions, we studied in detail the neuropathological and biochemical features of the brain damage. Histopathological studies showed that animals inoculated with RML-Ha prions exhibit the typical brain lesions of scrapie, including spongiform degeneration, astroglyosis and PrP^Sc deposition (Fig. 3A, B and C). Quantitative studies of the vacuolation profile in different brain areas showed that RML-Ha infected hamsters showed the largest extent of spongiosis in medulla and cerebellum, and less damage in hippocampus, cortex and colliculum (Fig. 3D). This pattern of brain damage was similar to that observed in 263K inoculated animals and statistically different from that obtained in hamsters injected with HY and DY (Fig. 3D). However, the extent of both astroglyosis (Fig. 3B) and PrP^Sc accumulation (Fig. 3C) in the medulla of RML-Ha infected animals was lower than in 263K sick animals and similar to that observed in HY injected hamsters (Fig. 3B and C). This data suggest again that the RML-Ha prions are a new strain with properties intermediate between the previously known 263K and HY hamster strains.

Figure 3

Histopathological features of the disease induced by inoculation of hamsters with PMCA-generated RML-Ha PrP^Sc

Comparative studies of the biochemical characteristics of PrP^Sc obtained from the brain of sick animals after inoculation with RML-Ha, 263K, HY and DY were done by analyzing the electrophoretical pattern of the protein, its susceptibility to proteolytic degradation and its resistance to denaturation. To compare the protease resistance profile, similar quantities of PrP^Sc from the new RML-Ha prions and PrP^Sc obtained from the brain of sick hamsters inoculated with the prion strains 263K, HY and DY were treated for 60 min with various concentration of proteinase K (PK) (Fig. 4A). RML-Ha PrP^Sc was highly resistant to large PK concentrations. The misfolded protein associated to the newly generated strain was more resistant than HY or DY, and similarly (but still significantly higher) susceptible to PK digestion than 263K PrP^Sc (Fig. 4A). The PK concentration in which 50% of the protein was degraded (PK50) was highest for PrP^Sc associated to RML-Ha, followed by 263K, HY, DY and RML (Table 1).

Figure 4

Biochemical characterization of RML-Ha PrP^Sc

Another characteristic we studied was the electrophoretic mobility and glycosylation pattern of PrP^Sc associated to distinct strains. The predominant glycoform for the hamsters strains (including the newly generated RML-Ha) is the di-glycosylated band, whereas mouse RML PrP^Sc shows a more even distribution of the three bands with the main one being the mono-glycosylated form. To assess the size of the protein after PK cleavage we performed endo-glycosidase treatment to remove the glycosylated chains (Fig. 4B). Whereas PrP^Sc associated to the DY strain has a higher electrophoretical mobility, no significant differences were observed among the other proteins. Another biochemical property of misfolded PrP often used to differentiate prion strains is its resistance to chemical denaturation (Safar et al., 1998). Clear differences were observed on the guanidine concentrations required to denature PrP^Sc associated to different strains (Fig, 4C). The concentration of the chaotropic agent needed to denature 50% of PrP^Sc RML-Ha was 1.11 M, substantially different from the 1.69, 1.56, and 1.72M required for the proteins associated to HY, DY and RML, respectively (Table 1).

Crossing the Hamsters-Mice species barrier to generate and stabilize new mouse prions

To study the barrier between these rodent species in the opposite direction we mixed 263K hamster prions with mouse healthy brain homogenate. As before, when a standard PMCA assay was done by diluting 263K brain homogenate 1000-folds into mouse healthy brain material we did not see detectable generation of mouse PrP^Sc (data not shown). However, when a higher quantity of hamster PrP^Sc was added we were able to generate new mouse PrP^Sc (termed 263K-Mo) that could be propagated by serial rounds of PMCA to reach a dilution of the hamster brain homogenate equivalent to 10⁻¹⁷ (Fig. 5A). Since there are not available antibodies capable to recognize mouse PrP and not hamster PrP, we could not compare the electrophoretical pattern of PrP^Sc generated in the first rounds of PMCA with the profile of PrP^Sc typically observed in mouse and hamster strains. However, the western blot pattern of 263K-Mo after 15 rounds of PMCA (when no more molecules of 263K PrP^Sc are present) is similar to the one observed for RML or other ovine-derived mouse strains, despite a slightly faster migration (Supplementary figure 1A) that will be investigated in more detail later. To assess whether newly generated PrP^Sc was indeed coming from conversion of mouse PrP^C induced by 263K hamster PrP^Sc, and not just spontaneous “de novo” formation of PrP^Sc in mice, we did an experiment to analyze the possibility for spontaneous generation of PrP^Sc and infectivity under our experimental conditions. Samples of healthy brain homogenate from 10 different mice were subjected to serial rounds of PMCA amplification in the absence of PrP^Sc seed. The result showed that up to 20 serial rounds of PMCA we did not observed de novo formation of PrP^Sc in any of the samples (Supplementary figure 1B).

