Report

Human Prion Protein with Valine 129 Prevents Expression of Variant CJD Phenotype

See allHide authors and affiliations

Science  03 Dec 2004:
Vol. 306, Issue 5702, pp. 1793-1796
DOI: 10.1126/science.1103932

Abstract

Variant Creutzfeldt-Jakob disease (vCJD) is a unique and highly distinctive clinicopathological and molecular phenotype of human prion disease associated with infection with bovine spongiform encephalopathy (BSE)–like prions. Here, we found that generation of this phenotype in transgenic mice required expression of human prion protein (PrP) with methionine 129. Expression of human PrP with valine 129 resulted in a distinct phenotype and, remarkably, persistence of a barrier to transmission of BSE-derived prions on subpassage. Polymorphic residue 129 of human PrP dictated propagation of distinct prion strains after BSE prion infection. Thus, primary and secondary human infection with BSE-derived prions may result in sporadic CJD-like or novel phenotypes in addition to vCJD, depending on the genotype of the prion source and the recipient.

Distinct prion strains are associated with biochemically distinct forms of disease-related prion protein (PrPSc). Four PrPSc types have been observed in brain tissue from patients with distinct Creutzfeldt-Jakob disease (CJD) phenotypes: types 1 to 3 in classical (sporadic or iatrogenic) CJD and type 4 in vCJD (13). Polymorphism at residue 129 of human PrP (where either methionine or valine can be encoded) powerfully affects genetic susceptibility to human prion diseases (47) and appears to critically influence the propagation of these human PrPSc types. So far, types 1 and 4 PrPSc have been found only in humans homozygous for Met129; type 3 PrPSc is seen almost exclusively in individuals with at least one valine allele; and type 2 PrPSc has been commonly observed in all codon 129 genotypes (13). BSE and vCJD prion infection in transgenic mice expressing human PrP, but not mouse PrP (1, 810), indicates that codon 129 polymorphism determines the ability of human PrP to form type 4 PrPSc and to generate the neuropathological phenotype of vCJD.

Challenge of transgenic mice expressing human PrP Met129 (129MM Tg35 and 129MM Tg45 mice) with BSE and vCJD prions (11) resulted in faithful propagation of type 4 PrPSc (10) (Figs. 1 and 2) accompanied by the key neuropathological hallmark of vCJD, the presence of abundant florid PrP plaques (10). However, transgenic mice expressing human PrP Val129 (129VV Tg152 mice) responded quite differently. Although these 129VV Tg152 mice lack a transmission barrier to classical forms of CJD, regardless of the codon 129 genotype of the inoculum (1, 8, 9), the primary challenge with vCJD prions was characterized by a substantial transmission barrier to infection (only ∼50% of inoculated mice were infected, compared with 100% of 129MM Tg35 and 129MM Tg45 mice) (Fig. 1; table S1). In addition, rather than type 4 PrPSc, vCJD-inoculated 129VV Tg152 mice propagated type 5 PrPSc (9), which shares the same predominance of the diglycosylated glycoform seen in type 4 PrPSc but is distinguished by proteinase K digestion products of greater molecular mass (Fig. 2A), which closely resemble those seen in human type 2 PrPSc (9). Type 5 PrPSc is associated with very weak diffuse PrP deposition in the brain (9), which contrasts markedly with the florid PrP plaques associated with the propagation of type 4 PrPSc in humans (12) or transgenic mice (10). Similar diffuse deposition of PrP is also observed in clinically affected BSE-inoculated 129VV Tg152 mice; however, type 5 PrPSc is undetectable in brain homogenate (9).

Fig. 1.

Summary of transmissions of vCJD and BSE prions to transgenic mice. The total number of prion-affected mice (both clinical and subclinical) is reported for each inoculated group: 129MM Tg45 mice (black), 129MM Tg35 mice (gray), 129VV Tg152 mice (white). Animals were scored by clinical signs, immunoblotting, and/or immunohistochemistry. Primary transmission data have been published previously (9, 10). In transmissions that result in bifurcation of propagated PrPSc type, the number of samples positive for type 2 or type 4 PrPSc is reported as a proportion of the total number of samples examined by immunoblotting. (★), The occurrence of subclinical prion infection only.

Fig. 2.

