Molecular Model of Prion Transmission to Humans

To assess interspecies barriers to transmission of transmissible spongiform encephalopathies, we investigated the ability of disease-associated prion proteins (PrPd) to initiate conversion of the human normal cellular form of prion protein of the 3 major PRNP polymorphic variants in vitro. Protein misfolding cyclic amplification showed that conformation of PrPd partly determines host susceptibility.

To assess interspecies barriers to transmission of transmissible spongiform encephalopathies (TSEs), we investigated the ability of disease-associated prion proteins (PrP d ) to initiate conversion of the human normal cellular form of prion protein of the 3 major PRNP polymorphic variants in vitro. Protein misfolding cyclic amplifi cation showed that conformation of PrP d partly determines host susceptibility.
T he agents responsible for the transmissible spongiform encephalopathies (TSEs) are called prions. Although their precise biochemical composition is a matter of debate, they are known to occur in a series of strains, each with a characteristic disease phenotype and host range (1). A central event in neuropathogenesis of TSEs is conversion of the normal cellular form of the prion protein (PrP C ) to the pathognomonic disease-associated isoform (PrP d ) (2). In the absence of a known nucleic acid genome, it has been proposed that the strain-like properties of different TSE agents are encoded by distinct self-propagating conformational variants (conformers) of PrP d (3). The best developed method available for typing these PrP d isoforms uses limited proteolysis and classifi cation of the protease-resistant prion protein (PrP res ) in terms of the sizes of the nonglycosylated fragment(s) produced and the ratio of the 3 possible glycoforms (3). If distinct conformers and glycotypes of PrP d are responsible for diversity of prion strains, then they would be expected to be able to impose these molecular characteristics onto PrP C of the same amino acid sequence (when transmitted or replicating within a species) and onto PrP C of a different primary sequence (when transmitted between species). In support of this theory, the agent responsible for the TSE of cattle, called bovine spongiform encephalopathy (BSE), the accepted cause of variant Creutzfeldt-Jakob disease (vCJD) in humans (4), has been shown to be trans-missible to at least 7 species (1), resulting in propagation of PrP d that retains the characteristic molecular signature of the original BSE prion strain (5)(6)(7).
Current thinking favors a seeded polymerization model for the conversion of PrP C into PrP d , which has led to the development of several cell-free in vitro conversion model systems (8). One such system is protein misfolding cyclic amplifi cation (PMCA) (9), in which small amounts of PrP d introduced (seeded) into substrate containing excess PrP C and other essential conversion cofactors can be amplifi ed to readily detectable levels by sequential cycles of sonication and incubation. We have previously reported that the molecular characteristics, electrophoretic mobility, and glycoform ratio of the PrP res associated with the vCJD PrP d conformer were faithfully reproduced by PMCA (10). However, the effi ciency of amplifi cation achieved depended on the substrate's prion protein gene codon 129 (PRNP-129) genotype. The most effi cient amplifi cation was achieved in a methionine homozygous (PRNP-129MM) substrate; the least effi cient, in a valine homozygous (PRNP-129VV) substrate. To estimate the molecular component of transmission barriers for particular TSE agents between species, we used PMCA reactions to amplify PrP d associated with vCJD (10), bovine BSE (11), ovine scrapie (12), and experimental ovine BSE (13) and substrates prepared from humanized transgenic mouse brain tissue expressing each of the 3 main PRNP polymorphic variants found in Caucasian human populations (PRNP-129MM, MV, and VV) (14).

The Study
We prepared seed and substrate homogenates as 10% (wt/vol) homogenates in PMCA conversion buffer (10). Seed homogenates were diluted into substrate homogenates so that all PMCA reactions contained equivalent amounts of PrP d based on the PrP res levels in each seed homogenate. PrP res levels were determined by Western blot titration that used monoclonal antibody (MAb) 6H4 after limited proteinase K digestion. The reaction mixes were split into 2 aliquots; 1 aliquot was stored immediately at -80°C (−PMCA), and the other was subjected to 48 cycles of PMCA (+PMCA) (10). To assess the degree of PrP d amplifi cation achieved from each seed in each substrate, the samples −/+ PMCA were subjected to limited proteinase K digestion, and PrP res was detected by Western blotting with MAb 6H4 (which recognizes human, bovine, and ovine PrP) and MAb 3F4 (which selectively recognizes only human PrP and would therefore specifi cally identify PrP res formed from human PrP C ).
Using MAb 6H4 to probe Western blots, we noted amplifi cation of vCJD, bovine BSE, and ovine BSE PrP res in the PRNP-129MM substrate ( blots by densitometry showed that the degree of amplification of vCJD PrP res was considerably greater than that of bovine or ovine BSE in the PRNP-129MM substrate ( Figure 2, panel A). A more sensitive and discriminatory Western blot conducted by using MAb 3F4 confi rmed effi cient amplifi cation of vCJD, bovine BSE, and ovine BSE PrP res in the PRNP-129MM substrate (Figure 1, panel B,  top), weaker amplifi cation in the PRNP-129MV substrate ( Figure 1, panel B, middle), and little, if any, amplifi cation in the PRNP-129VV substrate (Figure 1, panel B,  bottom). In all substrates, the amplifi ed PrP res retained the electrophoretic mobility and glycoform ratio associated with BSE-related PrP res . No amplifi cation of ovine scrapie PrP res was evident after PMCA in any of the PRNP humanized transgenic mouse brain substrates (Figure 1, panels A,   B). The difference between ovine scrapie and ovine BSE in ability to seed amplifi cation in PRNP-129MM substrate was a robust phenomenon evident in brain samples from 3 different ARQ/ARQ sheep with each disease (Figure 2,  panel B). However, failure of the ovine scrapie seed to amplify was not caused by a general lack of competence to do so or by inappropriate amplifi cation conditions because robust amplifi cation of ovine scrapie PrP res was evident after PMCA in a substrate prepared from normal ARQ/ARQ sheep brain (Figure 2, panel C).