Figure 5

In vitro conversion of mouse PrP^C induced by hamster 263K PrP^Sc generates infectious prions

To assess if mouse PrP^Sc generated in vitro from hamster 263K is infectious to wild type mice, and to determine whether the infectious properties are being adapted upon serial PMCA passages, several rounds of in vitro generated material were inoculated into mice (Fig. 5A). Despite the fact that the same amount of PrP^Sc was inoculated (as determined by western blot), striking differences in the infectious properties were seen among in vitro generated prions in distinct rounds of PMCA (Fig. 5B). Only 2 of the 6 mice inoculated with material produced in the first round of PMCA showed disease symptoms, which appear at a very long time after inoculation (around 500 days) (Fig. 5A and B). A complete attack rate was observed when animals were inoculated with material produced after 3 serial rounds of PMCA. However, the incubation period was long (around 310 days in average) and with a large dispersion among animals (Fig 5A and B). The incubation period became stable, short (around 165 days) and with little dispersion after the six serial round of PMCA. These findings indicate that upon successive rounds of PMCA the newly generated prion after crossing the species barrier is becoming adapted and stabilized to the new host, a process very similar to what is seen after several passages in vivo. The large dispersion of incubation times observed in the third round of PMCA suggests that more than one strain has been generated upon crossing the species barrier and that successive in vitro amplification leads to the selection and cloning of the most efficient of these strains. The incubation time for 263K-Mo after 15 rounds of PMCA (equivalent to a 10⁻¹⁷ dilution of the 263K inoculum) was around 165 days, similar to the one produced by scrapie-adapted mouse strains, such as RML, but different from the bovine strain 301C (Fig. 5C). In vitro replication of the mouse strains RML and 301C at expenses of mouse PrP^C produced a PrP^Sc with identical properties as the brain-derived material, reflected as an indistinguishable incubation period (Fig. 5C). As expected mouse inoculated with hamster 263K prions did not develop disease during the time of the experiment (>500 days). No disease was also observed in animals inoculated with mouse brain homogenate subjected to 20 rounds of PMCA in the absence of PrP^Sc, which correspond to the control experiment for the de novo generation of PrP^Sc (Fig. 5C).

To analyze whether the newly generated 263K-Mo infectious material corresponded to a new strain of mouse prions, we studied the histopathological and biochemical features of the brain damage. Animals affected with the disease produced by inoculation of 263K-Mo showed extensive vacuolation in the medulla and hypocampus and moderate, but clearly detectable damage in the cerebellum (Fig. 6A and D). The pattern of spongiform degeneration does not correspond with any of the previously known mouse strains studied and indeed is statistically significantly different to the vacuolation profile produced by RML and 301C prions (Fig. 6D). Differences were also detected in the extent of brain inflammation produced by 263K-Mo, since the degree of astroglyosis was less prominent than the one observed in animals inoculated with RML or 301C prions (Fig. 6B). The profile of PrP^Sc accumulation consisted mostly of diffuse deposition and was not clearly different from the one observed in the other strains (Fig. 6C). Then, we studied the biochemical characteristics of PrP^Sc obtained from the brain of animals infected with 263K-Mo. Electrophoretical migration was assessed after PK digestion and endoglycosidase treatment to remove glycosylation chains. The PK-resistant core of PrP^Sc migrated slightly faster than RML, but slightly slower than 301C, with an estimated molecular weight of 20 KDa (Fig. 7A and B). These results indicate that the cleavage site after PK digestion is different from all the currently known mouse strains. This is important because it is thought that differences in the PK cleavage site reflect disparities in the folding or aggregation of the protein (Chen et al., 2000; Collinge et al., 1996). To further search for biochemical differences we subjected the protein to proteolytic degradation using various concentrations of PK. 263K-Mo PrP^Sc was much more resistant to PK than RML (Fig. 7B), with a PK₅₀ (the PK concentration needed to degrade half of the protein) of 1450 ug/ml (Fig. 7C), much larger than the values obtained for RML (240 ug/ml) and 301C (430 ug/ml) (Supplementary Table 2). Interestingly, the high resistance of PrP^Sc is typical of the hamster prions (Supplementary Table 1), and indeed 263K, the parental strain of the newly generated mouse prions, has a PK₅₀ of around 1700 ug/ml.