Molecular strain typing of vCJD and BSE prion transmissions in transgenic mice. (A to D) Immunoblots of proteinase K–treated brain homogenates from variant and sporadic CJD (PRNP 129 MM genotype) and transgenic mice were analyzed by enhanced chemiluminescence with either monoclonal antibody 3F4 against PrP (A) or biotinylated monoclonal antibody ICSM 35 against PrP (B to D). The identity of the brain sample is designated above each lane with the type of PrPSc present in the sample designated below. Transmissions that result in the propagation of either type 2 or type 4 PrPSc.

To further evaluate the molecular and neuropathological phenotype of vCJD- or BSE-inoculated 129VV Tg152 mice, we performed a second passage in the same breed of mice. Primary transmission of prions from one species to another is associated with a species or transmission barrier that is largely or completely abrogated on second and subsequent passage in the second species as the prions adapt to the new host. Second passage then resembles within-species transmission with a high (typically 100%) attack rate and much shortened and more consistent incubation period. It was remarkable, however, that such adaptation did not occur on second passage of BSE or vCJD prions in 129VV Tg152 mice. Brain inocula derived from four clinically affected BSE-inoculated 129VV Tg152 mice failed to transmit clinical disease or asymptomatic prion infection to additional 129VV Tg152 mice (Figs. 1 and 3; table S2). However, two of these inocula produced clinical prion disease (Fig. 3; table S2) with abundant PrPSc accumulation (fig. S1) on inoculation of wild-type FVB mice with incubation periods that are not compatible with persistence of the original BSE inoculum [supporting online material (SOM) text]. The prion strain generated in BSE-inoculated 129VV Tg152 mice was thus infectious in wild-type FVB mice, but not in additional 129VV Tg152 mice.

Fig. 3.

Summary of transmissions of vCJD and BSE prions to transgenic and wild-type FVB mice. The total number of prion-affected mice (both clinical and subclinical) is reported for each inoculated group: 129VV Tg152 mice (white), wild-type FVB mice (gray). Animals were scored by clinical signs, immunoblotting, and/or immunohistochemistry. Data are derived from tables S1 and S2. (★), The occurrence of subclinical prion infection only.

Valine 129 is unique to human PrP, and the failure of BSE prions to adapt in 129VV Tg152 mice on second passage contrasts sharply with the marked adaptation of BSE prions in FVB mice (Fig. 3; table S2) or in other murine (13) or primate (14) hosts that encode methionine at the corresponding position of PrP. BSE prions also efficiently adapt on second passage in 129MM Tg35 transgenic mice. Primary transmission of BSE prions in 129MM Tg35 mice resulted in bifurcation of propagated strain type and produced either type 2 or 4 PrPSc (Figs. 1 and 2) and neuropathology consistent with human sporadic CJD or vCJD, respectively (10). These distinct molecular and neuropathological phenotypes consistently “breed true”with very high efficiency on second passage in additional 129MM Tg35 transgenic mice (15). These findings contrast sharply with the complete lack of prion transmission on second passage of the same BSE inocula in 129VV Tg152 mice, supporting the interpretation that human PrP Val129 severely restricts propagation of the BSE prion strain.

This conclusion was further reinforced by study of the parameters of second passage of vCJD prions. As seen with second passage of BSE prions, clinical disease was observed only in FVB, and not in 129VV Tg152, recipients. Brain inocula from clinically affected, type 5 PrPSc positive, primary vCJD-inoculated 129VV Tg152 mice produced clinical prion disease (Fig. 3; table S2) and PrPSc accumulation (fig. S1) on subpassage in FVB mice, but produced only subclinical infection (with PrPSc accumulation) in 7 out of 11 inoculated 129VV Tg152 mice (Figs. 1 and 3). Notably, in four of these, high-sensitivity methods (16) were required to detect PrPSc in brain homogenate (table S2). Type 5 PrPSc was faithfully propagated on second passage in 129VV Tg152 mice (Fig. 2A). In the three mice containing the highest levels of type 5 PrPSc, extensive spongiosis was also observed (Fig. 4), and in one of these, in contrast to the pathology seen on first passage, numerous PrP plaques were seen (Fig. 4). Type 4 PrPSc is invariably associated with prominent florid plaques in the cortex of human vCJD brain (12) and in vCJD- or BSE-prion inoculated 129MM Tg35 and Tg45 transgenic mice (10), whereas plaques associated with type 5 PrPSc were restricted to the corpus callosum and had a nonflorid morphology (Fig. 4). The lack of adaptation of vCJD prions on second passage in 129VV Tg152 mice contrasted sharply with the behavior of vCJD prions in wild-type FVB mice, where typical adaptation was observed on second passage with 100% clinical prion disease with abundant PrPSc accumulation (fig. S1) at markedly reduced incubation periods (Fig. 3; table S2).