Conclusions
Our results are best appreciated in terms of the molecular interaction between seed PrP d and substrate PrP C , specifi cally the species-specifi c amino acid sequence and PRNP polymorphic status of PrP C and PrP d and the PrP d conformers involved (Table). Regardless of the seed PrP amino acid sequence, the PrP d conformers associated with bovine BSE, ovine BSE, and human vCJD were amplifi ed in the humanized mouse substrate and displayed similar PRNP-129 genotype preferences (PRNP-129MM >PRNP-129MV >PRNP-129VV). In contrast, the PrP d conformer associated with the ovine scrapie strain, although sharing the same PrP amino acid sequence as the PrP d in ovine BSE, could not be amplifi ed in any of the PRNP humanized mouse substrates but could be amplifi ed in a sheep brain substrate. These observations are consistent with conformation of a TSE agent's PrP d (rather than solely its amino acid sequence) having a role in determining the susceptibility of a host's PrP C to conversion. They similarly suggest that these molecular factors could in turn have a powerful infl uence on disease susceptibility and incubation time.
To date, all clinical cases of vCJD have occurred in persons with the PRNP-129MM genotype, as might be predicted from the effi ciency of amplifi cation of BSE-related PrP d shown here. Extrapolating from these results, one would predict that the next genotypic group most likely to show susceptibility to the BSE agent would be heterozygous (MV) at codon 129 of the PRNP gene, as previously suggested from the corresponding in vivo transmission studies (14).
In the wake of BSE epidemics in the United Kingdom and elsewhere, enhanced surveillance has identifi ed apparently new TSEs (15), raising concerns regarding animal and human health. PMCA with suitable substrate sources could provide a rapid way to estimate the molecular component of transmission barriers for particular TSE agents between species, including humans. These estimates could thus indicate whether, like classical scrapie, the agents rep- Substrate was seeded with brain homogenates prepared from sheep with confi rmed scrapie and BSE such that each PMCA reaction mix contained an equivalent amount of PrP d according to detection of PrP res by Western blot titration after limited proteinase K digestion. PRNP-129MM substrate seeded with vCJD brain homogenate was included as a positive control in each experiment. C) Amplifi cation of PrP d associated with ovine scrapie and BSE in substrates prepared from PRNP-129 methionine homozygous humanized transgenic mouse brain tissue (MM substrate) and NSB substrate. Substrates were prepared as 10% (wt/vol) homogenates in PMCA conversion buffer (10). Each substrate was seeded with brain homogenates prepared from sheep with confi rmed scrapie and BSE so that each PMCA reaction mix contained an equivalent amount of PrP d as determined by detection of PrP res by Western blot titration after limited proteinase K digestion. Reaction mixes were divided into 2 lots: 1 was stored immediately at -80°C (−PMCA) and the other was subjected to 48 cycles of PMCA (+PMCA) by using standard conditions (10). After limited proteinase K digestion, PrP res in samples −/+PMCA was detected by Western blotting using MAb 6H4. PMCA, protein misfolding cyclic amplifi cation; BSE, bovine spongiform encephalopathy; vCJD, variant Creutzfeldt-Jakob disease; MM, methionine homozygous; PrP d , disease-associated prion protein; MAb, monoclonal antibody; PrP res , protease-resistant prion protein; NSB, normal ARQ/ARQ sheep brain tissue. Values on the left in panels B and C are in kilodaltons. resent little risk for human health or whether, like classical BSE, they represent cause for concern. This work was funded by the European Network of Excellence NeuroPrion (FOOD-CT-2004-506579), the Scottish National Blood Transfusion Services, and the Chief Scientist Offi ce of the Scottish Government (CZB/4/357). The National CJD Surveillance Unit is funded by the Department of Health and the Scottish Government. Dr Jones is a postdoctoral research fellow at the National CJD Surveillance Unit, University of Edinburgh. His primary research interests are the application of in vitro PrP d amplifi cation techniques, such as PMCA, to prion disease research in general and incorporation of these techniques into a confi rmatory screening assay to detect vCJD-associated PrP d in human plasma as a surrogate marker of vCJD infectivity in blood.