Figure 6

Histopathological features of the disease induced by inoculation of mice with PMCA-generated 263K-Mo PrP^Sc

Figure 7

Biochemical characteristcs of 263K-Mo PrP^Sc

Discussion

The phenomenon of the species barrier, by which the agent coming from one species can infect only a limited number of other species, is a typical feature of prion diseases. The molecular basis of this process is not well-understood, but it is thought to be controlled by the structure and folding of the prion protein (Moore et al., 2005; Vanik et al., 2004). As with the related phenomenon of prion strains, it is difficult to imagine how an infectious agent lacking genetic material and composed by a single protein can encode the structural diversity and specificity required to control strains variability and species selectivity (Soto and Castilla, 2004).

In addition to the intriguing molecular mechanism behind the species barrier, understanding this phenomenon has profound implications for public health. Indeed, one of the scariest medical problems of the last decades has been the emergence of a new and fatal human prion disease (variant CJD) originated by cross species transmission of BSE from cattle (Collinge, 1999; Will et al., 1996). BSE has not only been transmitted to humans. The extensive use of cow-derived material for feeding other animals led to the generation of new diseases in exotic felines, non human primates, and domestic cats (Doherr, 2003). Worrisomely, the transmission of BSE into these different species could create new prion strains with novel biological and biochemical characteristics and thus a potentially new hazard for human health. More frightening is perhaps the possibility that BSE has been passed into sheep and goats. Studies have already shown that this transmission is possible and actually relatively easy (Foster et al., 1993). The disease produced is clinically similar to scrapie, but since it comes from BSE it has the potential to be infectious to humans. One interesting new prion disease is CWD, a disorder affecting farm and wild species of cervids (Sigurdson and Aguzzi, 2006; Williams, 2005). The origin of CWD and its potential to transmit to humans are currently unknown. This is worrisome, considering that CWD has became endemic in some parts of USA and the number of cases continues to increase (Williams, 2005). CWD transmissibility studies have been performed in many species in order to predict how this disease could be spread by consumption of CWD meat (Sigurdson and Aguzzi, 2006). Transmission of CWD to humans cannot be ruled out at present and a similar infective episode to BSE involving CWD could result in catastrophic consequences.

The exciting scientific problem coupled with the relevant public health issue prompted us to develop strategies to reproduce the species barrier phenomenon in the test tube. We reported previously the generation of infectious prions in vitro by cyclic replication of the protein misfolding process featuring the pathogenesis of prion diseases (Castilla et al., 2005). These results were reproduced and extended by other groups to better dissect the elements required for prion replication (Deleault et al., 2007; Weber et al., 2007). The PMCA technology has been adapted to replicate prions from various species (Deleault et al., 2005; Jones et al., 2007; Kurt et al., 2007; Murayama et al., 2007; Sarafoff et al., 2005; Soto et al., 2005) and even to use bacterially produced recombinant PrP as substrate (Atarashi et al., 2007). The conclusion of these studies together with the findings reported in this manuscript is that propagation of the PrP^Sc misfolding results in formation of infectious material, which maintain the strains and species barrier properties of the original prions. Qualitatively similar conclusions have been obtained for yeast prions, which are a group of “infectious proteins” that behave as a non-Mendelian genetic element and transmit biological information in the absence of nucleic acid (Wickner et al., 1995). Recent studies showed that bacterially produced N-terminal fragments of the yeast prions Sup35p and Ure2p when transformed into amyloid fibrils were able to propagate the prion phenotype to yeast cells (Brachmann et al., 2005; King and Diaz-Avalos, 2004; Tanaka et al., 2004). Infection of yeast with different conformers led to generation of distinct prion strains in vivo (Brachmann et al., 2005; Tanaka et al., 2004), indicating that differences in the conformation of the infectious protein determine prion strain variation. Remarkably, yeast prions also show the species barrier phenomenon and recent data indicate that strain conformation is the critical determinant of cross-species prion transmission (Tanaka et al., 2005).