Fig. 4.

Neuropathological analysis of transgenic mouse brain. Primary transmission of vCJD prions in 129VV Tg152 mice produces type 5 PrPSc that is maintained after secondary passage in 129VV Tg152 mice but induces propagation of either type 2 or type 4 PrPSc after passage in 129MM Tg35 mice. Immunohistochemistry (PrP) shows abnormal PrP immunoreactivity, including PrP-positive plaques, stained with monoclonal antibody 3F4 against PrP. Sections stained with hematoxylin- and-eosin (H&E) show spongiform neurodegeneration (left, corpus callosum; middle and right, parietal cortex). Scale bar, 100 μm. Lower panels show the regional distribution of abnormal PrP deposition. Green boxes in the sketches denote the area from which PrP-stained sections are derived.

Both BSE and vCJD prions failed to propagate efficiently on either primary or, remarkably, second passage, in 129VV Tg152 mice in sharp contrast to 129MM Tg mice or wild-type animals, and where detectable, infection was associated with a distinct PrPSc type and pathological phenotype. Thus, human PrP Val129 appears not to be a compatible substrate for propagation of the prion strain seen in vCJD. This interpretation was supported by the transmission properties of 129VV Tg152–passaged vCJD prions in 129MM Tg35 mice. Here, 14 out of 15 129MM Tg35 mice inoculated with isolates containing type 5 PrPSc showed PrPSc accumulation (Fig. 1; table S3), typically to much higher levels than seen in 129VV Tg152 mice receiving the same inocula. However, the PrPSc seen was not of the type 5 pattern but instead these transmissions mirrored the behavior of BSE prions in 129MM Tg35 mice (10), where, instead, either type 4 or type 2 PrPSc were seen (Figs. 1 and 2, B to D). Thus human PrPSc types 4 and 5 are restricted to propagating in mice expressing human PrP Met129 or Val129, respectively.

Neuropathologically, type 4 PrPSc was associated with relatively little spongiosis and abundant florid plaques (Fig. 4) typical of vCJD in humans (12), whereas type 2 PrPSc was associated with much higher levels of vacuolation in many areas of the brain, accompanied by generally diffuse PrP deposition and occasional small, nonflorid plaques (Fig. 4) that closely resembled human sporadic CJD with type 2 PrPSc PRNP (human prion protein gene) 129MM (3). Clinical prion disease was observed in all 129MM Tg35 mice propagating type 2 PrPSc, whereas mice propagating type 4 PrPSc were subclinically infected (table S3).

In conclusion, we have demonstrated that BSE and vCJD prion infection in transgenic mice can result in the propagation of distinct molecular and neuropathological phenotypes dependent on host PrP residue 129 and possibly other, as yet unidentified, disease modifying loci (10). These data predict a critical role for PRNP codon 129 in governing the thermodynamic permissibility of human PrPSc conformation that can be interpreted within a conformational selection model of prion transmission barriers (1719) (SOM text) and suggest that there is no overlapping preferred conformation for Val129 and Met129 human PrP that can be generated as a result of exposure to the vCJD/BSE prion strain. Biophysical measurements suggest that this powerful effect of residue 129 on prion strain selection is likely to be mediated by means of its effect on the conformation of PrPSc or its precursors or on the kinetics of their formation, as it has no measurable effect on the folding, dynamics, or stability of the normal cellular prion protein PrPC (20).

Although caution must be exercised in extrapolating from animal models, even where, as here, faithful recapitulation of molecular and pathological phenotypes is possible, our findings argue that primary human BSE prion infection, as well as secondary infection with vCJD prions by iatrogenic routes, may not be restricted to a single disease phenotype. These data, together with the recent recognition of probable iatrogenic transmission of vCJD prions to recipients of blood (21, 22), including a PRNP codon 129 Met/Val heterozygous individual (22), reiterate the need to stratify all human prion disease patients by PrPSc type. This surveillance will facilitate rapid recognition of novel PrPSc types and of any change in relative frequencies of particular PrPSc subtypes in relation to either BSE exposure patterns or iatrogenic sources of vCJD prions.

Supporting Online Material

www.sciencemag.org/cgi/content/full/1103932/DC1

Materials and Methods

SOM Text

Fig. S1

Tables S1 to S3

References and Notes

References and Notes

View Abstract

Stay Connected to Science

Navigate This Article