In the current study we demonstrate the generation of novel infectious prions across the species barriers. For this purpose we mixed PrP^Sc from one species with PrP^C from a different animal species and subjected the mixture to serial rounds of PMCA to generate, propagate and stabilize new prion strains. Hamster PrP^Sc generated from mouse RML prions was infectious to wild-type hamsters. Detailed analysis of the disease characteristics and comparison with the illness produced by several known hamster prion strains indicate that the newly generated infectious material across the species barrier corresponds to a new prion strain in hamsters (termed RML-Ha). The main differences of the RML-Ha were on the incubation times after i.p. inoculation, the extremely high resistance to PK degradation and the pattern of brain damage (Supplementary Table 1). Similarly, PrP^Sc generated by converting mouse PrP^C with hamster PrP^Sc from the 263K strain showed to be infectious to wild type mice, with an incubation period comparable to that obtained after inoculation with some of the mouse-adapted scrapie strains, such as RML. Again, the disease produced by the new prions (termed 263K-Mo) was clearly distinguishable from the one produced by some of the currently known mouse prion strains. The major differences were seen on the electrophoretical migration, extremely high resistance to proteolytic degradation and pattern of brain spongiform degeneration (Supplementary Table 2). To rule out that newly generated PrP^Sc in these experiments was coming from “de novo” spontaneous conversion of PrP^C into PrP^Sc during PMCA, we used samples of healthy brain homogenate from 10 different mice and hamsters that were subjected to serial rounds of PMCA amplification in the absence of PrP^Sc seed. The result showed that up to 20 serial rounds of PMCA we did not observed de novo formation of PrP^Sc in any of the samples. This material was inoculated into wild-type animals and no disease has been observed >400 days after inoculation. These results strongly indicate that the generation of PrP^Sc reported in the present study when normal brain homogenate from one species was mixed with sick brain homogenate from the different species was due to inter-species conversion. Nevertheless, we would like to highlight that recently we have been able to generate in vitro PrP^Sc de novo without addition of PrP^Sc seed (data not shown). However, to reach this aim, the PMCA conditions need to be modified. The modifications include changes on the PMCA parameters (length of incubation and potency of sonication), pre-incubation or pre-treatment of the normal brain homogenate to induce/stabilize PrP misfolding prior to PMCA. These findings suggest that de novo formation of PrP^Sc can be experimentally distinguished from replication of pre-formed PrP^Sc, indicating that the biochemical, conformational or stability properties of the PrP structures involved in both processes are probably different.Standard PMCA conditions, as those used in the current study do not result in spontaneous PrP^Sc formation.

Interestingly, in our serial PMCA amplifications of RML PrP^Sc into hamster PrP^C we observed a progressive change on the western blot profile of the newly generated RML-Ha PrP^Sc. Indeed, in the first round of PMCA the glycoform distribution pattern was reminiscent of RML and later switched to a profile typical of the hamster strains, characterized by the predomination of the di-glycosylated form (Fig 1D). Our interpretation of this result was that consecutive rounds of PMCA may enable to adapt and stabilize the new prion strain. To further study this possibility in our experiments in which mouse prions were generated from 263K hamster prions, we inoculated the material generated after various rounds of PMCA. Strikingly, similar amounts of PrP^Sc generated after 1 and 3 rounds of PMCA produced disease with incomplete attack rates and/or very long incubation periods (Fig. 5B and C). Incubation time stabilized after 6 rounds of serial PMCA, suggesting that at this point the new strain is fully adapted. These findings are of high interest, because they suggest that PMCA not only enable to reproduce the inter-species transmission of prions, but also to mimic the strain adaptation process observed in vivo. In vivo adaptation and stabilization of prions generated after crossing species barrier takes at least 4 consecutive passages, which requires several years of work (Race et al., 2001; Race et al., 2002). Conversely, strain adaptation by PMCA takes only 2 or 3 weeks. Importantly, the kinetics of adaptation in vitro and in vivo, as well as the characteristics of the stabilized material are very similar. Indeed, it has been reported that 3 serial passages of 263K in mice produce disease in all animals with an incubation time of around 300 days (Race et al., 2002). This result is very similar to the data obtained with the material generated in vitro after 3 successive rounds on PMCA replication (Fig. 5B and C). Moreover, less than 3 in vivo passages produced an incomplete attack rate and more than 3 passages are needed to obtain a stable and low incubation period (Race et al., 2002), which is in the same range of our 263K-Mo infectious material. Finally, similarly to our in vitro data, the in vivo cross-species transmission between hamster and mice also led to generation of novel prion strains (Race et al., 2001; Race et al., 2002). Although we are tempted to speculate that each PMCA round has the same effect on strain adaptation than each in vivo passage, more experiments with other species combinations are needed to reach this conclusion.

In summary our results show that all elements controlling inter-species transmission of prions are contained in a cell-free system and that new prion strains can be generated, adapted and stabilized upon crossing species barrier in vitro by PMCA. These findings provide additional support for the prion hypothesis, suggesting that species barrier transmission and strain generation are determined by the propagation of PrP misfolding. Furthermore, the data demonstrate that PMCA is a valuable tool to investigate the strength of the barrier between diverse species, its molecular determinants and the expected features of the new infectious material produced. Finally, our findings suggest that the universe of possible prions is not restricted to those currently known, but that likely many new infectious foldings of the prion protein may be produced and one of the sources for this is cross-species transmission.

Experimental Procedures

Preparation of Tissue Homogenates

Healthy and sick animals were perfused with phosphate-buffered saline (PBS) plus 5 mM EDTA previous to harvesting the tissue. Ten percent brain homogenates (w/v) were prepared in conversion buffer (PBS containing NaCl 150mM, 1.0% Triton X-100, 4mM EDTA and the complete™ cocktail of protease inhibitors from Boehringer Mannheim, Mannheim, Germany). The samples were clarified by a brief, low-speed centrifugation (1500 rpm for 30s) using an Eppendorf centrifuge (Hamburg, Germany), model 5414.

Serial replication of prions in vitro by PMCA

Aliquots of 10% brain homogenate from clinically sick mice infected with RML or 301C and hamsters infected with 263K, HY or DY prions were diluted into 10% hamster or mice healthy brain homogenate. Samples were loaded onto 0.2-ml PCR tubes and positioned on an adaptor placed on the plate holder of a microsonicator (Misonix Model 3000, Farmingdale, NY). Each PMCA cycle consisted of 30 min incubation at 37°C followed by a 20 sec pulse of sonication set at potency of 7. Samples were incubated without shaking immersed in the water of the sonicator bath. After a round of 96 cycles, a 10 ul aliquot of the amplified material was diluted into 90 ul of more normal brain homogenate and a new round of 96 PMCA cycles was performed. This procedure was repeated several times to reach the final dilutions indicated in the text. The detailed protocol for PMCA, including reagents, solutions and troubleshooting, has been published elsewhere (Castilla et al., 2006; Saa et al., 2005).

Proteinase K degradation assay

The standard procedure to digest PrP^Sc consists on subjecting the samples to incubation in the presence of PK (50 µg/ml) during 60 min at 37°C. The digestion was stopped by adding electrophoresis sample buffer and the protease-resistant PrP was revealed by western blotting. To study the profile of PK sensitivity for in vitro- and in vivo-generated PrP^Sc, the samples were incubated for 60 min at 37°C with different concentrations of PK ranging from 0 to 2500 µg/ml. The PK₅₀ values represent the concentration of PK needed to digest half of the protein and these values are estimated based on the densitometric analysis of 3 replicated western blots.

Guanidine Denaturation Assay

Samples were incubated with different concentrations of guanidine hydrochloride for 2 hr at room temperature with shaking. Thereafter, samples were analyzed for PrP solubility. Samples were incubated in the presence of 10% sarkosyl for 30 min at 4°C and centrifuged at 100,000 × g for 1 hr in a Biosafe Optima MAX ultracentrifuge (Beckman Coulter, Fullerton, CA). The pellet of the centrifugation was resuspended in conversion buffer plus electrophoresis sample buffer and treated with PK as described above. Equivalent aliquots of pellet were analyzed by western blot. The Gdn₅₀ value corresponds to the concentration of guanidine hydrochloride required to denaturate 50% of the protein and these values were estimated based on the densitometric analysis of 3 replicated western blots.

Protein deglycosylation assay

PrP^Sc samples were first digested with PK as describe above. After addition of 10% sarkosyl, samples were centrifuged at 100,000×g for 1h at 4°C, supernatant was discarded and pellet resuspended in 100µl of glycoprotein denaturing buffer (New England Biolabs, Beverly, MA) and incubated for 10 min at 100°C. Thereafter, 26µl of 50 mM sodium phosphate, pH 7.5 containing 1% nonidet P-40 and 3µl of peptide N-glycosidase F (New England Biolabs, Beverly, MA) were added. Samples were incubated for 2h at 37°C, the reaction was stopped by adding electrophoresis buffer and samples analyzed by Western blot.

Western blot

Proteins were fractionated by sodium dodecyl sulphate-polyacrylamide gel electrophoresis (SDS-PAGE) under reducing conditions, electroblotted into nitrocellulose membrane and probed with 6H4 (for mouse samples) and 3F4 (for hamster samples) antibodies at a 1:5,000 dilution. The immunoreactive bands were visualized by enhanced chemoluminesence assay (Amersham, Piscataway, NJ) using an UVp image analysis system. To assess the quantity of PrP^Sc in the western blot, densitometric analyses were done by triplicate.

PrP^Sc quantification

To inject the same quantity of PrP^Sc from each preparation, the samples were compared by western blotting after PK digestion. To obtain a reliable and robust quantification, we ran several different dilutions of the sample in the same gel, to avoid artifacts due to saturation of the signal or to a too weak signal.

Infectivity Studies

In vivo infectivity studies were done in C57Bl6 female mice or Golden Syrian female hamsters, purchased from Charles river. Animals were 4- to 6-weeks old at the time of inoculation. Anesthesized animals were injected stereotaxically into the right hyppocampus with 2 or 4 ul of the mice or hamster infectious material, respectively. For the intra-peritoneal (i.p.) infectivity studies, 100 µl of the sample were injected into the peritoneal cavity. The quantity of infectious material injected corresponds to the plateau portion of the incubation period, therefore small differences in the amount of infectivity should not change incubation period unless there are strain differences. We know we are at plateau, because the injected material is equivalent to a 10% brain homogenate. This material can be diluted 10–50 times and still the incubation periods will be similar. The onset of clinical disease was measured by scoring the animals twice a week. For mice the following scale was used: 1, Normal animal; 2, roughcoat on limbs; 3, extensive roughcoat, hunckback and visible motor abnormalities; 4, urogenital lesions; 5, terminal stage of the disease in which the animal present cachexia and lies in the cage with little movement. For hamsters the following scoring scale was used: 1, Normal animal; 2, Mild behavioral abnormalities including hyperactivity and hypersensitivity to noise; 3, Moderate behavioral problems including tremor of the head, ataxia, wobbling gait, head bobbing, irritability and aggressiveness (or lethargy in case of the DY strain); 4, Severe behavioral abnormalities including all of the above plus jerks of the head and body and spontaneous backrolls; 5, Terminal stage of the disease in which the animal lies in the cage and is no longer able to stand up.

Animals scoring level 4 during two consecutive weeks were considered sick and were sacrificed to avoid excessive pain using exposition to carbonic dioxide. Brains were extracted and analyzed histologically and biochemically. The right cerebral hemisphere was frozen and stored at −70°C for biochemical examination of PrP^Sc using Western blot analysis and the left hemisphere was used for histology analysis.

Histopathological studies

Brain tissue was fixed in 10% formaldehyde solution, cutted in sections and embedded in paraffin. Serial sections (6µm thick) from each block were stained with hematoxylin-eosin, or incubated with monoclonal antibodies recognizing PrP or the glial fibrillary acidic protein, using our previously described protocols (Castilla et al., 2005). Samples were visualized with a Zeiss microscope. The vacuolation profile was estimated by considering both number and size of spongiform degeneration in 5 different brain areas: occipital cortex, cerebellum (mostly white matter), medulla (spinal 5 nucleus, interpolar part), inferior colliculum and hippocampus (CA1 and CA2 regions). Each analyzed brain area was scored from 0 to 4 according to the extent of vacuolation in slides stained with haematoxilin-eosin and visualized at a 40X magnification. Samples were analyzed blindly by two different persons and the scores represent the average of the two determinations.

Statistical analysis

The differences on incubation periods, histopathological profile of brain damage and biochemical characteristics of PrP^Sc were analyzed by ANOVA, followed by the Dunnett Multiple Comparison post-test to estimate the significance of the differences between the newly generated strains and each of the other hamster and mouse prion strains studied. For these studies the data was analyzed using the GraphPad Instat, version 3.05 software.

Supplementary Material

Click here to view.^{(60K, pdf)}

Acknowledgements

This research was supported in part by NIH grants R01NS049173 and P01AI77774 to CS.

Footnotes

This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final citable form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

Reference List

Atarashi R, Moore RA, Sim VL, Hughson AG, Dorward DW, Onwubiko HA, Priola SA, Caughey B. Ultrasensitive detection of scrapie prion protein using seeded conversion of recombinant prion protein. Nat. Methods. 2007;4:645–650. [PubMed]
Brachmann A, Baxa U, Wickner RB. Prion generation in vitro: amyloid of Ure2p is infectious. EMBO J. 2005;24:3082–3092. [PubMed]
Bruce ME. 'New variant' Creutzfeldt-Jakob disease and bovine spongiform encephalopathy. Nat. Med. 2000;6:258–259. [PubMed]
Bruce ME. TSE strain variation. Br. Med. Bull. 2003;66:99–108. [PubMed]
Buschmann A, Groschup MH. TSE eradication in small ruminants--quo vadis? Berl Munch. Tierarztl. Wochenschr. 2005;118:365–371. [PubMed]
Caramelli M, Ru G, Acutis P, Forloni G. Prion diseases: current understanding of epidemiology and pathogenesis, and therapeutic advances. CNS. Drugs. 2006;20:15–28. [PubMed]
Castilla J, Saá P, Hetz C, Soto C. In vitro generation of infectious scrapie prions. Cell. 2005;121:195–206. [PubMed]
Castilla J, Saa P, Morales R, Abid K, Maundrell K, Soto C. Protein misfolding cyclic amplification for diagnosis and prion propagation studies. Methods Enzymol. 2006;412:3–21. [PubMed]
Chen SG, Gambetti P. A journey through the species barrier. Neuron. 2002;34:854–856. [PubMed]
Chen SG, Zou W, Parchi P, Gambetti P. PrP(Sc) typing by N-terminal sequencing and mass spectrometry. Arch. Virol. 2000 Suppl:209–216. [PubMed]
Collinge J. Variant Creutzfeldt-Jakob disease. Lancet. 1999;354:317–323. [PubMed]
Collinge J. Prion diseases of humans and animals: their causes and molecular basis. Annu. Rev. Neurosci. 2001;24:519–550. [PubMed]
Collinge J, Sidle KC, Meads J, Ironside J, Hill AF. Molecular analysis of prion strain variation and the aetiology of 'new variant' CJD. Nature. 1996;383:685–690. [PubMed]
Deleault NR, Geoghegan JC, Nishina K, Kascsak R, Williamson RA, Supattapone S. Protease-resistant prion protein amplification reconstituted with partially purified substrates and synthetic polyanions. J. Biol. Chem. 2005;280:26873–26879. [PubMed]
Deleault NR, Harris BT, Rees JR, Supattapone S. Formation of native prions from minimal components in vitro. Proc. Natl. Acad. Sci. U. S. A. 2007;104:9741–9746. [PubMed]
Doherr MG. Bovine spongiform encephalopathy (BSE)--infectious, contagious, zoonotic or production disease? Acta Vet. Scand. 2003;98 Suppl:33–42. [PubMed]
Foster JD, Hope J, Fraser H. Transmission of bovine spongiform encephalopathy to sheep and goats. Vet. Rec. 1993;133:339–341. [PubMed]
Hill AF, Collinge J. Prion strains and species barriers. Contrib. Microbiol. 2004;11:33–49. [PubMed]
Hunter N. Scrapie and experimental BSE in sheep. Br. Med. Bull. 2003;66:171–183. [PubMed]
Hunter N. Scrapie. Mol. Biotechnol. 1998;9:225–234. [PubMed]
Ironside JW, Head MW. Neuropathology and molecular biology of variant Creutzfeldt-Jakob disease. Curr. Top. Microbiol. Immunol. 2004;284:133–159. [PubMed]
Jones M, Peden AH, Prowse CV, Groner A, Manson JC, Turner ML, Ironside JW, MacGregor IR, Head MW. In vitro amplification and detection of variant Creutzfeldt-Jakob disease PrPSc. J. Pathol. 2007;213:21–26. [PubMed]
Kimberlin RH, Walker CA. Pathogenesis of experimental scrapie. Ciba Found. Symp. 1988;135:37–62. [PubMed]
Kimberlin RH, Walker CA, Fraser H. The genomic identity of different strains of mouse scrapie is expressed in hamsters and preserved on reisolation in mice. J. Gen. Virol. 1989;70:2017–2025. [PubMed]
King CY, Diaz-Avalos R. Protein-only transmission of three yeast prion strains. Nature. 2004;428:319–323. [PubMed]
Kurt TD, Perrott MR, Wilusz CJ, Wilusz J, Supattapone S, Telling GC, Zabel MD, Hoover EA. Efficient in vitro amplification of chronic wasting disease PrPRES. J. Virol. 2007;81:9605–9608. [PMC free article] [PubMed]
Lund C, Olsen CM, Tveit H, Tranulis MA. Characterization of the prion protein 3F4 epitope and its use as a molecular tag. J. Neurosci. Methods. 2007;165:183–190. [PubMed]
Moore RA, Vorberg I, Priola SA. Species barriers in prion diseases--brief review. Arch. Virol. 2005 Suppl:187–202. [PubMed]
Morales R, Abid K, Soto C. The prion strain phenomenon: molecular basis and unprecedented features. Biochim. Biophys. Acta. 2007;1772:681–691. [PMC free article] [PubMed]
Murayama Y, Yoshioka M, Yokoyama T, Iwamaru Y, Imamura M, Masujin K, Yoshiba S, Mohri S. Efficient in vitro amplification of a mouse-adapted scrapie prion protein. Neurosci. Lett. 2007;413:270–273. [PubMed]
Prusiner SB. Prions. Proc. Natl. Acad. Sci. U.S.A. 1998;95:13363–13383. [PubMed]
Prusiner SB. Novel proteinaceous infectious particles cause scrapie. Science. 1982;216:136–144. [PubMed]
Race R, Meade-White K, Raines A, Raymond GJ, Caughey B, Chesebro B. Subclinical scrapie infection in a resistant species: persistence, replication, and adaptation of infectivity during four passages. J. Infect. Dis. 2002;186:S166–S170. [PubMed]
Race R, Raines A, Raymond GJ, Caughey B, Chesebro B. Long-term subclinical carrier state precedes scrapie replication and adaptation in a resistant species: analogies to bovine spongiform encephalopathy and variant Creutzfeldt-Jakob disease in humans. J. Virol. 2001;75:10106–10112. [PMC free article] [PubMed]
Saa P, Castilla J, Soto C. Cyclic amplification of protein misfolding and aggregation. Methods Mol. Biol. 2005;299:53–65. [PubMed]
Saborio GP, Permanne B, Soto C. Sensitive detection of pathological prion protein by cyclic amplification of protein misfolding. Nature. 2001;411:810–813. [PubMed]
Safar J, Wille H, Itri V, Groth D, Serban H, Torchia M, Cohen FE, Prusiner SB. Eight prion strains have PrP(Sc) molecules with different conformations. Nat. Med. 1998;4:1157–1165. [PubMed]
Sarafoff NI, Bieschke J, Giese A, Weber P, Bertsch U, Kretzschmar HA. Automated PrPres amplification using indirect sonication. J. Biochem. Biophys. Methods. 2005;63:213–221. [PubMed]
Sigurdson CJ, Aguzzi A. Chronic wasting disease. Biochim. Biophys. Acta. 2006;1772:610–618. [PMC free article] [PubMed]
Soto C, Anderes L, Suardi S, Cardone F, Castilla J, Frossard MJ, Peano S, Saá P, Limido L, Carbonatto M, Ironside J, Torres JM, Pocchiari M, Tagliavini F. Pre-symptomatic detection of prions by cyclic amplification of protein misfolding. FEBS Lett. 2005;579:638–642. [PubMed]
Soto C, Castilla J. The controversial protein-only hypothesis of prion propagation. Nat. Med. 2004;10:S63–S67. [PubMed]
Soto C, Saborio GP, Anderes L. Cyclic amplification of protein misfolding: application to prion-related disorders and beyond. Trends Neurosci. 2002;25:390–394. [PubMed]
Tanaka M, Chien P, Naber N, Cooke R, Weissman JS. Conformational variations in an infectious protein determine prion strain differences. Nature. 2004;428:323–328. [PubMed]
Tanaka M, Chien P, Yonekura K, Weissman JS. Mechanism of cross-species prion transmission: an infectious conformation compatible with two highly divergent yeast prion proteins. Cell. 2005;121:49–62. [PubMed]
Vanik DL, Surewicz KA, Surewicz WK. Molecular basis of barriers for interspecies transmissibility of mammalian prions. Mol. Cell. 2004;14:139–145. [PubMed]
Weber P, Giese A, Piening N, Mitteregger G, Thomzig A, Beekes M, Kretzschmar HA. Generation of genuine prion infectivity by serial PMCA. Vet. Microbiol. 2007;123:346–357. [PubMed]
Wickner RB, Masison DC, Edskes HK. [PSI] and [URE3] as yeast prions. Yeast. 1995;11:1671–1685. [PubMed]
Will RG, Ironside JW, Zeidler M, Cousens SN, Estibeiro K, Alperovitch A, Poser S, Pocchiari M, Hofman A, Smith PG. A new variant of Creutzfeldt-Jakob disease in the UK. Lancet. 1996;347:921–925. [PubMed]
Williams ES. Chronic wasting disease. Vet. Pathol. 2005;42:530–549. [PubMed]
Xie Z, O'Rourke KI, Dong Z, Jenny AL, Langenberg JA, Belay ED, Schonberger LB, Petersen RB, Zou W, Kong Q, Gambetti P, Chen SG. Chronic wasting disease of elk and deer and Creutzfeldt-Jakob disease: Comparative analysis of the scrapie prion protein. J. Biol. Chem. 2005;281:4199–4206. [PubMed]